| Author |
Message |
|---|
 RogerVolunteer developer
Volunteer tester
Send message
Joined: 27 Nov 11
Posts: 1138
ID: 120786
Credit: 268,668,824
RAC: 0
                    
|
|
Welcome to the Conjunction of Venus & Jupiter Challenge!
The first Challenge of the 2019 Challenge series is a 15 day challenge to celebrate the Conjunction of Venus and Jupiter. The challenge is being offered on the Seventeen or Bust LLR (SOB) application, working towards solving the Sierpinski Problem.
A conjunction of Venus and Jupiter will be visible on January 22. The two bright planets will be visible within 2.4 degrees of each other in the early morning sky. Look for this impressive sight in the east just before sunrise.
At around the same time, the two objects will also make a close approach, technically called an appulse.
Venus will be at mag -4.3, and Jupiter at mag -1.9, both in the constellation Ophiuchus.
The pair will be too widely separated to fit within the field of view of a telescope, but will be visible to the naked eye or through a pair of binoculars.
To participate in the Challenge, please select only the Seventeen or Bust LLR (SOB) project in your PrimeGrid preferences section. The challenge will begin 7th January 2019 05:43 UTC and end at 22nd January 2019 05:43 UTC, time of closest angular separation. Note the non-standard start time.
Application builds are available for Linux 32 and 64 bit, Windows 32 and 64 bit and MacIntel. Intel CPUs with AVX capabilities (Sandy Bridge, Ivy Bridge, Haswell, Broadwell, Skylake, Kaby Lake, Coffee Lake) will have a very large advantage, and Intel CPUs with FMA3 (Haswell, Broadwell, Skylake, Kaby Lake, Coffee Lake) will be the fastest.
ATTENTION: The primality program LLR is CPU intensive; so, it is vital to have a stable system with good cooling. It does not tolerate "even the slightest of errors." Please see this post for more details on how you can "stress test" your computer. Tasks on one CPU core will take ~2 days on fast/newer computers and 10+ days on slower/older computers. If your computer is highly overclocked, please consider "stress testing" it. Sieving is an excellent alternative for computers that are not able to LLR. :)
Highly overclocked Haswell, Broadwell, Skylake, Kaby Lake or Coffee Lake (i.e., Intel Core i7, i5, and i3 -4xxx or better) computers running the application will see fastest times. Note that SOB is running the latest FMA3 version of LLR which takes full advantage of the features of these newer CPUs. It's faster than the previous LLR app and draws more power and produces more heat. If you have a Haswell, Broadwell, Skylake, Kaby Lake or Coffee Lake CPU, especially if it's overclocked or has overclocked memory, and haven't run the new FMA3 LLR before, we strongly suggest running it before the challenge while you are monitoring the temperatures.
Please, please, please make sure your machines are up to the task.
Multi-threading optimisation instructions
Those looking to maximise their computer's performance during this challenge, or when running LLR in general, may find this information useful.
- Your mileage may vary. Before the challenge starts, take some time and experiment and see what works best on your computer.
- If you have an Intel CPU with hyperthreading, either turn off the hyperthreading in the BIOS, or set BOINC to use 50% of the processors.
- If you're using a GPU for other tasks, it may be beneficial to leave hyperthreading on in the BIOS and instead tell BOINC to use 50% of the CPU's. This will allow one of the hyperthreads to service the GPU.
- Use LLR's multithreaded mode. It requires a little bit of setup, but it's worth the effort. Follow these steps:
Time zone converter:
The World Clock - Time Zone Converter
NOTE: The countdown clock on the front page uses the host computer time. Therefore, if your computer time is off, so will the countdown clock. For precise timing, use the UTC Time in the data section at the very top, above the countdown clock.
Scoring Information
Scores will be kept for individuals and teams. Only tasks issued AFTER 7th January 2019 05:43 UTC and received BEFORE 22nd January 2019 05:43 UTC will be considered for credit. We will be using the same scoring method as we currently use for BOINC credits. A quorum of 2 is NOT needed to award Challenge score - i.e. no double checker. Therefore, each returned result will earn a Challenge score. Please note that if the result is eventually declared invalid, the score will be removed.
At the Conclusion of the Challenge
We kindly ask users "moving on" to ABORT their tasks instead of DETACHING, RESETTING, or PAUSING.
ABORTING tasks allows them to be recycled immediately; thus a much faster "clean up" to the end of an LLR Challenge. DETACHING, RESETTING, and PAUSING tasks causes them to remain in limbo until they EXPIRE. Therefore, we must wait until tasks expire to send them out to be completed.
Please consider either completing what's in the queue or ABORTING them. Thank you. :)
About the Sierpinski Problem
Wacław Franciszek Sierpiński (14 March 1882 — 21 October 1969), a Polish mathematician, was known for outstanding contributions to set theory, number theory, theory of functions and topology. It is in number theory where we find the Sierpinski problem.
Basically, the Sierpinski problem is "What is the smallest Sierpinski number?"
First we look at Proth numbers (named after the French mathematician François Proth). A Proth number is a number of the form k*2^n+1 where k is odd, n is a positive integer, and 2^n>k.
A Sierpinski number is an odd k such that the Proth number k*2^n+1 is not prime for all n. For example, 3 is not a Sierpinski number because n=2 produces a prime number (3*2^2+1=13). In 1962, John Selfridge proved that 78,557 is a Sierpinski number...meaning he showed that for all n, 78557*2^n+1 was not prime.
Most number theorists believe that 78,557 is the smallest Sierpinski number, but it hasn't yet been proven. In order to prove it, it has to be shown that every single k less than 78,557 is not a Sierpinski number, and to do that, some n must be found that makes k*2^n+1 prime.
Seventeen or Bust
Seventeen or Bust was a distributed computing project attempting to solve the Sierpinski problem. The name of the project is due to the fact that, when founded, there were seventeen values of k < 78,557 for which no primes were known.
The project was conceived in March of 2002 by two college undergraduates. After some planning and a lot of programming, the first public client was released on April 1. Seventeen or Bust ceased operations in 2016. The project was administered by:
• Louis Helm, a computer engineer in Austin, Texas.
• David Norris, a software engineer in Urbana, Illinois.
• Michael Garrison, a Computer Science undergraduate at Eastern Michigan University in Ypsilanti, Michigan.
Starting in 2010, PrimeGrid partnered with Seventeen or Bust to work towards solving the Sierpinski Problem. After the demise of the original Seventeen or Bust project in 2016, PrimeGrid is continuing by itself to continue the Seventeen or Bust project in looking to solve the Sierpinski Problem.
As of December of 2017, PrimeGrid and Seventeen or Bust have eliminated twelve of those seventeen candidates. The project might now be styled "Five or Bust," but the original name will be kept for consistency. Current remaining k's:
21181, 22699, 24737, 55459, 67607
SOB is currently 54% of the way through a double check of old search ranges. A brief word about the SOB tasks that will be running in the challenge:
- Around 30% of the tasks will be double checks against one or more residues from the original Seventeen or Bust project. Expect these tasks to validate almost immediately, unless your result doesn't match the residue. (This should also help with the clean-up)
- We’ve currently in the n=26-29M range, with most of the tasks around 8.6-8.7M digits long. Any prime found would come in as the 11th largest ever.
- Next tasks after that are n=29-30M, with a small bump in length to 8.7-9.0M digits.
- We're well beyond the point where the Seventeen or Bust project thoroughly double checked their work, so finding a missed prime is a possibility. Two of their primes were found while double checking their prior work. Will we find a third?
Last Prime found at PrimeGrid
10223*2^31172165+1 by Szabolcs Péter (SyP) on 31 October 2016. Official Announcement.
Previous to that the last SOB candidate was eliminated in 2007.
What is LLR?
The Lucas-Lehmer-Riesel (LLR) test is a primality test for numbers of the form N = k*2^n − 1, with 2^n > k. Also, LLR is a program developed by Jean Penne that can run the LLR-tests. It includes the Proth test to perform +1 tests and PRP to test non base 2 numbers. See also:
(Edouard Lucas: 1842-1891, Derrick H. Lehmer: 1905-1991, Hans Riesel: 1929-2014).
____________
|
|
|
|
|
|
Hooray!
That's what I've been waiting for!
Forwarded the information to the Rechenkraft.net forums.
German-speaking (or at least -reading) people might want to help themselves with my translation of those parts of the announcement I considered important. ^^
____________
Greetings, Jens
147433824^131072+1 |
|
|
|
|
|
Thank you Roger.
As for the rest of the challenge series for 2019, the duration is far too long (as is this one). 3-7 days should be the norm, less those who only participate in your project to be a team player for their team get burned out, or just ignore the challenges altogether. |
|
|
Ken_g6 Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 937
ID: 3110
Credit: 258,164,203
RAC: 78,071
                      
|
|
Can you tell us what the current FFT size is for this project? It would help with figuring out the best multithreaded configuration. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
                       
|
Can you tell us what the current FFT size is for this project? It would help with figuring out the best multithreaded configuration.
The answer to your question will be different for each K, so I'll have to give you 5 sets of values.
The FFT sizes (for each k) are the last ones we'll be using in the double check and won't change. That makes it simple. There are two different FFT sizes currently in use, depending on k:
+-------+-----------+----------+----------+
| k | FFT | min(n) | max(n) |
+-------+-----------+----------+----------+
| 21181 | 2560.0000 | 28626140 | 29999900 |
| 22699 | 2560.0000 | 28626958 | 29999638 |
| 24737 | 2560.0000 | 28626223 | 29998951 |
| 55459 | 2880.0000 | 28626106 | 29999998 |
| 67607 | 2880.0000 | 28626731 | 29539491 |
+-------+-----------+----------+----------+
There's a small number of 2304K and 2400K SoB tasks currently being crunched. If any of those tasks fail, they'll be recycled and sent out again, so it's possible some of those could be sent out during the challenge. It shouldn't be a lot of tasks, however. The vast majority will be 2560K or 2880K.
Also, the general caveat about FFT sizes: your mileage may vary. Your computer may choose a different FFT size than what the server's calculations expect.
____________
My lucky number is 75898524288+1 |
|
|
mackerelVolunteer tester
Send message
Joined: 2 Oct 08
Posts: 2639
ID: 29980
Credit: 568,393,769
RAC: 3,323
              
|
|
Based on those FFT sizes we're looking at needing 20MB+ cache at minimum to make a good dent in times. Otherwise, hope you have fast ram... |
|
|
ktoSend message
Joined: 13 Jan 09
Posts: 27
ID: 34164
Credit: 41,469,365
RAC: 0
            
|
|
I tried the instructions for "multithreading optimization instructions". When I select Option->Read Config Files it doesn't indicate (via a check mark or anything else) thait is selected.
The app_config.xml has the exact contents cited and is placed in the cited location, C:\ProgramData\BOINC\projects\www.primegrid.com
This is on a quad-core. On the off chance that there is no visual indicator, I have placed three tasks in "suspend" mode and the remaining task is only using one core.
Any suggestions? thanks!
____________
 |
|
|
|
|
|
There's no visual indicator, it will tell you in the event log.
If I remember correctly, changing number of threads requires you to close and restart BOINC to have it take effect immediately. |
|
|
ktoSend message
Joined: 13 Jan 09
Posts: 27
ID: 34164
Credit: 41,469,365
RAC: 0
|
|
I tried that. Multiple times. I feel like I've missed something else?
____________
|
|
|
AzmodesVolunteer tester
Send message
Joined: 30 Dec 16
Posts: 184
ID: 479275
Credit: 2,197,504,179
RAC: 18,428
           
|
|
Are you sure you closed BOINC for good and not just the GUI? Make sure the exit dialogue pops up and check the box that stops computation along with closing the client.
In other news, here are some SoB times from some of my CPUs. Estimations taken after 13-15 hours:
Threadripper 1950X: 124,000 secs / 1.4 days (8 cores) (Windows 10, DDR4 at 2666 MHz)
e6850: 1,395,000 secs / 16.1 days (1 core) (Ubuntu, DDR2)
i5 750: 693,000 secs / 8 days (2 cores) (Windows 10, DDR3)
Athlon X4 760k: 714,000 secs / 8.3 days (2 cores) (Ubuntu, DDR3)
X3370: 816,000 secs / 9.4 days (2 cores) (Windows 7, DDR2)
i5 2400S: 174,000 secs / 2 days (4 cores) (Ubuntu, DDR3)
i5 3470S: 144,000 secs / 1.7 days (4 cores) (Ubuntu, DDR3)
R5 2600: 186,000 secs / 2.1 days (6 cores) (Ubuntu, DDR4 at 2933 MHz)
Q6600: 643,000 secs / 7.4 days (3 cores) (Ubuntu, DDR2)
Yeah, I usually stick to GPU apps or sieving...
____________
Long live the sievers.
+ Encyclopaedia Metallum: The Metal Archives + |
|
|
|
|
|
kto wrote: When I select Option->Read Config Files it doesn't indicate (via a check mark or anything else) thait is selected.
This lets <avg_ncpus> take effect immediately. I.e. boinc-client will assume that this type of tasks will occupy avg_ncpus when running.
But as for <cmdline>, i.e. the more critical part which hands command line parameters to the LLR application when it is started:
recoil44 wrote: If I remember correctly, changing number of threads requires you to close and restart BOINC to have it take effect immediately.
Right, or in other words: A shutdown and restart of the boinc client makes the <cmdline> part take effect on already started tasks.
Alternatively, you can switch "Leave non-GPU tasks in memory while suspended" off, suspend an LLR task which you want to reconfigure, just to be sure check in task manager that the task was indeed suspended, then resume the task. |
|
|
|
|
I tried that. Multiple times. I feel like I've missed something else?
OK, start with the basics
Open the Event Log
Click on Read Config Files
You should now have a line reading: Found app_config.xml
Did this work? |
|
|
|
|
|
While it is appreciated that the only participants likely to have any chance of winning are those with highly performant machines, it is still fun for those of us with more modest setups to participate. The problem with the last few challenges is that the assigned tasks are so large and take so long on our machines that we cannot complete even one task in the time allocated.
It is more pleasurable for us lowly amateurs to participate if we have the chance of getting *some* placement on the scoreboard. We would hate to feel we were somehow a nuisance, getting in the way of the serious competitors, in the way that small children at a skateboard park get in the way of the big boys. However, if the admins decide that our efforts are not really needed on these challenges, we can stay clear if that is best.
____________
No matter how fast or how far you run, you're still in the space where you are. -- Russell |
|
|
|
|
|
so true...
____________
 |
|
|
JaySend message
Joined: 27 Feb 10
Posts: 131
ID: 56067
Credit: 63,534,968
RAC: 10,618
              
|
While it is appreciated that the only participants likely to have any chance of winning are those with highly performant machines, ...
I think that's going to be true regardless of the challenge particulars. The more / bigger hardware you throw at it, the more / higher you will score.
It is more pleasurable for us lowly amateurs to participate if we have the chance of getting *some* placement on the scoreboard.
I'm running a 9-year old CPU. I will complete at least one unit in this challenge, and likely more (barring a days long power outage, major hardware failure, meteorite hitting my home, etc). I honestly don't care about my placement in the challenge, but I do look at the scores for my name when it's done just to see where I'm at.
We would hate to feel we were somehow a nuisance, getting in the way of the serious competitors, in the way that small children at a skateboard park get in the way of the big boys. However, if the admins decide that our efforts are not really needed on these challenges, we can stay clear if that is best.
I don't understand how anyone could "get in the way" of anyone else. I believe there are way more than enough tasks to go around. We're not going to run out of them at any point during the challenge (unless we find all 5 remaining primes and prove the conjecture before it ends which is unlikely). My opinion is that every bit of effort is beneficial to the goal. The workunits all need to be checked eventually. Maybe you only check one while the next person checks five. That's fine, they all need to be checked eventually, and as a group, checking six is more than checking five.
