Wrong runtimes
log in

Advanced search

Message boards : Number crunching : Wrong runtimes

Author Message
Werinbert
Send message
Joined: 9 Dec 14
Posts: 18
Credit: 16,238,357
RAC: 28,211
Message 11121 - Posted: 15 Oct 2025, 16:38:10 UTC

The runtimes (not cputimes) presented in the task lists are wildly wrong and the app should be fixed to correct this data.

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7992
Credit: 44,538,964
RAC: 1
Message 11122 - Posted: 15 Oct 2025, 16:47:06 UTC - in response to Message 11121.
Last modified: 15 Oct 2025, 16:56:54 UTC

The runtimes (not cputimes) presented in the task lists are wildly wrong and the app should be fixed to correct this data.


As written in the FAQ these values are based on an i7-4930k@AVX. It is different on your hosts. You also can see a note in brackets AVX@3.8Ghz. AVX2 is 2 times faster, AVX512 3 times.

Werinbert
Send message
Joined: 9 Dec 14
Posts: 18
Credit: 16,238,357
RAC: 28,211
Message 11123 - Posted: 15 Oct 2025, 21:48:51 UTC
Last modified: 15 Oct 2025, 21:49:10 UTC

Runtime should be the time that my computer takes to run the task not some theoretical value.

Your faq is about credit and estimated runtime for a specific computer not actual runtime for my (or anyone else's) computer. So in the runtime column of the list of tasks have you made it to show the theoretical value rather than the (Boinc default) actual runtime?
____________
35376214*3^107944-1 is prime! (20 Dec 2014)

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7992
Credit: 44,538,964
RAC: 1
Message 11124 - Posted: 15 Oct 2025, 23:19:27 UTC - in response to Message 11123.

Runtime should be the time that my computer takes to run the task not some theoretical value.

Your faq is about credit and estimated runtime for a specific computer not actual runtime for my (or anyone else's) computer. So in the runtime column of the list of tasks have you made it to show the theoretical value rather than the (Boinc default) actual runtime?


Not really. It is only an estimated value of a based host. BOINC calculates the runtime of your host but can be different because the tests changes.

Werinbert
Send message
Joined: 9 Dec 14
Posts: 18
Credit: 16,238,357
RAC: 28,211
Message 11130 - Posted: 19 Oct 2025, 6:45:37 UTC

A snippet from my job logs:

