gpuOwL: an OpenCL program for Mersenne primality testing - Page 228

Xyzzy · 2020-09-28, 00:40

Quote:

Code:

2020-06-05 17:13:16 Radeon Pro W5500-0 OpenCL compilation in 3.10 s
2020-06-05 17:13:17 Radeon Pro W5500-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-06-05 17:13:21 Radeon Pro W5500-0 77936867 OK      800   0.00%; 2982 us/it; ETA 2d 16:34; 1579c241dc63eca6 (check 1.27s)
2020-06-05 17:23:18 Radeon Pro W5500-0 77936867 OK   200000   0.26%; 2991 us/it; ETA 2d 16:35; f0b04b45b0855bd2 (check 1.28s)
2020-06-05 17:33:15 Radeon Pro W5500-0 77936867 OK   400000   0.51%; 2979 us/it; ETA 2d 16:10; c03f94396a5aa29e (check 1.27s)
2020-06-05 17:43:17 Radeon Pro W5500-0 77936867 OK   600000   0.77%; 3004 us/it; ETA 2d 16:32; b9decd65ca71b629 (check 1.28s)

2020-09-04 13:24:28 GeForce GTX 1080 Ti-0 OpenCL compilation in 2.02 s
2020-09-04 13:24:29 GeForce GTX 1080 Ti-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-09-04 13:24:32 GeForce GTX 1080 Ti-0 77936867 OK      800   0.00%; 2481 us/it; ETA 2d 05:43; 1579c241dc63eca6 (check 1.04s)
2020-09-04 13:32:54 GeForce GTX 1080 Ti-0 77936867 OK   200000   0.26%; 2514 us/it; ETA 2d 06:18; f0b04b45b0855bd2 (check 1.04s)
2020-09-04 13:41:12 GeForce GTX 1080 Ti-0 77936867 OK   400000   0.51%; 2483 us/it; ETA 2d 05:29; c03f94396a5aa29e (check 1.05s)
2020-09-04 13:49:27 GeForce GTX 1080 Ti-0 77936867 OK   600000   0.77%; 2473 us/it; ETA 2d 05:07; b9decd65ca71b629 (check 1.06s)

2020-09-04 17:42:56 GeForce GTX 980 Ti-0 OpenCL compilation in 1.83 s
2020-09-04 17:42:58 GeForce GTX 980 Ti-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-09-04 17:43:04 GeForce GTX 980 Ti-0 77936867 OK      800   0.00%; 4221 us/it; ETA 3d 19:23; 1579c241dc63eca6 (check 1.73s)
2020-09-04 17:57:13 GeForce GTX 980 Ti-0 77936867 OK   200000   0.26%; 4258 us/it; ETA 3d 19:56; f0b04b45b0855bd2 (check 1.75s)
2020-09-04 18:11:28 GeForce GTX 980 Ti-0 77936867 OK   400000   0.51%; 4263 us/it; ETA 3d 19:49; c03f94396a5aa29e (check 1.75s)
2020-09-04 18:25:42 GeForce GTX 980 Ti-0 77936867 OK   600000   0.77%; 4262 us/it; ETA 3d 19:34; b9decd65ca71b629 (check 1.75s)

2020-09-25 01:11:43 GeForce RTX 2060 SUPER-0 OpenCL compilation in 1.35 s
2020-09-25 01:11:45 GeForce RTX 2060 SUPER-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-09-25 01:11:50 GeForce RTX 2060 SUPER-0 77936867 OK      800   0.00%; 3966 us/it; ETA 3d 13:51; 1579c241dc63eca6 (check 1.65s)
2020-09-25 01:34:08 GeForce RTX 2060 SUPER-0 77936867 OK   200000   0.26%; 5188 us/it; ETA 4d 16:01; f0b04b45b0855bd2 (check 2.18s)
2020-09-25 01:51:15 GeForce RTX 2060 SUPER-0 77936867 OK   400000   0.51%; 5123 us/it; ETA 4d 14:21; c03f94396a5aa29e (check 2.12s)
2020-09-25 02:08:30 GeForce RTX 2060 SUPER-0 77936867 OK   600000   0.77%; 5164 us/it; ETA 4d 14:56; b9decd65ca71b629 (check 2.13s)

