gpuOwL: an OpenCL program for Mersenne primality testing

[axn]

Originally Posted by BigDukeSix

Are there any recommendations for a bored nVidia RTX3080? TF or PRP?

TF. They're not particularly great at PRP.

[BigDukeSix]

Thank you very much. I will have a lot of reads to start there.

[moebius]

Originally Posted by BigDukeSix

Thank you very much. I will have a lot of reads to start there.

Yes, the card should be better in TF, but it is still one of the useful consumer cards in terms of PRP. Read this post for comparison values https://mersenneforum.org/showpost.p...postcount=2603

[storm5510]

Originally Posted by axn

TF. They're not particularly great at PRP.

Not so great at P-1 either. My 1080 can run them in about the same amount of time as my 2080.

[M344587487]

For no practical reason at all:


intel HD530 from an i7-6700HQ. 2133MT/s dual channel DDR4. PC slow to respond during test despite CPU practically idling.

./gpuowl -prp 77936867 -iters 20000
2020-12-27 17:52:10 GpuOwl VERSION v7.2-21-g28dbf88-dirty
2020-12-27 17:52:10 GpuOwl VERSION v7.2-21-g28dbf88-dirty
2020-12-27 17:52:10 Note: not found 'config.txt'
2020-12-27 17:52:10 config: -prp 77936867 -iters 20000 
2020-12-27 17:52:10 device 0, unique id ''
2020-12-27 17:52:12 Intel(R) Gen9 HD Graphics NEO-0 77936867 FFT: 4M 1K:8:256 (18.58 bpw)
2020-12-27 17:52:12 Intel(R) Gen9 HD Graphics NEO-0 77936867 OpenCL args "-DEXP=77936867u -DWIDTH=1024u -DSMALL_HEIGHT=256u -DMIDDLE=8u -DMM_CHAIN=1u -DMM2_CHAIN=2u -DMAX_ACCURACY=1 -DWEIGHT_STEP_MINUS_1=0.33644726404543274 -DIWEIGHT_STEP_MINUS_1=-0.25174750481886216 -DIWEIGHTS={0,-0.25174750481886216,-0.44011820345520131,-0.16213409745771243,-0.37306474779553728,-0.061788266441989627,-0.29798072935699788,-0.47471232907613115,-0.21390437908665341,-0.41180199020062258,-0.11975874301407295,-0.3413572830988989,-0.014337887291734644,-0.26247586476052853,-0.44814572555075455,-0.17414732433395128,}  -cl-std=CL2.0 -cl-finite-math-only "
2020-12-27 17:52:20 Intel(R) Gen9 HD Graphics NEO-0 77936867 OpenCL compilation in 7.82 s
2020-12-27 17:52:20 Intel(R) Gen9 HD Graphics NEO-0 77936867 maxAlloc: 0.0 GB
2020-12-27 17:52:20 Intel(R) Gen9 HD Graphics NEO-0 77936867 You should use -maxAlloc if your GPU has more than 4GB memory. See help '-h'
2020-12-27 17:52:20 Intel(R) Gen9 HD Graphics NEO-0 77936867 P1(0) 0 bits
2020-12-27 17:52:20 Intel(R) Gen9 HD Graphics NEO-0 77936867 PRP starting from beginning
2020-12-27 17:52:36 Intel(R) Gen9 HD Graphics NEO-0 77936867 OK         0 on-load: blockSize 400, 0000000000000003
2020-12-27 17:52:36 Intel(R) Gen9 HD Graphics NEO-0 77936867 validating proof residues for power 8
2020-12-27 17:52:36 Intel(R) Gen9 HD Graphics NEO-0 77936867 Proof using power 8
2020-12-27 17:53:19 Intel(R) Gen9 HD Graphics NEO-0 77936867 OK       800   0.00% 1579c241dc63eca6 35951 us/it + check 14.46s + save 0.19s; ETA 32d 10:18
2020-12-27 17:58:50 Intel(R) Gen9 HD Graphics NEO-0 77936867        10000   0.01% fc4f135f7cf4ad29 35923 us/it
2020-12-27 18:04:49 Intel(R) Gen9 HD Graphics NEO-0 77936867 Stopping, please wait..
2020-12-27 18:05:04 Intel(R) Gen9 HD Graphics NEO-0 77936867 OK     20000   0.03% 3cd1bd9d5e09cbc5 35930 us/it + check 14.46s + save 0.16s; ETA 32d 09:39
2020-12-27 18:05:04 Intel(R) Gen9 HD Graphics NEO-0 Exiting because "stop requested"
2020-12-27 18:05:04 Intel(R) Gen9 HD Graphics NEO-0 Bye

sudo dmidecode --type 17
# dmidecode 3.2
Getting SMBIOS data from sysfs.
SMBIOS 3.0.0 present.