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
Matt Westwood wrote: While it is appreciated that the only participants likely to have any chance of winning are those with highly performant machines, it is still fun for those of us with more modest setups to participate. The problem with the last few challenges is that the assigned tasks are so large and take so long on our machines that we cannot complete even one task in the time allocated..
This is your computer: http://www.primegrid.com/hosts_user.php?userid=483941
It's an Intel Desktop Haswell CPU with AVX and FMA3, 2 cores (4 threads, running at 3.7 GHz
This is my Haswell laptop: https://www.primegrid.com/show_host_detail.php?hostid=531731
The laptop is a Haswell Core i3, like your desktop, but it runs at 1.7 Ghz -- less than half as fast as your computer.
This is the challenge task it ran: https://www.primegrid.com/result.php?resultid=953413367 The run time was 523,528 seconds, or about 6 days. I was able to run only one task on that computer. Your computer is more than twice as fast as my laptop, and you should have been able to run at least two tasks.
Those numbers are correct as long as you took the time and put in the effort to configure your computer for maximum performance. It is a "challenge", after all.
For GFN21 CPU tasks, such as the last challenge, and large LLR tasks, like the next challenge, that means configuring BOINC to not use the hyperthreads and to use the multi-threading capabilities available in the latest versions of those apps. Instructions for setting up multi-threading are included in the introductory message in each forum challenge thread.
However, while you personally were using a computer that is fast enough to have completed some tasks in the last challenge (and even more in the next challenge), it's indeed true that there are some computers that are too slow. Not everyone is going to be able to participate in every challenge.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Point taken. I'll leave the games to the big boys.
____________
No matter how fast or how far you run, you're still in the space where you are. -- Russell |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
Point taken. I'll leave the games to the big boys.
That was not my point at all. In fact, it's the polar opposite of what I said.
I participated with a computer less than half as fast as your computer. And this was the most difficult challenge we've ever run.
While there are indeed some computers that are too slow -- that would be true no matter where we set the bar -- the bar is actually set very low. Almost any computer with a discreet GPU *OR* with AVX *OR* with at least 4 cores should have been able to participate. Probably a lot of computers that don't meet any of those conditions could have participated as well, but any machine with at least one of those should be able to have made the 10 day deadline.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
How does one properly benchmark SOB? For instance, I had a WU set to 4 cores. The remaining time was 3d 11h. Then I changed that same WU to 36 cores and it still says the exact same time is left, 3d 11h and counting.
I expected at-least a couple of hours shaved off with such an increase in cores...? |
|
|
|
|
How does one properly benchmark SOB? For instance, I had a WU set to 4 cores. The remaining time was 3d 11h. Then I changed that same WU to 36 cores and it still says the exact same time is left, 3d 11h and counting.
I expected at-least a couple of hours shaved off with such an increase in cores...?
That does bring a question up. If you change the cores - I'm guessing via the app_config file but use the same work unit, isn't the workunit still looking at only using the number of cores it was originally downloaded as?
Also in your case, going from 4 to 36 is quite a leap - must be nice :) |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
How does one properly benchmark SOB? For instance, I had a WU set to 4 cores. The remaining time was 3d 11h. Then I changed that same WU to 36 cores and it still says the exact same time is left, 3d 11h and counting.
I expected at-least a couple of hours shaved off with such an increase in cores...?
First of all, to get the app to actually switch, it needs to stop. The simplest way to do that, and be certain you're doing it right, is to reboot your computer. Yes, there's easier ways to do it, but if you're not familiar with what you're doing, you might end up thinking you restarted it without actually doing so. If you reboot, you know for certain the app was stopped and started.
But there's a bigger problem. The estimate BOINC shows you is based on its run time so far, so after you change the number of cores, it's basing its time estimate on a hybrid of the old setting and the new setting. That number is therefore worthless to you.
To see how long it's going to take, I run LLR by hand from a command line and look at the iteration timings that it displays. It's expressed in milliseconds, so you need to divide that number by 1000 and then multiply by the total number of iterations (which is displayed by LLR), and that gives you an estimate of the total runtime, in seconds. Important: don't use the first iteration timing since it includes initialization processing, and is artificially too high. Wait till at least the second timing at the 20,000 iteration mark.
If you're not comfortable with that, you can set up app_config and start a brand new task. Let it run for about a half hour or so to let the timing settle down. Then, abort the task, change the app_config settings, tell BOINC to read the new settings, and test with a new task. Rinse and repeat as necessary. IMPORTANT: If using this method, make sure each test uses the same FFT size. The FFT size can been seen in the file stderr.txt (while the task is running) or in the task page on the website (after the task has finished or been aborted.)
____________
My lucky number is 75898524288+1 |
|
|
|
|
How does one properly benchmark SOB?
I prefer "offline" testing with a fixed WU, i.e. running the LLR program stand-alone outside of BOINC with the very same input data in all tests. And since one SOB task takes far too much time = inappropriately many kWh per test, I stop the program after a fairly low percentage of completion.
Using different random WUs in a test series, or making up a test series by suspending & resuming a task with varying configs, is too imprecise and error prone IMO. |
|
|
mackerelVolunteer tester
Send message
Joined: 2 Oct 08
Posts: 2639
ID: 29980
Credit: 568,393,769
RAC: 3,323
|
|
Maybe an area for future research, but Prime95 29.x benchmark at the appropriate FFT size might be a quick indicator of relative performance since LLR uses the same math library to do the hard work. You can choose how many workers (tasks) and cores, HT on and off. I don't know what the practical implications are but looking at my SoB test units they are using complex FFTs, so that option probably should be selected. I don't know how they differ from non-complex ones.
Edit: I note in the WU output LLR version use 28.14. I do know there were some FFT selection optimisations at some point in 29.x so there may be differences due to that. I'd still lean to using 29.x over 28.x as the updated interface makes it a LOT easier to test at will without having to manually edit config files.
Edit 2: I further note LLR gives the gwnum version, which I don't know if it tracks the Prime95 version... well, test at your own risk! |
|
|
ktoSend message
Joined: 13 Jan 09
Posts: 27
ID: 34164
Credit: 41,469,365
RAC: 0
|
|
Thanks for the suggestions to get multithreaded LLR going. I thought I'd let y'all know I was able to get it working. In case anyone else is wondering what I did...
First and foremost, Windows 10 will by default add a ".txt" extension when you select New->Create Text Document within Windows Explorer. That means the created file has a ".txt" extension. So when I created app_config.xml I was actually creating app_config.xml.txt. Doh!
Regarding restarting PrimeGrid: just reboot. Keeps things simple.
Regarding checking the Event Log: good suggestion. It resolved the first mystery I noted above. Just look for the entry "Found app_config.xml".
If I had my druthers I'd configure the app_config.xml to always run every LLR with as many (real) cores as on my machine. Doing it one-by-one - and specifying the number of cores every time - is a bit cumbersome. I wonder if anyone has any easy method for doing this?
I'm currently running three different boxes with SOB tasks using all available (real) cores. I'll post my run-times when I get a few under my belt, in case anyone is curious.
Thanks again for the help.
____________
|
|
|
dukebgVolunteer tester
Send message
Joined: 21 Nov 17
Posts: 242
ID: 950482
Credit: 23,670,125
RAC: 0
            
|
How does one properly benchmark SOB?
I prefer "offline" testing with a fixed WU, i.e. running the LLR program stand-alone outside of BOINC with the very same input data in all tests. And since one SOB task takes far too much time = inappropriately many kWh per test, I stop the program after a fairly low percentage of completion.
llr has a -d command line option that shows the measurement of "time per bit". You have to wait for 20000 bits (second update it gives) for a better measurement as the first one has init/windup time. Or maybe the third update.
the whole command would look something like this (note that this one is for 4 threads)
llr64 -t4 -d -q"21181*2^28659332+1" "-oForcePRP=1" "-oFermatPRPtest=1"
(the number is the max-in-progress n for one of the SOB's k taken just now from the statistics page.)
Then, to test multi-threading options, make a bat file that launches multiple instances of this at the same time. Make sure to launch them from separate folders and clean up the previous checkpoints (deleting files z* and llr.ini)
|
|
|
|
|
|
mackerel wrote: Maybe an area for future research, but Prime95 29.x benchmark at the appropriate FFT size might be a quick indicator of relative performance since LLR uses the same math library to do the hard work.
While this method has its drawbacks, I guess it is one element of Czech National Team's recent good successes.
kto wrote: I'm currently running three different boxes with SOB tasks using all available (real) cores.
I see that you have one dual-socket machine. On a dual-socket machine with Linux, it is better to run an even number of SOB tasks concurrently, instead of an odd number. (On Windows, all bets are off.)
dukebg wrote: llr has a -d command line option that shows the measurement of "time per bit". You have to wait for 20000 bits (second update it gives) for a better measurement as the first one has init/windup time. Or maybe the third update.
I am patient enough to run a lot more iterations than these, for better precision. Such short tests vary too much to provide insight into cases where the performance does not differ a lot between different settings.
dukebg wrote: Then, to test multi-threading options, make a bat file that launches multiple instances of this at the same time. Make sure to launch them from separate folders and clean up the previous checkpoints (deleting files z* and llr.ini)
Yep, this is pretty much exactly my test environment. (I add that I use input data from a PG WU of which I know credits; this enables me to calculate PPD.) |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
In each chart below, count is the number of validated tasks and elapsed and cpu are the fastest times in hours, with the fastest at the top. Each table is for a particular number of threads. I checked from 100 to 2 threads. If a thread count isn't shown, there's no validated tasks that were run with that number of threads.
The table only counts tasks sent out in the last 60 days, which will primarily be tasks in the current n=28M range.
Note 1: a task that was stopped and continued with a different number of threads will show up in both tables.
Note 2: You'll notice that some lines have a CPU time that is identical to the elapsed time. This is probably due to the way that host's BOINC client is configured.
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
| 24 | Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz [Family 6 Model 62 Stepping 4] | 1 | 33.638888888888886 | 297.51743922222227 |
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-------------------------------------------------------------------------+-------+---------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------+-------+---------+--------------------+
| 20 | Intel(R) Core(TM) i9-7920X CPU @ 2.90GHz [Family 6 Model 85 Stepping 4] | 4 | 12.38 | 182.01802083333334 |
+---------+-------------------------------------------------------------------------+-------+---------+--------------------+
+---------+--------------------------------------------------------------------------+-------+-------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------+-------+-------------------+--------------------+
| 18 | Intel(R) Core(TM) i9-7980XE CPU @ 2.60GHz [Family 6 Model 85 Stepping 4] | 118 | 9.227222222222222 | 130.32901041666668 |
+---------+--------------------------------------------------------------------------+-------+-------------------+--------------------+
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 16 | AMD Ryzen Threadripper 1950X 16-Core Processor [Family 23 Model 1 Stepping 1] | 39 | 15.210833333333333 | 109.03533419444445 |
| 16 | Genuine Intel(R) CPU @ 2.70GHz [Family 6 Model 45 Stepping 6] | 2 | 15.560277777777777 | 189.6831987777778 |
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-----------------------------------------------------------------------+-------+---------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-----------------------------------------------------------------------+-------+---------+--------------------+
| 15 | AMD Ryzen 7 2700X Eight-Core Processor [Family 23 Model 8 Stepping 2] | 1 | 50.865 | 301.50666666666666 |
+---------+-----------------------------------------------------------------------+-------+---------+--------------------+
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| 12 | Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz [Family 6 Model 63 Stepping 2] | 22 | 17.483888888888888 | 174.44923177777775 |
| 12 | Intel(R) Xeon(R) CPU E5-2620 0 @ 2.00GHz [Family 6 Model 45 Stepping 7] | 6 | 26.43888888888889 | 274.3007552222222 |
| 12 | AMD Ryzen 7 1800X Eight-Core Processor [Family 23 Model 1 Stepping 1] | 6 | 27.829722222222223 | 224.33454427777775 |
| 12 | AMD Ryzen 7 1700 Eight-Core Processor [Family 23 Model 1 Stepping 1] | 1 | 40.36277777777778 | 311.6942881944444 |
| 12 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 135.2975303888889 | 135.2975303888889 |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| 11 | Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz [Family 6 Model 158 Stepping 10] | 4 | 44.863055555555555 | 222.67544444444445 |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+---------------------------------------------------------------------------+-------+--------------------+-------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+---------------------------------------------------------------------------+-------+--------------------+-------------------+
| 10 | Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz [Family 6 Model 79 Stepping 1] | 28 | 15.580555555555556 | 136.8384915 |
| 10 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 135.2975303888889 | 135.2975303888889 |
+---------+---------------------------------------------------------------------------+-------+--------------------+-------------------+
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 8 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 2 | 20.833055555555557 | 20.833055555555557 |
