mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Hardware > GPU Computing > GpuOwl

Reply
 
Thread Tools
Old 2020-10-23, 09:10   #2531
Viliam Furik
 
Jul 2018
Martin, Slovakia

1000000012 Posts
Default

Quote:
Originally Posted by aheeffer View Post
This is weird! Just finished a PRP test for

108980089

and the result was refused, though it was assigned to me through Primenet. As I have seen the name of the PRP tester mentioned before, did this proof certification succeed?
It looks like it wasn't assigned to you because there is no visible assignment history or expiration date. Either that or you somehow dropped the assignment while testing.

It is marked as a suspect result and based on the second CERT run from George, it did not succeed at all.
Viliam Furik is online now   Reply With Quote
Old 2020-10-23, 11:25   #2532
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

7,159 Posts
Default

Quote:
Originally Posted by aheeffer View Post
This is weird! Just finished a PRP test for

108980089

and the result was refused, though it was assigned to me through Primenet. As I have seen the name of the PRP tester mentioned before, did this proof certification succeed?
please email your results.json.txt file. I may need the proof file too.
Prime95 is online now   Reply With Quote
Old 2020-10-23, 11:44   #2533
moebius
 
moebius's Avatar
 
Jul 2009
Germany

2·229 Posts
Default

A first draft of the 77936867 top list for OpenOffice Calc. Does not contain macros. Presorted according to (µs / it). Unfortunately, I only have 15 values for now. So I'm happy about every halfway useful gpuowl-benchmark. Just post in this thread or via PM to me.
Attached Files
File Type: zip Toplist.zip (17.7 KB, 26 views)
moebius is online now   Reply With Quote
Old 2020-10-24, 02:18   #2534
moebius
 
moebius's Avatar
 
Jul 2009
Germany

2·229 Posts
Default

Quote:
Originally Posted by Neutron3529 View Post
Code:
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
With -carry short I was able to increase the iteration times of my RX Vega 64 to 1161 us / it. With the Nvidia P100-16GB I was able to achieve 686 us / it with -carry short -use CARRY32, ORIG_SLOWTRIG, IN_WG = 128, IN_SIZEX = 16, IN_SPACING = 4, OUT_WG = 128, OUT_SIZEX = 16, OUT_SPACING = 4 -nospin -block 100 -maxAlloc 10000 -B1 750 000 -rB2 20.
Maybe something is still possible with the RTX 3090?
moebius is online now   Reply With Quote
Old 2020-10-24, 08:07   #2535
M344587487
 
M344587487's Avatar
 
"Composite as Heck"
Oct 2017

2×349 Posts
Default

Quote:
Originally Posted by moebius View Post
The AMD RX Vega 64 remains the second winner in the consumer card category.

Benchmarks of 77936867 for the following graphics cards would be interesting.
AMD Radeon RX 6900/6800/6700 XT when available
Nvidia GeForce RTX 3080/3070, 3060 TI, 2080 Ti
Nvidia Titan RTX, Titan V
Nvidia A100 SXM4
intel's Xe iGPU in the 1165G7 would also be interesting, it had a mediocre showing on launch but mere weeks later patches have come in for it to pull way ahead in some workloads: https://www.phoronix.com/scan.php?pa...graphics&num=1

That said maybe not, 1.77 TFLOPS of SP in an iGPU is nice but DP may be omitted altogether ( https://www.sisoftware.co.uk/tag/1165g7/ ). It's promising for intel's dGPU line at least that their iGPU is so performant in SP, but how the power draw compares to AMD's Vega iGPUs and how performance/watt compares to the upcoming RDNA2 iGPUs is unknown. I wouldn't mind intel's consumer Xe (there are many Xe variants in the works apparently) omitting DP entirely if it means they can produce an SP monster.
M344587487 is online now   Reply With Quote
Old 2020-10-24, 15:33   #2536
moebius
 
moebius's Avatar
 
Jul 2009
Germany

2·229 Posts
Default

Quote:
Originally Posted by M344587487 View Post
intel's Xe iGPU in the 1165G7 would also be interesting..
Sure, but since it won't be able to keep up with most of the current dGPU's, I thought of a separate benchmark list.
This card f.e. have 4,113 TLOPS FP32 (SP) and 0,137 TLOPS FP64 (DP)
Code:
2020-10-24 15:45:24 Tesla K80-0 OpenCL compilation in 2.80 s
2020-10-24 15:45:25 Tesla K80-0 77936867 OK        0 loaded: blockSize 100, 0000000000000003
2020-10-24 15:45:25 Tesla K80-0 validating proof residues for power 8
2020-10-24 15:45:25 Tesla K80-0 Proof using power 8
2020-10-24 15:45:26 Tesla K80-0 77936867 OK      200   0.00%; 2778 us/it; ETA 2d 12:09; 2619e0f0cb78fe50 (check 0.40s)
2020-10-24 15:55:02 Tesla K80-0 77936867 OK   200000   0.26%; 2883 us/it; ETA 2d 14:15; f0b04b45b0855bd2 (check 0.42s)
2020-10-24 16:04:39 Tesla K80-0 77936867 OK   400000   0.51%; 2884 us/it; ETA 2d 14:07; c03f94396a5aa29e (check 0.41s)
2020-10-24 16:14:16 Tesla K80-0 77936867 OK   600000   0.77%; 2883 us/it; ETA 2d 13:56; b9decd65ca71b629 (check 0.42s)

