No, you have CUDA 6.5 where 8.0 is minimum, you need to update your CUDA driver.

Why is the plan class cuda23 if cuda80 is needed? Just call it cuda65 or cuda80 to match the requirement.

So just AVX? You said AVX256 and AVX and AVX2 added some 256 bit commands so it wasn't really specific enough.

Cuda 2.3 is "Compute capabilty" where Cuda 8.0 is driver version used in Gromacs compilation. Just CUDA capabilities are not same as CUDA runtime version...

For this app CUDA 2.3 version is need with CUDA 8.0 version driver.

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home project team and Universe@Home admin.

No, you have CUDA 6.5 where 8.0 is minimum, you need to update your CUDA driver.

Why is the plan class cuda23 if cuda80 is needed? Just call it cuda65 or cuda80 to match the requirement.

So just AVX? You said AVX256 and AVX and AVX2 added some 256 bit commands so it wasn't really specific enough.

Cuda 2.3 is "Compute capabilty" where Cuda 8.0 is driver version used in Gromacs compilation. Just CUDA capabilities are not same as CUDA runtime version...

For this app CUDA 2.3 version is need with CUDA 8.0 version driver.

Other projects put in the 6.5, 8.0, 9.0 number in the plan class. SETI and GPUGrid do. Thought it was that number.

Other projects put in the 6.5, 8.0, 9.0 number in the plan class. SETI and GPUGrid do. Thought it was that number.

Yes, that is the way I see it on the other projects. I think more people would understand it if they said "CUDA 8.0" in the name. I don't think the "compute capability" has much meaning to most people, unless they look it up for some reason.

Progress is lost if the task becomes paused

Progress is lost if the task becomes paused

No, isn't, just at the moment checkpoints are not reported to Manager.

I will update application for proper progress bar and checkpoints reporting later today.

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home project team and Universe@Home admin.

Progress is lost if the task becomes paused

No, isn't, just at the moment checkpoints are not reported to Manager.

I will update application for proper progress bar and checkpoints reporting later today.

Progress bar is the thing that seems to work. At least I've seen it progress.

Points were cut by a 1/10?

I'm having extreme difficulty getting my hands on a Gromacs application for my GPUs. Are maxwell series cards supported? If not my 980ti's would dramatically speed things up if you can include them.

In version 0.26 of gmx (for Windows) checkpoint and fraction done values are exported to wrapper.
Checkpoints are created from first version just wrapper doesn't know about it.

Checkpoints are done every 15 minutes. Also progress is reported every 15 minutes.

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home project team and Universe@Home admin.

I'm having extreme difficulty getting my hands on a Gromacs application for my GPUs. Are maxwell series cards supported? If not my 980ti's would dramatically speed things up if you can include them.

Downloading it is easy:
http://manual.gromacs.org/documentation/#latest-releases

The problem is getting it installed. I get an error message that home directory does not contain "CMakeLists.txt" when I follow the instructions provided on this forum.
I think someone needs to do some work to get these applications ready for prime time. I have a GTX 980 that I am hoping to use too; it should work OK.

Are you exactly do what is in my instruction?
This error says that probably you miss step where you need to go inside freshly created build directory (before do cmake command).

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home project team and Universe@Home admin.

Are you exactly do what is in my instruction?
This error says that probably you miss step where you need to go inside freshly created build directory (before do cmake command).

In my home directory, I created a folder called "GROMACS", and extracted the "gromacs-2016.4" file into that folder.
Then (from inside the GROMACS folder) I ran the command "cmake .. -DGMX_BUILD_OWN_FFTW=ON -DGMX_GPU=on -DREGRESSIONTEST_DOWNLOAD=ON -DGMX_BUILD_SHARED_EXE=OFF" and got that error message.

It appears to me that I should be inserting the name of the gromacs file somewhere in the command, but where?

EDIT: I think I have found the problem. I was not extracting the "gromacs-2016.4" file properly in Ubuntu. So I did it using 7zip on my Wn7 machine, and am copying it to the Ubuntu machine. If that does not work, I will post again, with whatever error I may get, but this should get me past that problem. Thanks.