2080 Super

Code:

2020-09-27 15:56:43 GeForce RTX 2080 SUPER-0 OpenCL compilation in 1.85 s
2020-09-27 15:56:45 GeForce RTX 2080 SUPER-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-09-27 15:56:48 GeForce RTX 2080 SUPER-0 77936867 OK      800   0.00%; 2623 us/it; ETA 2d 08:47; 1579c241dc63eca6 (check 1.10s)
2020-09-27 16:05:42 GeForce RTX 2080 SUPER-0 77936867 OK   200000   0.26%; 2675 us/it; ETA 2d 09:46; f0b04b45b0855bd2 (check 1.11s)
2020-09-27 16:14:38 GeForce RTX 2080 SUPER-0 77936867 OK   400000   0.51%; 2678 us/it; ETA 2d 09:40; c03f94396a5aa29e (check 1.10s)
2020-09-27 16:23:45 GeForce RTX 2080 SUPER-0 77936867 OK   600000   0.77%; 2730 us/it; ETA 2d 10:39; b9decd65ca71b629 (check 1.14s)

M344587487 · 2020-09-28, 09:36

What's the power draw of the W5500 and how does a Radeon VII compare throughput-wise with the same exponent? If the RTX 5500 has similar numbers which it should (DP ratio looks the same), if the 4GB model runs as well as the 8GB model, and if the card can be made to work (last I checked that was a big if), then the results are not as terrible as I expected. The price of an RTX 5500 is currently £170, on a throughput-per-dollar basis it looks like it could at least see the R7 ahead in the distance instead of being in division 3 (ignoring power draw and density limitations which of course is a hard sell).

moebius · 2020-09-28, 10:12

Code:

2020-09-28 11:54:26 gfx900 RX Vega AMD OpenCL compilation in 2.52 s
2020-09-28 11:54:27 gfx900 RX Vega AMD 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-09-28 11:54:28 gfx900 RX Vega AMD 77936867 OK      800   0.00%; 1206 us/it; ETA 1d 02:07; 1579c241dc63eca6 (check 0.53s)
2020-09-28 11:58:31 gfx900 RX Vega AMD 77936867 OK   200000   0.26%; 1216 us/it; ETA 1d 02:16; f0b04b45b0855bd2 (check 0.54s)
2020-09-28 12:02:35 gfx900 RX Vega AMD 77936867 OK   400000   0.51%; 1219 us/it; ETA 1d 02:15; c03f94396a5aa29e (check 0.54s)
2020-09-28 12:06:40 gfx900 RX Vega AMD 77936867 OK   600000   0.77%; 1219 us/it; ETA 1d 02:11; b9decd65ca71b629 (check 0.54s)

Code:

2020-09-28 10:43:59 Tesla P100-PCIE-16GB-0 OpenCL compilation in 1.72 s
2020-09-28 10:43:59 Tesla P100-PCIE-16GB-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-09-28 10:44:00 Tesla P100-PCIE-16GB-0 77936867 OK      800   0.00%;  728 us/it; ETA 0d 15:45; 1579c241dc63eca6 (check 0.41s)
2020-09-28 10:46:25 Tesla P100-PCIE-16GB-0 77936867 OK   200000   0.26%;  728 us/it; ETA 0d 15:43; f0b04b45b0855bd2 (check 0.41s)
2020-09-28 10:48:51 Tesla P100-PCIE-16GB-0 77936867 OK   400000   0.51%;  728 us/it; ETA 0d 15:41; c03f94396a5aa29e (check 0.41s)
2020-09-28 10:51:17 Tesla P100-PCIE-16GB-0 77936867 OK   600000   0.77%;  728 us/it; ETA 0d 15:38; b9decd65ca71b629 (check 0.41s)