Last fiddled with by moebius on 2020-10-24 at 16:16
moebius is online now   Reply With Quote
Old 2020-10-26, 16:41   #2537
De Wandelaar
 
De Wandelaar's Avatar
 
"Yves"
Jul 2017
Belgium

3·17 Posts
Default

See below results for a P4000 Mobile version v6.11-380-g79ea0cc

Quote:
020-10-26 16:37:37 Quadro P4000-0 OpenCL compilation in 0.03 s
2020-10-26 16:37:38 Quadro P4000-0 77936867 OK 0 loaded: blockSize 100, 0000000000000003
2020-10-26 16:37:38 Quadro P4000-0 validating proof residues for power 8
2020-10-26 16:37:38 Quadro P4000-0 Proof using power 8
2020-10-26 16:37:41 Quadro P4000-0 77936867 OK 200 0.00%; 7759 us/it; ETA 6d 23:59; 2619e0f0cb78fe50 (check 0.94s)
2020-10-26 16:38:58 Quadro P4000-0 77936867 OK 10000 0.01%; 7753 us/it; ETA 6d 23:49; fc4f135f7cf4ad29 (check 0.94s)
2020-10-26 16:40:16 Quadro P4000-0 77936867 OK 20000 0.03%; 7756 us/it; ETA 6d 23:52; 3cd1bd9d5e09cbc5 (check 0.95s)
2020-10-26 16:41:35 Quadro P4000-0 77936867 OK 30000 0.04%; 7755 us/it; ETA 6d 23:50; c4e0ff35e3290d98 (check 0.94s)
2020-10-26 16:42:53 Quadro P4000-0 77936867 OK 40000 0.05%; 7755 us/it; ETA 6d 23:48; dffe1b1b0d748128 (check 0.94s)
2020-10-26 16:44:12 Quadro P4000-0 77936867 OK 50000 0.06%; 7753 us/it; ETA 6d 23:44; 52e286945371ed29 (check 0.94s)
2020-10-26 16:45:30 Quadro P4000-0 77936867 OK 60000 0.08%; 7753 us/it; ETA 6d 23:43; 0945da4dc08bdd95 (check 0.94s)
2020-10-26 16:46:49 Quadro P4000-0 77936867 OK 70000 0.09%; 7756 us/it; ETA 6d 23:46; 7131fa4eb77f4bb2 (check 0.99s)
2020-10-26 16:48:07 Quadro P4000-0 77936867 OK 80000 0.10%; 7756 us/it; ETA 6d 23:44; 8d76071d27ee4221 (check 0.94s)
2020-10-26 16:49:26 Quadro P4000-0 77936867 OK 90000 0.12%; 7754 us/it; ETA 6d 23:41; 0bacff453b2f470e (check 0.94s)
2020-10-26 16:50:44 Quadro P4000-0 77936867 OK 100000 0.13%; 7756 us/it; ETA 6d 23:42; 6d7296b9e2830f50 (check 0.94s)
2020-10-26 16:52:03 Quadro P4000-0 77936867 OK 110000 0.14%; 7754 us/it; ETA 6d 23:38; 8cbfd4435622bda7 (check 0.95s)
2020-10-26 16:53:21 Quadro P4000-0 77936867 OK 120000 0.15%; 7755 us/it; ETA 6d 23:38; 79ae5dad855057ad (check 0.94s)
2020-10-26 16:54:40 Quadro P4000-0 77936867 OK 130000 0.17%; 7755 us/it; ETA 6d 23:37; 50c97bcbf876231f (check 0.94s)
2020-10-26 16:55:58 Quadro P4000-0 77936867 OK 140000 0.18%; 7754 us/it; ETA 6d 23:34; e1db15f897271496 (check 0.94s)
2020-10-26 16:57:17 Quadro P4000-0 77936867 OK 150000 0.19%; 7753 us/it; ETA 6d 23:31; 127631386c6a9b17 (check 0.94s)
2020-10-26 16:58:35 Quadro P4000-0 77936867 OK 160000 0.21%; 7754 us/it; ETA 6d 23:31; 25b7b6206fc6f085 (check 0.95s)
2020-10-26 16:59:54 Quadro P4000-0 77936867 OK 170000 0.22%; 7752 us/it; ETA 6d 23:28; 416816b0d9f4bba8 (check 0.94s)
2020-10-26 17:01:12 Quadro P4000-0 77936867 OK 180000 0.23%; 7754 us/it; ETA 6d 23:29; 6bee5d054f770861 (check 0.94s)
2020-10-26 17:02:31 Quadro P4000-0 77936867 OK 190000 0.24%; 7754 us/it; ETA 6d 23:28; f37f068f014b18a0 (check 0.94s)
2020-10-26 17:03:47 Quadro P4000-0 Stopping, please wait..
2020-10-26 17:03:49 Quadro P4000-0 77936867 OK 200000 0.26%; 7754 us/it; ETA 6d 23:27; f0b04b45b0855bd2 (check 0.94s)
De Wandelaar is online now   Reply With Quote
Old 2020-10-26, 16:43   #2538
De Wandelaar
 
