| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: Round 3 - GPU Beta Test for Help Conquer Cancer |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 151
|
|
| Author |
|
|
johncmacalister2010@gmail.com
Veteran Cruncher Canada Joined: Nov 16, 2010 Post Count: 799 Status: Offline Project Badges: 
|
Are you still getting these HCC GPU Betas ? Or is the shower over already ? I just got one on 17 March with the following results: Result Name Device Name Status Sent Time Time Due / CPU Time (hours) Claimed/ Granted BOINC Credit Credits/h Return Time A_ X0000130860161201112120942_ 2-- John-DT100W64 Valid 17-03-12 0:50 17-03-12 1:10 0.03 193.6 / 193.6 6465.333333 OS: Windows 7 64 bit, Ultimate CPU: AMD Phenon II X6 1090T @ 3.79 GHz GPU: AMD Radeon HD 6570 Driver: 8.950.0.0 Driver date: 14-Feb-12 193.6 Credits in 0.03h....   crunching, crunching, crunching. AMD Ryzen 5 2600 6-core Processor with Windows 11 64 Pro. AMD Ryzen 7 3700X 8-Core Processor with Windows 11 64 Pro (part time) |
||
|
|
BladeD
Ace Cruncher USA Joined: Nov 17, 2004 Post Count: 28976 Status: Offline Project Badges: 
|
I guess no more GPU WUs until Monday at the earliest.
---------------------------------------- |
||
|
|
mikey
Veteran Cruncher Joined: May 10, 2009 Post Count: 821 Status: Offline Project Badges:
|
Why can I not keep WCG from downloading CPU tasks.. As BladeD said they do not have the option to NOT get them yet as they ARE part of the Beta test. The idea is to run BOTH types of units and see if the results come back the same, not just Valid but close enough to send a unit to either one and they will both Validate. That is the ideal situation, they can send a unit to whoever or whatever asks for it and it will validate against whoever or whatever else got it. It is easier on the project that way, they only have to create one kind of unit and send it out to us users. They have said they WILL put a gpu/cpu exclusion switch, but no date has been announced yet. I am guessing it will be after we testers prove the system works.   |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
There was an error in the server code that caused to accidentally exclude all NVIDIA cards. We have corrected this an we are running a new beta test that will include NVIDIA this time.
We are aware of a higher rate of invalid results and it is one of the items that we will be working on addressing as we work through beta testing. |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
Would not surprise me if the BETA limit still follows the same rule... 1 per CPU thread i.e. even if there's 1 GPU in the system, but 8 allowed CPU threads, you'd be getting a max of 8 on a machine. --//-- The rule (with the new server code) is 1 job per CPU core, 4 jobs per GPU for beta testing. However, there is also an initial limit of 2 jobs per day per host per application version. That limit is raised as success results are returned. |
||
|
|
Ingleside
Veteran Cruncher Norway Joined: Nov 19, 2005 Post Count: 974 Status: Offline Project Badges: 
|
The rule (with the new server code) is 1 job per CPU core, 4 jobs per GPU for beta testing. However, there is also an initial limit of 2 jobs per day per host per application version. That limit is raised as success results are returned. Assuming haven't changed the server-code, the quota isn't increased before the results are validated. Also, like with the old server-code, the quota is multiplied with #cores or #GPU's, so for cpu-apps it starts at 2 jobs per day per core per host per application version.  "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." [Edit 1 times, last edit by Ingleside at Mar 19, 2012 10:31:20 PM] |
||
|
|
Dataman
Ace Cruncher Joined: Nov 16, 2004 Post Count: 4865 Status: Offline Project Badges: 
|
I am completing them in about four minutes (clock). These are using enough CPU to put the CPU jobs that were running their threads to go into a wait state (no worries). Looks like I am going idle some threads to get both working more efficiently (not like I could improve much over 4 minutes
---------------------------------------- )Edit: The CPU jobs resumed running (@ 88%) after about 3 minutes ... I guess both got what they could get in those threads. Yep, am going to have to idle some threads.  [Edit 1 times, last edit by Dataman at Mar 19, 2012 11:15:47 PM] |
||
|
|
pirogue
Veteran Cruncher USA Joined: Dec 8, 2008 Post Count: 685 Status: Offline Project Badges:
|
I had 3 quick -1073741515 failures on a GPU that ran the previous set without issue.
----------------------------------------edit: It is a 5450. [Edit 1 times, last edit by pirogue at Mar 19, 2012 11:12:52 PM] |
||
|
|
honzyk
Cruncher Czech republic Joined: Dec 8, 2004 Post Count: 8 Status: Offline Project Badges:
|
I know NVS4200 produces more errors according to some lists here however is it possible that I get this error because I run on GPU Seti? At least first lines of error log shows "All pipe instances are busy ..."
Seti is running just on GPU and with no errors. Thanks <core_client_version>7.0.18</core_client_version> <![CDATA[ <message> All pipe instances are busy. (0xe7) - exit code 231 (0xe7) </message> <stderr_txt> Commandline: projects/www.worldcommunitygrid.org/wcg_beta3_img_6.44_windows_intelx86__cuda_opencl X0900060940053200511301512.jp2 --device 0 Boinc requested NVIDIA gpu device number 0 Found compute platform NVIDIA Corporation Selecting this platform CL_DEVICE_NAME: NVS 4200M CL_DEVICE_VENDOR: NVIDIA Corporation CL_DEVICE_VERSION: 268.02 CL_DEVICE_MAX_COMPUTE_UNITS: CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS: 3 CL_DEVICE_MAX_WORK_ITEM_SIZES: 1024 / 1024 / 64 CL_DEVICE_MAX_WORK_GROUP_SIZE: 1024 CL_DEVICE_MAX_CLOCK_FREQUENCY: 1620 MHz CL_DEVICE_ADDRESS_BITS: 32 CL_DEVICE_MAX_MEM_ALLOC_SIZE: 248 MByte CL_DEVICE_GLOBAL_MEM_SIZE: 993 MByte CL_DEVICE_ERROR_CORRECTION_SUPPORT: no CL_DEVICE_LOCAL_MEM_TYPE: local CL_DEVICE_LOCAL_MEM_SIZE: 48 KByte CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE: 64 KByte CL_DEVICE_QUEUE_PROPERTIES: CL_QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE CL_DEVICE_QUEUE_PROPERTIES: CL_QUEUE_PROFILING_ENABLE CL_DEVICE_EXTENSIONS: cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll 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_fp64 CL_DEVICE_COMPUTE_CAPABILITY_NV: 2.1 CL_DEVICE_REGISTERS_PER_BLOCK_NV: 32768 CL_DEVICE_WARP_SIZE_NV: 32 CL_DEVICE_GPU_OVERLAP_NV: CL_TRUE CL_DEVICE_KERNEL_EXEC_TIMEOUT_NV: CL_TRUE CL_DEVICE_INTEGRATED_MEMORY_NV: CL_FALSE Estimated kernel execution time = 2.14864 [sec] ERROR: Kernel execution time estimate too high, exiting. 19:25:16 (8820): called boinc_finish </stderr_txt> ]]> |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
The GPU beta WUs use 100% of a CPU on Nvidia cards. I think it has something to do with the drivers. My ATI card only uses 36%.
---------------------------------------- They both use 95-97% of the GPU on my cards. @Dataman. Freeing 1 CPU core should do it. Worked for me on i7-2600k with a GTX 460. The other 7 cores are running non-stop @ 99.7%.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
----------------------------------------  [Edit 2 times, last edit by nanoprobe at Mar 20, 2012 12:19:28 AM] |
||
|
|
|