Xyzzy · 2020-09-28, 11:50

Quote:

Originally Posted by M344587487

What's the power draw of the W5500 and how does a Radeon VII compare throughput-wise with the same exponent? If the RTX 5500 has similar numbers which it should (DP ratio looks the same), if the 4GB model runs as well as the 8GB model, and if the card can be made to work (last I checked that was a big if), then the results are not as terrible as I expected. The price of an RTX 5500 is currently £170, on a throughput-per-dollar basis it looks like it could at least see the R7 ahead in the distance instead of being in division 3 (ignoring power draw and density limitations which of course is a hard sell).

The W5500 uses ~100W.

We haven't figured out a way to get a "normal" consumer card to work.

One nice feature of the W5500 is that it is a single slot card.

petrw1 · 2020-09-29, 04:56

Quote:

Originally Posted by petrw1

Why does it ignore
-log 50000
in config.txt.?

I still get progress updats every 10000.

Code:

-user petrw1 -cpu colab -device 0 -log 50000 -maxAlloc 4000

I faked it out; This is my Colab GPUOwl run script (most of it borrowed from others here; see the last couple lines.

Code:

import subprocess
import os
import os.path
from google.colab import drive
import sys
if not os.path.exists('/content/drive/My Drive'):
  drive.mount('/content/drive')
%cd '/content/drive/My Drive/HOOT/'
!cp libstdc* /usr/lib/x86_64-linux-gnu/

statinfo = os.stat('./worktodo.txt')
if statinfo.st_size < 50:
  print ('WARNING, small file size indicates little or no gpuowl work to do')
!LD_LIBRARY_PATH="lib:${LD_LIBRARY_PATH}" && chmod 777 gpuowl.exe && chmod 777 worktodo.txt
!./gpuowl.exe  >>gpuowllog.txt 2>&1 &
print('gpuowl launched in background')
!nvidia-smi -L
!tail -f gpuowl.log | egrep '00000|P2|factor'

2nd last line: Tells me which GPU I got; P100 is several times 5 faster than K80, P4 or T4
I "Interrupt Execution"; "Factory Reset" and try again until I get one.

Last line gives me only:
- Every 100,000 P1 iterations
- Each P2 iteration
- Factor/No Factor lines

DrobinsonPE · 2020-09-30, 04:42

Quote:

Originally Posted by Xyzzy

Somehow we got gpuowl to work on our integrated graphics. (It runs so slow that it isn't worth doing, but it is neat that it works at all!)

I saw your post and thought it would be an interesting experiment to try gpuowl on my Celeron J4105 with UHD 600 integrated graphics. I got mfakto working on it so why not try gpuowl as well. It looks like it failed on the OpenCL version.

Code:

