[QUOTE=kracker;556981]...99% sure I don't have any issues there.[/QUOTE]


Are you, by any chance, trying to run with an [U]overclocked[/U] GPU? In my experience, there is little gain, and risk of damage. I don't do it. :ermm:

[QUOTE=storm5510;557068]Are you, by any chance, trying to run with an [U]overclocked[/U] GPU? In my experience, there is little gain, and risk of damage. I don't do it. :ermm:[/QUOTE]

No. I never run computing programs on anything overclocked.

Can someone please point me to the post that describes config.txt
 

Thanks

[URL]https://www.mersenneforum.org/showthread.php?t=23391[/URL][QUOTE=petrw1;557094]Thanks[/QUOTE]Also the help file in the zip file of almost any posted Windows build; rerun gpuowl-win -h on your own hardware to get the correct device number to gpu model output for your hardware.

Error Message
 

[CODE]2020-09-16 02:40:26 gpuowl v6.11-180-g6e722e2-dirty
2020-09-16 02:40:26 config: -user petrw1 -cpu colab -device 1 -maxAlloc 7500
2020-09-16 02:40:26 device 1, unique id ''
2020-09-16 02:40:26 colab 24000577 FFT 1280K: Width 8x8, Height 256x4, Middle 10; 18.31 bits/word
2020-09-16 02:40:26 colab Exception gpu_error: clGetPlatformIDs(16, platforms, (unsigned *) &nPlatforms) at clwrap.cpp:71 getDeviceIDs
2020-09-16 02:40:26 colab Bye[/CODE]

Disregard for now...I forgot to tell the Runtime to use GPU

For real error message this time.
 

[CODE]2020-09-16 04:35:17 gpuowl v6.11-180-g6e722e2-dirty
2020-09-16 04:35:17 config: -user petrw1 -cpu colab -device 0 -maxAlloc 7500
2020-09-16 04:35:17 device 0, unique id ''
2020-09-16 04:35:17 colab 24000577 FFT 1280K: Width 8x8, Height 256x4, Middle 10; 18.31 bits/word
2020-09-16 04:35:18 colab OpenCL args "-DEXP=24000577u -DWIDTH=64u -DSMALL_HEIGHT=1024u -DMIDDLE=10u -DWEIGHT_STEP=0x1.9cb7d592d4171p+0 -DIWEIGHT_STEP=0x1.3d951748cc15ep-1 -DWEIGHT_BIGSTEP=0x1.d5818dcfba487p+0 -DIWEIGHT_BIGSTEP=0x1.172b83c7d517bp-1 -DPM1=1 -cl-fast-relaxed-math -cl-std=CL2.0"
2020-09-16 04:35:18 colab OpenCL compilation error -11 (args -DEXP=24000577u -DWIDTH=64u -DSMALL_HEIGHT=1024u -DMIDDLE=10u -DWEIGHT_STEP=0x1.9cb7d592d4171p+0 -DIWEIGHT_STEP=0x1.3d951748cc15ep-1 -DWEIGHT_BIGSTEP=0x1.d5818dcfba487p+0 -DIWEIGHT_BIGSTEP=0x1.172b83c7d517bp-1 -DPM1=1 -cl-fast-relaxed-math -cl-std=CL2.0 -DNO_ASM=1)
2020-09-16 04:35:18 colab <kernel>:382:16: error: use of undeclared identifier 'a'
return 2 * (a + b);
^
<kernel>:382:20: error: use of undeclared identifier 'b'
return 2 * (a + b);
^

2020-09-16 04:35:18 colab Exception gpu_error: BUILD_PROGRAM_FAILURE clBuildProgram at clwrap.cpp:247 build
2020-09-16 04:35:18 colab Bye[/CODE]

[QUOTE=petrw1;557106][CODE]2020-09-16 04:35:17 gpuowl v6.11-180-g6e722e2-dirty
[/QUOTE]
try to build new with this phyton notebook from ATH [URL="https://mersenneforum.org/showpost.php?p=555507&postcount=33"]https://mersenneforum.org/showpost.php?p=555507&postcount=33[/URL]

or alternatively use the binary [B]gpuowlv6.11-380-g79ea0cc[/B] for ubuntu/colab compiled by me [URL="https://mersenneforum.org/showpost.php?p=556311&postcount=40"]https://mersenneforum.org/showpost.php?p=556311&postcount=40[/URL]

[QUOTE=moebius;557107]
or alternatively use the binary [B]gpuowlv6.11-380-g79ea0cc[/B] for ubuntu/colab compiled by me [URL="https://mersenneforum.org/showpost.php?p=556311&postcount=40"]https://mersenneforum.org/showpost.php?p=556311&postcount=40[/URL][/QUOTE]

./gpuowl.exe: /usr/lib/x86_64-linux-gnu/libstdc++.so.6: version `GLIBCXX_3.4.26' not found (required by ./gpuowl.exe)

[QUOTE=petrw1;557184]./gpuowl.exe: /usr/lib/x86_64-linux-gnu/libstdc++.so.6: version `GLIBCXX_3.4.26' not found (required by ./gpuowl.exe)[/QUOTE]

What does [C]g++ --version[/C] say?

You might have to hardwire your compiler name and version in gpuOwl's makefile.

[QUOTE=moebius;557107]try to build new with this phyton notebook from ATH [URL="https://mersenneforum.org/showpost.php?p=555507&postcount=33"]https://mersenneforum.org/showpost.php?p=555507&postcount=33[/URL]
[/QUOTE]

Trying this....session is running; just not sure what it is doing or if it is actually running a P-1.

I'll keep you posted.

[QUOTE=paulunderwood;557185]What does [C]g++ --version[/C] say?

You might have to hardwire your compiler name and version in gpuOwl's makefile.[/QUOTE]

This???

[CODE]echo Version: `cat version.inc`
Version: "v6.11-380-g79ea0cc"[/CODE]