Join PrimeGrid
Returning Participants
Community
Leader Boards
Results
Other
drummers-lowrise
|
Message boards :
Project Staging Area :
Few fixes for ( windows 522) PRPNET package
Author |
Message |
Crun-chi Volunteer tester
 Send message
Joined: 25 Nov 09 Posts: 3250 ID: 50683 Credit: 152,648,376 RAC: 7,833
                         
|
first one "in masterprpnet.ini"
--remove old projects ( SGS are gone long time ago... ) and from today
PPSLow is also "dead"...
-- add .exe in line with genferCUDA
//geneferexe=geneferCUDA ---> geneferexe=geneferCUDA.exe
Also ( for 64 bit Windows hosts) please add cudart.dll, cudart32_40_17.dll,
cufft.dll and cufft32_40_17.dll ( to main package) so when new user download "new 522 prpclient for windows"
can immediately start run GenferCUDA on 64 bit host.
Instead he need to download main package and add-on package for 64 bit CUDA computing.
I am sure that make new "updated" archive, and upload to server is trivial task, but can save much
time for first-time user.
____________
92*10^1585996-1 NEAR-REPDIGIT PRIME :) :) :)
4 * 650^498101-1 CRUS PRIME
2022202116^131072+1 GENERALIZED FERMAT
Proud member of team Aggie The Pew. Go Aggie! | |
|
|
I strongly support these changes. I didn't even realize that the first was a problem, which is probably one of the reasons why 5.2.2 didn't work for me, and once I saw that cudart32_40_17.dll wasn't there in an error message, I just gave up as well.
____________
| |
|
Crun-chi Volunteer tester
 Send message
Joined: 25 Nov 09 Posts: 3250 ID: 50683 Credit: 152,648,376 RAC: 7,833
                         
|
Until there is official update of PRPNet package here is temporary link
This package include
- all cuda dll-s for 32 and 64 bit host-s
- new pfgw 3.7.0 ( for 32 and 64 bit hosts)
-fixed genefer line in master_prpnet.ini
https://www.dropbox.com/s/0864kz53kxmi2wf/prp-package.zip
Package have about 52 MB
____________
92*10^1585996-1 NEAR-REPDIGIT PRIME :) :) :)
4 * 650^498101-1 CRUS PRIME
2022202116^131072+1 GENERALIZED FERMAT
Proud member of team Aggie The Pew. Go Aggie! | |
|
Honza Volunteer moderator Volunteer tester Project scientist Send message
Joined: 15 Aug 05 Posts: 1963 ID: 352 Credit: 6,425,775,796 RAC: 2,439,389
                                      
|
PFGW 3.7.0 is quite recent. Was it properly tested?
Don't get me wrong...I'm for new version.
____________
My stats | |
|
Crun-chi Volunteer tester
 Send message
Joined: 25 Nov 09 Posts: 3250 ID: 50683 Credit: 152,648,376 RAC: 7,833
                         
|
As I say - until there is official update", so this is unofficial package :)
I made this zip only to help people when error from missing dll appear.
Also: if you even find prime with "new"pfgw then it will be verified , so dont hesitate to use new version :)
And at the end, using new version is best way to test it, and to declare it as official version :)
Honza, last llr 3.8.9 for windows declared ok and put as official, but still I and many others claim to find bug, and after few version bug was fixed. So there is thin line between official, and unofficial :)
____________
92*10^1585996-1 NEAR-REPDIGIT PRIME :) :) :)
4 * 650^498101-1 CRUS PRIME
2022202116^131072+1 GENERALIZED FERMAT
Proud member of team Aggie The Pew. Go Aggie! | |
|
Michael Goetz Volunteer moderator Project administrator
 Send message
Joined: 21 Jan 10 Posts: 14045 ID: 53948 Credit: 485,285,209 RAC: 665,777
                               
|
As I say - until there is official update", so this is unofficial package :)
I made this zip only to help people when error from missing dll appear.
Also: if you even find prime with "new"pfgw then it will be verified , so dont hesitate to use new version :)
And at the end, using new version is best way to test it, and to declare it as official version :)
Honza, last llr 3.8.9 for windows declared ok and put as official, but still I and many others claim to find bug, and after few version bug was fixed. So there is thin line between official, and unofficial :)
But the reverse is NOT true.
If there's a flaw in the software and it says a number is composite but it's really prime, nothing will catch that error because we don't do double checks on PRPNet.
____________
My lucky number is 75898524288+1 | |
|
Crun-chi Volunteer tester
 Send message
Joined: 25 Nov 09 Posts: 3250 ID: 50683 Credit: 152,648,376 RAC: 7,833
                         
