please add Vega 11 (Ryzen 2400G)

[plonk420]

i hope i'm doing this correctly...

0x1002:0x15dd:56:0:0:Advanced Micro Devices, Inc. [AMD/ATI]:AMD Radeon(TM) RX Vega 11 Graphics
AKA the graphics side of Ryzen 5 2400G



edit: oops, here's the log

Code: Select all

*********************** Log Started 2019-11-14T23:29:05Z ***********************
23:29:05:************************* Folding@home Client *************************
23:29:05:        Website: https://foldingathome.org/
23:29:05:      Copyright: (c) 2009-2018 foldingathome.org
23:29:05:         Author: Joseph Coffland <joseph@cauldrondevelopment.com>
23:29:05:           Args: --open-web-control
23:29:05:         Config: <none>
23:29:05:******************************** Build ********************************
23:29:05:        Version: 7.5.1
23:29:05:           Date: May 11 2018
23:29:05:           Time: 13:06:32
23:29:05:     Repository: Git
23:29:05:       Revision: 4705bf53c635f88b8fe85af7675557e15d491ff0
23:29:05:         Branch: master
23:29:05:       Compiler: Visual C++ 2008
23:29:05:        Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox /MT
23:29:05:       Platform: win32 10
23:29:05:           Bits: 32
23:29:05:           Mode: Release
23:29:05:******************************* System ********************************
23:29:05:            CPU: AMD Ryzen 5 2400G with Radeon Vega Graphics
23:29:05:         CPU ID: AuthenticAMD Family 23 Model 17 Stepping 0
23:29:05:           CPUs: 8
23:29:05:         Memory: 15.70GiB
23:29:05:    Free Memory: 8.79GiB
23:29:05:        Threads: WINDOWS_THREADS
23:29:05:     OS Version: 6.2
23:29:05:    Has Battery: false
23:29:05:     On Battery: false
23:29:05:     UTC Offset: -7
23:29:05:            PID: 10244
23:29:05:            CWD: C:\Users\ranma\AppData\Roaming\FAHClient
23:29:05:             OS: Windows 10 Enterprise
23:29:05:        OS Arch: AMD64
23:29:05:           GPUs: 0
23:29:05:           CUDA: Not detected: Failed to open dynamic library 'nvcuda.dll': The
23:29:05:                 specified module could not be found.
23:29:05:
23:29:05:OpenCL Device 0: Platform:0 Device:0 Bus:56 Slot:0 Compute:1.2 Driver:2906.10
23:29:05:  Win32 Service: false
23:29:05:***********************************************************************
23:29:05:<config>
23:29:05:  <!-- Folding Slots -->
23:29:05:</config>
23:29:05:Connecting to assign1.foldingathome.org:8080
23:29:05:Updated GPUs.txt
23:29:05:Read GPUs.txt
23:29:05:Trying to access database...
23:29:05:Successfully acquired database lock
23:29:05:Enabled folding slot 00: PAUSED cpu:6 (not configured)
23:29:05:Enabled folding slot 01: PAUSED gpu:0:Raven [Ryzen vega 8 moble] (not configured)
23:29:07:10:127.0.0.1:New Web connection
23:30:06:Saving configuration to config.xml
23:30:06:<config>
23:30:06:  <!-- Folding Slots -->
23:30:06:  <slot id='0' type='CPU'/>
23:30:06:  <slot id='1' type='GPU'/>
23:30:06:</config>
23:30:06:Set client configured
23:30:06:WU00:FS00:Connecting to 65.254.110.245:8080
23:30:06:WU00:FS00:Connecting to 65.254.110.245:8080
23:30:06:WU00:FS00:Assigned to work server 155.247.166.220
23:30:06:WU00:FS00:Requesting new work unit for slot 00: READY cpu:6 from 155.247.166.220
23:30:06:WU00:FS00:Connecting to 155.247.166.220:8080
23:30:06:WU01:FS01:Connecting to 65.254.110.245:8080
23:30:06:WU01:FS01:Assigned to work server 128.252.203.10
23:30:06:WU01:FS01:Requesting new work unit for slot 01: READY gpu:0:Raven [Ryzen vega 8 moble] from 128.252.203.10
23:30:06:WU01:FS01:Connecting to 128.252.203.10:8080