De Wandelaar's Avatar
 
"Yves"
Jul 2017
Belgium

1100112 Posts
Default

And results for a GTX 1080 Mobile (@ 1695 Mhz) :

Quote:
2020-10-26 16:40:14 GeForce GTX 1080-0 OpenCL compilation in 0.04 s
2020-10-26 16:40:15 GeForce GTX 1080-0 77936867 OK 0 loaded: blockSize 100, 0000000000000003
2020-10-26 16:40:15 GeForce GTX 1080-0 validating proof residues for power 8
2020-10-26 16:40:15 GeForce GTX 1080-0 Proof using power 8
2020-10-26 16:40:16 GeForce GTX 1080-0 77936867 OK 200 0.00%; 4433 us/it; ETA 3d 23:59; 2619e0f0cb78fe50 (check 0.63s)
2020-10-26 16:40:59 GeForce GTX 1080-0 77936867 OK 10000 0.01%; 4331 us/it; ETA 3d 21:45; fc4f135f7cf4ad29 (check 0.62s)
2020-10-26 16:41:43 GeForce GTX 1080-0 77936867 OK 20000 0.03%; 4307 us/it; ETA 3d 21:13; 3cd1bd9d5e09cbc5 (check 0.63s)
2020-10-26 16:42:27 GeForce GTX 1080-0 77936867 OK 30000 0.04%; 4307 us/it; ETA 3d 21:12; c4e0ff35e3290d98 (check 0.62s)
2020-10-26 16:43:10 GeForce GTX 1080-0 77936867 OK 40000 0.05%; 4302 us/it; ETA 3d 21:06; dffe1b1b0d748128 (check 0.63s)
2020-10-26 16:43:54 GeForce GTX 1080-0 77936867 OK 50000 0.06%; 4306 us/it; ETA 3d 21:10; 52e286945371ed29 (check 0.63s)
2020-10-26 16:44:38 GeForce GTX 1080-0 77936867 OK 60000 0.08%; 4306 us/it; ETA 3d 21:09; 0945da4dc08bdd95 (check 0.62s)
2020-10-26 16:45:22 GeForce GTX 1080-0 77936867 OK 70000 0.09%; 4308 us/it; ETA 3d 21:11; 7131fa4eb77f4bb2 (check 0.63s)
2020-10-26 16:46:06 GeForce GTX 1080-0 77936867 OK 80000 0.10%; 4366 us/it; ETA 3d 22:26; 8d76071d27ee4221 (check 0.63s)
2020-10-26 16:46:50 GeForce GTX 1080-0 77936867 OK 90000 0.12%; 4303 us/it; ETA 3d 21:03; 0bacff453b2f470e (check 0.62s)
2020-10-26 16:47:33 GeForce GTX 1080-0 77936867 OK 100000 0.13%; 4303 us/it; ETA 3d 21:02; 6d7296b9e2830f50 (check 0.62s)
2020-10-26 16:48:17 GeForce GTX 1080-0 77936867 OK 110000 0.14%; 4305 us/it; ETA 3d 21:04; 8cbfd4435622bda7 (check 0.63s)
2020-10-26 16:49:01 GeForce GTX 1080-0 77936867 OK 120000 0.15%; 4305 us/it; ETA 3d 21:03; 79ae5dad855057ad (check 0.61s)
2020-10-26 16:49:44 GeForce GTX 1080-0 77936867 OK 130000 0.17%; 4303 us/it; ETA 3d 21:01; 50c97bcbf876231f (check 0.63s)
2020-10-26 16:50:28 GeForce GTX 1080-0 77936867 OK 140000 0.18%; 4317 us/it; ETA 3d 21:17; e1db15f897271496 (check 0.61s)
2020-10-26 16:51:11 GeForce GTX 1080-0 77936867 OK 150000 0.19%; 4222 us/it; ETA 3d 19:14; 127631386c6a9b17 (check 0.62s)
2020-10-26 16:51:54 GeForce GTX 1080-0 77936867 OK 160000 0.21%; 4221 us/it; ETA 3d 19:11; 25b7b6206fc6f085 (check 0.62s)
2020-10-26 16:52:36 GeForce GTX 1080-0 77936867 OK 170000 0.22%; 4220 us/it; ETA 3d 19:09; 416816b0d9f4bba8 (check 0.62s)
2020-10-26 16:53:21 GeForce GTX 1080-0 77936867 OK 180000 0.23%; 4346 us/it; ETA 3d 21:53; 6bee5d054f770861 (check 0.62s)
2020-10-26 16:54:04 GeForce GTX 1080-0 77936867 OK 190000 0.24%; 4305 us/it; ETA 3d 20:58; f37f068f014b18a0 (check 0.62s)
2020-10-26 16:54:47 GeForce GTX 1080-0 Stopping, please wait..
2020-10-26 16:54:48 GeForce GTX 1080-0 77936867 OK 200000 0.26%; 4351 us/it; ETA 3d 21:57; f0b04b45b0855bd2 (check 0.63s)
De Wandelaar is online now   Reply With Quote
Old 2020-10-26, 20:53   #2539
moebius
 
