![]() |
|
|
#34 |
|
"James Heinrich"
May 2004
ex-Northern Ontario
2×13×131 Posts |
I'm still getting hard crashes of v25.5 on WinXP32 on my Northwood 3.0C. I ran a 2-thread torture test for 14 hours today, no problems at all, but tried running ECM in 2 threads for maybe 15 minutes and it died. No errors logged.
|
|
|
|
|
|
#35 |
|
Jan 2006
Tampa, Florida
2×97 Posts |
OK, I just increased my L2 cache to 32,768KB and the per iteration time decreased to 0.644 seconds. Is there still something I can do to reduce it to 0.500?
|
|
|
|
|
|
#36 |
|
Bemusing Prompter
"Danny"
Dec 2002
California
2×5×239 Posts |
How do you modify the L2 cache size?
|
|
|
|
|
|
#37 | |
|
P90 years forever!
Aug 2002
Yeehaw, FL
2·53·71 Posts |
Quote:
Xyzzy's guess is probably accurate. Your huge FFT is putting enormous pressure on your smallish L2 cache. Using another CPU increases the pressure on the L2 cache - negating much of the advantage of using a second CPU. What can you do? Not much short of buying new hardware. Sorry. |
|
|
|
|
|
|
#38 |
|
P90 years forever!
Aug 2002
Yeehaw, FL
2×53×71 Posts |
Is it dying because of overcommitting memory? I have a fix for that particular bug. I've uploaded a new Win32 version for you to try.
Last fiddled with by Prime95 on 2007-09-27 at 03:08 |
|
|
|
|
|
#39 | |
|
Jan 2006
Tampa, Florida
3028 Posts |
Quote:
I just increased the cache to 256,000KB and it reduced interation time to 0.636 seconds. However, I noticed that if I even move the mouse the iteration time increases to 0.650 - 0.660 seconds, even though I have the priority set to 10! Another thing that I am wondering about is that I only have 1GB of RAM on my computer. I am allotting 920MB (the maximum allowed) to this system, but Windows Task Manager says Prime95 is only using 367MB (it uses about 20MB for a 10 million digit number) and 95% of the CPU. Would adding more RAM increase the speed at all, or is 0.636 iterations per second the best I can do with my current machine using both cores of the Pentium D? |
|
|
|
|
|
|
#40 | |
|
1976 Toyota Corona years forever!
"Wayne"
Nov 2006
Saskatchewan, Canada
22×7×167 Posts |
Quote:
I am still running the current production version 24.14 and LL testing a number in the 36M range. I tried CPUL2CACHESIZE anywhere from 128 to 32768 and the per iteration time on my P4 2.4 Ghz was always 0.082 secs. |
|
|
|
|
|
|
#41 |
|
Jan 2006
Tampa, Florida
2·97 Posts |
It doesn't improve your per iteration time all that much (increasing from 128MB to 256000MB improved mine from 0.660 seconds to 0.636 seconds). The only way to check if it worked or not is to open up Prime95, click on Options, then CPU. It will tell you your L2 Cache size in this window at the bottom.
|
|
|
|
|
|
#42 | |
|
Jul 2007
101002 Posts |
Quote:
|
|
|
|
|
|
|
#43 | |||||
|
Jun 2003
22218 Posts |
Quote:
Quote:
Quote:
EDIT: the limit for ECM factoring is 20,500,000, which your exponent is obviously way beyond. Quote:
Quote:
Last fiddled with by Mr. P-1 on 2007-09-27 at 10:10 |
|||||
|
|
|
|
|
#44 |
|
Bemusing Prompter
"Danny"
Dec 2002
California
2·5·239 Posts |
Yes, the L2 cache and physical memory are two different things. I tried increasing my "L2 cache" to 32,768 kb like you described, and nothing happened.
|
|
|
|
![]() |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Prime95 version 29.4 | Prime95 | Software | 441 | 2020-02-16 15:18 |
| Prime95 version 27.3 | Prime95 | Software | 148 | 2012-03-18 19:24 |
| Prime95 version 26.3 | Prime95 | Software | 76 | 2010-12-11 00:11 |
| Prime95 version 25.4 | Prime95 | PrimeNet | 143 | 2007-09-24 21:01 |
| When the next prime95 version ? | pacionet | Software | 74 | 2006-12-07 20:30 |