| 8 | Intel(R) Xeon(R) CPU E5-2623 v3 @ 3.00GHz [Family 6 Model 63 Stepping 2] | 27 | 23.274722222222223 | 169.2438281111111 |
| 8 | Intel(R) Core(TM) i7-5775C CPU @ 3.30GHz [Family 6 Model 71 Stepping 1] | 1 | 25.46361111111111 | 198.3506111111111 |
| 8 | Intel(R) Core(TM) i7-7820X CPU @ 3.60GHz [Family 6 Model 85 Stepping 4] | 2 | 28.47222222222222 | 56.94444444444444 |
| 8 | AMD Ryzen 7 1700 Eight-Core Processor [Family 23 Model 1 Stepping 1] | 5 | 32.376666666666665 | 207.74805122222222 |
| 8 | AMD Ryzen Threadripper 1950X 16-Core Processor [Family 23 Model 1 Stepping 1] | 2 | 34.99611111111111 | 270.3683941111111 |
| 8 | Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz [Family 6 Model 60 Stepping 3] | 4 | 37.795833333333334 | 266.3890929444444 |
| 8 | Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz [Family 6 Model 158 Stepping 10] | 5 | 44.863055555555555 | 167.8102213611111 |
| 8 | AMD Ryzen 7 2700X Eight-Core Processor [Family 23 Model 8 Stepping 2] | 1 | 50.87388888888889 | 219.75235677777778 |
| 8 | Intel(R) Xeon(R) CPU E5-2667 v2 @ 3.30GHz [Family 6 Model 62 Stepping 4] | 4 | 52.401944444444446 | 188.65073783333332 |
| 8 | Intel(R) Xeon(R) CPU E5520 @ 2.27GHz [Family 6 Model 26 Stepping 5] | 4 | 54.61638888888889 | 416.74198141666665 |
| 8 | Intel(R) Xeon(R) CPU D-1540 @ 2.00GHz [Family 6 Model 86 Stepping 2] | 4 | 68.19194444444445 | 297.1747222222222 |
| 8 | Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz [Family 6 Model 60 Stepping 3] | 1 | 81.37416666666667 | 190.07163194444445 |
| 8 | Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz [Family 6 Model 60 Stepping 3] | 1 | 81.9136100425 | 78.36344444444445 |
| 8 | Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz [Family 6 Model 94 Stepping 3] | 1 | 83.12861111111111 | 114.28796441666667 |
| 8 | AMD Ryzen 7 1800X Eight-Core Processor [Family 23 Model 1 Stepping 1] | 3 | 210.78704427777777 | 210.78704427777777 |
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------+-------+--------------------+--------------------+
| 7 | Intel(R) Core(TM) i9-7940X CPU @ 3.10GHz [Family 6 Model 85 Stepping 4] | 1 | 17.586666666666666 | 100.39038627777778 |
| 7 | Intel(R) Core(TM) i9-7920X CPU @ 2.90GHz [Family 6 Model 85 Stepping 4] | 6 | 60.98444444444444 | 219.05364583333332 |
| 7 | AMD Ryzen 7 1700X Eight-Core Processor [Family 23 Model 1 Stepping 1] | 2 | 335.32059894444444 | 335.32059894444444 |
+---------+-------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| 6 | Intel(R) Core(TM) i7-7800X CPU @ 3.50GHz [Family 6 Model 85 Stepping 4] | 4 | 15.590833333333334 | 16.034444444444443 |
| 6 | Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz [Family 6 Model 158 Stepping 10] | 5 | 15.837777777777777 | 93.0715625 |
| 6 | Intel(R) Core(TM) i7-6900K CPU @ 3.20GHz [Family 6 Model 79 Stepping 1] | 4 | 17.324166666666667 | 97.7238888888889 |
| 6 | Intel(R) Core(TM) i9-7940X CPU @ 3.10GHz [Family 6 Model 85 Stepping 4] | 3 | 17.586666666666666 | 98.48595486111111 |
| 6 | Intel(R) Core(TM) i7-6850K CPU @ 3.60GHz [Family 6 Model 79 Stepping 1] | 16 | 17.86 | 99.0555946111111 |
| 6 | Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz [Family 6 Model 63 Stepping 2] | 19 | 19.141111111111112 | 99.70429688888888 |
| 6 | Intel(R) Core(TM) i5-8600K CPU @ 3.60GHz [Family 6 Model 158 Stepping 10] | 3 | 22.465555555555557 | 22.465555555555557 |
| 6 | Intel(R) Core(TM) i5-8600 CPU @ 3.10GHz [Family 6 Model 158 Stepping 10] | 1 | 28.398888888888887 | 164.73358333333334 |
| 6 | Intel(R) Core(TM) i7-8086K CPU @ 4.00GHz [Family 6 Model 158 Stepping 10] | 1 | 29.89611111111111 | 167.80257377777778 |
| 6 | Intel(R) Core(TM) i7-6800K CPU @ 3.40GHz [Family 6 Model 79 Stepping 1] | 3 | 31.433333333333334 | 130.3205487222222 |
| 6 | Intel(R) Core(TM) i7-4930K CPU @ 3.40GHz [Family 6 Model 62 Stepping 4] | 1 | 46.647777777777776 | 120.76230555555556 |
| 6 | Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz [Family 6 Model 45 Stepping 7] | 6 | 46.8575 | 227.68264755555555 |
| 6 | Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz [Family 6 Model 63 Stepping 2] | 5 | 56.026944444444446 | 189.99004775 |
| 6 | AMD Ryzen 5 1600 Six-Core Processor [Family 23 Model 1 Stepping 1] | 3 | 74.52416666666667 | 176.64426216666666 |
| 6 | Intel(R) Xeon(R) CPU E5-2695 v2 @ 2.40GHz [Family 6 Model 62 Stepping 4] | 1 | 233.1325 | 894.3326085 |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-------------------------------------------------------------------------------+-------+--------------------+-------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------------+-------+--------------------+-------------------+
| 5 | Intel(R) Core(TM) i7-6900K CPU @ 3.20GHz [Family 6 Model 79 Stepping 1] | 1 | 79.2675 | 108.9948611111111 |
| 5 | Intel(R) Core(TM) i7-6850K CPU @ 3.60GHz [Family 6 Model 79 Stepping 1] | 1 | 91.44307725 | 91.44307725 |
| 5 | AMD Ryzen Threadripper 2920X 12-Core Processor [Family 23 Model 8 Stepping 2] | 4 | 117.21416666666667 | 316.0363498333333 |
+---------+-------------------------------------------------------------------------------+-------+--------------------+-------------------+
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 4 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 51 | 18.147471944444444 | 72.58988777777778 |
| 4 | Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz [Family 6 Model 94 Stepping 3] | 17 | 19.819166666666668 | 78.13075955555556 |
| 4 | Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz [Family 6 Model 94 Stepping 3] | 72 | 23.436666666666667 | 85.78393227777778 |
| 4 | Intel(R) Xeon(R) CPU E3-1240 v6 @ 3.70GHz [Family 6 Model 158 Stepping 9 | 10 | 25.268055555555556 | 98.66650000000001 |
| 4 | Intel(R) Core(TM) i5-6500 CPU @ 3.20GHz [Family 6 Model 94 Stepping 3] | 19 | 28.98138888888889 | 86.94416666666666 |
| 4 | Intel(R) Core(TM) i5-6600K CPU @ 3.50GHz [Family 6 Model 94 Stepping 3] | 4 | 30.07111111111111 | 107.93157116666667 |
| 4 | Intel(R) Core(TM) i5-7400 CPU @ 3.00GHz [Family 6 Model 158 Stepping 9] | 1 | 30.335277777777776 | 30.335277777777776 |
| 4 | Intel(R) Core(TM) i7-7700 CPU @ 3.60GHz [Family 6 Model 158 Stepping 9] | 11 | 30.432777777777776 | 101.5621571111111 |
| 4 | Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz [Family 6 Model 60 Stepping 3] | 19 | 31.134166666666665 | 80.9071111111111 |
| 4 | Intel(R) Xeon(R) CPU E3-1226 v3 @ 3.30GHz [Family 6 Model 60 Stepping 3] | 28 | 31.720833333333335 | 109.35075563888888 |
| 4 | Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz [Family 6 Model 60 Stepping 3] | 71 | 31.919444444444444 | 121.89534288888889 |
| 4 | Intel(R) Core(TM) i5-4670K CPU @ 3.40GHz [Family 6 Model 60 Stepping 3] | 1 | 34.173611111111114 | 118.77926538888889 |
| 4 | Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz [Family 6 Model 63 Stepping 2] | 2 | 35.46805555555556 | 140.10922308333332 |
| 4 | Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz [Family 6 Model 58 Stepping 9] | 8 | 35.700833333333335 | 125.28943413888888 |
| 4 | Intel(R) Core(TM) i7-7820HQ CPU @ 2.90GHz [x86 Family 6 Model 158 Stepping 9] | 8 | 35.78811111111111 | 35.78811111111111 |
| 4 | Intel(R) Core(TM) i5-4590 CPU @ 3.30GHz [Family 6 Model 60 Stepping 3] | 18 | 36.21194444444444 | 133.62274738888888 |
| 4 | Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz [Family 6 Model 42 Stepping 7] | 3 | 36.38527777777778 | 143.2641388888889 |
| 4 | Intel(R) Core(TM) i5-3470S CPU @ 2.90GHz [Family 6 Model 58 Stepping 9] | 1 | 40.02111111111111 | 155.52744444444446 |
| 4 | Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz [Family 6 Model 42 Stepping 7] | 18 | 40.39555555555555 | 149.08126363888888 |
| 4 | Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz [Family 6 Model 58 Stepping 9] | 15 | 40.84111111111111 | 153.74727955555556 |
| 4 | Intel(R) Xeon(R) CPU E5-2687W v3 @ 3.10GHz [Family 6 Model 63 Stepping 2] | 7 | 42.52055555555555 | 112.3628611111111 |
| 4 | Intel(R) Xeon(R) Platinum 8168 CPU @ 2.70GHz [Family 6 Model 85 Stepping 4] | 15 | 42.9725 | 165.78449999999998 |
| 4 | Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz [Family 6 Model 58 Stepping 9] | 3 | 43.0975 | 155.48871005555557 |
| 4 | Intel(R) Core(TM) i7-4771 CPU @ 3.50GHz [x86 Family 6 Model 60 Stepping 3] | 13 | 43.49263888888889 | 43.49263888888889 |
| 4 | Intel(R) Xeon(R) CPU E5-2650 v3 @ 2.30GHz [Family 6 Model 63 Stepping 2] | 29 | 46.11416666666667 | 154.21372222222223 |
| 4 | Intel(R) Core(TM) i7-3770T CPU @ 2.50GHz [Family 6 Model 58 Stepping 9] | 1 | 46.378055555555555 | 182.52863888888888 |
| 4 | Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz [Family 6 Model 45 Stepping 7] | 14 | 47.74861111111111 | 176.5703333333333 |
| 4 | Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz [Family 6 Model 62 Stepping 4] | 6 | 48.57416666666666 | 187.50247222222222 |
| 4 | Intel(R) Core(TM) i7-4750HQ CPU @ 2.00GHz [x86 Family 6 Model 70 Stepping 1] | 1 | 48.881499999999996 | 48.881499999999996 |
| 4 | Intel(R) Core(TM) i5-2500 CPU @ 3.30GHz [Family 6 Model 42 Stepping 7] | 3 | 49.46722222222222 | 188.93355555555556 |
| 4 | Intel(R) Core(TM) i5-2400S CPU @ 2.50GHz [Family 6 Model 42 Stepping 7] | 1 | 50.72611111111111 | 195.7026111111111 |
| 4 | Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz [Family 6 Model 42 Stepping 7] | 5 | 51.489444444444445 | 179.8108593888889 |
| 4 | Intel(R) Core(TM) i5-2550K CPU @ 3.40GHz [Family 6 Model 42 Stepping 7] | 5 | 52.44638888888889 | 189.02893663888887 |
| 4 | Intel(R) Core(TM) i7-3615QM CPU @ 2.30GHz [x86 Family 6 Model 58 Stepping 9] | 1 | 60.685762375555555 | 60.68494444444444 |
| 4 | Intel(R) Core(TM) i9-7960X CPU @ 2.80GHz [Family 6 Model 85 Stepping 4] | 1 | 61.6575 | 131.58113280555554 |
| 4 | Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz [Family 6 Model 60 Stepping 3] | 1 | 81.37416666666667 | 190.07163194444445 |
| 4 | Intel(R) Core(TM) i7-4510U CPU @ 2.00GHz [Family 6 Model 69 Stepping 1] | 1 | 86.33333333333333 | 174.7771111111111 |
| 4 | Intel(R) Core(TM) i7 CPU 950 @ 3.07GHz [Family 6 Model 26 Stepping 5] | 8 | 88.72416666666666 | 351.060829 |
| 4 | Intel(R) Core(TM) i3-2100 CPU @ 3.10GHz [Family 6 Model 42 Stepping 7] | 2 | 93.67777777777778 | 360.19399358333334 |
| 4 | Intel(R) Xeon(R) CPU X3460 @ 2.80GHz [Family 6 Model 30 Stepping 5] | 2 | 98.35444444444444 | 387.31583333333333 |
| 4 | Intel(R) Core(TM) i5-4300U CPU @ 1.90GHz [Family 6 Model 69 Stepping 1] | 1 | 114.49138888888889 | 370.1404093333333 |
| 4 | AMD Ryzen Threadripper 2920X 12-Core Processor [Family 23 Model 8 Stepping 2] | 1 | 117.21416666666667 | 316.0363498333333 |
| 4 | Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz [Family 6 Model 45 Stepping 7] | 2 | 125.19572222222222 | 125.19572222222222 |
| 4 | Intel(R) Xeon(R) CPU E3-1270 V2 @ 3.50GHz [Family 6 Model 58 Stepping 9] | 1 | 134.43972222222223 | 176.34381944444445 |
| 4 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 135.2975303888889 | 135.2975303888889 |
| 4 | Genuine Intel(R) CPU @ 2.90GHz [Family 6 Model 45 Stepping 6] | 1 | 167.4863611111111 | 167.4863611111111 |
| 4 | Intel(R) Core(TM) i7-3840QM CPU @ 2.80GHz [Family 6 Model 58 Stepping 9] | 2 | 223.00308161111113 | 223.00308161111113 |
| 4 | AMD Opteron(tm) Processor 3280 [Family 21 Model 1 Stepping 2] | 1 | 244.60861111111112 | 692.8864582222222 |
| 4 | Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz [Family 6 Model 142 Stepping 9] | 1 | 249.23037327777777 | 249.23037327777777 |
| 4 | Intel(R) Core(TM) i5 CPU 750 @ 2.67GHz [Family 6 Model 30 Stepping 5] | 1 | 412.47811441666664 | 412.47811441666664 |
| 4 | Intel(R) Pentium(R) CPU G4560 @ 3.50GHz [Family 6 Model 158 Stepping 9] | 1 | 436.3376085 | 436.3376085 |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 3 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 3 | 22.672313055555556 | 68.01693916666666 |
| 3 | Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz [Family 6 Model 158 Stepping 10] | 1 | 23.270555555555557 | 98.43414930555555 |
| 3 | Intel(R) Core(TM) i5-6500 CPU @ 3.20GHz [Family 6 Model 94 Stepping 3] | 1 | 28.98138888888889 | 86.94416666666666 |
| 3 | Intel(R) Core(TM) i5-6600K CPU @ 3.50GHz [Family 6 Model 94 Stepping 3] | 1 | 36.751666666666665 | 101.6596441111111 |
| 3 | Intel(R) Core(TM) i5-4670K CPU @ 3.40GHz [Family 6 Model 60 Stepping 3] | 14 | 38.03111111111111 | 111.63954597222222 |
| 3 | Intel(R) Core(TM) i7-7700 CPU @ 3.60GHz [Family 6 Model 158 Stepping 9] | 13 | 41.51166666666666 | 121.97755555555555 |
| 3 | Intel(R) Xeon(R) CPU E5-2687W v3 @ 3.10GHz [Family 6 Model 63 Stepping 2] | 1 | 44.57388888888889 | 112.3628611111111 |
| 3 | Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz [Family 6 Model 58 Stepping 9] | 8 | 48.596944444444446 | 140.56565005555555 |
| 3 | Intel(R) Core(TM) i9-7960X CPU @ 2.80GHz [Family 6 Model 85 Stepping 4] | 5 | 48.943888888888885 | 129.66131077777777 |
| 3 | AMD Ryzen Threadripper 1950X 16-Core Processor [Family 23 Model 1 Stepping 1] | 18 | 54.95666666666666 | 161.9469661388889 |
| 3 | AMD Ryzen 5 1600X Six-Core Processor [Family 23 Model 1 Stepping 1] | 19 | 66.01944444444445 | 165.37591666666668 |
| 3 | Intel(R) Core(TM) i9-7940X CPU @ 3.10GHz [Family 6 Model 85 Stepping 4] | 1 | 108.93088974999999 | 108.93088974999999 |
| 3 | Intel(R) Core(TM) i7 CPU 940 @ 2.93GHz [Family 6 Model 26 Stepping 4] | 1 | 123.32611111111112 | 353.81 |
| 3 | Intel(R) Xeon(R) CPU E5-2620 0 @ 2.00GHz [Family 6 Model 45 Stepping 7] | 8 | 257.8011111111111 | 309.7438681944444 |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 2 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 4 | 35.270833333333336 | 67.75619444444445 |
| 2 | Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz [Family 6 Model 94 Stepping 3] | 19 | 48.70527777777778 | 95.69704861111111 |
| 2 | Intel Core Processor (Broadwell) [Family 6 Model 61 Stepping 2] | 17 | 49.021388888888886 | 96.74547222222222 |
| 2 | Intel Core Processor (Haswell, no TSX) [Family 6 Model 60 Stepping 1] | 28 | 49.1675 | 97.40852777777778 |
| 2 | Intel(R) Core(TM) i5-6600 CPU @ 3.30GHz [Family 6 Model 94 Stepping 3] | 8 | 49.431666666666665 | 95.87002169444445 |
| 2 | Intel(R) Core(TM) i5-7200U CPU @ 2.50GHz [Family 6 Model 142 Stepping 9] | 1 | 59.8175 | 117.0059808888889 |
| 2 | Intel(R) Core(TM) i9-7960X CPU @ 2.80GHz [Family 6 Model 85 Stepping 4] | 3 | 61.6575 | 131.58113280555554 |
| 2 | Intel(R) Core(TM) i7-2700K CPU @ 3.50GHz [Family 6 Model 42 Stepping 7] | 1 | 72.40666666666667 | 142.6563888888889 |
| 2 | Intel(R) Xeon(R) CPU E5-1660 v2 @ 3.70GHz [Family 6 Model 62 Stepping 4] | 1 | 77.45666666666666 | 152.97315058333334 |
| 2 | Intel(R) Core(TM) i5-3330 CPU @ 3.00GHz [Family 6 Model 58 Stepping 9] | 6 | 78.25083333333333 | 151.70339777777778 |
| 2 | Intel(R) Core(TM) i7-4510U CPU @ 2.00GHz [Family 6 Model 69 Stepping 1] | 4 | 86.33333333333333 | 173.84366666666665 |
| 2 | Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz [Family 6 Model 58 Stepping 9] | 5 | 92.14 | 153.24448833333335 |
| 2 | Intel(R) Core(TM) i5-3320M CPU @ 2.60GHz [Family 6 Model 58 Stepping 9] | 4 | 131.59416666666667 | 188.2761378888889 |
| 2 | Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz [Family 6 Model 42 Stepping 7] | 1 | 165.96548586111112 | 165.96548586111112 |
| 2 | Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz [Family 6 Model 58 Stepping 9] | 1 | 169.67777777777778 | 269.8104036388889 |
| 2 | Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz [Family 6 Model 94 Stepping 3] | 1 | 180.11103733333331 | 180.11103733333331 |
| 2 | AMD A6-6400B APU with Radeon(tm) HD Graphics [Family 21 Model 19 Stepping 1] | 1 | 542.6242144166666 | 542.6242144166666 |
+---------+------------------------------------------------------------------------------+-------+--------------------+--------------------+
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Really good data as usual and in spite of various minor shortcomings. |
|
|
|
|
|
Michael Goetz wrote: In each chart below ...