23:30:06:WU00:FS00:Downloading 1.84MiB
23:30:06:ERROR:WU01:FS01:Exception: Server did not assign work unit
23:30:07:WU01:FS01:Connecting to 65.254.110.245:8080
23:30:07:WU01:FS01:Assigned to work server 155.247.166.220
23:30:07:WU01:FS01:Requesting new work unit for slot 01: READY gpu:0:Raven [Ryzen vega 8 moble] from 155.247.166.220
23:30:07:WU01:FS01:Connecting to 155.247.166.220:8080
23:30:07:WU01:FS01:Downloading 15.86MiB
23:30:08:WU00:FS00:Download complete
23:30:08:WU00:FS00:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:13794 run:16 clone:172 gen:32 core:0xa7 unit:0x000000280002894c5d5f3406323dbc12
23:30:08:WU00:FS00:Downloading core from http://cores.foldingathome.org/v7/win/64bit/avx/Core_a7.fah
23:30:08:WU00:FS00:Connecting to cores.foldingathome.org:80
23:30:08:WU00:FS00:FahCore a7: Downloading 6.71MiB
23:30:09:WU00:FS00:FahCore a7: Download complete
23:30:09:WU00:FS00:Valid core signature
23:30:09:WU00:FS00:Unpacked 19.85MiB to cores/cores.foldingathome.org/v7/win/64bit/avx/Core_a7.fah/FahCore_a7.exe
23:30:09:WU00:FS00:Unpacked 2.64MiB to cores/cores.foldingathome.org/v7/win/64bit/avx/Core_a7.fah/libfftw3f-3.dll
23:30:09:WU00:FS00:Starting
23:30:09:WU00:FS00:Running FahCore: C:\internet\FAHClient/FAHCoreWrapper.exe C:\Users\ranma\AppData\Roaming\FAHClient\cores/cores.foldingathome.org/v7/win/64bit/avx/Core_a7.fah/FahCore_a7.exe -dir 00 -suffix 01 -version 705 -lifeline 10244 -checkpoint 15 -np 6
23:30:09:WU00:FS00:Started FahCore on PID 5456
23:30:09:WU00:FS00:Core PID:8604
23:30:09:WU00:FS00:FahCore 0xa7 started
23:30:10:WU00:FS00:0xa7:*********************** Log Started 2019-11-14T23:30:09Z ***********************
23:30:10:WU00:FS00:0xa7:************************** Gromacs Folding@home Core ***************************
23:30:10:WU00:FS00:0xa7:       Type: 0xa7
23:30:10:WU00:FS00:0xa7:       Core: Gromacs
23:30:10:WU00:FS00:0xa7:       Args: -dir 00 -suffix 01 -version 705 -lifeline 5456 -checkpoint 15 -np 6
23:30:10:WU00:FS00:0xa7:************************************ CBang *************************************
23:30:10:WU00:FS00:0xa7:       Date: Oct 26 2019
23:30:10:WU00:FS00:0xa7:       Time: 01:38:25
23:30:10:WU00:FS00:0xa7:   Revision: c46a1a011a24143739ac7218c5a435f66777f62f
23:30:10:WU00:FS00:0xa7:     Branch: master
23:30:10:WU00:FS00:0xa7:   Compiler: Visual C++ 2008
23:30:10:WU00:FS00:0xa7:    Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox /MT
23:30:10:WU00:FS00:0xa7:   Platform: win32 10
23:30:10:WU00:FS00:0xa7:       Bits: 64
23:30:10:WU00:FS00:0xa7:       Mode: Release
23:30:10:WU00:FS00:0xa7:************************************ System ************************************
23:30:10:WU00:FS00:0xa7:        CPU: AMD Ryzen 5 2400G with Radeon Vega Graphics
23:30:10:WU00:FS00:0xa7:     CPU ID: AuthenticAMD Family 23 Model 17 Stepping 0
23:30:10:WU00:FS00:0xa7:       CPUs: 8
23:30:10:WU00:FS00:0xa7:     Memory: 15.70GiB
23:30:10:WU00:FS00:0xa7:Free Memory: 8.69GiB
23:30:10:WU00:FS00:0xa7:    Threads: WINDOWS_THREADS
23:30:10:WU00:FS00:0xa7: OS Version: 6.2
23:30:10:WU00:FS00:0xa7:Has Battery: false
23:30:10:WU00:FS00:0xa7: On Battery: false
23:30:10:WU00:FS00:0xa7: UTC Offset: -7
23:30:10:WU00:FS00:0xa7:        PID: 8604
23:30:10:WU00:FS00:0xa7:        CWD: C:\Users\ranma\AppData\Roaming\FAHClient\work
23:30:10:WU00:FS00:0xa7:******************************** Build - libFAH ********************************
23:30:10:WU00:FS00:0xa7:    Version: 0.0.18
23:30:10:WU00:FS00:0xa7:     Author: Joseph Coffland <joseph@cauldrondevelopment.com>