2020-09-29 21:11:18 gpuowl v6.11-380-g79ea0cc
2020-09-29 21:11:18 Note: not found 'config.txt'
2020-09-29 21:11:18 device 0, unique id ''
2020-09-29 21:11:18 Intel(R) UHD Graphics 600-0 58279579 FFT: 3M 1K:6:256 (18.53 bpw)
2020-09-29 21:11:18 Intel(R) UHD Graphics 600-0 Expected maximum carry32: 48550000
2020-09-29 21:11:19 Intel(R) UHD Graphics 600-0 OpenCL args "-DEXP=58279579u -DWIDTH=1024u -DSMALL_HEIGHT=256u -DMIDDLE=6u -DPM1=0 -DMM_CHAIN=1u -DMM2_CHAIN=1u -DMAX_ACCURACY=1 -DWEIGHT_STEP_MINUS_1=0xc.6dc0cf04225b8p-5 -DIWEIGHT_STEP_MINUS_1=-0x8.f3ac43205c59p-5  -cl-unsafe-math-optimizations -cl-std=CL2.0 -cl-finite-math-only "
2020-09-29 21:11:20 Intel(R) UHD Graphics 600-0 ASM compilation failed, retrying compilation using NO_ASM
2020-09-29 21:11:21 Intel(R) UHD Graphics 600-0 OpenCL compilation error -11 (args -DEXP=58279579u -DWIDTH=1024u -DSMALL_HEIGHT=256u -DMIDDLE=6u -DPM1=0 -DMM_CHAIN=1u -DMM2_CHAIN=1u -DMAX_ACCURACY=1 -DWEIGHT_STEP_MINUS_1=0xc.6dc0cf04225b8p-5 -DIWEIGHT_STEP_MINUS_1=-0x8.f3ac43205c59p-5  -cl-unsafe-math-optimizations -cl-std=CL2.0 -cl-finite-math-only  -DNO_ASM=1)
2020-09-29 21:11:21 Intel(R) UHD Graphics 600-0 -cl-std OpenCLC version greater than OpenCL (API) version
-cl-std OpenCLC version greater than OpenCL (API) version
2020-09-29 21:11:21 Intel(R) UHD Graphics 600-0 Exception gpu_error: BUILD_PROGRAM_FAILURE clBuildProgram at clwrap.cpp:246 build
2020-09-29 21:11:21 Intel(R) UHD Graphics 600-0 Bye

This was the Windows compiled gpuowl from this thread https://www.mersenneforum.org/showthread.php?t=25624. I also attached a picture of the GPUZ screen for the UHD 600.

I have an I3-9100 and an I5-8250U that I might experiment on as well if/went I have the time. Did gpuowl work in Windows or Linux?

M344587487 · 2020-10-13, 14:04

Quote:

Originally Posted by Xyzzy

Somehow we got gpuowl to work on our integrated graphics. (It runs so slow that it isn't worth doing, but it is neat that it works at all!)

Code:

2020-09-25 13:26:07 Intel(R) UHD Graphics 630-1 OpenCL compilation in 6.71 s
2020-09-25 13:26:21 Intel(R) UHD Graphics 630-1 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-09-25 13:26:56 Intel(R) UHD Graphics 630-1 77936867 OK      800   0.00%; 29530 us/it; ETA 26d 15:17; 1579c241dc63eca6 (check 11.88s)
 2020-09-25 15:07:23 Intel(R) UHD Graphics 630-1 77936867 OK   200000   0.26%; 30198 us/it; ETA 27d 04:05; f0b04b45b0855bd2 (check 11.97s)

Here's the Vega iGPU from a 4700u (7nm, 7 GPU cores, supposedly 1600MHz):

Code:

pn50@pn50:~/Documents/git/gpuowlv6$ ./gpuowl -prp 77936867 -maxAlloc 4000
2020-10-13 13:28:29 gpuowl v6.11-380-g79ea0cc-dirty
2020-10-13 13:28:29 Note: not found 'config.txt'
2020-10-13 13:28:29 config: -prp 77936867 -maxAlloc 4000 
2020-10-13 13:28:29 device 0, unique id ''
2020-10-13 13:28:29 gfx900-0 77936867 FFT: 4M 1K:8:256 (18.58 bpw)
2020-10-13 13:28:29 gfx900-0 Expected maximum carry32: 583B0000
2020-10-13 13:28:29 gfx900-0 OpenCL args "-DEXP=77936867u -DWIDTH=1024u -DSMALL_HEIGHT=256u -DMIDDLE=8u -DPM1=0 -DAMDGPU=1 -DMM_CHAIN=1u -DMM2_CHAIN=2u -DMAX_ACCURACY=1 -DWEIGHT_STEP_MINUS_1=0x1.5885a1af9d807p-2 -DIWEIGHT_STEP_MINUS_1=-0x1.01ca19102fbbfp-2  -cl-unsafe-math-optimizations -cl-std=CL2.0 -cl-finite-math-only "
2020-10-13 13:28:32 gfx900-0 OpenCL compilation in 2.41 s
2020-10-13 13:29:24 gfx900-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-10-13 13:29:24 gfx900-0 validating proof residues for power 8
2020-10-13 13:29:24 gfx900-0 Proof using power 8
2020-10-13 13:29:50 gfx900-0 77936867 OK      800   0.00%; 11753 us/it; ETA 10d 14:26; 1579c241dc63eca6 (check 16.44s)
 2020-10-13 14:15:10 gfx900-0 77936867 OK   200000   0.26%; 13572 us/it; ETA 12d 05:04; f0b04b45b0855bd2 (check 16.44s)