EDIT 2: Also, even though I was executing the command from inside the GROMACS folder, I still had to include that folder path in the command, as follows:
"cmake ../GROMACS -DGMX_BUILD_OWN_FFTW=ON -DGMX_GPU=on -DREGRESSIONTEST_DOWNLOAD=ON -DGMX_BUILD_SHARED_EXE=OFF"

It seems to be compiling OK now, though I have not run it yet.

1. Unpack Gromacs2016.4
2. Inside unpacket directory create "build" folder.
3. Go inside folder created above.
4. Execute "cmake .. -DGMX_BUILD_OWN_FFTW=ON -DGMX_GPU=on -DREGRESSIONTEST_DOWNLOAD=ON -DGMX_BUILD_SHARED_EXE=OFF".
6. Execute "make" command.

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home project team and Universe@Home admin.

See my comment #2 below. I had to include the path to the GROMACS folder in the command, then it worked.

Not inside Gromacs folder... Inside "Gromacs2016.4/build" folder...

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home project team and Universe@Home admin.

Not inside Gromacs folder... Inside "Gromacs2016.4/build" folder...

I was in the GROMACS folder which contained all the extracted files when I opened the terminal. I don't see a "build" folder inside of it. So that is the problem. I should have created a "build" folder.

At any rate, it has found the files and is compiling OK now.

Well still no luck. Everything seemed to compile and install correctly, at least I did not see any obvious error messages.
But the first two work units failed, with a similar error message as with the earlier version of gromacs (or even before gromacs was installed?):

<core_client_version>7.8.3</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
15:41:39 (2486): wrapper (7.7.26016): starting (812 +dev#, cuda_fermi)
Device number read from command line: -1
15:41:40 (2486): wrapper (7.7.26016): starting (812 +dev#, cuda_fermi)
15:41:40 (2486): wrapper: running gmx (mdrun -nt 1 -deffnm md -gpu_id 0  0 )
                      :-) GROMACS - gmx mdrun, 2016.4 (-:

                            GROMACS is written by:
     Emile Apol      Rossen Apostolov  Herman J.C. Berendsen    Par Bjelkmar   
 Aldert van Buuren   Rudi van Drunen     Anton Feenstra    Gerrit Groenhof  
 Christoph Junghans   Anca Hamuraru    Vincent Hindriksen Dimitrios Karkoulis
    Peter Kasson        Jiri Kraus      Carsten Kutzner      Per Larsson    
  Justin A. Lemkul   Magnus Lundborg   Pieter Meulenhoff    Erik Marklund   
   Teemu Murtola       Szilard Pall       Sander Pronk      Roland Schulz   
  Alexey Shvetsov     Michael Shirts     Alfons Sijbers     Peter Tieleman  
  Teemu Virolainen  Christian Wennberg    Maarten Wolf   
                           and the project leaders:
        Mark Abraham, Berk Hess, Erik Lindahl, and David van der Spoel

Copyright (c) 1991-2000, University of Groningen, The Netherlands.
Copyright (c) 2001-2017, The GROMACS development team at
Uppsala University, Stockholm University and
the Royal Institute of Technology, Sweden.
check out http://www.gromacs.org for more information.

GROMACS is free software; you can redistribute it and/or modify it
under the terms of the GNU Lesser General Public License
as published by the Free Software Foundation; either version 2.1
of the License, or (at your option) any later version.

GROMACS:      gmx mdrun, version 2016.4
Executable:   gmx

-------------------------------------------------------
Program:     gmx mdrun, version 2016.4
Source file: src/gromacs/utility/path.cpp (line 241)
Function:    static bool gmx::Path::isEquivalent(const string&, const string&)

System I/O error:
Path::isEquivalent called with two invalid files
  Reason: No such file or directory
  (call to stat() returned error code 2)

For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------
15:41:41 (2486): gmx exited; CPU time 0.004000
15:41:41 (2486): app exit status: 0x1
15:41:41 (2486): called boinc_finish(195)

</stderr_txt>
]]>

You just get failed batch, it's should fix in few minutes, when you update project (I'm working now on progress bar and checkpoint reporting).

---

Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home project team and Universe@Home admin.

I have just disabled the use of GPU in DDH prefs as getting fed up of errors. Please advise when or if we can enable GPUs ever again for DDH WUs. Thanks.