23:30:10:WU00:FS00:0xa7:  Copyright: 2019 foldingathome.org
23:30:10:WU00:FS00:0xa7:   Homepage: https://foldingathome.org/
23:30:10:WU00:FS00:0xa7:       Date: Oct 26 2019
23:30:10:WU00:FS00:0xa7:       Time: 01:52:30
23:30:10:WU00:FS00:0xa7:   Revision: c1e3513b1bc0c16013668f2173ee969e5995b38e
23:30:10:WU00:FS00:0xa7:     Branch: master
23:30:10:WU00:FS00:0xa7:   Compiler: Visual C++ 2008
23:30:10:WU00:FS00:0xa7:    Options: /TP /nologo /EHa /wd4297 /wd4103 /Ox /MT
23:30:10:WU00:FS00:0xa7:   Platform: win32 10
23:30:10:WU00:FS00:0xa7:       Bits: 64
23:30:10:WU00:FS00:0xa7:       Mode: Release
23:30:10:WU00:FS00:0xa7:************************************ Build *************************************
23:30:10:WU00:FS00:0xa7:       SIMD: avx_256
23:30:10:WU00:FS00:0xa7:********************************************************************************
23:30:10:WU00:FS00:0xa7:Project: 13794 (Run 16, Clone 172, Gen 32)
23:30:10:WU00:FS00:0xa7:Unit: 0x000000280002894c5d5f3406323dbc12
23:30:10:WU00:FS00:0xa7:Reading tar file core.xml
23:30:10:WU00:FS00:0xa7:Reading tar file frame32.tpr
23:30:10:WU00:FS00:0xa7:Digital signatures verified
23:30:10:WU00:FS00:0xa7:Calling: mdrun -s frame32.tpr -o frame32.trr -cpt 15 -nt 6
23:30:10:WU00:FS00:0xa7:Steps: first=80000000 total=2500000
23:30:11:WU00:FS00:0xa7:Completed 1 out of 2500000 steps (0%)
23:30:11:WU01:FS01:Download complete
23:30:11:WU01:FS01:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:14180 run:10 clone:131 gen:144 core:0x21 unit:0x000000dc0002894c5d3b55e336886bde
23:30:11:WU01:FS01:Downloading core from http://cores.foldingathome.org/v7/win/64bit/amd/Core_21.fah
23:30:11:WU01:FS01:Connecting to cores.foldingathome.org:80
23:30:11:WU01:FS01:FahCore 21: Downloading 3.48MiB
23:30:12:WU01:FS01:FahCore 21: Download complete
23:30:12:WU01:FS01:Valid core signature
23:30:12:WU01:FS01:Unpacked 11.81MiB to cores/cores.foldingathome.org/v7/win/64bit/amd/Core_21.fah/FahCore_21.exe
23:30:12:WU01:FS01:Starting
23:30:12:WU01:FS01:Running FahCore: C:\internet\FAHClient/FAHCoreWrapper.exe C:\Users\ranma\AppData\Roaming\FAHClient\cores/cores.foldingathome.org/v7/win/64bit/amd/Core_21.fah/FahCore_21.exe -dir 01 -suffix 01 -version 705 -lifeline 10244 -checkpoint 15 -gpu-vendor amd -opencl-platform 0 -opencl-device 0 -gpu 0
23:30:12:WU01:FS01:Started FahCore on PID 952
23:30:12:WU01:FS01:Core PID:2340
23:30:12:WU01:FS01:FahCore 0x21 started
23:30:13:WU01:FS01:0x21:*********************** Log Started 2019-11-14T23:30:12Z ***********************
23:30:13:WU01:FS01:0x21:Project: 14180 (Run 10, Clone 131, Gen 144)
23:30:13:WU01:FS01:0x21:Unit: 0x000000dc0002894c5d3b55e336886bde
23:30:13:WU01:FS01:0x21:CPU: 0x00000000000000000000000000000000
23:30:13:WU01:FS01:0x21:Machine: 1
23:30:13:WU01:FS01:0x21:Reading tar file core.xml
23:30:13:WU01:FS01:0x21:Reading tar file integrator.xml
23:30:13:WU01:FS01:0x21:Reading tar file state.xml
23:30:13:WU01:FS01:0x21:Reading tar file system.xml
23:30:13:WU01:FS01:0x21:Digital signatures verified
23:30:13:WU01:FS01:0x21:Folding@home GPU Core21 Folding@home Core
23:30:13:WU01:FS01:0x21:Version 0.0.20
23:31:07:Saving configuration to config.xml
23:31:07:<config>
23:31:07:  <!-- Folding Slot Configuration -->
23:31:07:  <cause v='CANCER'/>
23:31:07:
23:31:07:  <!-- User Information -->
23:31:07:  <team v='86565'/>
23:31:07:  <user v='plonk420'/>
23:31:07:
23:31:07:  <!-- Folding Slots -->
23:31:07:  <slot id='0' type='CPU'/>
23:31:07:  <slot id='1' type='GPU'/>
23:31:07:</config>
23:36:25:WU00:FS00:0xa7:Completed 25000 out of 2500000 steps (1%)