Thank you !
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment. |
|
|
|
|
In each chart below...
Thank you, Michael!
This is so helpful as most of us have varied CPUs crunching. Could Kaby Lake be that more efficient on a slower CPU? I wonder if the memory configuration influenced these samples:
| 4 | Intel(R) Core(TM) i5-7400 CPU @ 3.00GHz [Family 6 Model 158 Stepping 9] | 1 | 30.335277777777776 | 30.335277777777776 |
| 4 | Intel(R) Core(TM) i7-7700 CPU @ 3.60GHz [Family 6 Model 158 Stepping 9] | 11 | 30.432777777777776 | 101.5621571111111 |
Regardless, very useful benchmarks and your efforts are tremendously appreciated!
|
|
|
|
|
|
FishFry wrote: Could Kaby Lake be that more efficient on a slower CPU?
1.) There is only a single task for the i5 in the sample. Remember that there is a considerable variation in size of the tasks that were in circulation during the last 60 days.
2.) We don't know what the i7 owners ran concurrently with the 4-threaded SoB tasks. For all we know, they had GPU tasks going in addition.
So the answer is: We cannot conclude anything at all from the two records which you quoted.
But yes, memory performance does have effect on SoB-LLR performance. I tested with different RAM configs, but this was long ago with the v7 application.
--------
(edit)
And remember: If the boinc client fetched more than one task in a single request for more work, then the run time of multithreaded tasks which gets recorded at the server when finished gets semi-randomly corrupted. |
|
|
dukebgVolunteer tester
Send message
Joined: 21 Nov 17
Posts: 242
ID: 950482
Credit: 23,670,125
RAC: 0
|
In each chart below, count is the number of validated tasks and elapsed and cpu are the fastest times in hours, with the fastest at the top. Each table is for a particular number of threads.
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 4 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 51 | 18.147471944444444 | 72.58988777777778 |
| 4 | Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz [Family 6 Model 94 Stepping 3] | 17 | 19.819166666666668 | 78.13075955555556 |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
To reiterate, the values in elapsed and cpu-time are the single best runs out of the corresponding 51 and 17, correct (may be different ones for two columns)? Not the average of these 51 and 17 runs and then the whole table is sorted by elapsed?
I ran benchmarks in preparation to the challenge on my (recently built) i7-9700K and I just don't see getting lower than about 20-21 hours. And these "older" cpus got 18-19?.. Being just single "lucky" best values (rather than the average) can be explained by BOINC messing up run times in some cases (it's possible sometimes – with tasks being suspended and client restarted and stuff – to have run time "restart" and then boinc reports the value which was lower than the real one). The memory is a bottleneck for SOB, so my explanation#2 is maybe these people had faster memory sticks than i do? (frequency and/or latency). I didn't go crazy price max top existing, but I thought I still got fairly "competitive" ones. My explanation #3 is that I ran the test with a leading edge (max in progress) number from the stats and the "lucky" runs could probably be for numbers from n=26M range (which still had active tasks 60 days ago, so might've got into Michael's tables).
It would ease my worries if you could look up average elapsed for these two rows (in case it was indeed single best values) and tell me that the averages there are more like 23-25 hours.
(Also, if you could check if runs for n=26M or lower range numbers got into these stats too – that's some skew, better to know if it's there) |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
To reiterate, the values in elapsed and cpu-time are the single best runs out of the corresponding 51 and 17, correct (may be different ones for two columns)? Not the average of these 51 and 17 runs and then the whole table is sorted by elapsed?
Correct. Definitely NOT the average. The average includes so much noise that it's absolutely worthless. Consider the elapsed time value being inflated either because something else is eating 99% of the CPU time, or the "bug" whereby BOINC downloads multiple tasks when multitasking and multiplies the elapsed time by N. The fastest elapsed time is likely to be correct. The average elapsed time is garbage.
I ran benchmarks in preparation to the challenge on my (recently built) i7-9700K and I just don't see getting lower than about 20-21 hours. And these "older" cpus got 18-19?..
It's not clear the Kaby Lake is really any slower than the latest chips. Certainly not significantly slower. My understanding is that the big advantage with the newer chips is more cores.
Also, you have no clue at all what clock speeds and memory speeds are being used on anybody else's computers.
Finally, read again what I wrote at the top of the tables. In particular, the part about tasks that were run with varying number of threads count in ALL thread counts. A task than ran for 10 hours with 18 threads and for 2 minutes with 2 threads would show up in both the 18 thread and the 2 thread tables.
There's a LOT of noise and unknowns in the data, and you should not worry about any numbers that don't seem to make sense.
(Also, if you could check if runs for n=26M or lower range numbers got into these stats too – that's some skew, better to know if it's there)
I picked the calendar window (63 days) such that it started after we we moved to the 28M range, but it's possible some resends of smaller tasks were included in the data set. Even within the 28M range, however, there's going to be a difference in runtimes between 28.0M and 28.5M, which is about where we are today.
And, of course, depending on which K you have, there are two different FFT sizes. That changes the run time significantly. When you did your benchmarks, were you using the 2560K FFT or the 2880K FFT? I benchmarked my systems with both sizes.
I will try running it again and explicitly excluding anything below 28M. If the results are different, I'll post a new table.
____________
My lucky number is 75898524288+1 |
|
|
|
|
1.) There is only a single task for the i5 in the sample. Remember that there is a considerable variation in size of the tasks that were in circulation during the last 60 days.
2.) We don't know what the i7 owners ran concurrently with the 4-threaded SoB tasks. For all we know, they had GPU tasks going in addition.
So the answer is: We cannot conclude anything at all from the two records which you quoted.
But yes, memory performance does have effect on SoB-LLR performance. I tested with different RAM configs, but this was long ago with the v7 application.
--------
(edit)
And remember: If the boinc client fetched more than one task in a single request for more work, then the run time of multithreaded tasks which gets recorded at the server when finished gets semi-randomly corrupted.
Good points to consider. Thank you for adding some more color to the charts.
Good luck, all! |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
There were some slight differences, so here's the data explicitly limited to n=28M:
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
| 24 | Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz [Family 6 Model 62 Stepping 4] | 1 | 33.638888888888886 | 297.51743922222227 |
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-------------------------------------------------------------------------+-------+---------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------+-------+---------+--------------------+
| 20 | Intel(R) Core(TM) i9-7920X CPU @ 2.90GHz [Family 6 Model 85 Stepping 4] | 4 | 12.38 | 182.01802083333334 |
+---------+-------------------------------------------------------------------------+-------+---------+--------------------+
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
| 18 | Intel(R) Core(TM) i9-7980XE CPU @ 2.60GHz [Family 6 Model 85 Stepping 4] | 123 | 9.227222222222222 | 130.32901041666668 |
| 18 | Intel(R) Xeon(R) CPU E5-2686 v3 @ 2.00GHz [Family 6 Model 63 Stepping 2] | 1 | 59.327222222222225 | 180.79346355555558 |
+---------+--------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 16 | AMD Ryzen Threadripper 1950X 16-Core Processor [Family 23 Model 1 Stepping 1] | 40 | 15.210833333333333 | 109.03533419444445 |
| 16 | Genuine Intel(R) CPU @ 2.70GHz [Family 6 Model 45 Stepping 6] | 2 | 15.560277777777777 | 189.6831987777778 |
| 16 | AMD Ryzen 7 1700X Eight-Core Processor [Family 23 Model 1 Stepping 1] | 1 | 58.27611111111111 | 440.18444444444447 |
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-----------------------------------------------------------------------+-------+---------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-----------------------------------------------------------------------+-------+---------+--------------------+
| 15 | AMD Ryzen 7 2700X Eight-Core Processor [Family 23 Model 8 Stepping 2] | 1 | 50.865 | 301.50666666666666 |
+---------+-----------------------------------------------------------------------+-------+---------+--------------------+
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| 12 | Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz [Family 6 Model 63 Stepping 2] | 24 | 17.483888888888888 | 174.44923177777775 |
| 12 | Intel(R) Xeon(R) CPU E5-2620 0 @ 2.00GHz [Family 6 Model 45 Stepping 7] | 7 | 26.43888888888889 | 274.3007552222222 |
| 12 | AMD Ryzen 7 1800X Eight-Core Processor [Family 23 Model 1 Stepping 1] | 10 | 27.829722222222223 | 224.33454427777775 |
| 12 | AMD Ryzen 7 1700 Eight-Core Processor [Family 23 Model 1 Stepping 1] | 1 | 40.36277777777778 | 311.6942881944444 |
| 12 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 135.2975303888889 | 135.2975303888889 |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| 11 | Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz [Family 6 Model 158 Stepping 10] | 3 | 44.863055555555555 | 225.04433333333333 |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+---------------------------------------------------------------------------+-------+--------------------+-------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+---------------------------------------------------------------------------+-------+--------------------+-------------------+
| 10 | Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz [Family 6 Model 79 Stepping 1] | 27 | 15.580555555555556 | 136.8384915 |
| 10 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 2 | 135.2975303888889 | 135.2975303888889 |
+---------+---------------------------------------------------------------------------+-------+--------------------+-------------------+
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 8 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 3 | 20.833055555555557 | 20.833055555555557 |
| 8 | Intel(R) Xeon(R) CPU E5-2623 v3 @ 3.00GHz [Family 6 Model 63 Stepping 2] | 26 | 23.274722222222223 | 169.2438281111111 |
| 8 | AMD Ryzen 7 1700 Eight-Core Processor [Family 23 Model 1 Stepping 1] | 7 | 25.37388888888889 | 196.48741319444446 |
| 8 | Intel(R) Core(TM) i7-7820X CPU @ 3.60GHz [Family 6 Model 85 Stepping 4] | 3 | 28.47222222222222 | 56.94444444444444 |
| 8 | AMD Ryzen Threadripper 1950X 16-Core Processor [Family 23 Model 1 Stepping 1] | 4 | 33.55888888888889 | 242.2691536388889 |
| 8 | Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz [Family 6 Model 60 Stepping 3] | 4 | 37.795833333333334 | 266.3890929444444 |
| 8 | Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz [Family 6 Model 158 Stepping 10] | 5 | 44.863055555555555 | 167.8102213611111 |
| 8 | Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz [Family 6 Model 45 Stepping 6] | 1 | 45.96666666666667 | 183.86666666666667 |
| 8 | AMD Ryzen 7 2700X Eight-Core Processor [Family 23 Model 8 Stepping 2] | 1 | 50.87388888888889 | 219.75235677777778 |
| 8 | Intel(R) Xeon(R) CPU E5-2667 v2 @ 3.30GHz [Family 6 Model 62 Stepping 4] | 3 | 52.401944444444446 | 188.65073783333332 |
| 8 | Intel(R) Xeon(R) CPU E5520 @ 2.27GHz [Family 6 Model 26 Stepping 5] | 4 | 54.61638888888889 | 416.74198141666665 |
| 8 | AMD Ryzen 7 1700X Eight-Core Processor [Family 23 Model 1 Stepping 1] | 1 | 57.36833333333333 | 114.73666666666666 |
| 8 | Intel(R) Xeon(R) CPU D-1540 @ 2.00GHz [Family 6 Model 86 Stepping 2] | 4 | 68.19194444444445 | 297.1747222222222 |
| 8 | Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz [Family 6 Model 60 Stepping 3] | 1 | 81.37416666666667 | 190.07163194444445 |
| 8 | Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz [Family 6 Model 60 Stepping 3] | 1 | 81.9136100425 | 78.36344444444445 |
| 8 | Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz [Family 6 Model 94 Stepping 3] | 1 | 83.12861111111111 | 114.28796441666667 |
| 8 | AMD Opteron(TM) Processor 6276 [Family 21 Model 1 Stepping 2] | 2 | 94.59805555555556 | 728.9463888888889 |
| 8 | AMD Ryzen 7 1800X Eight-Core Processor [Family 23 Model 1 Stepping 1] | 3 | 210.78704427777777 | 210.78704427777777 |
+---------+-------------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------+-------+--------------------+--------------------+
| 7 | Intel(R) Core(TM) i9-7940X CPU @ 3.10GHz [Family 6 Model 85 Stepping 4] | 1 | 17.586666666666666 | 100.39038627777778 |
| 7 | Intel(R) Core(TM) i9-7920X CPU @ 2.90GHz [Family 6 Model 85 Stepping 4] | 6 | 60.98444444444444 | 219.05364583333332 |
| 7 | AMD Ryzen 7 1700X Eight-Core Processor [Family 23 Model 1 Stepping 1] | 2 | 335.32059894444444 | 335.32059894444444 |
+---------+-------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
| 6 | Intel(R) Core(TM) i7-7800X CPU @ 3.50GHz [Family 6 Model 85 Stepping 4] | 4 | 15.590833333333334 | 16.034444444444443 |
| 6 | Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz [Family 6 Model 158 Stepping 10] | 5 | 15.837777777777777 | 93.0715625 |
| 6 | Intel(R) Core(TM) i7-6900K CPU @ 3.20GHz [Family 6 Model 79 Stepping 1] | 4 | 17.324166666666667 | 97.7238888888889 |
| 6 | Intel(R) Core(TM) i9-7940X CPU @ 3.10GHz [Family 6 Model 85 Stepping 4] | 3 | 17.586666666666666 | 98.48595486111111 |
