Strange behavior from i7-12700K

liontaur
liontaur
Joined: 15 Feb 22
Posts: 2
Credit: 12546235
RAC: 0
Topic 227007

I've just started with Einstein@home so i'm not overly familiar with it. I've decided to start with a new build and i'm seeing some strange behavior. I've got an i7-12700K CPU with 32GB of DDR5 RAM and just a crappy GeForce 210 card as I don't plan to do anything GPU with this system. It's running Windows 10 with all updates.

As for the weird behavior, when BOINC starts up, it runs 100% across all 20 threads for about 50 seconds and then just four threads continue at 100%. The rest all go down to pretty much zero. BOINC still says that all 20 threads are running but task manager shows just four going, the last four - CPUS 16-19. I've rebooted a few times and get the same thing.

I've changed my E@h computing preferences to the maxmimum preset already.

Not sure if there's something up with the CPU scheduler or what.

Any help would be most appreciated!

Here's my BOINC event log:

2/18/2022 8:18:27 AM |  | cc_config.xml not found - using defaults
2/18/2022 8:18:27 AM |  | Starting BOINC client version 7.16.20 for windows_x86_64
2/18/2022 8:18:27 AM |  | log flags: file_xfer, sched_ops, task
2/18/2022 8:18:27 AM |  | Libraries: libcurl/7.47.1 OpenSSL/1.0.2s zlib/1.2.8
2/18/2022 8:18:27 AM |  | Data directory: C:\ProgramData\BOINC
2/18/2022 8:18:27 AM |  | Running under account <redacted>
2/18/2022 8:18:27 AM |  | CUDA: NVIDIA GPU 0: GeForce 210 (driver version 341.74, CUDA version 6.5, compute capability 1.2, 1024MB, 871MB available, 67 GFLOPS peak)
2/18/2022 8:18:27 AM |  | OpenCL: NVIDIA GPU 0: GeForce 210 (driver version 341.74, device version OpenCL 1.0 CUDA, 1024MB, 871MB available, 67 GFLOPS peak)
2/18/2022 8:18:27 AM |  | Windows processor group 0: 20 processors
2/18/2022 8:18:27 AM |  | Host name: ALIB-ML11Z
2/18/2022 8:18:27 AM |  | Processor: 20 GenuineIntel 12th Gen Intel(R) Core(TM) i7-12700K [Family 6 Model 151 Stepping 2]
2/18/2022 8:18:27 AM |  | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx smx tm2 pbe fsgsbase bmi1 smep bmi2
2/18/2022 8:18:27 AM |  | OS: Microsoft Windows 10: Professional x64 Edition, (10.00.19044.00)
2/18/2022 8:18:27 AM |  | Memory: 31.79 GB physical, 36.54 GB virtual
2/18/2022 8:18:27 AM |  | Disk: 1.82 TB total, 1.73 TB free
2/18/2022 8:18:27 AM |  | Local time is UTC -8 hours
2/18/2022 8:18:27 AM |  | No WSL found.
2/18/2022 8:18:27 AM |  | VirtualBox version: 6.0.14
2/18/2022 8:18:27 AM | Einstein@Home | General prefs: from Einstein@Home (last modified 15-Feb-2022 09:46:29)
2/18/2022 8:18:27 AM | Einstein@Home | Host location: none
2/18/2022 8:18:27 AM | Einstein@Home | General prefs: using your defaults
2/18/2022 8:18:27 AM |  | Reading preferences override file
2/18/2022 8:18:27 AM |  | Preferences:
2/18/2022 8:18:27 AM |  | max memory usage when active: 24417.70 MB
2/18/2022 8:18:27 AM |  | max memory usage when idle: 29301.24 MB
2/18/2022 8:18:27 AM |  | max disk usage: 1676.15 GB
2/18/2022 8:18:27 AM |  | (to change preferences, visit a project web site or select Preferences in the Manager)
2/18/2022 8:18:27 AM |  | Setting up project and slot directories
2/18/2022 8:18:27 AM |  | Checking active tasks
2/18/2022 8:18:27 AM | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12919290; resource share 100
2/18/2022 8:18:27 AM |  | Setting up GUI RPC socket
2/18/2022 8:18:27 AM |  | Checking presence of 68 project files
2/18/2022 8:18:30 AM | Einstein@Home | Sending scheduler request: To fetch work.
2/18/2022 8:18:30 AM | Einstein@Home | Requesting new tasks for NVIDIA GPU
2/18/2022 8:18:32 AM | Einstein@Home | Scheduler request completed: got 0 new tasks
2/18/2022 8:18:32 AM | Einstein@Home | No work sent
2/18/2022 8:18:32 AM | Einstein@Home | see scheduler log messages on https://einsteinathome.org/host/12919290/log
2/18/2022 8:18:32 AM | Einstein@Home | Binary Radio Pulsar Search (Arecibo) is not available for your type of computer.
2/18/2022 8:18:32 AM | Einstein@Home | Project requested delay of 60 seconds

