![]() |
![]() |
#1 |
Apr 2008
Oslo, Norway
7·31 Posts |
![]()
Hi everyone!
![]() I have been discussing something with ltd and want your input. We are currently looking for primes for 9 k's and running second pass (double check) on 1 k. There are approx. 1000 second pass tests in the queue now meaning that if a client misbehaves, we will not know about it for quite a while. A possible solution: With prpclient comes a very welcome new function; The function to work with more than one server automatically. If we change these lines in our configs: Code:
server=PSPtest:100:1:www.psp-project.de:7100 server=PSPtestdc:0:1:www.psp-project.de:7101 Code:
server=PSPtest:50:1:www.psp-project.de:7100 server=PSPtestdc:50:1:www.psp-project.de:7101 This will help us find problematic clients faster, do less additional work and waste less CPU. If you want the project to go this way, ltd will make combined statistics of first and second pass scores, and for those of you who are running this project to get a better DC-Vault rank, I'll make sure the DC-Vault starts using the new stats. And of course, running second pass tests would still be optional and noone will be required to alter their ini files. ![]() Let me know what you think, and also let me know if anything is unclear in this post. |
![]() |
![]() |
![]() |
#2 |
Oct 2008
Trondheim, Norway
238 Posts |
![]()
I think it is a really good plan to speed up second pass, and this seems like a good idea.
Getting a combined score is the other way to encourage people to run more second pass (for those concerned about DC-Vault). I give this a ![]() .R |
![]() |
![]() |
![]() |
#3 |
Dec 2004
1001010112 Posts |
![]()
Well since the vault has never acknowledged secondpass as a project even though the scores are seperate... this has been several years now...
I now fully support combining second and first pass scoring. They are combining scores in SoB for first and secondpass just makes sence to do that here as well. |
![]() |
![]() |
![]() |
#4 | |
Aug 2002
3·52·7 Posts |
![]()
Well Opyrt, I agree with most of your ideas. We will never get anywhere as long as the second pass is not in the vault. So either vault it, or combine it.
Quote:
LTD should put a "sprinkling" of larger n into the queue(s). If the second pass is vaulted, we need to build some excitement around it and push it. If the second pass is combined with the first pass, then LTD could put a small percentage of second pass into the first pass queue. The scoring is the same, so no one should object. Your idea of having people change their ini files, is a good one but will not work. People will not change their ini files. They "set and forget". LTD should change the example ini, so that new people would have the new settings when they start. So, what do you say? |
|
![]() |
![]() |
![]() |
#5 | ||
Apr 2008
Oslo, Norway
7·31 Posts |
![]()
With current progress on the second pass effort, it should catch up some time this year.
Quote:
Quote:
I agree that the examples will have to be altered, but I actually think asking participants to alter their files will win some over also. |
||
![]() |
![]() |
![]() |
#6 | ||
Aug 2002
3×52×7 Posts |
![]() Quote:
Quote:
|
||
![]() |
![]() |
![]() |
#7 | |
Apr 2008
Oslo, Norway
21710 Posts |
![]() Quote:
![]() |
|
![]() |
![]() |
![]() |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Recommendations for cutting down input | fivemack | Msieve | 1 | 2017-12-12 17:22 |
For the amusement of the record prime hunters | a1call | Miscellaneous Math | 11 | 2017-02-05 07:19 |
Hello fellow prime number hunters. | zampa | Information & Answers | 5 | 2016-04-22 02:04 |
GMP-ECM and the Cunningham Input List | M0CZY | GMP-ECM | 10 | 2006-12-21 14:13 |
Welcome to Lone Mersenne Hunters | tha | Lone Mersenne Hunters | 23 | 2005-07-15 12:18 |