|
If there's a flaw in the software and it says a number is composite but it's really prime, nothing will catch that error because we don't do double checks on PRPNet.
I agree with you, but you also must agree with me that chance is so small: since there was no big changes in code.
____________
92*10^1585996-1 NEAR-REPDIGIT PRIME :) :) :)
4 * 650^498101-1 CRUS PRIME
2022202116^131072+1 GENERALIZED FERMAT
Proud member of team Aggie The Pew. Go Aggie! | |
|
|
C - TheDawgz hope that your banking institution doesn't feel the way you do.
"Yes, our ATM can give you your money 15 seconds faster - sometimes you might get a 10 instead of a 20 - but it's in our favour and it doesn't happen very often - so it's no big problem."
If there is a rationale for using a newer version of LLR or PFGW, etc.; TheDawgz are confident that the folks here will implement the new version when and if they are ready to support it's use. | |
|
Crun-chi Volunteer tester
 Send message
Joined: 25 Nov 09 Posts: 3250 ID: 50683 Credit: 152,648,376 RAC: 7,833
                         
|
C - TheDawgz hope that your banking institution doesn't feel the way you do.
"Yes, our ATM can give you your money 15 seconds faster - sometimes you might get a 10 instead of a 20 - but it's in our favour and it doesn't happen very often - so it's no big problem."
If there is a rationale for using a newer version of LLR or PFGW, etc.; TheDawgz are confident that the folks here will implement the new version when and if they are ready to support it's use.
It is strange how people using own time to do "better thing" are ridiculed, not respected :(
When I ask moderators to do any thing, I ALWAYS get same answer: we will do it, we will look at it, we will this or that. And that is answer in many cases , that is ALL that moderators do ( in case of MY QUESTIONS). If I ask again then I got answer : we do it in our free time etc etc etc... ( if you need list of opened, unanswered question just say it , I will be happy to ask again)
If I do some good thing , and compress all files in one place, write it is unofficial package , then I do "bad" thing.
It is so unfair, I repeat so unfair...
But that will teach me a lesson: so now lesson is learned:will not repeat same mistake twice...
Crun-chi
____________
92*10^1585996-1 NEAR-REPDIGIT PRIME :) :) :)
4 * 650^498101-1 CRUS PRIME
2022202116^131072+1 GENERALIZED FERMAT
Proud member of team Aggie The Pew. Go Aggie! | |
|
|
Condensing everything into one place was good, they are taking issue with you including a new version that hasn't been QA'd. As someone who QA quant models for equity trading, DEVs would get fired as fast as lightning if they got caught sneaking a change in that was not approved by QA.
____________
| |
|
|
Thanks for the suggestions Crun-chi!
I've uploaded new packages just now containing the following changes:
* fixed genefercude entry in windows master_prpclient.ini
* added TRPDC to windows master_prpclient.ini
* removed SGS and PPSELow from all master_prpclient.ini
* Set ESP as default project for all master_prpclient.ini
* Added CUDA DLLs to Windows package
I have not updated to the new PFGW yet, since I haven't had time to test it. Also, the changes since 3.6.7 are not relevant to the types of tests we do using PRPNet, so there is no pressing need.
Of course you are very welcome to run this version yourself, and assuming no problems are found we will switch to it in due course (probably when I next update the prpclient packages for some other reason).
Cheers
- Iain
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! | |
|
|
Thanks Iain, I will try out the new packages after the GFN challenge ends.
____________
| |
|
Crun-chi Volunteer tester
 Send message
Joined: 25 Nov 09 Posts: 3250 ID: 50683 Credit: 152,648,376 RAC: 7,833
                         
|
Since Iain make all changes in prpnet package and upload it Net, there is no need to download my package, so I deleted package from dropbox, and that link is no more valid.
For newest version go to http://uwin.mine.nu/PRPNet/
____________
92*10^1585996-1 NEAR-REPDIGIT PRIME :) :) :)
4 * 650^498101-1 CRUS PRIME
2022202116^131072+1 GENERALIZED FERMAT
Proud member of team Aggie The Pew. Go Aggie! | |
|
Message boards :
Project Staging Area :
Few fixes for ( windows 522) PRPNET package |