[bruce]

Its not clear to me whether this GPU supports Double Precision math. (The current FAHCores now require DP math.) If you can find documentation that it does, I'll add it.

(Before DP math was required, I had experience with a similar configuration and found that using the GPU on my APU was slower than using the CPU. That may or may not apply to your system.)

I recommend you upgrade to a discrete GPU.

[_r2w_ben]

(Before DP math was required, I had experience with a similar configuration and found that using the GPU on my APU was slower than using the CPU. That may or may not apply to your system.)

As a counterpoint, I've been folding on a Kaveri based A8-7600 APU for years. It produces 3-5K PPD on the CPU side and 12-18K PPD on the GPU side.
This 2400G has nearly twice the shaders and a 520MHz clock speed advantage. I estimate it could produce 40-75K PPD.

TechPowerUp mentions 1:16 DP performance for Vega 11.

[bruce]

The Double Precision support is critical. My experience is not.

[Jetster220]

Hi Bruce, is this the info you need about the double precision?
https://www.techpowerup.com/gpu-specs/r ... a-11.c3054
Shows that Vega 11 supports FP64 (double) performance
109.1 GFLOPS (1:16)

[BobWilliams757]

Hi Bruce, is this the info you need about the double precision?
https://www.techpowerup.com/gpu-specs/r ... a-11.c3054
Shows that Vega 11 supports FP64 (double) performance
109.1 GFLOPS (1:16)

The Vega 11 onboard the 2400G worked fine from the initial time I set up F@H. I haven't looked but I've been told it's already on the whitelist. In the logs and client windows it shows up just as in the initial post on this thread:
Raven [Ryzen vega 8 moble]

For an onboard chip it actually does reasonably well. GPU alone would pull 16-18K without QRB. I didn't pay much attention to the CPU numbers as I kept tweaking around with my fan profiles and such to get things set up properly for long term folding.

