We know that, as it is a Turing nVidia GeForce MX550, here: https://www.techpowerup.com/gpu-specs/geforce-mx550.c3860
...
Well, there is a comment here saying it does not support CUDA
George was incorrect in saying that. He's probably confusing it with the much older 500 series cards (which still support CUDA, but a much older version). the MX line just means it's a mobile GPU.
the MX550 will work fine here, it's Turing generation similar to something like a GTX 1650, with the caveats of laptop cooling issues. and yes with only 2GB, the O3AS GW tasks are likely out of the question, BRP7 only.
Might setup a separate e@h project profile for this system. Then you can try out different settings without breaking all your other systems.
Select brp7 only. No test or non-requested apps. GPU processing only. Nvidia only.
If it comes back with no tasks available message rather than no tasks sent there is a different issue.
Tom M
A Proud member of the O.F.A. (Old Farts Association). Be well, do good work, and keep in touch.® (Garrison Keillor) I want some more patience. RIGHT NOW!
Just a heads-up. Anyone with recent NVIDIA GPU (RTX 30- and 40-series) might want to check their GPU hotspot temperature. I don't think the single temperature reported by nvidia-smi is it.
Bouncing off thermal throttling sounds like a good way to kill them fast with more-or-less constant load like E@H, though it might take longer to get to that point in the first place than typical game load would, when half of the problem is thermal paste pump-out.
A Proud member of the O.F.A. (Old Farts Association). Be well, do good work, and keep in touch.® (Garrison Keillor) I want some more patience. RIGHT NOW!
San-Fernando-Valley
)
We know that, as it is a Turing nVidia GeForce MX550, here: https://www.techpowerup.com/gpu-specs/geforce-mx550.c3860
That should work, but problem is:
So Restart & go to BOINC Manager -> open Tools in menu -> click Event log
Go to top of the BOINC Event log window & post first 50 lines in this topic. ;)
non-profit org. Play4Life in Zagreb, Croatia, EU
KLiK wrote:...We know
)
Well, there is a comment here saying it does not support CUDA
San-Fernando-Valley
)
George was incorrect in saying that. He's probably confusing it with the much older 500 series cards (which still support CUDA, but a much older version). the MX line just means it's a mobile GPU.
the MX550 will work fine here, it's Turing generation similar to something like a GTX 1650, with the caveats of laptop cooling issues. and yes with only 2GB, the O3AS GW tasks are likely out of the question, BRP7 only.
_________________________________________________________________________
San-Fernando-Valley
)
No, check again here: https://www.techpowerup.com/gpu-specs/geforce-mx550.c3860
It says (hardware) CUDA 7.5 on chip.
non-profit org. Play4Life in Zagreb, Croatia, EU
KLiK
)
Well, we know that.
Also please check what IAN&STEVE said.
Thanks to all for comments.
)
Thanks to all for comments. Still trying to digest them. To TOM M, is this what you want?
7/3/2024 6:41:58 PM | | cc_config.xml not found - using defaults
7/3/2024 6:41:58 PM | | Starting BOINC client version 8.0.2 for windows_x86_64
7/3/2024 6:41:58 PM | | log flags: file_xfer, sched_ops, task
7/3/2024 6:41:58 PM | | Libraries: libcurl/8.7.0-DEV Schannel zlib/1.3.1
7/3/2024 6:41:58 PM | | Data directory: C:\ProgramData\BOINC
7/3/2024 6:41:58 PM | | Running under account bradb
7/3/2024 6:42:01 PM | | CUDA: NVIDIA GPU 0: NVIDIA GeForce MX550 (driver version 546.59, CUDA version 12.3, compute capability 7.5, 2048MB, 2048MB available, 3072 GFLOPS peak)
7/3/2024 6:42:01 PM | | OpenCL: NVIDIA GPU 0: NVIDIA GeForce MX550 (driver version 546.59, device version OpenCL 3.0 CUDA, 2048MB, 2048MB available, 3072 GFLOPS peak)
7/3/2024 6:42:01 PM | | OpenCL: Intel GPU 0: Intel(R) Iris(R) Xe Graphics (driver version 31.0.101.5382, device version OpenCL 3.0 NEO, 14806MB, 14806MB available, 998 GFLOPS peak)
7/3/2024 6:42:01 PM | | Windows processor group 0: 12 processors
7/3/2024 6:42:01 PM | | Host name: Charlie1
7/3/2024 6:42:01 PM | | Processor: 12 GenuineIntel 13th Gen Intel(R) Core(TM) i7-1355U [Family 6 Model 186 Stepping 3]
7/3/2024 6:42:01 PM | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush acpi mmx fxsr sse sse2 ss htt tm pni pclmulqdq ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 adx gfni vaes vpclmulqdq tm2 pbe fsgsbase bmi1 smep bmi2
7/3/2024 6:42:01 PM | | OS: Microsoft Windows 11: Core x64 Edition, (10.00.22631.00)
7/3/2024 6:42:01 PM | | Memory: 31.70 GB physical, 33.70 GB virtual
7/3/2024 6:42:01 PM | | Disk: 927.75 GB total, 801.54 GB free
7/3/2024 6:42:01 PM | | Local time is UTC -7 hours
7/3/2024 6:42:01 PM | | No WSL found.