moebius · 2020-10-19, 23:57

Code:

2020-10-19 23:45:38 Tesla V100-SXM2-16GB-0 OpenCL compilation in 1.79 s
2020-10-19 23:45:38 Tesla V100-SXM2-16GB-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2020-10-19 23:45:38 Tesla V100-SXM2-16GB-0 validating proof residues for power 8
2020-10-19 23:45:38 Tesla V100-SXM2-16GB-0 Proof using power 8
2020-10-19 23:45:39 Tesla V100-SXM2-16GB-0 77936867 OK      800   0.00%;  445 us/it; ETA 0d 09:39; 1579c241dc63eca6 (check 0.30s)
2020-10-19 23:47:07 Tesla V100-SXM2-16GB-0 77936867 OK   200000   0.26%;  444 us/it; ETA 0d 09:35; f0b04b45b0855bd2 (check 0.30s)
2020-10-19 23:48:37 Tesla V100-SXM2-16GB-0 77936867 OK   400000   0.51%;  444 us/it; ETA 0d 09:34; c03f94396a5aa29e (check 0.30s)
2020-10-19 23:50:05 Tesla V100-SXM2-16GB-0 77936867 OK   600000   0.77%;  443 us/it; ETA 0d 09:31; b9decd65ca71b629 (check 0.30s)

kriesel · 2020-10-20, 15:58

Quote:

Originally Posted by DrobinsonPE

I have an I3-9100 and an I5-8250U that I might experiment on as well if/went I have the time. Did gpuowl work in Windows or Linux?

Xyzzy is a Linux fan.
There's a reference thread for gpuowl and mfakto on integrated graphics. With considerable Wndows experience. https://www.mersenneforum.org/showthread.php?t=25717
Older (MUCH older) versions of gpuowl might work where recent versions don't, perhaps allowing LLDC or PRP. Running PRP tests on igp in V1.9's 8M fft would take probably months each, and lack the PRP proof capability.

Neutron3529 · 2020-10-21, 09:39

I found that my GPU does not reach 350W power limit(~295W, but reach 1965Mhz which is ~200 Mhz higher than it is in mfaktc.)

Code:

[neutron@neutron-GPU gpuowl]$ ./gpuowl -prp 77936867 -maxAlloc 20000 2020-10-21 17:29:12 gpuowl v7.0-64-g85e1f5e-dirty
2020-10-21 17:29:12 Note: not found 'config.txt'
2020-10-21 17:29:12 config: -prp 77936867 -maxAlloc 20000 
2020-10-21 17:29:12 device 0, unique id ''
2020-10-21 17:29:13 GeForce RTX 3090-0 77936867 FFT: 4M 1K:8:256 (18.58 bpw)
2020-10-21 17:29:13 GeForce RTX 3090-0 77936867 OpenCL args "-DEXP=77936867u -DWIDTH=1024u -DSMALL_HEIGHT=256u -DMIDDLE=8u -DCARRY64=1 -DCARRYM64=1 -DMM_CHAIN=1u -DMM2_CHAIN=2u -DMAX_ACCURACY=1 -DWEIGHT_STEP_MINUS_1=0x1.5885a1af9d807p-2 -DIWEIGHT_STEP_MINUS_1=-0x1.01ca19102fbbfp-2  -cl-unsafe-math-optimizations -cl-std=CL2.0 -cl-finite-math-only "
2020-10-21 17:29:13 GeForce RTX 3090-0 77936867 

