1)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 138566)
Posted 93 days ago by  Luigi R.
Well you need a bit of luck here and I had zero for TDP 2020.
I was pumping out plenty of GFN 16 with my RTX2080 and 65-70 second run times.
Nothing, not even a double check.
I feel you, bro. I found 1 Proth prime (PPSE, luckily first) after ~29k tests over the years.
~15k GFN16 and nothing too.
|
2)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 138511)
Posted 95 days ago by Luigi R.
Miraculously, I was able to discover prime on the mountain stage.
And it was a megaprime, the best outcome. Congratz!
Nice to see the jerseys go to four different persons.
Nice! :)
|
3)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 138458)
Posted 95 days ago by Luigi R.
...and you can just throw money at this project, it leaves anyone out who doesn't have the money or want to spend thousands on 50+ computers and GPUs spending hundreds a month on electricity...
The real wastefullness in challenges is to run multithreading when it's not the most efficent solution.
|
4)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 138393)
Posted 97 days ago by Luigi R.
@Nita: you won the lottery by that MEGA. :D
|
5)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 138381)
Posted 97 days ago by Luigi R.
Argh, I'm always doublechecker on SR5.
|
6)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 137872)
Posted 111 days ago by Luigi R.
If your host is very fast and you are 80% first, you will observe that your first rate tends to 50% though.
|
7)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 137869)
Posted 111 days ago by Luigi R.
I don't understand your premise. Even in your example, the next guy in line wins. There is no harm to anyone. If the first guy returns first, he wins. If the second guy is the first to return because the first guy fails, the second guy wins. And so on.[
As to getting original task vs resends, we have already been told by the admin here that it is random. I think you are claiming grievance about being issued resends vs new tasks. No way to avoid that, regardless of the amounts of aborts of previous tasks.
Yeah, number of resends is higher than it should be because someone is aborting a bunch of tasks. Fair or not? Are those aborts justified?
Race is killed.
|
8)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 137866)
Posted 111 days ago by Luigi R.
Is it not good to be second, when the first are abandoned? In other words, what is the harm? I am obviously missing the point of the complaint. Why the anger?
WUs are sent to 2 users. When he aborts it, you are the third and you almost always do double check.
Maybe you will be 50% first on average cause of this. If your machine can get more than 50% first (not my case), it's not nice.
E.g. https://www.primegrid.com/workunit.php?wuid=641872022
|
9)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 137864)
Posted 111 days ago by Luigi R.
I found 1 user who downloaded tons of SR5. I don't know if it's good news.
He aborted the most work yesterday and he downloaded 885 SR5 tasks on 1 single host today. I guess he is going to abort hundreds of tasks again. Is this pratice tolerated?
|
10)
Message boards :
Number crunching :
Tour de Primes 2020
(Message 137830)
Posted 112 days ago by Luigi R.
* Polk-a-dot Jersey -- on the 29th (Leap Day) of February we'll have a "Mountain Stage" and award the Polk-a-dot Jersey to the one who finds the most primes on that day (tiebreaker will be prime score for that day).
I didn't notice that before. So does this year have a different "Mountain Stage" day?
|
