Task in question: https://boinc.bakerlab.org/rosetta/result.php?resultid=1431763806
So what was Rosetta doing with all that computer time allocated to it if not for crunching?

If the app was actually crunching during that time why are the credits so low?

Task in question: https://boinc.bakerlab.org/rosetta/result.php?resultid=1431763806
So what was Rosetta doing with all that computer time allocated to it if not for crunching?

It wasn't doing anything- some other programme on your computer was using that extra time. That's why it took over 9 hours (Run time) to do 8 hours worth of work (CPU Time)- which is the case for all of your Rosetta Tasks- it takes much more Run time than CPU to complete a Task. Check the Run time v CPU time for my systems for a comparison.
It is the same for Prime Grid- it takes that system 19 hours to do 10 hours worth of work on a Task there.


Have a look in Task Manager & see what it is that's eating up all of your CPU time.

If the app was actually crunching during that time why are the credits so low?

No idea why that is.
It is a new host, and it can take a while for Credit to settle down (usually 10 or so Valid Tasks), but even so that's an exceptionally low amount of Credit.

It is a new host, and it can take a while for Credit to settle down (usually 10 or so Valid Tasks), but even so that's an exceptionally low amount of Credit.

Your one task that gave you 3 credits only generated 16 models. That's pretty odd. I ran one task for 7hrs, generated 1 model and got only 107 credits.

To run 9 hrs for 8hrs work or 7hrs for 6hrs work is just normal with our systems.

Run time 8 hours  0 min 41 sec
CPU time 7 hours 59 min  5 sec

It is a new host, and it can take a while for Credit to settle down (usually 10 or so Valid Tasks), but even so that's an exceptionally low amount of Credit.

Yes, it is a new host but all the rest of the task have been ball park consistent in the credits granted. And the typical task's cr/hr is what I have come to expect from this host. But this one task was granted only 1/100th of the credit expected.

(I am not worried about the runtime vs CPU time....I am pushing that computer pretty hard with other projects, so just look at the CPU time used.)

To run 9 hrs for 8hrs work or 7hrs for 6hrs work is just normal with our systems.

No it isn't.
It just means that something else is making significant use of the CUPU while Rosetta is running.


Checked task manager currently 85-86% BOINC, FAH (CPU to support GPU) - 6.1 and 6.2 and something cmdock 5.6%. Browsers high fluctuations. Between .1 and 5% every second, but mostly .1%
I would think FAH is using the free system core to manage itself since only 15 are in use for BOINC.

Why is alk using 82% of a core and degrader is using 81% and another degrader is using 84%?
I check my SiDock tasks and they use 90%+ of a core with VM. ( How much does VM use in CPU?)

And:


System specs:

Me -Measured floating point speed 5365.31 million ops/sec
Measured integer speed 19865.92 million ops/sec
Memory 24501.59 MB

Other - Measured floating point speed 5279.76 million ops/sec
Measured integer speed 19321.67 million ops/sec
Memory 16310.63 MB

-------------

eg This Ryzen 7 3700X system has good CPU time v Run time numbers.

Run time 8 hours  0 min 41 sec
CPU time 7 hours 59 min  5 sec

tau_site11_7nrq_graft_bcov_v1_xad_SAVE_ALL_OUT_IGNORE_THE_REST_2jt0hl2g_1824323_1_0

To run 9 hrs for 8hrs work or 7hrs for 6hrs work is just normal with our systems.

No it isn't.
It just means that something else is making significant use of the CUPU while Rosetta is running.


Checked task manager currently 85-86% BOINC, FAH (CPU to support GPU) - 6.1 and 6.2 and something cmdock 5.6%. Browsers high fluctuations. Between .1 and 5% every second, but mostly .1%
I would think FAH is using the free system core to manage itself since only 15 are in use for BOINC.

Why is alk using 82% of a core and degrader is using 81% and another degrader is using 84%?
I check my SiDock tasks and they use 90%+ of a core with VM. ( How much does VM use in CPU?)

And:


System specs:

Me -Measured floating point speed 5365.31 million ops/sec
Measured integer speed 19865.92 million ops/sec
Memory 24501.59 MB

Other - Measured floating point speed 5279.76 million ops/sec
Measured integer speed 19321.67 million ops/sec
Memory 16310.63 MB

-------------

eg This Ryzen 7 3700X system has good CPU time v Run time numbers.

Run time 8 hours  0 min 41 sec
CPU time 7 hours 59 min  5 sec

tau_site11_7nrq_graft_bcov_v1_xad_SAVE_ALL_OUT_IGNORE_THE_REST_2jt0hl2g_1824323_1_0

And how can CPU's be "slowed" if each core is doing its own thing?
All RAH tasks have their own cores as do the other projects.
The GPU stuff has it own core to monitor its work and one core is free for system and web.
So what exactly is "slowing" things down?