 

And here's my scheduler log messages:

2022-02-18 16:16:55.5731 [PID=13714]   Request: [USER#xxxxx] [HOST#12919290] [IP xxx.xxx.xxx.243] client 7.16.20
2022-02-18 16:16:55.6782 [PID=13714] [debug]   [HOST#12919290] Resetting nresults_today
2022-02-18 16:16:55.6782 [PID=13714] [debug]   have_master:1 have_working: 1 have_db: 1
2022-02-18 16:16:55.6783 [PID=13714] [debug]   using working prefs
2022-02-18 16:16:55.6783 [PID=13714] [debug]   have db 1; dbmod 1644947189.000000; global mod 1644947189.000000
2022-02-18 16:16:55.6783 [PID=13714]    [send] DCF=10.415557; setting to 10
2022-02-18 16:16:55.6783 [PID=13714]    [send] effective_ncpus 20 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2022-02-18 16:16:55.6783 [PID=13714]    [send] effective_ngpus 1 max_jobs_on_host_gpu 999999
2022-02-18 16:16:55.6783 [PID=13714]    [send] Not using matchmaker scheduling; Not using EDF sim
2022-02-18 16:16:55.6783 [PID=13714]    [send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00
2022-02-18 16:16:55.6784 [PID=13714]    [send] CUDA: req 51840.00 sec, 1.00 instances; est delay 0.00
2022-02-18 16:16:55.6784 [PID=13714]    [send] work_req_seconds: 0.00 secs
2022-02-18 16:16:55.6784 [PID=13714]    [send] available disk 99.96 GB, work_buf_min 8640
2022-02-18 16:16:55.6784 [PID=13714]    [send] active_frac 0.999974 on_frac 0.999151 DCF 10.415557
2022-02-18 16:16:55.6798 [PID=13714]    [mixed] sending locality work first (0.2022)
2022-02-18 16:16:55.6814 [PID=13714]    [send] send_old_work() no feasible result older than 336.0 hours
2022-02-18 16:16:55.6824 [PID=13714]    [send] send_old_work() no feasible result younger than 171.2 hours and older than 168.0 hours
2022-02-18 16:16:55.6854 [PID=13714]    [version] Checking plan class 'GW-opencl-ati'
2022-02-18 16:16:55.6901 [PID=13714]    [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2022-02-18 16:16:55.6902 [PID=13714]    [version] parsed project prefs setting 'gpu_util_gw': 0.000000
2022-02-18 16:16:55.6902 [PID=13714]    [version] No ATI devices found
2022-02-18 16:16:55.6902 [PID=13714]    [version] Checking plan class 'GW-opencl-nvidia'
2022-02-18 16:16:55.6902 [PID=13714]    [version] parsed project prefs setting 'gpu_util_gw': 0.000000
2022-02-18 16:16:55.6902 [PID=13714]    [version] GPU RAM calculated: min: 512 MB, use: 300 MB, WU#608218757 CPU: 300 MB
2022-02-18 16:16:55.6902 [PID=13714]    [version] device (or driver) doesn't support double precision fp math
2022-02-18 16:16:55.6903 [PID=13714]    [version] no app version available: APP#56 (einstein_O3AS) PLATFORM#9 (windows_x86_64) min_version 0
2022-02-18 16:16:55.6903 [PID=13714]    [version] no app version available: APP#56 (einstein_O3AS) PLATFORM#2 (windows_intelx86) min_version 0
2022-02-18 16:16:55.6905 [PID=13714]    [mixed] sending non-locality work second
2022-02-18 16:16:55.7193 [PID=13714]    [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#9 (windows_x86_64) min_version 0
2022-02-18 16:16:55.7194 [PID=13714]    [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#2 (windows_intelx86) min_version 0
2022-02-18 16:16:55.7195 [PID=13714]    [version] Checking plan class 'FGRPopencl-ati'
2022-02-18 16:16:55.7195 [PID=13714]    [version] parsed project prefs setting 'gpu_util_fgrp': 0.000000
2022-02-18 16:16:55.7195 [PID=13714]    [version] No ATI devices found
2022-02-18 16:16:55.7195 [PID=13714]    [version] Checking plan class 'FGRPopencl-intel_gpu'
2022-02-18 16:16:55.7195 [PID=13714]    [version] parsed project prefs setting 'gpu_util_fgrp': 0.000000
2022-02-18 16:16:55.7195 [PID=13714]    [version] No Intel GPU devices found
2022-02-18 16:16:55.7195 [PID=13714]    [version] Checking plan class 'FGRPopencl-nvidia'
2022-02-18 16:16:55.7195 [PID=13714]    [version] parsed project prefs setting 'gpu_util_fgrp': 0.000000
2022-02-18 16:16:55.7195 [PID=13714]    [version] GPU RAM calculated: min: 766 MB, use: 750 MB, WU#608208701 CPU: 429 MB
2022-02-18 16:16:55.7196 [PID=13714]    [version] OpenCL device version required min: 101, supplied: 100
2022-02-18 16:16:55.7196 [PID=13714]    [version] Checking plan class 'FGRPopencl1K-ati'
2022-02-18 16:16:55.7196 [PID=13714]    [version] parsed project prefs setting 'gpu_util_fgrp': 0.000000
2022-02-18 16:16:55.7196 [PID=13714]    [version] No ATI devices found