moebius's Avatar
 
Jul 2009
Germany

45810 Posts
Default

Quote:
Originally Posted by De Wandelaar View Post
And results for a GTX 1080 Mobile (@ 1695 Mhz) :
Product Name GPU Chip Released Bus Memory GPU clock Memory clock Shaders / TMUs / ROPs
GeForce GTX 1080 Max-Q GP104 Jun 27th, 2017 PCIe 3.0 x16 8 GB, GDDR5X, 256 bit 1290 MHz 1251 MHz 2560 / 160 / 64
GeForce GTX 1080 Max-Q GP104B Jun 27th, 2017 PCIe 3.0 x16 8 GB, GDDR5X, 256 bit 1290 MHz 1251 MHz 2560 / 160 / 64
GeForce GTX 1080 Mobile GP104B Aug 15th, 2016 PCIe 3.0 x16 8 GB, GDDR5X, 256 bit 1582 MHz 1251 MHz 2560 / 160 / 64
GeForce GTX 1080 Mobile GP104 Aug 15th, 2016 PCIe 3.0 x16 8 GB, GDDR5X, 256 bit 1556 MHz 1251 MHz 2560 / 160 / 64

Please enter the exact chipset and product name.
moebius is online now   Reply With Quote
Old 2020-10-27, 07:00   #2540
De Wandelaar
 
De Wandelaar's Avatar
 
"Yves"
Jul 2017
Belgium

3×17 Posts
Default

Quote:
Originally Posted by moebius View Post
Product Name GPU Chip Released Bus Memory GPU clock Memory clock Shaders / TMUs / ROPs
GeForce GTX 1080 Max-Q GP104 Jun 27th, 2017 PCIe 3.0 x16 8 GB, GDDR5X, 256 bit 1290 MHz 1251 MHz 2560 / 160 / 64
GeForce GTX 1080 Max-Q GP104B Jun 27th, 2017 PCIe 3.0 x16 8 GB, GDDR5X, 256 bit 1290 MHz 1251 MHz 2560 / 160 / 64
GeForce GTX 1080 Mobile GP104B Aug 15th, 2016 PCIe 3.0 x16 8 GB, GDDR5X, 256 bit 1582 MHz 1251 MHz 2560 / 160 / 64
GeForce GTX 1080 Mobile GP104 Aug 15th, 2016 PCIe 3.0 x16 8 GB, GDDR5X, 256 bit 1556 MHz 1251 MHz 2560 / 160 / 64

Please enter the exact chipset and product name.
See GPU-Z screen copy in attach.
Attached Thumbnails
Click image for larger version

Name:	Capture d’écran 2020-10-27 075541.jpg
Views:	21
Size:	83.4 KB
ID:	23632  
De Wandelaar is online now   Reply With Quote
Old 2020-10-27, 07:09   #2541
De Wandelaar
 
De Wandelaar's Avatar
 
"Yves"
Jul 2017
Belgium

3·17 Posts
Default

And the GPU-Z attach for the P4000 :
Attached Thumbnails
Click image for larger version

Name:	Capture d’écran 2020-10-27 080733.jpg
Views:	20
Size:	145.5 KB
ID:	23634  
De Wandelaar is online now   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
mfakto: an OpenCL program for Mersenne prefactoring Bdot GPU Computing 1657 2020-10-27 01:23
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

All times are UTC. The time now is 19:30.

Sun Nov 29 19:30:48 UTC 2020 up 80 days, 16:41, 4 users, load averages: 1.03, 1.31, 1.32

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.