![]() |
|
|
#12 |
|
Mar 2004
72·11 Posts |
I actually have four sticks of single-channel 1GB RAM stick. I upgraded from four sticks of dual-channel 512K RAM.
|
|
|
|
|
|
#13 | |
|
Jun 2003
22218 Posts |
Quote:
It might be different if you intend to more than incidental P-1 work, or if you're testing huge exponents. |
|
|
|
|
|
|
#14 |
|
Sep 2006
Odenton, MD, USA
22·41 Posts |
|
|
|
|
|
|
#15 |
|
P90 years forever!
Aug 2002
Yeehaw, FL
752610 Posts |
|
|
|
|
|
|
#16 |
|
Mar 2004
72×11 Posts |
|
|
|
|
|
|
#17 |
|
Sep 2006
Odenton, MD, USA
22×41 Posts |
Ok, then we might have a problem. Earlier I had set the amount to memory to 1792 MB but when P-1 Stage 2 started it got the memory error and went to approx 1400ish. My system is running Windows XP SP3 with 2.5GB of memory (2 X 1GB chips and 2 x 256 MB chips). I'm currently in the middle of stage 2, do you want me to up the memory now to see what happens or wait for the next P-1 assignment to change the memory settings?
|
|
|
|
|
|
#18 |
|
Tribal Bullet
Oct 2004
67258 Posts |
Unless you boot with the /3GB switch, 1400MB of memory is about the most that a single win32 process can reasonably allocate for itself. At least I haven't seen one allocate more than that, the theoretical limit is 2GB.
|
|
|
|
|
|
#19 |
|
Mar 2004
72×11 Posts |
So is there a way to get the software to allocate more memory than the 1100/1300 MB that it went down to?
|
|
|
|
|
|
#20 |
|
Sep 2006
Odenton, MD, USA
101001002 Posts |
On my 32-bit system, I normally run 1536MB for P-1 testing and I don't get any memory allocation errors.
|
|
|
|
|
|
#21 |
|
Mar 2004
53910 Posts |
I just tried using two threads to do P-1 tests on two different exponents. No dice. It says "Out of memory!", which it's NOT. I can't even get my ram usage past 2GB with prime95!!
|
|
|
|
|
|
#22 |
|
Jun 2003
116910 Posts |
I used to use that exact same amount.
There's no reason to use a binary-round number. If you want to optimise your memory allocation, look at the number of relative primes processed each pass, and make sure that you never have a small number left over for the final pass. |
|
|
|
![]() |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| "lanczos error: only trivial dependencies found" with massive oversieving | eigma | Msieve | 21 | 2015-05-28 03:27 |
| "Hybrid Memory Cube" offers 1 Tb/s memory bandwith at just 1.4 mW/Gb/s | ixfd64 | Hardware | 4 | 2011-12-14 21:24 |
| Memory Allocation Error | joel893 | Software | 0 | 2008-06-03 14:54 |
| "Feature" request: Keep new B2 value when memory allotment increased | JuanTutors | Software | 3 | 2006-12-18 06:31 |
| Would Minimizing "iterations between results file" may reveal "is not prime" earlier? | nitai1999 | Software | 7 | 2004-08-26 18:12 |