2022-02-18 16:16:55.7196 [PID=13714]    [version] Checking plan class 'FGRPopencl1K-nvidia'
2022-02-18 16:16:55.7196 [PID=13714]    [version] parsed project prefs setting 'gpu_util_fgrp': 0.000000
2022-02-18 16:16:55.7196 [PID=13714]    [version] GPU RAM calculated: min: 1000 MB, use: 750 MB, WU#608208701 CPU: 429 MB
2022-02-18 16:16:55.7196 [PID=13714]    [version] OpenCL device version required min: 101, supplied: 100
2022-02-18 16:16:55.7196 [PID=13714]    [version] Checking plan class 'FGRPopenclTV-nvidia'
2022-02-18 16:16:55.7196 [PID=13714]    [version] parsed project prefs setting 'gpu_util_fgrp': 0.000000
2022-02-18 16:16:55.7197 [PID=13714]    [version] GPU RAM calculated: min: 1000 MB, use: 750 MB, WU#608208701 CPU: 429 MB
2022-02-18 16:16:55.7197 [PID=13714]    [version] OpenCL device version required min: 101, supplied: 100
2022-02-18 16:16:55.7197 [PID=13714]    [version] Checking plan class 'FGRPopencl2-ati'
2022-02-18 16:16:55.7197 [PID=13714]    [version] parsed project prefs setting 'gpu_util_fgrp': 0.000000
2022-02-18 16:16:55.7197 [PID=13714]    [version] No ATI devices found
2022-02-18 16:16:55.7197 [PID=13714]    [version] Checking plan class 'FGRPopencl2Pup-nvidia'
2022-02-18 16:16:55.7197 [PID=13714]    [version] parsed project prefs setting 'gpu_util_fgrp': 0.000000
2022-02-18 16:16:55.7197 [PID=13714]    [version] GPU RAM calculated: min: 1000 MB, use: 750 MB, WU#608208701 CPU: 429 MB
2022-02-18 16:16:55.7197 [PID=13714]    [version] OpenCL device version required min: 200, supplied: 100
2022-02-18 16:16:55.7197 [PID=13714]    [version] no app version available: APP#40 (hsgamma_FGRPB1G) PLATFORM#9 (windows_x86_64) min_version 0
2022-02-18 16:16:55.7197 [PID=13714]    [version] no app version available: APP#40 (hsgamma_FGRPB1G) PLATFORM#2 (windows_intelx86) min_version 0
2022-02-18 16:16:55.7198 [PID=13714]    [version] Checking plan class 'opencl-intel_gpu'
2022-02-18 16:16:55.7198 [PID=13714]    [version] parsed project prefs setting 'gpu_util_brp': 0.000000
2022-02-18 16:16:55.7198 [PID=13714]    [version] No Intel GPU devices found
2022-02-18 16:16:55.7198 [PID=13714]    [version] Checking plan class 'opencl-intel_gpu-Beta'
2022-02-18 16:16:55.7198 [PID=13714]    [version] beta test app versions not allowed in project prefs.
2022-02-18 16:16:55.7198 [PID=13714]    [version] Checking plan class 'opencl-intel_gpu-new'
2022-02-18 16:16:55.7198 [PID=13714]    [version] parsed project prefs setting 'gpu_util_brp': 0.000000
2022-02-18 16:16:55.7199 [PID=13714]    [version] No Intel GPU devices found
2022-02-18 16:16:55.7199 [PID=13714]    [version] Checking plan class 'opencl-intel_gpu'
2022-02-18 16:16:55.7199 [PID=13714]    [version] parsed project prefs setting 'gpu_util_brp': 0.000000
2022-02-18 16:16:55.7199 [PID=13714]    [version] No Intel GPU devices found
2022-02-18 16:16:55.7199 [PID=13714]    [version] no app version available: APP#25 (einsteinbinary_BRP4G) PLATFORM#9 (windows_x86_64) min_version 0
2022-02-18 16:16:55.7199 [PID=13714]    [version] no app version available: APP#25 (einsteinbinary_BRP4G) PLATFORM#2 (windows_intelx86) min_version 0
2022-02-18 16:16:55.7200 [PID=13714]    [version] Checking plan class 'FGRPSSE'
2022-02-18 16:16:55.7200 [PID=13714]    [version] plan class ok
2022-02-18 16:16:55.7200 [PID=13714]    [version] Don't need CPU jobs, skipping version 108 for hsgamma_FGRP5 (FGRPSSE)
2022-02-18 16:16:55.7200 [PID=13714]    [version] no app version available: APP#46 (hsgamma_FGRP5) PLATFORM#9 (windows_x86_64) min_version 0
2022-02-18 16:16:55.7200 [PID=13714]    [version] no app version available: APP#46 (hsgamma_FGRP5) PLATFORM#2 (windows_intelx86) min_version 0
2022-02-18 16:16:55.7269 [PID=13714] [debug]   [HOST#12919290] MSG(high) No work sent
2022-02-18 16:16:55.7270 [PID=13714] [debug]   [HOST#12919290] MSG(high) see scheduler log messages on https://einsteinathome.org/host/12919290/log
2022-02-18 16:16:55.7270 [PID=13714] [debug]   [HOST#12919290] MSG(high) Binary Radio Pulsar Search (Arecibo) is not available for your type of computer.
2022-02-18 16:16:55.7270 [PID=13714]    Sending reply to [HOST#12919290]: 0 results, delay req 60.00
2022-02-18 16:16:55.7271 [PID=13714]    Scheduler ran 0.158 seconds