Handle 0x0044, DMI type 17, 40 bytes
Memory Device
    Array Handle: 0x0043
    Error Information Handle: Not Provided
    Total Width: 64 bits
    Data Width: 64 bits
    Size: 8192 MB
    Form Factor: SODIMM
    Set: None
    Locator: ChannelA-DIMM0
    Bank Locator: BANK 0
    Type: DDR4
    Type Detail: Synchronous
    Speed: 2133 MT/s
    Manufacturer: SK Hynix
    Serial Number: 65436018
    Asset Tag: 9876543210
    Part Number: HMA41GS6AFR8N-TF    
    Rank: 2
    Configured Memory Speed: 2133 MT/s
    Minimum Voltage: Unknown
    Maximum Voltage: Unknown
    Configured Voltage: 1.2 V

Handle 0x0045, DMI type 17, 40 bytes
Memory Device
    Array Handle: 0x0043
    Error Information Handle: Not Provided
    Total Width: 64 bits
    Data Width: 64 bits
    Size: 8192 MB
    Form Factor: SODIMM
    Set: None
    Locator: ChannelB-DIMM0
    Bank Locator: BANK 2
    Type: DDR4
    Type Detail: Synchronous
    Speed: 2133 MT/s
    Manufacturer: SK Hynix
    Serial Number: 65435418
    Asset Tag: 9876543210
    Part Number: HMA41GS6AFR8N-TF    
    Rank: 2
    Configured Memory Speed: 2133 MT/s
    Minimum Voltage: Unknown
    Maximum Voltage: Unknown
    Configured Voltage: 1.2 V

[moebius]

Originally Posted by Xyzzy

How many "R7 units" will an MI100 replace?

Does anyone here work for the European Union and get access to Lumi with AMD Instinct gpu's at the end of 2021? The peak performance of LUMI is an astonishing 552 petaflop/s. The global bandwidth of the LUMI-GPU partition is 160 TB/s.
Then I could answer the question exactly.
https://www.lumi-supercomputer.eu/lu...upercomputers/

[Xyzzy]

RX 6800

2021-01-07 13:33:05 gfx1030-0 OpenCL compilation in 2.54 s
2021-01-07 13:33:05 gfx1030-0 77936867 OK        0 loaded: blockSize 400, 0000000000000003
2021-01-07 13:33:06 gfx1030-0 77936867 OK      800   0.00%;  871 us/it; ETA 0d 18:51; 1579c241dc63eca6 (check 0.39s)
2021-01-07 13:36:00 gfx1030-0 77936867 OK   200000   0.26%;  870 us/it; ETA 0d 18:47; f0b04b45b0855bd2 (check 0.39s)
2021-01-07 13:38:55 gfx1030-0 77936867 OK   400000   0.51%;  873 us/it; ETA 0d 18:48; c03f94396a5aa29e (check 0.39s)
2021-01-07 13:41:49 gfx1030-0 77936867 OK   600000   0.77%;  869 us/it; ETA 0d 18:40; b9decd65ca71b629 (check 0.38s)
2021-01-07 13:44:43 gfx1030-0 77936867 OK   800000   1.03%;  866 us/it; ETA 0d 18:33; 21ebf3636148f663 (check 0.41s)
2021-01-07 13:47:35 gfx1030-0 77936867 OK  1000000   1.28%;  862 us/it; ETA 0d 18:25; 9bf9d9e6bff4286e (check 0.39s)
2021-01-07 13:50:29 gfx1030-0 77936867 OK  1200000   1.54%;  868 us/it; ETA 0d 18:30; da42731132f00140 (check 0.38s)
2021-01-07 13:53:23 gfx1030-0 77936867 OK  1400000   1.80%;  864 us/it; ETA 0d 18:22; 56d6868b11382d6c (check 0.38s)

[moebius]

Originally Posted by Xyzzy

RX 6800

Thx for the benchmark, your fan speed is at 49%, shouldn't it run a little bit faster with more rpm.

[Xyzzy]

That is the default fan curve. They probably tuned it to be quiet. The temps are reasonable so we'll leave it be.

[M344587487]

As long as it doesn't thermal throttle it shouldn't matter. Technically lower temps does mean slightly lower power consumption all other things equal, but increasing power to the fans to lower the temps is robbing Peter to pay Paul.

[moebius]

Originally Posted by M344587487

As long as it doesn't thermal throttle it shouldn't matter. Technically lower temps does mean slightly lower power consumption all other things equal, but increasing power to the fans to lower the temps is robbing Peter to pay Paul.

I actually only meant that, for example with GPU Tweak II, the temperatures could be briefly reduced manually a little and the memory clock could be increased by e.g. 150 MHz for for the benchmark . But I think the card won't reach values below 800 µs/it.