| Author |
Message |
|---|
|
|
|
Just thought I'd see what everyone is currently working on. Just last week I built two more rigs running AMD Phenom II X6 cpus, so I've gotten LOTS of new cores up and running :-)
My focus recently has been almost entirely PPS in hopes of getting some primes, (although almost 20000 WUs in and only 1 prime so far, which i got around the 8000 WU)
I may start dedicating at least one of my new boxes entirely to badges. Not sure yet.
SO yeah, just wanted to see what everyone else was currently crunching or focusing their energy. I currently am not running any GPU at primegrid, I may switch it back here soon..its over at DNETC right now getting me some credits. :-) I have a amd 6990 running. but no nvidia / cuda cards here :-( so not as useful for primegrid. |
|
|
|
|
|
I am doing PPS at the moment, Mark. I've been a bit unlucky with them recently so I reckon I'm due to find a few.
Cheers!
T |
|
|
|
|
|
I am working towards my PSA Silver - I think I have enough credits in the bank to make it when they next get converted to BOINC credit. In the meantime I'm still crunching Reisel LLR on BOINC, looking for my silver, and hopefully that elusive prime!
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! |
|
|
|
|
|
CW Sieve, mostly, on GPU (some Collatz). TRP Sieve on CPU, driving towards gold, though I switch a few cores to PRPNet's five-or-bust now and then... oh yeah I still run SETI on one core...
--Gary |
|
|
|
|
|
Working on the 121 project at PRPNet.
I am working towards my next badge for PSA at the moment and making Aggie the number 1 team for the 121 project. After this I will try to get some of my PG silver badges
____________
Welcome to Holland
 |
|
|
 Neo Volunteer tester
Send message
Joined: 28 Oct 10
Posts: 710
ID: 71509
Credit: 91,178,992
RAC: 0
                   
|
|
Welcome to the team Mark!
As for what I'm working on, it constantly changes.
Right now I'm working via prpnet with a project breakdown of:
25% SGS - 10 w/u
25% PPSEHigh - 7 w/u
25% SR5 - 2 w/u
25% PPSElow - 25 w/u
I have an AMD 1055T :) Glad to hear about your new additions Mark. I'm green with envy :)
As for GPU - I have a GT430 and it spends most of its time sieving using tpsieve via Primesearchteam
FYI: I believe that most all BOINC Proth LLR w/u's are doublechecks until N reaches 1M. So, chances of finding a prime in this project is going to be ssssllliiiimmmmm.
|
|
|
|
|
FYI: I believe that most all BOINC Proth LLR w/u's are doublechecks until N reaches 1M. So, chances of finding a prime in this project is going to be ssssllliiiimmmmm.
Wow! That's information I would have loved to have a week ago when I first got my new PCs running. :-) On the plus side, I cranked out a lot of WUs since then. So good credit / badge getting.
Maybe I'll do some SGS for a bit instead then and still pull in a prime. |
|
|
NeoVolunteer tester
Send message
Joined: 28 Oct 10
Posts: 710
ID: 71509
Credit: 91,178,992
RAC: 0
|
|
On the Primegrid homepage, I'm seeing some Proth primes listed as newly discovered... n=~945K...
Maybe the double checking has stopped....??
Anywh00t, I have five TRP's crunching right now. Going to eliminate a K.... :)
Neo |
|
|
|
|
|
I am starting to get PPS's building up in my pending credit list Neo so yep not all double checking at the moment. All composite mind you! Been ages since I found a PPS prime and my average will soon be below 0.02%. I hope you are luckier than me.
Cheers!
T |
|
|
NeoVolunteer tester
Send message
Joined: 28 Oct 10
Posts: 710
ID: 71509
Credit: 91,178,992
RAC: 0
|
|
I just read about the new addition of the MEGA prime project on prpnet.
I'm finishing my BOINC w/u's and giving this project a try.
A MILLION DIGITS. Wh00t! |
|
|
|
|
I just read about the new addition of the MEGA prime project on prpnet.
I'm finishing my BOINC w/u's and giving this project a try.
A MILLION DIGITS. Wh00t!
Ha beat ya to it.. already got some done but no stats on the page yet. Not sure when the port will be added. |
|
|
NeoVolunteer tester
Send message
Joined: 28 Oct 10
Posts: 710
ID: 71509
Credit: 91,178,992
RAC: 0
|
|
Ha beat ya to it.. already got some done but no stats on the page yet. Not sure when the port will be added.
http://prpnet.mine.nu:12010/all.html |
|
|
|
|
|
You should give the 121 project a go.
It starts at 121*2^k with k>4330000. That has got to be a mega prime if found.
Also the 27 project starts at 27*2^k with k> 3760000. If a prime is found in these 2 projects it will be bigger than that found at the mega project surely
____________
Welcome to Holland
|
|
|
|
|
|
I've committed all but 2 pc's to PG and the 1 on prp is running the mega only 12 hrs a day. Neo is kicking my butt on the mega port.