 

GWGeorge007
GWGeorge007
Joined: 8 Jan 18
Posts: 3075
Credit: 4975337686
RAC: 1433217

liontaur wrote:I've just

Welcome to Einstein!!

liontaur wrote:

I've just started with Einstein@home so i'm not overly familiar with it. I've decided to start with a new build and i'm seeing some strange behavior. I've got an i7-12700K CPU with 32GB of DDR5 RAM and just a crappy GeForce 210 card as I don't plan to do anything GPU with this system. It's running Windows 10 with all updates.

As for the weird behavior, when BOINC starts up, it runs 100% across all 20 threads for about 50 seconds and then just four threads continue at 100%. The rest all go down to pretty much zero. BOINC still says that all 20 threads are running but task manager shows just four going, the last four - CPUS 16-19. I've rebooted a few times and get the same thing.

I'm not familiar with the new 12 Gen Intel CPUs such as yours, but when looking at the Intel spec sheet (below), it appears that your CPU reverts to the 4 Efficient-Cores after the initial 50 secs of run time.  This may be a quirk of this (these?) CPUs.

https://www.intel.com/content/www/us/en/products/sku/134594/intel-core-i712700k-processor-25m-cache-up-to-5-00-ghz/specifications.html

George

Proud member of the Old Farts Association

Ian&Steve C.
Ian&Steve C.
Joined: 19 Jan 20
Posts: 3965
Credit: 47220032642
RAC: 65374860

Probably because Windows10

Probably because Windows10 doesn’t know what to do with the differing P and E cores. 
 

moving to Windows11 might help. But disabling the E cores in the BIOS will probably provide the best solution. 

_________________________________________________________________________

liontaur
liontaur
Joined: 15 Feb 22
Posts: 2
Credit: 12546235
RAC: 0

That's it. Thanks to both of

That's it. Thanks to both of you for the direction! I've disabled the E cores now and the 16 remaining threads are maxed out for the last 30 minutes.

 

Thanks again folks!

treaclepumpkin
treaclepumpkin
Joined: 22 May 17
Posts: 1
Credit: 144370334
RAC: 108597

I've had the same problem for

I've had the same problem for weeks. Been searching for a solution, ovbviously using the wrong key words but finally found this thread. So disabled the E-cores and everything running at 100% across the remaining 6 P-cores.

Strange, thanks for the tip! Now to dig further though, maybe it is only a Windows 10 thing and 11 can handle the E-cores correctly, not ready for that move yet though.

Ian&Steve C.
Ian&Steve C.
Joined: 19 Jan 20
Posts: 3965
Credit: 47220032642
RAC: 65374860

yes, it's windows 10 vs

yes, it's windows 10 vs windows 11.

Windows 11 can handle the E cores
Windows 10 cannot. so you have to disable E-cores in BIOS.

_________________________________________________________________________

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.