| 6 | Intel(R) Core(TM) i7-6850K CPU @ 3.60GHz [Family 6 Model 79 Stepping 1] | 16 | 17.86 | 99.0555946111111 |
| 6 | Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz [Family 6 Model 63 Stepping 2] | 19 | 19.141111111111112 | 99.70429688888888 |
| 6 | Intel(R) Core(TM) i5-8600K CPU @ 3.60GHz [Family 6 Model 158 Stepping 10] | 3 | 23.86 | 23.86 |
| 6 | Intel(R) Core(TM) i5-8600 CPU @ 3.10GHz [Family 6 Model 158 Stepping 10] | 1 | 28.398888888888887 | 164.73358333333334 |
| 6 | Intel(R) Core(TM) i7-8086K CPU @ 4.00GHz [Family 6 Model 158 Stepping 10] | 1 | 29.89611111111111 | 167.80257377777778 |
| 6 | Intel(R) Core(TM) i7-6800K CPU @ 3.40GHz [Family 6 Model 79 Stepping 1] | 2 | 31.433333333333334 | 136.42853819444443 |
| 6 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 42.59638888888889 | 98.74000433333333 |
| 6 | Intel(R) Core(TM) i7-4930K CPU @ 3.40GHz [Family 6 Model 62 Stepping 4] | 1 | 46.647777777777776 | 120.76230555555556 |
| 6 | Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz [Family 6 Model 45 Stepping 7] | 7 | 46.8575 | 227.68264755555555 |
| 6 | Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz [Family 6 Model 63 Stepping 2] | 5 | 56.026944444444446 | 189.99004775 |
| 6 | Intel(R) Core(TM) i7-9700K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 67.24638888888889 | 107.05632377777778 |
| 6 | AMD Ryzen 5 1600 Six-Core Processor [Family 23 Model 1 Stepping 1] | 3 | 74.52416666666667 | 176.64426216666666 |
| 6 | Intel(R) Xeon(R) CPU E5-2695 v2 @ 2.40GHz [Family 6 Model 62 Stepping 4] | 1 | 233.1325 | 894.3326085 |
+---------+---------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+-------------------------------------------------------------------------------+-------+--------------------+-------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+-------------------------------------------------------------------------------+-------+--------------------+-------------------+
| 5 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 42.59638888888889 | 98.74000433333333 |
| 5 | Intel(R) Core(TM) i7-6900K CPU @ 3.20GHz [Family 6 Model 79 Stepping 1] | 1 | 79.2675 | 108.9948611111111 |
| 5 | Intel(R) Core(TM) i7-6850K CPU @ 3.60GHz [Family 6 Model 79 Stepping 1] | 1 | 91.44307725 | 91.44307725 |
| 5 | AMD Ryzen Threadripper 2920X 12-Core Processor [Family 23 Model 8 Stepping 2] | 4 | 117.21416666666667 | 316.0363498333333 |
+---------+-------------------------------------------------------------------------------+-------+--------------------+-------------------+
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 4 | Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz [Family 6 Model 94 Stepping 3] | 15 | 19.819166666666668 | 78.13075955555556 |
| 4 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 41 | 21.38294027777778 | 85.53176111111112 |
| 4 | Intel(R) Xeon(R) CPU E3-1240 v6 @ 3.70GHz [Family 6 Model 158 Stepping 9 | 12 | 25.268055555555556 | 98.66650000000001 |
| 4 | Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz [Family 6 Model 94 Stepping 3] | 69 | 26.74111111111111 | 97.15629775000001 |
| 4 | Intel(R) Core(TM) i5-6500 CPU @ 3.20GHz [Family 6 Model 94 Stepping 3] | 17 | 28.98138888888889 | 86.94416666666666 |
| 4 | Intel(R) Core(TM) i5-6600K CPU @ 3.50GHz [Family 6 Model 94 Stepping 3] | 4 | 30.07111111111111 | 107.93157116666667 |
| 4 | Intel(R) Core(TM) i5-7400 CPU @ 3.00GHz [Family 6 Model 158 Stepping 9] | 2 | 30.335277777777776 | 30.335277777777776 |
| 4 | Intel(R) Core(TM) i7-7700 CPU @ 3.60GHz [Family 6 Model 158 Stepping 9] | 12 | 30.432777777777776 | 101.5621571111111 |
| 4 | Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz [Family 6 Model 60 Stepping 3] | 19 | 31.134166666666665 | 80.9071111111111 |
| 4 | Intel(R) Xeon(R) CPU E3-1226 v3 @ 3.30GHz [Family 6 Model 60 Stepping 3] | 27 | 31.720833333333335 | 124.72346250000001 |
| 4 | Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz [Family 6 Model 60 Stepping 3] | 70 | 32.88055555555555 | 127.52136283333334 |
| 4 | Intel(R) Core(TM) i5-4670K CPU @ 3.40GHz [Family 6 Model 60 Stepping 3] | 1 | 34.173611111111114 | 118.77926538888889 |
| 4 | Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz [Family 6 Model 63 Stepping 2] | 2 | 35.46805555555556 | 140.10922308333332 |
| 4 | Intel(R) Core(TM) i7-7820HQ CPU @ 2.90GHz [x86 Family 6 Model 158 Stepping 9] | 9 | 35.78811111111111 | 35.78811111111111 |
| 4 | Intel(R) Core(TM) i5-4590 CPU @ 3.30GHz [Family 6 Model 60 Stepping 3] | 17 | 36.21194444444444 | 133.62274738888888 |
| 4 | Intel(R) Core(TM) i5-3470S CPU @ 2.90GHz [Family 6 Model 58 Stepping 9] | 1 | 40.02111111111111 | 155.52744444444446 |
| 4 | Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz [Family 6 Model 58 Stepping 9] | 7 | 40.49 | 143.2937608888889 |
| 4 | Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz [Family 6 Model 58 Stepping 9] | 15 | 40.84111111111111 | 153.74727955555556 |
| 4 | Intel(R) Xeon(R) CPU E5-2687W v3 @ 3.10GHz [Family 6 Model 63 Stepping 2] | 7 | 42.52055555555555 | 112.3628611111111 |
| 4 | Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz [Family 6 Model 42 Stepping 7] | 16 | 42.58722222222222 | 149.34919969444445 |
| 4 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 2 | 42.59638888888889 | 98.74000433333333 |
| 4 | Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz [Family 6 Model 58 Stepping 9] | 3 | 43.0975 | 155.48871005555557 |
| 4 | Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz [Family 6 Model 42 Stepping 7] | 2 | 43.25416666666667 | 169.41463330555555 |
| 4 | Intel(R) Core(TM) i7-4771 CPU @ 3.50GHz [x86 Family 6 Model 60 Stepping 3] | 12 | 43.49263888888889 | 43.49263888888889 |
| 4 | Intel(R) Xeon(R) Platinum 8168 CPU @ 2.70GHz [Family 6 Model 85 Stepping 4] | 14 | 45.33361111111111 | 174.86227777777776 |
| 4 | Intel(R) Xeon(R) CPU E5-2650 v3 @ 2.30GHz [Family 6 Model 63 Stepping 2] | 29 | 46.11416666666667 | 154.21372222222223 |
| 4 | Intel(R) Core(TM) i7-3770T CPU @ 2.50GHz [Family 6 Model 58 Stepping 9] | 1 | 46.378055555555555 | 182.52863888888888 |
| 4 | Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz [Family 6 Model 45 Stepping 7] | 15 | 47.74861111111111 | 176.5703333333333 |
| 4 | Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz [Family 6 Model 62 Stepping 4] | 7 | 48.57416666666666 | 187.50247222222222 |
| 4 | Intel(R) Core(TM) i7-4750HQ CPU @ 2.00GHz [x86 Family 6 Model 70 Stepping 1] | 1 | 48.881499999999996 | 48.881499999999996 |
| 4 | Intel(R) Core(TM) i5-2500 CPU @ 3.30GHz [Family 6 Model 42 Stepping 7] | 4 | 49.46722222222222 | 188.93355555555556 |
| 4 | Intel(R) Core(TM) i5-2400S CPU @ 2.50GHz [Family 6 Model 42 Stepping 7] | 1 | 50.72611111111111 | 195.7026111111111 |
| 4 | Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz [Family 6 Model 42 Stepping 7] | 5 | 51.489444444444445 | 179.8108593888889 |
| 4 | Intel(R) Core(TM) i5-2550K CPU @ 3.40GHz [Family 6 Model 42 Stepping 7] | 5 | 52.44638888888889 | 189.02893663888887 |
| 4 | Intel(R) Xeon(R) CPU E5-2686 v3 @ 2.00GHz [Family 6 Model 63 Stepping 2] | 1 | 59.327222222222225 | 180.79346355555558 |
| 4 | Intel(R) Core(TM) i7-3615QM CPU @ 2.30GHz [x86 Family 6 Model 58 Stepping 9] | 1 | 60.685762375555555 | 60.68494444444444 |
| 4 | Intel(R) Core(TM) i9-7960X CPU @ 2.80GHz [Family 6 Model 85 Stepping 4] | 1 | 61.6575 | 131.58113280555554 |
| 4 | Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz [Family 6 Model 60 Stepping 3] | 1 | 81.37416666666667 | 190.07163194444445 |
| 4 | Intel(R) Core(TM) i7-4510U CPU @ 2.00GHz [Family 6 Model 69 Stepping 1] | 1 | 86.33333333333333 | 174.7771111111111 |
| 4 | Intel(R) Core(TM) i7 CPU 950 @ 3.07GHz [Family 6 Model 26 Stepping 5] | 8 | 88.72416666666666 | 351.060829 |
| 4 | Intel(R) Core(TM) i3-2100 CPU @ 3.10GHz [Family 6 Model 42 Stepping 7] | 2 | 93.67777777777778 | 360.19399358333334 |
| 4 | Intel(R) Xeon(R) CPU X3460 @ 2.80GHz [Family 6 Model 30 Stepping 5] | 2 | 98.35444444444444 | 387.31583333333333 |
| 4 | Intel(R) Core(TM) i5-4300U CPU @ 1.90GHz [Family 6 Model 69 Stepping 1] | 1 | 114.49138888888889 | 370.1404093333333 |
| 4 | AMD Ryzen Threadripper 2920X 12-Core Processor [Family 23 Model 8 Stepping 2] | 1 | 117.21416666666667 | 316.0363498333333 |
| 4 | Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz [Family 6 Model 45 Stepping 7] | 2 | 125.19572222222222 | 125.19572222222222 |
| 4 | Intel(R) Xeon(R) CPU E3-1270 V2 @ 3.50GHz [Family 6 Model 58 Stepping 9] | 1 | 134.43972222222223 | 176.34381944444445 |
| 4 | Genuine Intel(R) CPU @ 2.90GHz [Family 6 Model 45 Stepping 6] | 2 | 167.4863611111111 | 167.4863611111111 |
| 4 | Intel(R) Core(TM) i7-3840QM CPU @ 2.80GHz [Family 6 Model 58 Stepping 9] | 2 | 223.00308161111113 | 223.00308161111113 |
| 4 | Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz [Family 6 Model 142 Stepping 9] | 2 | 223.06042100000002 | 223.06042100000002 |
| 4 | AMD Opteron(tm) Processor 3280 [Family 21 Model 1 Stepping 2] | 2 | 244.60861111111112 | 674.4487693611111 |
| 4 | Intel(R) Core(TM) i5 CPU 750 @ 2.67GHz [Family 6 Model 30 Stepping 5] | 1 | 412.47811441666664 | 412.47811441666664 |
| 4 | Intel(R) Pentium(R) CPU G4560 @ 3.50GHz [Family 6 Model 158 Stepping 9] | 1 | 436.3376085 | 436.3376085 |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 3 | Intel(R) Core(TM) i5-6500 CPU @ 3.20GHz [Family 6 Model 94 Stepping 3] | 1 | 28.98138888888889 | 86.94416666666666 |
| 3 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 3 | 34.13527777777778 | 70.16058333333334 |
| 3 | Intel(R) Core(TM) i5-6600K CPU @ 3.50GHz [Family 6 Model 94 Stepping 3] | 1 | 36.751666666666665 | 101.6596441111111 |
| 3 | Intel(R) Core(TM) i5-4670K CPU @ 3.40GHz [Family 6 Model 60 Stepping 3] | 15 | 38.03111111111111 | 111.63954597222222 |
| 3 | Intel(R) Core(TM) i7-7700 CPU @ 3.60GHz [Family 6 Model 158 Stepping 9] | 13 | 41.51166666666666 | 121.97755555555555 |
| 3 | Intel(R) Core(TM) i9-7960X CPU @ 2.80GHz [Family 6 Model 85 Stepping 4] | 6 | 42.420833333333334 | 121.59132811111112 |
| 3 | Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 42.59638888888889 | 98.74000433333333 |
| 3 | Intel(R) Xeon(R) CPU E5-2687W v3 @ 3.10GHz [Family 6 Model 63 Stepping 2] | 1 | 44.57388888888889 | 112.3628611111111 |
| 3 | Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz [Family 6 Model 58 Stepping 9] | 7 | 48.596944444444446 | 140.56565005555555 |
| 3 | AMD Ryzen Threadripper 1950X 16-Core Processor [Family 23 Model 1 Stepping 1] | 19 | 54.95666666666666 | 161.9469661388889 |
| 3 | AMD Ryzen 5 1600X Six-Core Processor [Family 23 Model 1 Stepping 1] | 20 | 66.01944444444445 | 165.37591666666668 |
| 3 | Intel(R) Core(TM) i7-9700K CPU @ 3.60GHz [Family 6 Model 158 Stepping 12] | 1 | 67.24638888888889 | 107.05632377777778 |
| 3 | Intel(R) Core(TM) i9-7940X CPU @ 3.10GHz [Family 6 Model 85 Stepping 4] | 1 | 108.93088974999999 | 108.93088974999999 |
| 3 | Intel(R) Core(TM) i7 CPU 940 @ 2.93GHz [Family 6 Model 26 Stepping 4] | 1 | 123.32611111111112 | 353.81 |
| 3 | Intel(R) Xeon(R) CPU E5-2620 0 @ 2.00GHz [Family 6 Model 45 Stepping 7] | 9 | 144.60916666666665 | 203.2086966111111 |
| 3 | Intel(R) Xeon(R) CPU E3-1226 v3 @ 3.30GHz [Family 6 Model 60 Stepping 3] | 1 | 147.68637202777776 | 147.68637202777776 |
| 3 | Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz [Family 6 Model 158 Stepping 10] | 1 | 158.12305555555557 | 168.12388022222223 |
+---------+--------------------------------------------------------------------------------+-------+--------------------+--------------------+
+---------+------------------------------------------------------------------------------+-------+--------------------+--------------------+
| threads | CPU | count | elapsed | cpu time |
+---------+------------------------------------------------------------------------------+-------+--------------------+--------------------+
| 2 | Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9] | 4 | 35.270833333333336 | 67.75619444444445 |
| 2 | Intel Core Processor (Broadwell) [Family 6 Model 61 Stepping 2] | 19 | 49.021388888888886 | 96.74547222222222 |
| 2 | Intel Core Processor (Haswell, no TSX) [Family 6 Model 60 Stepping 1] | 28 | 49.1675 | 97.40852777777778 |
| 2 | Intel(R) Core(TM) i5-6600 CPU @ 3.30GHz [Family 6 Model 94 Stepping 3] | 10 | 49.431666666666665 | 95.87002169444445 |
| 2 | Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz [Family 6 Model 94 Stepping 3] | 19 | 56.577777777777776 | 108.64893227777779 |
| 2 | Intel(R) Core(TM) i9-7960X CPU @ 2.80GHz [Family 6 Model 85 Stepping 4] | 3 | 61.6575 | 131.58113280555554 |
| 2 | Intel(R) Core(TM) i7-2700K CPU @ 3.50GHz [Family 6 Model 42 Stepping 7] | 1 | 72.40666666666667 | 142.6563888888889 |
| 2 | Intel(R) Xeon(R) CPU E5-1660 v2 @ 3.70GHz [Family 6 Model 62 Stepping 4] | 1 | 77.45666666666666 | 152.97315058333334 |
| 2 | Intel(R) Core(TM) i5-3330 CPU @ 3.00GHz [Family 6 Model 58 Stepping 9] | 6 | 78.25083333333333 | 151.70339777777778 |
| 2 | Intel(R) Core(TM) i7-4510U CPU @ 2.00GHz [Family 6 Model 69 Stepping 1] | 4 | 86.33333333333333 | 173.84366666666665 |
| 2 | Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz [Family 6 Model 58 Stepping 9] | 5 | 92.14 | 153.24448833333335 |
| 2 | Intel(R) Core(TM) i5-3320M CPU @ 2.60GHz [Family 6 Model 58 Stepping 9] | 4 | 131.59416666666667 | 188.2761378888889 |
| 2 | Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz [Family 6 Model 42 Stepping 7] | 1 | 165.96548586111112 | 165.96548586111112 |
| 2 | Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz [Family 6 Model 58 Stepping 9] | 1 | 169.67777777777778 | 269.8104036388889 |
| 2 | Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz [Family 6 Model 94 Stepping 3] | 1 | 180.11103733333331 | 180.11103733333331 |
| 2 | AMD A6-6400B APU with Radeon(tm) HD Graphics [Family 21 Model 19 Stepping 1] | 1 | 542.6242144166666 | 542.6242144166666 |
+---------+------------------------------------------------------------------------------+-------+--------------------+--------------------+
Note that some time has passed between the previous tables and these, and it's possible some tasks have been purged from the database during that time. It's not a certainty that changes are due to excluding tasks below 28M.