With a passkey and QRB, the Vega 11 will pull 50-70K PPD, somewhat variable with WU and your memory speed. CPU will pull 20+K PPD as well. Right now with both running (CPU at full 6 threads) my estimated PPD is 82.8K PPD.

Due to the graphics being on die, there is some overlap as for how these things will use memory and the various pipelines when used together. They lose a little bit of efficiency when both running at once. But more than capable of folding, just be prepared for some GPU WU's that take a looooooonng time. I had one that was almost two days. But on the positive side using GPU only creates minimal power draw and heat. You can leave it running in the background and probably never know it's there unless you look for it.

[Jetster220]

Thanks, I got it running, had to manually add GPU slot, and set the slider to full for it to run and not stay idle, it's chugging along well now.

[Rev]

I second the add of this APU under Linux with standard open source Mesa drivers and mesa-opencl. Since I use the latest kernels I had to remove my folding NVidia 1030 (also saves power) and now I cannot fold on my 24/7 system anymore. I think folding doesn't even recognize the APU as working opencl device. That should be basic on a APU that is as old as the 2400G(E) or the 2200G(E). Right now it doesn't even recognize the APU as vega11 but as vega8. And I feel that is just sloppy.

[JimboPalmer]

I second the add of this APU under Linux with standard open source Mesa drivers and mesa-opencl.

My understanding is that F@H needs 'real' AMD Pro drivers and OpenCL.

[Joe_H]

Someone has also posted about getting the ROCm drivers to work and support OpenCL processing for F@h. But the Mesa drivers do not work for folding.

As for being recognized as Vega 8 versus Vega 11, at best the description field may get updated to be something like Vega 8/11. The GPUs.txt entries and hardware detection is based on the Device ID, and AMD used the same ID number for both. Essentially they are the same chip, AMD has disabled some of the shaders to create an "8" out of an "11".

[MeeLee]

Wasn't AMD/Vega GPUs supported in Core 22?
With the low core 22 WU count, it might be possible the IGP might not immediately get WUs assigned...
Just asking if this is still the case?

[Joe_H]

You are thinking of Radeon RX 5000 series such as the 5700 XT which are Navi chip based and use the RDNA microarchitecture. Those required updated software support, only Core_22 supports those. The Vega GPUs are GCN 5th generation based cards, and should be usable on both Core_21 and Core_22 projects.

[Rev]

A lot of time has past.
Look, I do not want to look greedy to help, but I can at least tell you my opinions:
Mesa drivers are the standard on the most used Linux, Ubuntu, with AMD graphics components. And there is a working opencl driver in the official repo of ubuntu.
I have a (daily) mesa driver version running and after installing the opencl driver I also get an positive message about if this combination works.

But it still does not work with Folding@home.

Can you at least try to involve a majority of Ubuntu users with AMD graphics components in ubuntus default driver settings? I mean, how hard can it be to implement the installation of the opencl driver from the ubuntu repo and make it compatible to Folding@home?

It has been nearly one year now, and you still consider windows users your default clienteel. Or users who update their AMD pro drivers more often than their kernels. And that is the biggest issue here - If you use most recent release candidate kernels, you are most likely excluded from using AMD pro drivers.

And this one year later, my Ubuntu still has all the prerequisites to run Folding, but Folding did not move a centimeter in this year to solve anything in this regard.

So can I please ask you again to make folding "compatible" with standard ubuntu and APU's from AMD?

