Since Astropulse work units started going out again, I have noticed that the task is frequently aborted. Soluto tells me Astropulse has crashed, and the BOINC Manager tells me that the aborted task has had a "computation error." BOINC will start a new Astropulse task until it has four tasks labeled "computation error." Then BOINC stops receiving Astropulse units.

Does anyone have any suggestions? For now I just won't let BOINC use my GPU.

More information:

I recently updated my graphics card driver. Before I did that, Astropulse caused my computer to restart a few minutes after starting the task.

If I tell BOINC not to use my GPU, there are no problems.

I have monitored my temperatures with HWMonitor, and CPU and GPU temperatures are within reasonable limits.

I have stress tested my system with Prime95 and had no errors after 12 hours.

Processor: AMD Phenom II X4 945
Graphics card: ATI Radeon HD 5770
Motherboard: Gigabyte GA-770TA-UD3
RAM: 4GB DDR3
Operating System: Windows 7

Somehow your host managed to get 2 OpenCL ATi platfoms instead of one:
Remove older driver and reinstal newer one.

Number of OpenCL platforms: 2


OpenCL Platform Name: ATI Stream
Number of devices: 1
Max compute units: 10
Max work group size: 256
Max clock frequency: 850Mhz
Max memory allocation: 134217728
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 536870912
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: Juniper
Vendor: Advanced Micro Devices, Inc.
Driver version: CAL 1.4.1848
Version: OpenCL 1.1 ATI-Stream-v2.3 (451)
Extensions: 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_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_printf cl_amd_media_ops cl_amd_popcnt cl_khr_d3d10_sharing


OpenCL Platform Name: AMD Accelerated Parallel Processing
Number of devices: 1
Max compute units: 10
Max work group size: 256
Max clock frequency: 850Mhz
Max memory allocation: 536870912
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: Juniper
Vendor: Advanced Micro Devices, Inc.
Driver version: 1642.5 (VM)
Version: OpenCL 1.2 AMD-APP (1642.5)
Extensions: 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_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_khr_dx9_media_sharing cl_amd_image2d_from_buffer_read_only cl_khr_spir cl_khr_gl_event

Hi,
from looking at one of your AstroPulse results your graphics driver update did not work correct. The graphics engine is newer then the OpenCL runtime (necessary part of graphics driver to crunch work_units).

First try to uninstall all parts of the graphics driver currently installed, before reinstalling the new driver.

That should fix your current problems.

---

_\|/_
U r s

First try to uninstall all parts of the graphics driver currently installed, before reinstalling the new driver.

Although you may not want to do that on the latest 14.12 Omega drivers. I've been fighting my TV server half the day due to a problem.

The HD4850 in it had something corrupt, due to which all kinds of scan lines were showing on the TV. So I took it out and put in the HD6850 I had still lying about. Previously I had uninstalled the 12.1 drivers that were on the system and I installed Omega 14.12 after installing the 6850... only to be met by Windows not recognizing the videocard, due to the drivers not being digitally signed. VGASave was running the show, the 42 inch TV could at max run at 1600x1200p, where it ran at 1920x1080p without trouble on the 4850.

Looking around the net, I found that it's a usual problem when updating something on Windows, that for some unknown reason it forgets that drivers are digitally signed. The advice by Microsoft is to start with F8 and then choose the "Run without digital checking" option, but you have to do that each and every time you start the computer... Another option is to reformat and reinstall Windows.

So no, didn't go that way.
Instead I reverted to the restore point I made last night due to Windows Update, and then just installed Omega 14.12 directly over the older drivers. That fixed the whole digitally signed problem for me, as immediately the 6850 is recognized and the TV is also recognized. Plus Holly may find it nice that Raptr optimized some of the games on that system.

---

Thanks a lot for your responses.

I uninstalled my graphics card driver in the Windows control panel. My resolution got really low. I tried to install the driver I downloaded from the AMD website (it turns out it was Catalyst, which then auto-detected which driver I needed, and then I installed that), but my resolution was still really low and my graphics card didn't show up in the device manager. I began to fear that I was in Ageless's position. But then Windows Update automatically installed a driver dated from 2011 that brought my resolution back to normal. I installed the driver I had downloaded from AMD again.

I enabled my GPU in BOINC and had SETI@Home update. But no Astropulse tasks showed up. So I had Einstein@Home update (which had been on "no new tasks"), and it ran a task using my GPU. That crashed. I tried again, and BOINC crashed.

Is there something I should have done differently? I don't know the difference between the graphics engine and OpenCL runtime....

---

Its important to reboot before uninstalling the drivers.
OpenCL runtime is locked by the system so it wont update properly.

Reboot
Uninstall drivers
reboot
Install new drivers.

A good tool http://www.wagnardmobile.com/DDU/

---

With each crime and every kindness we birth our future.

Just to clear something up, as I was talking about the Omega drivers. There aren't downloaded from that Omegadrivers.net, but it's the official name of the Catalyst 14.12 drivers that you can get from the AMD site.

Name	      File Size    Revision Number	Release Date
Catalyst        288 MB	    Omega (14.12) 	12/9/2014
Software Suite

It is very probable that AMD worked together with the guys from www.omegadrivers.net to get this end result, also because of the addition of AMD Gaming Evolved (the Raptr app I mentioned).

For the number crunchers among us, who wonder about the version number:
12/12/2014 17:05:41 | | CAL: ATI GPU 0: AMD Radeon HD 7850/7870 series (Pitcairn) (CAL version 1.4.1848, 2048MB, 2008MB available, 6400 GFLOPS peak)
12/12/2014 17:05:41 | | OpenCL: AMD/ATI GPU 0: AMD Radeon HD 7850/7870 series (Pitcairn) (driver version 1642.5 (VM), device version OpenCL 1.2 AMD-APP (1642.5), 2048MB, 2008MB available, 6400 GFLOPS peak)

---

Just to clear something up, as I was talking about the Omega drivers. There aren't downloaded from that Omegadrivers.net, but it's the official name of the Catalyst 14.12 drivers that you can get from the AMD site.

Name	      File Size    Revision Number	Release Date
Catalyst        288 MB	    Omega (14.12) 	12/9/2014
Software Suite

It is very probable that AMD worked together with the guys from www.omegadrivers.net to get this end result, also because of the addition of AMD Gaming Evolved (the Raptr app I mentioned).

For the number crunchers among us, who wonder about the version number:
12/12/2014 17:05:41 | | CAL: ATI GPU 0: AMD Radeon HD 7850/7870 series (Pitcairn) (CAL version 1.4.1848, 2048MB, 2008MB available, 6400 GFLOPS peak)
12/12/2014 17:05:41 | | OpenCL: AMD/ATI GPU 0: AMD Radeon HD 7850/7870 series (Pitcairn) (driver version 1642.5 (VM), device version OpenCL 1.2 AMD-APP (1642.5), 2048MB, 2008MB available, 6400 GFLOPS peak)

I normally just like to use version numbers.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[