7/3/2024 6:42:01 PM | | VirtualBox version: 7.0.18
7/3/2024 6:42:01 PM | | Last CPU benchmark was 30 days 01:32:12 ago
7/3/2024 6:42:06 PM | Einstein@Home | General prefs: from Einstein@Home (last modified 27-May-2024 16:33:32)
7/3/2024 6:42:06 PM | Einstein@Home | Host location: none
7/3/2024 6:42:06 PM | Einstein@Home | General prefs: using your defaults
7/3/2024 6:42:06 PM | | Preferences:
7/3/2024 6:42:06 PM | | - When computer is in use
7/3/2024 6:42:06 PM | | - 'In use' means mouse/keyboard input in last 1.00 minutes
7/3/2024 6:42:06 PM | | - max CPUs used: 10
7/3/2024 6:42:06 PM | | - Use at most 100% of the CPU time
7/3/2024 6:42:06 PM | | - suspend if non-BOINC CPU load exceeds 99%
7/3/2024 6:42:06 PM | | - max memory usage: 28.53 GB
7/3/2024 6:42:06 PM | | - When computer is not in use
7/3/2024 6:42:06 PM | | - max CPUs used: 10
7/3/2024 6:42:06 PM | | - Use at most 100% of the CPU time
7/3/2024 6:42:06 PM | | - suspend if non-BOINC CPU load exceeds 99%
7/3/2024 6:42:06 PM | | - max memory usage: 31.70 GB
7/3/2024 6:42:06 PM | | - Leave apps in memory if not running
7/3/2024 6:42:06 PM | | - Store at least 0.50 days of work
7/3/2024 6:42:06 PM | | - Store up to an additional 0.00 days of work
7/3/2024 6:42:06 PM | | - max disk usage: 100.00 GB
7/3/2024 6:42:06 PM | | - (to change preferences, visit a project web site or select Preferences in the Manager)
7/3/2024 6:42:06 PM | | Setting up project and slot directories
7/3/2024 6:42:06 PM | | Checking active tasks
7/3/2024 6:42:06 PM | | Using account manager BOINCstatsBAM!
7/3/2024 6:42:06 PM | Einstein@Home | URL https://einstein.phys.uwm.edu/; Computer ID 13185787; resource share 100
7/3/2024 6:42:06 PM | Einstein@Home | Not using CPU: project preferences
7/3/2024 6:42:06 PM | Milkyway@home | URL https://milkyway.cs.rpi.edu/milkyway/; Computer ID 1025780; resource share 100
7/3/2024 6:42:06 PM | | Setting up GUI RPC socket
7/3/2024 6:42:06 PM | | Checking presence of 30 project files
7/3/2024 6:42:06 PM | | Running CPU benchmarks
7/3/2024 6:42:06 PM | | Suspending computation - CPU benchmarks in progress
7/3/2024 6:42:37 PM | | Benchmark results:
7/3/2024 6:42:37 PM | | Number of CPUs: 10
7/3/2024 6:42:37 PM | | 3039 floating point MIPS (Whetstone) per CPU
7/3/2024 6:42:37 PM | | 18341 integer MIPS (Dhrystone) per CPU
7/3/2024 6:42:38 PM | | Contacting account manager at https://bam.boincstats.com/
7/3/2024 6:42:39 PM | | Account manager: BAM! User: 202077, James Bradshaw
7/3/2024 6:42:39 PM | | Account manager: BAM! Host: 1302054
7/3/2024 6:42:39 PM | | Account manager: Number of BAM! connections for this host: 281
7/3/2024 6:42:39 PM | | Account manager contact succeeded
7/3/2024 6:42:59 PM | Einstein@Home | update requested by user
7/3/2024 6:43:01 PM | Einstein@Home | Sending scheduler request: Requested by user.
7/3/2024 6:43:01 PM | Einstein@Home | Requesting new tasks for NVIDIA GPU and Intel GPU
7/3/2024 6:43:03 PM | Einstein@Home | Scheduler request completed: got 0 new tasks
7/3/2024 6:43:03 PM | Einstein@Home | No work sent
7/3/2024 6:43:03 PM | Einstein@Home | Project requested delay of 60 seconds
7/3/2024 6:43:08 PM | Einstein@Home | update requested by user
7/3/2024 6:43:08 PM | Einstein@Home | Sending scheduler request: Requested by user.