I'm having extreme difficulty getting my hands on a Gromacs application for my GPUs. Are maxwell series cards supported? If not my 980ti's would dramatically speed things up if you can include them.

Read this message: Message 5045

The guy has 4 980's crunching just fine, he's probably using Linux though, I didn't check but I'm not sure you need to install 'gromacs' in Windows. I didn't install it on my Windows pc's and it's crunching for me anyway,

Now I may use the app-config.xml file the same guy in the above message talks about to see if it makes mine faster, some of my workunits are taking MUCH longer.

Gromacs 0.26 for GPU didn't use checkpoint after paused (WIN10)!

After resuming a task it starts from begining, not from a checkpoint! I checked md.log and checkpoint.txt.
Sample checkpoint.txt in slot 10:
Writing checkpoint, step 232600 at Wed Oct 25 13:04:07 2017
After resuming at 16:30, now at 17:23:
Writing checkpoint, step 198320 at Wed Oct 25 17:23:49 2017

md.log 13:04 when paused:
Step Time Lambda
230000 460.00000 0.00000
...
Writing checkpoint, step 232600 at Wed Oct 25 13:04:07 2017

md.log 17:23 after one hours runtime:
Step Time Lambda
195000 390.00000 0.00000
...
Writing checkpoint, step 198320 at Wed Oct 25 17:23:49 2017

Boinc-Logfile at 13:04:
25.10.2017 13:04:10 | DrugDiscovery@Home | [task] result gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 checkpointed
25.10.2017 13:04:11 | DrugDiscovery@Home | [task] result gmx_GMX-T025_XXXX_TTTTT_tasks.gmx_393_1000.in_3 checkpointed
25.10.2017 13:04:16 | DrugDiscovery@Home | [task] result gmx_GMX-T030_XXXX_TTTTT_tasks.gmx_79_1000.in_3 checkpointed
25.10.2017 13:04:18 | DrugDiscovery@Home | [task] result gmx_GMX-T030_XXXX_TTTTT_tasks.gmx_27_1000.in_3 checkpointed
25.10.2017 13:08:17 | DrugDiscovery@Home | [task] result gmx_GMX-T034_XXXX_TTTTT_tasks.gmx_662_1000.in_2 checkpointed
25.10.2017 13:10:19 | DrugDiscovery@Home | task gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 suspended by user
25.10.2017 13:10:19 | DrugDiscovery@Home | [cpu_sched] Preempting gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 (removed from memory)
25.10.2017 13:10:19 | DrugDiscovery@Home | [task] task_state=QUIT_PENDING for gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 from request_exit()
25.10.2017 13:10:19 | | request_exit(): PID 12900 has 3 descendants
25.10.2017 13:10:19 | | PID 3492
25.10.2017 13:10:19 | | PID 460
25.10.2017 13:10:19 | | PID 11624
25.10.2017 13:10:20 | DrugDiscovery@Home | [task] Process for gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 exited, exit code 0, task state 8
25.10.2017 13:10:20 | DrugDiscovery@Home | [task] task_state=UNINITIALIZED for gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 from handle_exited_app
25.10.2017 13:10:20 | DrugDiscovery@Home | [task] task_state=COPY_PENDING for gmx_GMX-T2008_XXXX_TTTTT_tasks.gmx_47_50.in_2 from start
25.10.2017 13:10:20 | DrugDiscovery@Home | Starting task gmx_GMX-T2008_XXXX_TTTTT_tasks.gmx_47_50.in_2


Logfile from 16:38
25.10.2017 16:38:41 | | PID 688
25.10.2017 16:38:45 | DrugDiscovery@Home | [task] task_state=COPY_PENDING for gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 from start
25.10.2017 16:38:45 | DrugDiscovery@Home | [cpu_sched] Restarting task gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 using gmx version 26 (cuda23) in slot 10
25.10.2017 16:38:47 | DrugDiscovery@Home | [task] task_state=EXECUTING for gmx_GMX-T028_XXXX_TTTTT_tasks.gmx_20_100.in_3 from start
25.10.2017 16:39:26 | DrugDiscovery@Home | Sending scheduler request: Requested by project.
25.10.2017 16:39:26 | DrugDiscovery@Home | Not requesting tasks: "no new tasks" requested via Manager

Why it doesn´t use the checkpoint? Lose many hours of computing time.