1760745897 ue 6826.734223 ct 21387.940000 fe 80000000000 nm R24_800-900k_wu_40862_0 et 5502.368664 es 0 1760750353 ue 7147.987346 ct 27522.200000 fe 80000000000 nm R37_700-900k_wu_26787_0 et 7064.325409 es 0 1760751401 ue 6846.506881 ct 21371.000000 fe 80000000000 nm R24_800-900k_wu_42090_0 et 5503.180866 es 0 1760755748 ue 6865.931812 ct 20969.510000 fe 80000000000 nm R24_800-900k_wu_42903_0 et 5394.826078 es 0 1760758365 ue 7179.557473 ct 27139.210000 fe 80000000000 nm R37_700-900k_wu_28176_0 et 6962.318829 es 0 1760761122 ue 6877.702494 ct 20957.250000 fe 80000000000 nm R24_800-900k_wu_44061_0 et 5372.188438 es 0 1760764159 ue 7204.821898 ct 22469.450000 fe 80000000000 nm R37_700-900k_wu_29496_0 et 5793.388176 es 0 1760766515 ue 6893.025469 ct 21053.350000 fe 80000000000 nm R24_800-900k_wu_44835_0 et 5392.735672 es 0 1760768700 ue 6893.025469 ct 17592.300000 fe 80000000000 nm R24_800-900k_wu_45467_0 et 4540.006973 es 0 1760772823 ue 7204.821898 ct 24564.900000 fe 80000000000 nm R37_900k-1M_wu_30876_0 et 6306.978279 es 0 1760774099 ue 6918.784162 ct 21007.240000 fe 80000000000 nm R24_800-900k_wu_46954_0 et 5397.936483 es 0 1760777857 ue 6936.246320 ct 19607.160000 fe 80000000000 nm R24_800-900k_wu_47801_0 et 5032.471640 es 0 1760779147 ue 6936.246320 ct 19650.430000 fe 80000000000 nm R24_800-900k_wu_48223_0 et 5047.087866 es 0 1760783378 ue 6962.883019 ct 21314.480000 fe 80000000000 nm R24_800-900k_wu_49029_0 et 5519.772696 es 0 1760786639 ue 7246.576842 ct 29153.690000 fe 80000000000 nm R37_900k-1M_wu_32844_0 et 7490.541756 es 0 1760788881 ue 6980.384189 ct 21365.970000 fe 80000000000 nm R24_800-900k_wu_50145_0 et 5502.374666 es 0 1760791175 ue 6980.384189 ct 17562.840000 fe 80000000000 nm R24_800-900k_wu_50776_0 et 4534.669765 es 0 1760794308 ue 7004.674443 ct 21174.060000 fe 80000000000 nm R24_800-900k_wu_52744_0 et 5426.305931 es 0 1760796548 ue 7004.674443 ct 20962.900000 fe 80000000000 nm R24_800-900k_wu_53121_0 et 5372.908035 es 0 1760798055 ue 7004.674443 ct 14486.850000 fe 80000000000 nm R24_800-900k_wu_54039_0 et 3745.433180 es 0 1760801617 ue 7037.258293 ct 19653.680000 fe 80000000000 nm R24_800-900k_wu_54482_0 et 5067.213902 es 0 1760805601 ue 7281.561074 ct 29440.040000 fe 80000000000 nm R37_900k-1M_wu_36027_0 et 7544.829853 es 0 1760807160 ue 7049.625710 ct 21523.930000 fe 80000000000 nm R24_800-900k_wu_55501_0 et 5542.228735 es 0 1760811184 ue 7067.519453 ct 21725.350000 fe 80000000000 nm R24_800-900k_wu_56412_0 et 5582.575405 es 0 1760812750 ue 7067.519453 ct 21725.570000 fe 80000000000 nm R24_800-900k_wu_56714_0 et 5589.303019 es 0 1760816748 ue 7102.532615 ct 21675.560000 fe 80000000000 nm R24_800-900k_wu_57670_0 et 5563.151371 es 0 1760817850 ue 7102.532615 ct 19829.080000 fe 80000000000 nm R24_800-900k_wu_57966_0 et 5098.718954 es 0 1760822313 ue 7131.611816 ct 21657.280000 fe 80000000000 nm R24_800-900k_wu_58735_0 et 5563.570574 es 0 1760823422 ue 7131.611816 ct 21640.800000 fe 80000000000 nm R24_800-900k_wu_59018_0 et 5571.219587 es 0 1760827798 ue 7165.081141 ct 21328.640000 fe 80000000000 nm R24_800-900k_wu_60026_0 et 5483.656631 es 0 1760828941 ue 7165.081141 ct 21427.140000 fe 80000000000 nm R24_800-900k_wu_60354_0 et 5517.506691 es 0 1760834123 ue 7165.081141 ct 20112.740000 fe 80000000000 nm R24_800-900k_wu_61608_0 et 5181.458500 es 0 1760836929 ue 7317.245463 ct 35719.330000 fe 80000000000 nm R37_900k-1M_wu_40607_0 et 9129.877235 es 0 1760840944 ue 7317.245463 ct 26484.070000 fe 80000000000 nm R37_900k-1M_wu_41433_0 et 6819.400978 es 0 1760842426 ue 7205.575660 ct 21366.460000 fe 80000000000 nm R24_800-900k_wu_63093_0 et 5496.191055 es 0 1760846446 ue 7220.619904 ct 21387.750000 fe 80000000000 nm R24_800-900k_wu_64059_0 et 5501.186062 es 0 1760847970 ue 7220.619904 ct 21549.650000 fe 80000000000 nm R24_800-900k_wu_64603_0 et 5543.212536 es 0 1760852088 ue 7245.609768 ct 22010.630000 fe 80000000000 nm R24_800-900k_wu_65277_0 et 5641.496708 es 0 1760853165 ue 7245.609768 ct 20165.780000 fe 80000000000 nm R24_800-900k_wu_65539_0 et 5193.680323 es 0


The et value is the elapsed time and is what should be used for the runtime values (ct is the cputime). If comparing these data to those presented in tasks page you will find that the et value and runtimes do not match (cpu times however do match).
____________
35376214*3^107944-1 is prime! (20 Dec 2014)

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7992
Credit: 44,538,964
RAC: 1
Message 11131 - Posted: 19 Oct 2025, 8:13:45 UTC

To be more precise, BOINC use this formula

<rsc_fpops_est> / <flops> * <duration_correction_factor> //dcf=1

Werinbert
Send message
Joined: 9 Dec 14
Posts: 18
Credit: 16,238,357
RAC: 28,211
Message 11132 - Posted: 19 Oct 2025, 16:49:40 UTC

I believe that is for the estimated time (ue) not elapsed time (et).
____________
35376214*3^107944-1 is prime! (20 Dec 2014)

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7992
Credit: 44,538,964
RAC: 1
Message 11133 - Posted: 19 Oct 2025, 17:29:18 UTC - in response to Message 11132.

I believe that is for the estimated time (ue) not elapsed time (et).


yes, must be in client_state.xml


Post to thread

Message boards : Number crunching : Wrong runtimes


Main page · Your account · Message boards


Copyright © 2014-2025 BOINC Confederation / rebirther