Have enough woodall's to go silver once they all validate. Starting work on the cullen llr now. Man these are long wu's.
|
|
|
NeoVolunteer tester
Send message
Joined: 28 Oct 10
Posts: 710
ID: 71509
Credit: 91,178,992
RAC: 0
|
|
MEGA port runtime with 5 of 6 cores running on Phenom II 1055t (2.8 ghz)
AMD K10 type-3 FFT length 192K, Pass1=256, Pass2=768
Time per bit: 3.399 ms.
AMD K10 type-3 FFT length 256K, Pass1=256, Pass2=1K
Time per bit: 4.738 ms.
Averaging a 15,500 sec runtime per w/u.
I ran a w/u on my GT430 using llrCUDA and canceled it after 20 minutes because I was getting an 8.5 ms per bit time. I guess the reason is not enough cuda cores?
Anywh00t.
Neo |
|
|
|
|
|
Threw 1/2 of the i7 back at prpnet's "5oB" effort (other half running boinc trp sieve). Just for fun I started off with hyperthreading turned off in the BIOS and was getting times like 7.3-7.4 ms per bit. Ran that way for several hours then shut down momentarily to turn HT on. Time per bit now (with 4 5oB tasks running rather than 2, plus 4 sieves) is consistently less than 14. The sieving time with HT off is about 7100s vs. roughly 12000s (with twice as many running) with HT on. I conclude from this that HT "on" is more efficient, usually. Of course there are scenarios where it is better to turn HT off, like squeezing in a few last WUs at the end of a challenge, or if one is focused on being the first reporter of a big prime rather than the double-checker.
Wow there were a whole lot of numbers in that first paragraph. I apologize. Please feel free to poke holes in my logic. Or if you prefer, lean back and crack open a bottle of pale ale. That's what I just did.
Happy Weekend,
Gary |
|
|
|
|
|
Nope! No flaws in logic :-) (other than sadly I have no cpu with hyperthreading..waiting for next generation of i7s in Q4)
and i'm with ya on the pale ale! |
|
|
|
|
|
Thanks for that info Gary. I was curious about that myself. Now I'm free hyper-pale ale without that theory hanging over my head.
Mark, I think the release of the new i7 has been pushed back some. There was speculation that they would move that date forward to counter AMDs Bulldozer. But now it looks more like end of Q1-2012, probably to coincide with CES. The good news is that it looks like USB 3.0 will be supported natively.
Anyway, not that far off....
Enjoy the weekend.
Caravaggio |
|
|
|
|
|
Last night I rebooted my gpu pc and started 1 core on the prpnet low. My pps sieve jobs went from 1126 to 1141 seconds per wu. Amazing the difference running the cpu makes.
Speaking of ale - you know, when golfing it's a requirement that you drink at least 3 before finishing :) |
|
|
|
|
Speaking of ale - you know, when golfing it's a requirement that you drink at least 3 before finishing :)
My rule always is a 6-pack per 9 holes :-) |
|
|
|
|
Speaking of ale - you know, when golfing it's a requirement that you drink at least 3 before finishing :)
My rule always is a 6-pack per 9 holes :-)
lol if i did that, i'd never finish !
|
|
|
|
|
My rule always is a 6-pack per 9 holes :-)
I feel a joke coming on here but alas my missus is fed up with my constant wordplay jokes.
I asked her, "What can I do to stop my addiction?"
She said, "Whatever means necessary."
"No it doesn't." |
|
|
NeoVolunteer tester
Send message
Joined: 28 Oct 10
Posts: 710
ID: 71509
Credit: 91,178,992
RAC: 0
|
|
Hey Team,
For those that have had the opportunity to work on the AP26 project......
Does anyone know why this project is not still going? maybe an AP27+?
The number of primes Aggie the Pew has found just as a team (1,000 primes just on PPSelow alone) would, at least in my mind, justify re-opening the project, let alone all of the other primes discovered.
Wouldn't all these new primes allow for an AP27? AP28? a second AP26?
Am I just clueless??
Neo
|
|
|
|
|
|
Not just clueless Neo. :)
I think the argument was that as the length of the progression increased the probability of finding prime decreased. It certainly seems true from my results
AP22: 11188619077563023+22271281*23#*n for n=0..21
AP21: 47410499638738333+6076548*23#*n for n=0..20
AP21: 24496804147495859+28412*23#*n for n=0..20
AP20: 30816489451135919+30616432*23#*n for n=0..19
AP20: 10551203564694349+31548736*23#*n for n=0..19
AP20: 24676050684111929+23009527*23#*n for n=0..19
AP20: 60608976010062257+1343181*23#*n for n=0..19
AP20: 52950478197043523+2860001*23#*n for n=0..19
AP20: 13948786959390131+8517438*23#*n for n=0..19
AP20: 22863495571766891+1305074*23#*n for n=0..19
but there is no technical reason so far as I know why it couldn't be done.
I suppose if we made a good enough case for it in terms of committed resources it might be possible to get the project going again. But then again committing AtP is like herding rats.
Cheers!
T |
|
|