____________
My lucky number is 75898524288+1 |
|
|
dukebgVolunteer tester
Send message
Joined: 21 Nov 17
Posts: 242
ID: 950482
Credit: 23,670,125
RAC: 0
|
And, of course, depending on which K you have, there are two different FFT sizes. That changes the run time significantly. When you did your benchmarks, were you using the 2560K FFT or the 2880K FFT? I benchmarked my systems with both sizes.
Hmm, only 2560K. I'll consider rerunning everything with a 2880K FFT number, but I feel like the relative values of different combinations I tried (1x8, 1x4, 2x4, 8x1 etc.) would be the same, just scaled up. So my decision on how to run the challenge would stay the same.
Hah, I see the the i7-7700K 18.1h result that I was "worried" about disappeared in the new data. Might've been a purged old WU, but I'll hold the view that it was a smaller number (actually, an old WU would most likely be a smaller number either way). The other 19.8h one is more within my 20-21h expectations. Especially with 28.0M-28.5M difference you noted. So I'm relieved.
Good luck in the challenge, everyone! About 10 hours remaining. Let's uncover some primes in those no-res double checks. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
Hmm, only 2560K. I'll consider rerunning everything with a 2880K FFT number, but I feel like the relative values of different combinations I tried (1x8, 1x4, 2x4, 8x1 etc.) would be the same, just scaled up. So my decision on how to run the challenge would stay the same.
That would be my conclusion as well.
____________
My lucky number is 75898524288+1 |
|
|
RogerVolunteer developer
Volunteer tester
Send message
Joined: 27 Nov 11
Posts: 1138
ID: 120786
Credit: 268,668,824
RAC: 0
|
|
I've run a trial task with multithreaded on and aborted it. Ready to switch over to SoB when I come home from work tonight. Aiming to get 2, maybe 3 WU's done over the Challenge duration. Every little bit helps. |
|
|
|
|
|
ok thank you. sadly atm the main pc got stuck with 2 month cpu work units |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
The challenge has started. Good luck everyone!
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
I'm trying 1x4t SoB task by i7-4770k.
Temperatures are awful.
70-72°C at 3.00GHz.
I believe my 4-year-old AiO (Corsair H80i) is gone.
Did anyone else have a similar experience?
Maybe some water evaporated. |
|
|
|
|
|
Your temps are fine, I wouldn't have recommended an H80 but you're under-clocking the CPU so it works out.
I'm seeing the same temps on a 4770K @ 4.00 GHz running PSP using a Corsair H110. I would normally be running a higher OC and temps would be ~80C 24/7 only running @4GHz this winter because I got tired of fan noise.
|
|
|
|
|
|
Well, I have chosen the right frequency to get good temps.
This cooler was able to afford 3.5GHz some time ago, turbo boost ON (3.7GHz) too.
I recently changed thermal paste without improvements.
I think the cooler itself is getting worse.
On February 2018 I did a test:
PPS-MEGA @ 3.5GHz Turbo Boost OFF ===> 53-60°C
PPS-MEGA @ 3.5GHz Turbo Boost ON ===> 63-72°C
Now @3.5GHz reaches 84°C, then there is throttling.
Anyway I let the i7-4770k @ 3.0GHz (ram 1866MHz) run a bit. It should complete this task in ~35h at this "speed". FFT is 2560k.
However, I could have to decrease frequency when I'm close to this host because of noise. So final time will be higher. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
The first task has been returned in just over 7 hours!
____________
My lucky number is 75898524288+1 |
|
|
|
|
The first task has been returned in just over 7 hours!
Impressive.
The last SOB task I crunched took 30 hours.
24737*2^28655431+1 is not prime
FFT 2560 k
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment. |
|
|
JaySend message
Joined: 27 Feb 10
Posts: 131
ID: 56067
Credit: 63,534,968
RAC: 10,618
|
The first task has been returned in just over 7 hours!
Dang... What was the hardware? I hope he or she keeps going at that pace.
|
|
|
|
|
The first task has been returned in just over 7 hours!
Dang... What was the hardware? I hope he or she keeps going at that pace.
We do too ! |
|
|
|
|
The first task has been returned in just over 7 hours!
Dang... What was the hardware? I hope he or she keeps going at that pace.
GenuineIntel
Intel(R) Xeon(R) Platinum 8124M CPU @ 3.00GHz [Family 6 Model 85 Stepping 4] |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
Early statistics:
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-07 16:24:22 UTC)
5101 tasks have been sent out. [CPU/GPU/anonymous_platform: 5101 (100%) / 0 (0%) / 0 (0%)]
Of those tasks that have been sent out:
762 (15%) were aborted. [762 (15%) / 0 (0%) / 0 (0%)]
22 (0%) came back with some kind of an error. [22 (0%) / 0 (0%) / 0 (0%)]
7 (0%) have returned a successful result. [7 (0%) / 0 (0%) / 0 (0%)]
4310 (84%) are still in progress. [4310 (84%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
7 (100%) are pending validation. [7 (100%) / 0 (0%) / 0 (0%)]
0 (0%) have been successfully validated. [0 (0%) / 0 (0%) / 0 (0%)]
0 (0%) were invalid. [0 (0%) / 0 (0%) / 0 (0%)]
0 (0%) are inconclusive. [0 (0%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=28911812. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 0.69% as much as it had prior to the challenge!
I'm making a change in the statistics. Because BOINC tends to send extra tasks when you're multi-tasking, and many people abort those extra tasks, I decided to break aborted tasks in a separate line instead of including them in errors. As you can see above, instead of having 784 errors, we have 762 aborted tasks and only 22 errors. That paints a very different picture!
For those that are wondering, this is from last month's GFN21 challenge. GPU challenges always produce a lot of errors because driver issues stop the GPU tasks from even starting:
4607 (10%) were aborted. [3255 (7%) / 1352 (3%) / 0 (0%)]
18068 (41%) came back with some kind of an error. [840 (2%) / 17228 (39%) / 0 (0%)]
Notice how most of the aborted tasks are on CPU tasks (which could be multi-threaded), while most of the errors were on GPU tasks.
____________
My lucky number is 75898524288+1 |
|
|
|
|
I'm trying 1x4t SoB task by i7-4770k.
Temperatures are awful.
70-72°C at 3.00GHz.
I believe my 4-year-old AiO (Corsair H80i) is gone.
Did anyone else have a similar experience?
Maybe some water evaporated.
I have with my own H80 (v1) which was my first foray into the "wonderful" new world of CLCs. After many years of running my first 3930k OC at very reasonable temps I started noticing my task times being higher than that of my non-overclocked 3930k and looked to see it running well under speed at 90+C. Judging by the corrosion, I think it started a small leak around the copper plate and now hardly has any liquid in it (it's in my recycle pile).
My advice to you is replace that cooler ASAP before the worst happens. Since any computer in my house will run PG until it is no longer working or viable, I've abandoned CLCs and only use air cooling on my systems, mainly the biggest Noctua cooler that will fit into the case--I have 10 year old U12's still running cool and quiet 24/7. I'm looking at custom WC for my next-gen Core-X build, but that will sit on my desk where I can monitor it.
____________
Eating more cheese on Thursdays. |
|
|
|
|
The first task has been returned in just over 7 hours!
Dang... What was the hardware? I hope he or she keeps going at that pace.
We do too !
LOL !!! |
|
|
|
|
Because BOINC tends to send extra tasks when you're multi-tasking, ....
No. This depends on your configuration:
e.g. 6 cores with 12 threads and you want to run 2 tasks in parallel:
1. Set your buffer to zero.
2. in cc_config.xml: <ncpus>2</ncpus>
3. in app_config.xml:
<app>
<name>llrSOB</name>
<max_concurrent>2</max_concurrent>
</app>
<app_version>
<app_name>llrSOB</app_name>
<cmdline>-t 6</cmdline>
<avg_ncpus>1</avg_ncpus>
<max_ncpus>1</max_ncpus>
</app_version>
BOINC loads only 2 tasks and runs only 2 tasks...
____________
DeleteNull |
|
|
|
|
Because BOINC tends to send extra tasks when you're multi-tasking, ....
No. This depends on your configuration:
e.g. 6 cores with 12 threads and you want to run 2 tasks in parallel:
1. Set your buffer to zero.
2. in cc_config.xml: <ncpus>2</ncpus>
3. in app_config.xml:
<app>
<name>llrSOB</name>
<max_concurrent>2</max_concurrent>
</app>
<app_version>
<app_name>llrSOB</app_name>
<cmdline>-t 6</cmdline>
<avg_ncpus>1</avg_ncpus>
<max_ncpus>1</max_ncpus>
</app_version>
BOINC loads only 2 tasks and runs only 2 tasks...
There are times when you WILL get more than one task as defined by your config file downloaded. This happens almost every time I start a new project on which I've added it to my app_config file. Not sure why but I'm guessing the first time through the pc is not smart enough to have it only download the "requested" number of tasks. This has happened every time for me. Once I've actually run and sent back a task, the remainder of the time I only get what I think I should.
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
Because BOINC tends to send extra tasks when you're multi-tasking, ....
No. This depends on your configuration:
e.g. 6 cores with 12 threads and you want to run 2 tasks in parallel:
1. Set your buffer to zero.
2. in cc_config.xml: <ncpus>2</ncpus>
3. in app_config.xml:
<app>
<name>llrSOB</name>
<max_concurrent>2</max_concurrent>
</app>
<app_version>
<app_name>llrSOB</app_name>
<cmdline>-t 6</cmdline>
<avg_ncpus>1</avg_ncpus>
<max_ncpus>1</max_ncpus>
</app_version>
BOINC loads only 2 tasks and runs only 2 tasks...
That's certainly a viable option.
However, many people just abort the tasks, so... there's lots of aborted tasks, and it makes sense to report actual errors and aborts separately.
____________
My lucky number is 75898524288+1 |
|
|
|
|
No. This depends on your configuration:
e.g. 6 cores with 12 threads and you want to run 2 tasks in parallel:
1. Set your buffer to zero.
2. in cc_config.xml: <ncpus>2</ncpus>
3. in app_config.xml:
<app>
<name>llrSOB</name>
<max_concurrent>2</max_concurrent>
</app>
<app_version>
<app_name>llrSOB</app_name>
<cmdline>-t 6</cmdline>
<avg_ncpus>1</avg_ncpus>
<max_ncpus>1</max_ncpus>
</app_version>
BOINC loads only 2 tasks and runs only 2 tasks...
Don't you also need to have BOINC computing preferences set to "Use at most 16.67% of the CPUs"? |
|
|
TylerProject administrator
Volunteer tester
Send message
Joined: 4 Dec 12
Posts: 1078
ID: 183129
Credit: 1,376,122,338
RAC: 8,547
         
|
No. This depends on your configuration:
e.g. 6 cores with 12 threads and you want to run 2 tasks in parallel:
1. Set your buffer to zero.
2. in cc_config.xml: <ncpus>2</ncpus>
3. in app_config.xml:
<app>
<name>llrSOB</name>
<max_concurrent>2</max_concurrent>
</app>
<app_version>
<app_name>llrSOB</app_name>
<cmdline>-t 6</cmdline>
<avg_ncpus>1</avg_ncpus>
<max_ncpus>1</max_ncpus>
</app_version>
BOINC loads only 2 tasks and runs only 2 tasks...
Don't you also need to have BOINC computing preferences set to "Use at most 16.67% of the CPUs"?
No, since he set ncpus to 2. This tells BOINC to only use 2 CPU cores in total. |
|
|
Dave
Send message
Joined: 13 Feb 12
Posts: 3170
ID: 130544
Credit: 2,231,845,189
RAC: 766,099
          
|
|
I set BOINC prefs to 1% CPU & that works the same as 25%. |
|
|
|
|
No, since he set ncpus to 2. This tells BOINC to only use 2 CPU cores in total.
That's it. 2 (virtual) cores with 6 threads each.
____________
DeleteNull |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
No, since he set ncpus to 2. This tells BOINC to only use 2 CPU cores in total.
That's it. 2 (virtual) cores with 6 threads each.
When you do that, does the CPU time reported by BOINC correctly report the CPU time as if it's running on 6 cores? (I would guess that it does, although based on what I've seen in the database I'm not so sure.)
Also, if you use WuProp, does it record the time correctly as if you're running on 6 cores? (I would guess that it does not.)
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
The CPU time isn't correct:
If there are 2 WU's running the CPU time is just twice the runtime:
http://www.primegrid.com/results.php?hostid=916741&offset=0&show_names=0&state=0&appid=13
(PC with 10 cores/20 threads 2 WU with 10 threads each)
If there is only one WU running CPU time and runtime are the same:
http://www.primegrid.com/results.php?hostid=912667&offset=0&show_names=0&state=0&appid=13
(PC with 6 CPU/threads, one WU with 6 threads)
With the old server (Rackspace) the CPU time was calculated correctly. I don't use WU-Prop but "top" shows the CPU time correct.
____________
DeleteNull |
|
|
|
|
I have with my own H80 (v1) which was my first foray into the "wonderful" new world of CLCs. After many years of running my first 3930k OC at very reasonable temps I started noticing my task times being higher than that of my non-overclocked 3930k and looked to see it running well under speed at 90+C. Judging by the corrosion, I think it started a small leak around the copper plate and now hardly has any liquid in it (it's in my recycle pile).
Copper plate is a bit oxidized where it doesn't overlap the CPU IHS. Copper is shiny where there was thermal paste. Cracks in rubber connections appeared. Tubes are made of some plastic material instead, so they are still perfect. Luckily, I didn't find any important/visible leak.
My advice to you is replace that cooler ASAP before the worst happens. Since any computer in my house will run PG until it is no longer working or viable, I've abandoned CLCs and only use air cooling on my systems, mainly the biggest Noctua cooler that will fit into the case--I have 10 year old U12's still running cool and quiet 24/7. I'm looking at custom WC for my next-gen Core-X build, but that will sit on my desk where I can monitor it.
LOL! I have exactly a Noctua cooler. It's the Noctua NH-L12 which I substituted for Corsair H80i and finally I didn't sell. I should get a larger case to move to bigger coolers.
Thank everyone for feedback. ;) |
|
|
|
|
No, since he set ncpus to 2. This tells BOINC to only use 2 CPU cores in total.
I think this is where the disconnect is. All the "how to multi-thread" that I've seen (until now) has only dealt with app_config.
Using app_config only, BOINC knows how many CPUs you have and what % you want to use (to determine how many tasks to get). It doesn't look into app_config until it's ready to run a task that's already been downloaded.