Here are the libraries (on my system), that have opencl in the name:

mesa-opencl-icd 21.2~git2105081800.d81a6e~oibaf~f
ocl-icd-libopencl1 2.2.11-1ubuntu1
ocl-icd-libopencl1:i386 2.2.11-1ubuntu1
opencl-c-headers 2.2~2019.08.06-g0d5f18c-1

And clinfo:

$ clinfo
Number of platforms 1
Platform Name Clover
Platform Vendor Mesa
Platform Version OpenCL 1.1 Mesa 21.2.0-devel (git-d81a6e5 2021-05-08 focal-oibaf-ppa)
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd
Platform Extensions function suffix MESA

Platform Name Clover
Number of devices 1
Device Name AMD Radeon(TM) Vega 11 Graphics (RAVEN, DRM 3.40.0, 5.12.2-051202-generic, LLVM 12.0.0)
Device Vendor AMD
Device Vendor ID 0x1002
Device Version OpenCL 1.1 Mesa 21.2.0-devel (git-d81a6e5 2021-05-08 focal-oibaf-ppa)
Driver Version 21.2.0-devel
Device OpenCL C Version OpenCL C 1.1
Device Type GPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Max compute units 11
Max clock frequency 1250MHz
Max work item dimensions 3
Max work item sizes 256x256x256
Max work group size 256
Preferred / native vector sizes
char 16 / 16
short 8 / 8
int 4 / 4
long 2 / 2
half 0 / 0 (n/a)
float 4 / 4
double 2 / 2 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals No
Infinity and NANs Yes
Round to nearest Yes
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
Double-precision Floating-point support (cl_khr_fp64)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Address bits 64, Little-Endian
Global memory size 3221225472 (3GiB)
Error Correction support No
Max memory allocation 2254857830 (2.1GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 32768 bits (4096 bytes)
Global Memory cache type None
Image support No
Local memory type Local
Local memory size 32768 (32KiB)
Max number of constant args 16
Max constant buffer size 67108864 (64MiB)
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Profiling timer resolution 0ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Device Extensions cl_khr_byte_addressable_store cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_fp64 cl_khr_extended_versioning

NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Clover
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [MESA]
clCreateContext(NULL, ...) [default] Success [MESA]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
Platform Name Clover
Device Name AMD Radeon(TM) Vega 11 Graphics (RAVEN, DRM 3.40.0, 5.12.2-051202-generic, LLVM 12.0.0)
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1)
Platform Name Clover
Device Name AMD Radeon(TM) Vega 11 Graphics (RAVEN, DRM 3.40.0, 5.12.2-051202-generic, LLVM 12.0.0)
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1)
Platform Name Clover
Device Name AMD Radeon(TM) Vega 11 Graphics (RAVEN, DRM 3.40.0, 5.12.2-051202-generic, LLVM 12.0.0)

ICD loader properties
ICD loader Name OpenCL ICD Loader
ICD loader Vendor OCL Icd free software
ICD loader Version 2.2.11
ICD loader Profile OpenCL 2.1

[Joe_H]

A lot of time has past.
Look, I do not want to look greedy to help, but I can at least tell you my opinions:
Mesa drivers are the standard on the most used Linux, Ubuntu, with AMD graphics components. And there is a working opencl driver in the official repo of ubuntu.
I have a (daily) mesa driver version running and after installing the opencl driver I also get an positive message about if this combination works.

What exactly do you want F@h to do, write new Mesa driver code? That is what is needed, not a change to F@h. They have never supported Mesa drivers, and until they reliably give OpenCL 1.2 or better support won't.

As shown in the information in your own post, only OpenCL 1.1 is being shown:

Code: Select all

Platform Version OpenCL 1.1 Mesa 21.2.0-devel (git-d81a6e5 2021-05-08 focal-oibaf-ppa)
...
Device OpenCL C Version OpenCL C 1.1

I don't know who is working on updating the Mesa drivers to support later versions of OpenCL, but so far it is not showing any improvement.

[bruce]

I'd like to second Joe_H's response.

FAH is in the business of doing scientific research, not in the business of developing drivers or improving things like Ubuntu. TheFAH application were written for OpenCL 1.2. It's up to you or AMD to figure out how to get support for OpenCL 1.2 on your distro.

Asking for help here is the first step toward accomplishing that. Version 1.2 (or greater) is available from AMD and probably from Ubuntu.