2020-10-21 17:29:13 GeForce RTX 3090-0 77936867 OpenCL compilation in 0.01 s
2020-10-21 17:29:13 GeForce RTX 3090-0 77936867 maxAlloc: 19.5 GB
2020-10-21 17:29:13 GeForce RTX 3090-0 77936867 P1(0) 0 bits
2020-10-21 17:29:13 GeForce RTX 3090-0 77936867 PRP starting from beginning
2020-10-21 17:29:14 GeForce RTX 3090-0 77936867 OK         0 loaded: blockSize 400, 0000000000000003
2020-10-21 17:29:14 GeForce RTX 3090-0 77936867 validating proof residues for power 8
2020-10-21 17:29:14 GeForce RTX 3090-0 77936867 Proof using power 8
2020-10-21 17:29:16 GeForce RTX 3090-0 77936867 OK       800   0.00% 1579c241dc63eca6 1552 us/it + check 0.68s + save 0.12s; ETA 1d 09:36
2020-10-21 17:29:30 GeForce RTX 3090-0 77936867        10000   0.01% fc4f135f7cf4ad29 1564 us/it
2020-10-21 17:29:46 GeForce RTX 3090-0 77936867        20000   0.03% 3cd1bd9d5e09cbc5 1568 us/it
2020-10-21 17:30:02 GeForce RTX 3090-0 77936867        30000   0.04% c4e0ff35e3290d98 1568 us/it
2020-10-21 17:30:17 GeForce RTX 3090-0 77936867        40000   0.05% dffe1b1b0d748128 1569 us/it
2020-10-21 17:30:33 GeForce RTX 3090-0 77936867        50000   0.06% 52e286945371ed29 1569 us/it
2020-10-21 17:30:49 GeForce RTX 3090-0 77936867        60000   0.08% 0945da4dc08bdd95 1569 us/it
2020-10-21 17:31:04 GeForce RTX 3090-0 77936867        70000   0.09% 7131fa4eb77f4bb2 1569 us/it
2020-10-21 17:31:20 GeForce RTX 3090-0 77936867        80000   0.10% 8d76071d27ee4221 1569 us/it
2020-10-21 17:31:36 GeForce RTX 3090-0 77936867        90000   0.12% 0bacff453b2f470e 1569 us/it
2020-10-21 17:31:51 GeForce RTX 3090-0 77936867       100000   0.13% 6d7296b9e2830f50 1569 us/it
2020-10-21 17:32:07 GeForce RTX 3090-0 77936867       110000   0.14% 8cbfd4435622bda7 1569 us/it
2020-10-21 17:32:23 GeForce RTX 3090-0 77936867       120000   0.15% 79ae5dad855057ad 1569 us/it
2020-10-21 17:32:39 GeForce RTX 3090-0 77936867       130000   0.17% 50c97bcbf876231f 1569 us/it
2020-10-21 17:32:54 GeForce RTX 3090-0 77936867       140000   0.18% e1db15f897271496 1569 us/it
2020-10-21 17:33:10 GeForce RTX 3090-0 77936867       150000   0.19% 127631386c6a9b17 1569 us/it
2020-10-21 17:33:26 GeForce RTX 3090-0 77936867       160000   0.21% 25b7b6206fc6f085 1569 us/it
2020-10-21 17:33:41 GeForce RTX 3090-0 77936867       170000   0.22% 416816b0d9f4bba8 1569 us/it
2020-10-21 17:33:57 GeForce RTX 3090-0 77936867       180000   0.23% 6bee5d054f770861 1569 us/it
2020-10-21 17:34:13 GeForce RTX 3090-0 77936867       190000   0.24% f37f068f014b18a0 1569 us/it
2020-10-21 17:34:29 GeForce RTX 3090-0 77936867 OK    200000   0.26% f0b04b45b0855bd2 1569 us/it + check 0.69s + save 0.12s; ETA 1d 09:52
2020-10-21 17:34:45 GeForce RTX 3090-0 77936867       210000   0.27% 43eb2fc2424d8aac 1569 us/it
2020-10-21 17:35:01 GeForce RTX 3090-0 77936867       220000   0.28% a1081c6dc6a7689f 1585 us/it
2020-10-21 17:35:16 GeForce RTX 3090-0 77936867       230000   0.30% 2387818d3d3d0d01 1573 us/it
2020-10-21 17:35:32 GeForce RTX 3090-0 77936867       240000   0.31% a9deae45055e5216 1572 us/it
2020-10-21 17:35:48 GeForce RTX 3090-0 77936867       250000   0.32% 89fcab15218f7cac 1572 us/it
2020-10-21 17:36:04 GeForce RTX 3090-0 77936867       260000   0.33% 55da428da4cf928a 1576 us/it
2020-10-21 17:36:19 GeForce RTX 3090-0 77936867       270000   0.35% dc349756c5f05abf 1573 us/it
2020-10-21 17:36:35 GeForce RTX 3090-0 77936867       280000   0.36% 3564af24488443f4 1572 us/it
2020-10-21 17:36:51 GeForce RTX 3090-0 77936867       290000   0.37% 63fb281a06f78198 1576 us/it
2020-10-21 17:37:07 GeForce RTX 3090-0 77936867       300000   0.38% 990aa099aad5bf9c 1573 us/it
2020-10-21 17:37:10 GeForce RTX 3090-0 77936867 Stopping, please wait..
2020-10-21 17:37:11 GeForce RTX 3090-0 77936867 OK    302000   0.39% eeb2e0e45213c10c 1572 us/it + check 0.70s + save 0.12s; ETA 1d 09:54
2020-10-21 17:37:11 GeForce RTX 3090-0 Exiting because "stop requested"
2020-10-21 17:37:11 GeForce RTX 3090-0 Bye