So DeleteNull's "2. in cc_config.xml: <ncpus>2</ncpus>" is either a global PG setting or global BOINC setting and that would have to be edited every time you wanted to run a different amount of tasks. I do this via the "use x% CPUs" and vary it accordingly. For 2 tasks on a 12 thread CPU it's 2/12 or 16.67%
Not sure how this "fixes" itself for those leaving it at 50% or 100% CPUs. I suspect that after finishing a task it initially only asks for 1 task as a replacement (as opposed to the start-up having no tasks at all), new task starts, uses all the CPUs and it doesn't ask for any more. |
|
|
|
|
Also, if you use WuProp, does it record the time correctly as if you're running on 6 cores? (I would guess that it does not.)
Unfortunately WuProp does not recognize multi-thread, it's counted as straight time, but since it does take into account the number of instances; running 3 concurrent PPS Sieves on 1 GPU for an hour shows as 3 hours run. |
|
|
|
|
The first task has been returned in just over 7 hours!
Dang... What was the hardware? I hope he or she keeps going at that pace.
I saw that... damn. fast! |
|
|
|
|
|
Re avoiding download of too many tasks:
For some time now, I too have been using <avg_ncpus>1</avg_ncpus> for all MT applications at PrimeGrid and for some MT applications at other projects, in order to have the client download only the desired amount of tasks.
This method is only viable though if you don't run tasks with different thread counts at the same time (in the same client instance). But I never do this anyway, as I am not satisfied with how the client schedules tasks then.
I am not specifying <max_concurrent> nor <project_max_concurrent>. Neither of these effects how many tasks the client requests when fetching new work. I control the total number of concurrent tasks either with cc_config.options.ncpus or with global_preferences.max_ncpus_pct (usually set via boincmgr GUI), and I tend to prefer the latter.
Re CPU time reported at results.php:
I have a single computer running in this competition, which completed 6 tasks yesterday. It fetched them in requests of 2 + 1 + 1 + 1 + 1 tasks, and ran them as 2...4 tasks concurrently. (I reconfigured the host during the day from best run time to best throughput.) As it happens, results.php is showing "run time" as well as "CPU time" correctly for all six tasks.
I haven't seen "CPU time" misreported with any of my hosts yet, neither now nor whenever I checked last year, even though I set <avg_ncpus>1</avg_ncpus> and am running more than one task at once in the same client. |
|
|
dthononVolunteer tester
Send message
Joined: 6 Dec 17
Posts: 435
ID: 957147
Credit: 1,737,342,032
RAC: 25,049
                    
|
|
[quote]This method is only viable though if you don't run tasks with different thread counts at the same time (in the same client instance). But I never do this anyway, as I am not satisfied with how the client schedules tasks then./quote]
I use the same method, and the main issue is when switching from tasks with few threads to tasks with more threads, i.e. from SGS (-t 1) to SOB (-t 4). As soon as the first SGS finished, BOINC dowloads a single SOB and puts the 3 other SGS on hold. I usually suspend manually the SOB task for a few minutes, so as to let the remaining SGS finish. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
After one day (more or less):
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-08 13:43:40 UTC)
7109 tasks have been sent out. [CPU/GPU/anonymous_platform: 7109 (100%) / 0 (0%) / 0 (0%)]
Of those tasks that have been sent out:
1192 (17%) were aborted. [1192 (17%) / 0 (0%) / 0 (0%)]
74 (1%) came back with some kind of an error. [74 (1%) / 0 (0%) / 0 (0%)]
218 (3%) have returned a successful result. [218 (3%) / 0 (0%) / 0 (0%)]
5625 (79%) are still in progress. [5625 (79%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
186 (85%) are pending validation. [186 (85%) / 0 (0%) / 0 (0%)]
31 (14%) have been successfully validated. [31 (14%) / 0 (0%) / 0 (0%)]
0 (0%) were invalid. [0 (0%) / 0 (0%) / 0 (0%)]
1 (0%) are inconclusive. [1 (0%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=28989391. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 0.96% as much as it had prior to the challenge!
____________
My lucky number is 75898524288+1 |
|
|
TylerProject administrator
Volunteer tester
Send message
Joined: 4 Dec 12
Posts: 1078
ID: 183129
Credit: 1,376,122,338
RAC: 8,547
|
|
Made some pages that might be interesting to look at --
Challenge summary w/ top 10 users & teams: https://statsgr.id/challenge.php
Has links to other info, full user/team rank lists. Anywhere you see a user/teamname they're hyperlinked to see a graph of that user/team's score over time during the challenge.
One of the more interesting pages has the top 10 teams together or top 10 users together on a graph. Links to those also on that main page above, or here:
Teams: https://statsgr.id/top_team_graph.php
Users: https://statsgr.id/top_user_graph.php
Everything there should stay updated every 15 minutes, as the primegrid challenge pages are.
Suggestions/criticisms are welcome
~Golfer
____________
275*2^3585539+1 is prime!!! (1079358 digits)
 Proud member of Aggie the Pew
 |
|
|
|
|
Made some pages that might be interesting to look at --
Very nice !
Thanks.
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
The double check is now 57% complete, and we've started sending out tasks with n>29M.
____________
My lucky number is 75898524288+1 |
|
|
|
|
The first task has been returned in just over 7 hours!
Dang... What was the hardware? I hope he or she keeps going at that pace.
Under our preferences for this project it says the following
NOTE: We are double checking old SoB work and tasks may be much shorter than typical SoB tasks.
So a task are completing in 7 hours is possible by the way I interpret above message |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 937
ID: 3110
Credit: 258,164,203
RAC: 78,071
|
The double check is now 57% complete, and we've started sending out tasks with n>29M.
By number of tasks, or by estimated time to completion? I assume the former; can you estimate the CPU time percentage complete? |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
The first task has been returned in just over 7 hours!
Dang... What was the hardware? I hope he or she keeps going at that pace.
Under our preferences for this project it says the following
NOTE: We are double checking old SoB work and tasks may be much shorter than typical SoB tasks.
So a task are completing in 7 hours is possible by the way I interpret above message
Actually, that message is pretty old. I'll remove it. That tasks are back to being very long. Not quite as long as the leading edge tasks would be, but close.
EDIT: That message is now:
NOTE: Very long tasks.
____________
My lucky number is 75898524288+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
The double check is now 57% complete, and we've started sending out tasks with n>29M.
By number of tasks, or by estimated time to completion? I assume the former; can you estimate the CPU time percentage complete?
The latter.
That percentage is based on estimated total computing time.
____________
My lucky number is 75898524288+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
After 2 days:
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-09 06:12:08 UTC)
7984 tasks have been sent out. [CPU/GPU/anonymous_platform: 7984 (100%) / 0 (0%) / 0 (0%)]
Of those tasks that have been sent out:
1416 (18%) were aborted. [1416 (18%) / 0 (0%) / 0 (0%)]
129 (2%) came back with some kind of an error. [129 (2%) / 0 (0%) / 0 (0%)]
546 (7%) have returned a successful result. [546 (7%) / 0 (0%) / 0 (0%)]
5893 (74%) are still in progress. [5893 (74%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
447 (82%) are pending validation. [447 (82%) / 0 (0%) / 0 (0%)]
97 (18%) have been successfully validated. [97 (18%) / 0 (0%) / 0 (0%)]
1 (0%) were invalid. [1 (0%) / 0 (0%) / 0 (0%)]
1 (0%) are inconclusive. [1 (0%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=29019934. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 1.07% as much as it had prior to the challenge!
____________
My lucky number is 75898524288+1 |
|
|
|
|
I'm trying 1x4t SoB task by i7-4770k.
Temperatures are awful.
70-72°C at 3.00GHz.
I believe my 4-year-old AiO (Corsair H80i) is gone.
Did anyone else have a similar experience?
Maybe some water evaporated.
FWIW, I am running 6 threads on an 8700K @ 5 GHz under an H80iV2 and my temps are in the low to mid 70s.
____________
 |
|
|
|
|
I'm trying 1x4t SoB task by i7-4770k.
Temperatures are awful.
70-72°C at 3.00GHz.
I believe my 4-year-old AiO (Corsair H80i) is gone.
Did anyone else have a similar experience?
Maybe some water evaporated.
FWIW, I am running 6 threads on an 8700K @ 5 GHz under an H80iV2 and my temps are in the low to mid 70s.
I'm running 7 cores (hyperthreading turned off) on my 9900K @ 4.6Ghz under an H115i. My hottest core is at 95C, my coldest core in use is 90C. The unused core is at 80C, probably just from close proximity of the other cores in use. Current TDP is bouncing between 160watts to 165watts. So, yea... 70 is not hot. Also Intel's Tuning utility shows that there is no throttling on this system. And if anyone is interested my current SoB task is 24.9% complete at an elapsed time of 3 hour, 20 min. This task seems to be running much faster than my previous tasks, but this was after I tweaked some things. |
|
|
|
|
|
95C is kinda pushing it...
Those kind of temperatures make me think a delidded 8700K / 8086K is a better proposition in the long run.
____________
|
|
|
|
|
95C is kinda pushing it...
Those kind of temperatures make me think a delidded 8700K / 8086K is a better proposition in the long run.
Oh I totally agree it's pushing it. Stressing it for this task to see how fast I can complete it. Anyway, my old system was a delided 7700K. I wouldn't trade my 9900 for anything. It's a monster. |
|
|
compositeVolunteer tester
Send message
Joined: 16 Feb 10
Posts: 1140
ID: 55391
Credit: 1,014,113,687
RAC: 1,688,928
      
|
95C is kinda pushing it...
My i7-920 peaked at 101C with stock cooler before using a better one.
That system had weekly resets and freezes for a while.
The CPU figuratively gave up the smoke after 6 years of abuse on Primegrid.
It was time to get something better anyway.
|
|
|
|
|
95C is kinda pushing it...
My i7-920 peaked at 101C with stock cooler before using a better one.
That system had weekly resets and freezes for a while.
The CPU figuratively gave up the smoke after 6 years of abuse on Primegrid.
It was time to get something better anyway.
I'm seeing images of Beavis saying "fire! FIRE!"
____________
|
|
|
|
|
95C is kinda pushing it...
Those kind of temperatures make me think a delidded 8700K / 8086K is a better proposition in the long run.
Oh I totally agree it's pushing it. Stressing it for this task to see how fast I can complete it. Anyway, my old system was a delided 7700K. I wouldn't trade my 9900 for anything. It's a monster.
not even for a board of 8x Xeon Platinum 8180M?
https://en.wikichip.org/wiki/intel/xeon_platinum/8180m
____________
|
|
|
|
|
|
I'm running 7 SoB tasks (2 threaded) on a ryzen 1700X at 71°C. (I should have used a better cooler that da bequit pure rock slim with a noctua NF-A9, but in this ITX-Case wasn't enough space.) Its looking okay, the configured maximum for the CPU is 75°C, but it could be cooler/the fan could have less to do...maybe I buy a nother case and cpu cooler...
____________
|
|
|
|
|
I'm running 7 SoB tasks (2 threaded) on a ryzen 1700X at 71°C. (I should have used a better cooler that da bequit pure rock slim with a noctua NF-A9, but in this ITX-Case wasn't enough space.) Its looking okay, the configured maximum for the CPU is 75°C, but it could be cooler/the fan could have less to do...maybe I buy a nother case and cpu cooler...
I've never used Be Quiet! but I understand they are quality parts. Noctua is about as good as it gets, fan wise; my rig is decked out with the new A12s (I think thats the part, I'm too lazy to look it up). If you can find a copper heat sink suitable to your case / TDP needs, you will get significantly better heat dissipation than aluminium.
____________
|
|
|
|
|
I'm running 7 SoB tasks (2 threaded) on a ryzen 1700X at 71°C. (I should have used a better cooler that da bequit pure rock slim with a noctua NF-A9, but in this ITX-Case wasn't enough space.) Its looking okay, the configured maximum for the CPU is 75°C, but it could be cooler/the fan could have less to do...maybe I buy a nother case and cpu cooler...
I've never used Be Quiet! but I understand they are quality parts. Noctua is about as good as it gets, fan wise; my rig is decked out with the new A12s (I think thats the part, I'm too lazy to look it up). If you can find a copper heat sink suitable to your case / TDP needs, you will get significantly better heat dissipation than aluminium.
nah...cooper is much more expensive and dosn't bring that much. at least I can't find any good full cooper cooler for AM4...
I would prefer noctua NH-U14 (the best perfoming air cpu cooler for a reasonable price) or the bequiet dark rock pro 4 (10% cheaper, but only a bit behind the noctua, but bigger and heavier)
both can even cool a Threadripper 2990WX a stock clock(!). thats really crazy and can compete with some AiO water coolings. If you want something really better you would need a custom water cooling...
but for the dark rock und ne U14 I need a case that gives me more then 160mm above the cpu...
btw I have an unused I7-6850K on a MB. on top of the cpu is a dark rock pro 3 mounted. this little monster (it is as good as the dark rock pro 4) got the Boradwell-EP-cpu down to ~60°C @ stock clock. to bad that I have nor case and other spare parts to build something... (I could take the parts from the ryzen system and buy a case...but I like the ryzen^^)
____________
|
|
|
compositeVolunteer tester
Send message
Joined: 16 Feb 10
Posts: 1140
ID: 55391
Credit: 1,014,113,687
RAC: 1,688,928
|
|
Current aggregate score of participants in this Challenge without a team (9 individuals) is 9th place, at 2 163 651.99 representing 3.69% of the total score.
Go Team NULL!
EDIT: or is it, Go Team ""!
Hmmm, an empty string is not the same thing as NULL.
To be consistent with the Top Teams list, it would be 8.5th place. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
3 days in:
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-10 06:25:23 UTC)
9309 tasks have been sent out. [CPU/GPU/anonymous_platform: 9309 (100%) / 0 (0%) / 0 (0%)]
Of those tasks that have been sent out:
1658 (18%) were aborted. [1658 (18%) / 0 (0%) / 0 (0%)]
189 (2%) came back with some kind of an error. [189 (2%) / 0 (0%) / 0 (0%)]
1122 (12%) have returned a successful result. [1122 (12%) / 0 (0%) / 0 (0%)]
6340 (68%) are still in progress. [6340 (68%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
844 (75%) are pending validation. [844 (75%) / 0 (0%) / 0 (0%)]
273 (24%) have been successfully validated. [273 (24%) / 0 (0%) / 0 (0%)]
1 (0%) were invalid. [1 (0%) / 0 (0%) / 0 (0%)]
4 (0%) are inconclusive. [4 (0%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=29065090. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 1.23% as much as it had prior to the challenge!
____________
My lucky number is 75898524288+1 |
|
|
|
|
Current aggregate score of participants in this Challenge without a team (9 individuals) is 9th place, at 2 163 651.99 representing 3.69% of the total score.
Go Team NULL!
EDIT: or is it, Go Team ""!
Hmmm, an empty string is not the same thing as NULL.
To be consistent with the Top Teams list, it would be 8.5th place.
Haha! "Team Remainder," maybe? "Team != exists"?
____________
Proud member of Team Aggie the Pew
"Wir müssen wissen. Wir werden wissen."
"We must know, we shall know."