7/3/2024 6:43:08 PM | Einstein@Home | Requesting new tasks for NVIDIA GPU and Intel GPU
7/3/2024 6:43:10 PM | Einstein@Home | Scheduler request completed: got 0 new tasks
7/3/2024 6:43:10 PM | Einstein@Home | Not sending work - last request too recent: 8 sec
7/3/2024 6:43:10 PM | Einstein@Home | Project requested delay of 60 seconds
7/3/2024 6:43:33 PM | | Fetching configuration file from https://bam.boincstats.com/get_project_config.php
7/3/2024 6:43:36 PM | | Contacting account manager at https://bam.boincstats.com/
7/3/2024 6:43:37 PM | | Account manager: BAM! User: 202077, James Bradshaw
7/3/2024 6:43:37 PM | | Account manager: BAM! Host: 1302054
7/3/2024 6:43:37 PM | | Account manager: Number of BAM! connections for this host: 282
7/3/2024 6:43:37 PM | | Account manager contact succeeded
7/3/2024 6:44:11 PM | Einstein@Home | Sending scheduler request: To fetch work.
7/3/2024 6:44:11 PM | Einstein@Home | Requesting new tasks for NVIDIA GPU and Intel GPU
7/3/2024 6:44:14 PM | Einstein@Home | Scheduler request completed: got 0 new tasks
7/3/2024 6:44:14 PM | Einstein@Home | No work sent
7/3/2024 6:44:14 PM | Einstein@Home | Project requested delay of 60 seconds
7/3/2024 6:44:19 PM | Milkyway@home | Sending scheduler request: To fetch work.
7/3/2024 6:44:19 PM | Milkyway@home | Reporting 1 completed tasks
7/3/2024 6:44:19 PM | Milkyway@home | Requesting new tasks for CPU
7/3/2024 6:44:21 PM | Milkyway@home | Scheduler request completed: got 1 new tasks
7/3/2024 6:44:21 PM | Milkyway@home | Project requested delay of 91 seconds
7/3/2024 6:45:56 PM | Milkyway@home | Sending scheduler request: To fetch work.
7/3/2024 6:45:56 PM | Milkyway@home | Requesting new tasks for CPU
7/3/2024 6:45:57 PM | Milkyway@home | Scheduler request completed: got 1 new tasks
7/3/2024 6:45:57 PM | Milkyway@home | Project requested delay of 91 seconds
7/3/2024 6:47:21 PM | Milkyway@home | Computation for task de_nbody_07_01_2024_v186_pal5__data__13_1718187602_1531020_0 finished
7/3/2024 6:47:22 PM | Milkyway@home | Starting task de_nbody_07_01_2024_v186_pal5__data__15_1718187602_1531069_0
Yes. Thank you. Might
)
Yes. Thank you.
Might setup a separate e@h project profile for this system. Then you can try out different settings without breaking all your other systems.
Select brp7 only. No test or non-requested apps. GPU processing only. Nvidia only.
If it comes back with no tasks available message rather than no tasks sent there is a different issue.
Tom M
A Proud member of the O.F.A. (Old Farts Association). Be well, do good work, and keep in touch.® (Garrison Keillor) I want some more patience. RIGHT NOW!
Just a heads-up. Anyone with
)
Just a heads-up. Anyone with recent NVIDIA GPU (RTX 30- and 40-series) might want to check their GPU hotspot temperature. I don't think the single temperature reported by nvidia-smi is it.
Bouncing off thermal throttling sounds like a good way to kill them fast with more-or-less constant load like E@H, though it might take longer to get to that point in the first place than typical game load would, when half of the problem is thermal paste pump-out.
Related reads: Igor's Lab, Tom's Hardware
Tom M wrote: Yes. Thank
)
You do not need to make a separate E@h project profile.
Only a new set of Preferences under your profile & add that computer (which you are testing) under that Preferences (which has 4 settings).
Hope this helps.
non-profit org. Play4Life in Zagreb, Croatia, EU
https://www.pcgamer.com/hardw
)
https://www.pcgamer.com/hardware/graphics-cards/nvidias-third-party-rtx-40-series-gpus-are-losing-performance-over-time-thanks-to-rubbish-factory-installed-thermal-paste/
I hope we will not be afflicted.
A Proud member of the O.F.A. (Old Farts Association). Be well, do good work, and keep in touch.® (Garrison Keillor) I want some more patience. RIGHT NOW!