preda · 2020-10-21, 10:11

Quote:

Originally Posted by Neutron3529

I found that my GPU does not reach 350W power limit(~295W, but reach 1965Mhz which is ~200 Mhz higher than it is in mfaktc.)

Code:

[neutron@neutron-GPU gpuowl]$ ./gpuowl -prp 77936867 -maxAlloc 20000
1564 us/it

Thanks for the timing info!

In general I would recommend a smaller -maxAlloc, e.g. "-maxAlloc 18G" (just to be sure you don't overfill the GPU RAM which would produce a slowdown), but in this case you ran without P-1 so that doesn't matter.

The performance is not great, but that was expected given the heavy use of FP64 in the current implementation.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
mfakto: an OpenCL program for Mersenne prefactoring	Bdot	GPU Computing	1676	2021-06-30 21:23
GPUOWL AMD Windows OpenCL issues	xx005fs	GpuOwl	0	2019-07-26 21:37
Testing an expression for primality	1260	Software	17	2015-08-28 01:35
Testing Mersenne cofactors for primality?	CRGreathouse	Computer Science & Computational Number Theory	18	2013-06-08 19:12
Primality-testing program with multiple types of moduli (PFGW-related)	Unregistered	Information & Answers	4	2006-10-04 22:38

2020-09-28, 09:36	#2499
M344587487 "Composite as Heck" Oct 2017 2²×3²×23 Posts	What's the power draw of the W5500 and how does a Radeon VII compare throughput-wise with the same exponent? If the RTX 5500 has similar numbers which it should (DP ratio looks the same), if the 4GB model runs as well as the 8GB model, and if the card can be made to work (last I checked that was a big if), then the results are not as terrible as I expected. The price of an RTX 5500 is currently £170, on a throughput-per-dollar basis it looks like it could at least see the R7 ahead in the distance instead of being in division 3 (ignoring power draw and density limitations which of course is a hard sell).