- David Hilbert, 1930 |
|
|
Dave
Send message
Joined: 13 Feb 12
Posts: 3170
ID: 130544
Credit: 2,231,845,189
RAC: 766,099
|
|
Everyone who does not belong to a team should be made a member of "Team Null" by default. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
After 4-ish days:
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-11 05:29:32 UTC)
10280 tasks have been sent out. [CPU/GPU/anonymous_platform: 10272 (100%) / 0 (0%) / 8 (0%)]
Of those tasks that have been sent out:
1857 (18%) were aborted. [1857 (18%) / 0 (0%) / 0 (0%)]
222 (2%) came back with some kind of an error. [214 (2%) / 0 (0%) / 8 (0%)]
1713 (17%) have returned a successful result. [1713 (17%) / 0 (0%) / 0 (0%)]
6488 (63%) are still in progress. [6488 (63%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
1190 (69%) are pending validation. [1190 (69%) / 0 (0%) / 0 (0%)]
504 (29%) have been successfully validated. [504 (29%) / 0 (0%) / 0 (0%)]
2 (0%) were invalid. [2 (0%) / 0 (0%) / 0 (0%)]
17 (1%) are inconclusive. [17 (1%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=29094884. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 1.33% as much as it had prior to the challenge!
____________
My lucky number is 75898524288+1 |
|
|
robishVolunteer moderator
Volunteer tester
Send message
Joined: 7 Jan 12
Posts: 2195
ID: 126266
Credit: 7,299,570,840
RAC: 3,191,535
       
|
|
Wow, teams at 1st and 2nd sooooo close it's hard to call. Only 1/3rd into it and It's so evenly matched it's going to be, probably, the most interesting challenge in years!!!
____________
My lucky numbers 10590941048576+1 and 224584605939537911+81292139*23#*n for n=0..26 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
The double check is now 58% complete.
____________
My lucky number is 75898524288+1 |
|
|
|
|
After 4-ish days:
the leading edge has advanced 1.33% as much as it had prior to the challenge!
I am not quite sure what you are meaning by ^^ are you are saying the project is progressing at the same speed as before the challenge started?
Good luck to everybody in the challenge |
|
|
|
|
After 4-ish days:
the leading edge has advanced 1.33% as much as it had prior to the challenge!
I am not quite sure what you are meaning by ^^ are you are saying the project is progressing at the same speed as before the challenge started?
Nah - it's much faster.
____________
Greetings, Jens
147433824^131072+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
After 4-ish days:
the leading edge has advanced 1.33% as much as it had prior to the challenge!
I am not quite sure what you are meaning by ^^ are you are saying the project is progressing at the same speed as before the challenge started?
Good luck to everybody in the challenge
It's saying that, during the 4 days of the challenge, the project advanced the leading edge by 1.33% as much as it had been advanced in all the time prior to the challenge.
____________
My lucky number is 75898524288+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
After 5 days:
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-12 10:07:31 UTC)
11455 tasks have been sent out. [CPU/GPU/anonymous_platform: 11447 (100%) / 0 (0%) / 8 (0%)]
Of those tasks that have been sent out:
2161 (19%) were aborted. [2161 (19%) / 0 (0%) / 0 (0%)]
264 (2%) came back with some kind of an error. [256 (2%) / 0 (0%) / 8 (0%)]
2484 (22%) have returned a successful result. [2484 (22%) / 0 (0%) / 0 (0%)]
6546 (57%) are still in progress. [6546 (57%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
1577 (63%) are pending validation. [1577 (63%) / 0 (0%) / 0 (0%)]
877 (35%) have been successfully validated. [877 (35%) / 0 (0%) / 0 (0%)]
3 (0%) were invalid. [3 (0%) / 0 (0%) / 0 (0%)]
27 (1%) are inconclusive. [27 (1%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=29128366. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 1.45% as much as it had prior to the challenge!
____________
My lucky number is 75898524288+1 |
|
|
|
|
After 4-ish days:
the leading edge has advanced 1.33% as much as it had prior to the challenge!
I am not quite sure what you are meaning by ^^ are you are saying the project is progressing at the same speed as before the challenge started?
Good luck to everybody in the challenge
gemini8 said
Nah - it's much faster.
It's saying that, during the 4 days of the challenge, the project advanced the leading edge by 1.33% as much as it had been advanced in all the time prior to the challenge.
Thank you both for the answers. I still don't fully understand what you mean. I understand it is progressing faster.
Pleased it is processing along faster |
|
|
compositeVolunteer tester
Send message
Joined: 16 Feb 10
Posts: 1140
ID: 55391
Credit: 1,014,113,687
RAC: 1,688,928
|
After 4-ish days:
the leading edge has advanced 1.33% as much as it had prior to the challenge!
I am not quite sure what you are meaning by ^^ are you are saying the project is progressing at the same speed as before the challenge started?
Good luck to everybody in the challenge
It's saying that, during the 4 days of the challenge, the project advanced the leading edge by 1.33% as much as it had been advanced in all the time prior to the challenge.
It means that to accomplish the same feat %-wise next year, it would require much greater processing power than is being applied during this challenge because more work will have been done before the next challenge starts. |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 937
ID: 3110
Credit: 258,164,203
RAC: 78,071
|
After 4-ish days:
the leading edge has advanced 1.33% as much as it had prior to the challenge!
I am not quite sure what you are meaning by ^^ are you are saying the project is progressing at the same speed as before the challenge started?
Good luck to everybody in the challenge
It's saying that, during the 4 days of the challenge, the project advanced the leading edge by 1.33% as much as it had been advanced in all the time prior to the challenge.
Putting a number to "all the time prior to the challenge" might help. I assume in this case it means when the double-check started? When was that? |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
After 4-ish days:
the leading edge has advanced 1.33% as much as it had prior to the challenge!
I am not quite sure what you are meaning by ^^ are you are saying the project is progressing at the same speed as before the challenge started?
Good luck to everybody in the challenge
It's saying that, during the 4 days of the challenge, the project advanced the leading edge by 1.33% as much as it had been advanced in all the time prior to the challenge.
Putting a number to "all the time prior to the challenge" might help. I assume in this case it means when the double-check started? When was that?
There's actually two answers to this. The simpler one is what you said -- since the beginning of the challenge.
But it also could mean "since the beginning of the SoB project", although that's more complicated because that originally went further.
In the first case, It's comparing the last 5 days to the last 34 months. In the second case, it's comparing the 5 days to the years 2002 through approximately 2015.
EDIT: This is a a script I run for every challenge, and it only measure how far the leading edge advances. It's not a measurement of how much work is done. Tasks increase in size significantly as the leading edge advances, which means the amount of computation needed to advance the leading edge now is much greater than the amount of computation needed to advance the leading edge at the beginning of the challenge.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
For everyone thinking about WUprop-hours being miscalculated.
with no cc_config but following app_config it seems the hours get reported correctly:
<app>
<name>llrSOB</name>
<fraction_done_exact/>
<max_concurrent>1</max_concurrent>
</app>
<app_version>
<app_name>llrSOB</app_name>
<cmdline>-t 8</cmdline>
<avg_ncpus>8</avg_ncpus>
<max_ncpus>8</max_ncpus>
</app_version>
That is for a usage of 8 cores, and a single instance of SOB. Runtime ~31 Hours on a ryzen 1800x (reporting ~250 Hours on wuprop). seems the Thing is you have to tell BOINC how much cores it uses for WUPROP to properly read runtimes.
____________
|
|
|
JaySend message
Joined: 27 Feb 10
Posts: 131
ID: 56067
Credit: 63,534,968
RAC: 10,618
|
EDIT: This is a a script I run for every challenge, and it only measure how far the leading edge advances. It's not a measurement of how much work is done. Tasks increase in size significantly as the leading edge advances, which means the amount of computation needed to advance the leading edge now is much greater than the amount of computation needed to advance the leading edge at the beginning of the challenge.
Since, as you said, tasks are much larger than at the beginning of the project, and since computation power is generally greater than in the past, would it make since (and be easy to implement in the script) to say, "Since the challenge started # days ago, we've done x amount of work on the project. In the # days prior to the challenge we did y amount of work on the project."
I feel like comparing the more recent, non-challenge time, to the challenge time, would give me a better feeling for what the challenge is accomplishing vs not having the challenge.
I could be misunderstanding this completely though...
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
6 days done, 9 to go:
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-13 10:34:34 UTC)
12593 tasks have been sent out. [CPU/GPU/anonymous_platform: 12585 (100%) / 0 (0%) / 8 (0%)]
Of those tasks that have been sent out:
2312 (18%) were aborted. [2312 (18%) / 0 (0%) / 0 (0%)]
350 (3%) came back with some kind of an error. [342 (3%) / 0 (0%) / 8 (0%)]
3228 (26%) have returned a successful result. [3228 (26%) / 0 (0%) / 0 (0%)]
6703 (53%) are still in progress. [6703 (53%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
1866 (58%) are pending validation. [1866 (58%) / 0 (0%) / 0 (0%)]
1329 (41%) have been successfully validated. [1329 (41%) / 0 (0%) / 0 (0%)]
8 (0%) were invalid. [8 (0%) / 0 (0%) / 0 (0%)]
25 (1%) are inconclusive. [25 (1%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=29171407. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 1.60% as much as it had prior to the challenge!
____________
My lucky number is 75898524288+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
Since, as you said, tasks are much larger than at the beginning of the project, and since computation power is generally greater than in the past, would it make since (and be easy to implement in the script) to say, "Since the challenge started # days ago, we've done x amount of work on the project. In the # days prior to the challenge we did y amount of work on the project."
Indeed it would, but that's beyond the capability of that script.
____________
My lucky number is 75898524288+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
I usually say some encouraging words about how fantastic participation is during this challenge, but this time it's difficult.
Not because you'all are not doing a fantastic job -- you are -- but because it's particularly difficult to measure it this year for a variety of reasons.
So I found a different way.
During last year's challenge, it was easy to measure the effect of the challenge against the progress of the double check. I record each 1% increment in the amount of the double check that's completed, so we can measure the number of days between increments.
Before last year's SoB challenge: (days between 1% increments)
31
29
28
17
15
19
16
10
11
12
During the challenge, and slightly afterwards:
5
2
2
2
1
1
2
1
2
1
2
The effect of the challenge last year was crystal clear, but that metric won't work as well in 2019.
I can, however, compare this year to last year.
This is last year at day 6:
6280 (35%) have returned a successful result.
This is this year at day 6:
3228 (26%) have returned a successful result.
We've done half as many tasks this year... but the tasks are almost 4 times longer now than they were a year ago, so we're doing close to double the computations that we did in 2018's challenge!!!
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Would it be possible to add the number of tasks to the challenge stats?
Cheers Rick |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
Would it be possible to add the number of tasks to the challenge stats?
Cheers Rick
I'll look into it.
____________
My lucky number is 75898524288+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
Would it be possible to add the number of tasks to the challenge stats?
Cheers Rick
I'll look into it.
Done!
____________
My lucky number is 75898524288+1 |
|
|
robishVolunteer moderator
Volunteer tester
Send message
Joined: 7 Jan 12
Posts: 2195
ID: 126266
Credit: 7,299,570,840
RAC: 3,191,535
|
Would it be possible to add the number of tasks to the challenge stats?
Cheers Rick
I'll look into it.
Done!
That's cool, thanks Mike.
____________
My lucky numbers 10590941048576+1 and 224584605939537911+81292139*23#*n for n=0..26 |
|
|
Dave
Send message
Joined: 13 Feb 12
Posts: 3170
ID: 130544
Credit: 2,231,845,189
RAC: 766,099
|
Would it be possible to add the number of tasks to the challenge stats?
Cheers Rick
I'll look into it.
Done!
Wow ty. |
|
|
|
|
Would it be possible to add the number of tasks to the challenge stats?
Cheers Rick
I'll look into it.
Done!
Nice! |
|
|
|
|
Would it be possible to add the number of tasks to the challenge stats?
Cheers Rick
I'll look into it.
Done!
Very nice! Thank you very much!
Cheers |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
After (just under) 1 week:
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-14 04:08:15 UTC)
13332 tasks have been sent out. [CPU/GPU/anonymous_platform: 13324 (100%) / 0 (0%) / 8 (0%)]
Of those tasks that have been sent out:
2450 (18%) were aborted. [2450 (18%) / 0 (0%) / 0 (0%)]
365 (3%) came back with some kind of an error. [357 (3%) / 0 (0%) / 8 (0%)]
3764 (28%) have returned a successful result. [3764 (28%) / 0 (0%) / 0 (0%)]
6753 (51%) are still in progress. [6753 (51%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
2057 (55%) are pending validation. [2057 (55%) / 0 (0%) / 0 (0%)]
1659 (44%) have been successfully validated. [1659 (44%) / 0 (0%) / 0 (0%)]
9 (0%) were invalid. [9 (0%) / 0 (0%) / 0 (0%)]
39 (1%) are inconclusive. [39 (1%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=29199271. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 1.69% as much as it had prior to the challenge!
The double check just passed 59% complete!
____________
My lucky number is 75898524288+1 |
|
|
|
|
I'll look into it.
Done!
Nice! Will it stay "forever" after completion of this challenge, or just while the tasks are still on the server?
I see the December 2018 challenge stats also has task counts now, but other historical challenges do not.
/JeppeSN |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
I'll look into it.
Done!
Nice! Will it stay "forever" after completion of this challenge, or just while the tasks are still on the server?
I see the December 2018 challenge stats also has task counts now, but other historical challenges do not.
/JeppeSN
The reports are permanent.
We're sill in the cleanup phase for the December challenge, so those reports continue to be generated. That's why they have task counts. The reports prior to that are no longer generated, and it's a good thing they're not: those tasks are purged, so if those reports were ever generated again, all their contents would be erased.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Michael Goetz wrote: We've done half as many tasks this year... but the tasks are almost 4 times longer now than they were a year ago, so we're doing close to double the computations that we did in 2018's challenge!!!
When I compared a few days ago, it did not at all appear as if double the credit was earned in this challenge though.
(This will be easier to compare at the end of the challenge, obviously.) |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 391,496,192
RAC: 143,756
|
|
Half way!
Challenge: Conjunction of Venus & Jupiter
App: 13 (SoB-LLR)
(As of 2019-01-14 18:57:59 UTC)
13979 tasks have been sent out. [CPU/GPU/anonymous_platform: 13971 (100%) / 0 (0%) / 8 (0%)]
Of those tasks that have been sent out:
2612 (19%) were aborted. [2612 (19%) / 0 (0%) / 0 (0%)]
377 (3%) came back with some kind of an error. [369 (3%) / 0 (0%) / 8 (0%)]
4184 (30%) have returned a successful result. [4184 (30%) / 0 (0%) / 0 (0%)]
6806 (49%) are still in progress. [6806 (49%) / 0 (0%) / 0 (0%)]
Of the tasks that have been returned successfully:
2200 (53%) are pending validation. [2200 (53%) / 0 (0%) / 0 (0%)]
1925 (46%) have been successfully validated. [1925 (46%) / 0 (0%) / 0 (0%)]
10 (0%) were invalid. [10 (0%) / 0 (0%) / 0 (0%)]
49 (1%) are inconclusive. [49 (1%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is n=29227942. The leading edge was at n=28712732 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 1.79% as much as it had prior to the challenge!
____________
My lucky number is 75898524288+1 |
|
|
RafaelVolunteer tester
Send message
Joined: 22 Oct 14
Posts: 909
ID: 370496
Credit: 527,602,293
RAC: 304,931
 
|
Half way!
Half the challenge and I'm already 1~2 units away from breaking my ruby set and being OCD'd into getting my Turquoise set...
some1 help me. |
|
|
Dave
Send message
Joined: 13 Feb 12
Posts: 3170
ID: 130544
Credit: 2,231,845,189
RAC: 766,099
|
Half way!
Half the challenge and I'm already 1~2 units away from breaking my ruby set and being OCD'd into getting my Turquoise set...
some1 help me.
If it's in the bag there's nothing you can do. You have to move on eventually. At least it shows the right priority doing SoB first :). You could do PSP then have a red outline... |
|
|
RafaelVolunteer tester
Send message
Joined: 22 Oct 14
Posts: 909
ID: 370496
Credit: 527,602,293
RAC: 304,931
|