And how can CPU's be "slowed" if each core is doing its own thing?
All RAH tasks have their own cores as do the other projects.
The GPU stuff has it own core to monitor its work and one core is free for system and web.
So what exactly is "slowing" things down?

And how can CPU's be "slowed" if each core is doing its own thing?
All RAH tasks have their own cores as do the other projects.
The GPU stuff has it own core to monitor its work and one core is free for system and web.
So what exactly is "slowing" things down?

A task does not have exclusive access to a core, the work is threaded and a task can be interrupted at any time depending on its priority and Boinc runs at minimal priority thus the cpu time will be shorter than the elapsed time to the extent of the interruptions it experiences.

I don't know how to get the core view...the best I can guess is details. There is a column called CPU.

I don't know how to get the core view...the best I can guess is details. There is a column called CPU.

Each thread is considered a CPU by Task Manager.
Click on the top of the CPU tab, and it will sort things from highest to lowest CPU usage. Clicking on the top of the Name tab will take it back to the default sorting by name view.

I don't know how to get the core view...the best I can guess is details. There is a column called CPU.

Each thread is considered a CPU by Task Manager.
Click on the top of the CPU tab, and it will sort things from highest to lowest CPU usage. Clicking on the top of the Name tab will take it back to the default sorting by name view.

Well like I said before...85-86 BOINC, 2x 6.2 for FAH browsers and other things .02 to .01 around 8 times. All the other running processes are 0%

Browsers high fluctuations. Between .1 and 5% every second, but mostly .1%

I don't know how to get the core view...the best I can guess is details. There is a column called CPU.

Each thread is considered a CPU by Task Manager.
Click on the top of the CPU tab, and it will sort things from highest to lowest CPU usage. Clicking on the top of the Name tab will take it back to the default sorting by name view.

Well like I said before...85-86 BOINC, 2x 6.2 for FAH browsers and other things .02 to .01 around 8 times. All the other running processes are 0%

So it's not something running continuously using a bit of CPU time, but just for short periods using quite a bit of CPU time.


The numbers you posted above show just under 100% usage.
From an earlier post-

Browsers high fluctuations. Between .1 and 5% every second, but mostly .1%

The browser hitting 5% usage will be taking processing time from one of the other in use threads to drive that. If Rosetta are the lowest priority Tasks, then they will be the ones that loose that processing time.
Depending on how much of the time those browser tabs are open, and how often they need that 5% CPU time then that could be why. 5% every few seconds over 8 hours could easily account for an extra hour's time to actually finish 8 hours worth of Rosetta processing.
The other option is a 3rd party AV programme or similar doing regular daily scans- if Rosetta is the lowest priority Task, then it's processing will be stalled while the AV scan makes heavy use of the CPU. Or the AV programme set to scan all network traffic would result in heavy CPU usage each time a web page is loaded or refreshed.

I'm not saying that these are the cause of your longer Run times, but showing how they could be.
Intermittent issues are often the toughest to resolve.

I have a once a day drive/registry cleaning program (Wise365) that runs, usually around midnight for a few minutes, but that is it.
Tabs...yeah well that's entirely possible. I have a bunch open from articles out of my email to read, but other than that there is no demand on my system.
How much do you think FAH GPU control with CPU's causes delays?

I have a once a day drive/registry cleaning program (Wise365) that runs, usually around midnight for a few minutes, but that is it.
Tabs...yeah well that's entirely possible. I have a bunch open from articles out of my email to read, but other than that there is no demand on my system.
How much do you think FAH GPU control with CPU's causes delays?

Without actually seeing what else is using the CPU time when it's happening it's near impossible to say one way or the other.
Having said that- It's unlikely to be FAH, as it's running most of the time as well, and without a CPU core set aside for it's use the difference between Rosetta CPU time & Run time would be much, much greater.

Most likely it's some other process just using up CPU time now & then that results in the overall extended Runtime.
The quickest way to see if it's Wise365 would be to suspend it for a few days while Rosetta work is being done & see if the discrepancy between CPU time & Runtime is reduced significantly or not.


From the looks of it's blurb, it's an AV programme & system optimisation utility.
The fact is that even back in the days of WIn3.x memory, disk & registry optimisers made little if any difference to a system's performance (the only one that did make a difference was a disk defragmenter, and even then it's effect was generally only noticeable on a system that had a lot of heavy disk write/read/delete activity. And even windows own defrag was good enough to give a similar improvement). Yes, a 3rd party AV programme is a good idea if you frequent extremely risky sites (ie free software, movies/TVV show downloads etc), but the benefits of it's memory/disk/registry optimisation functions are probably minimal, if any.

BOINC uses 17.7% max.
Vbox is running now..so 24-25%
cmdock 6.2% max and around 5.7 averag
hsgama (Einstein GPU with CPU control) 6.2-6.3 x2
FAH as usual uses 6.3 x 2
RAH 6.2 (peak) to 4.8 (minimum) x6
WCG 6.2 to 4.7% x 2

BOINC uses 17.7% max.
Vbox is running now..so 24-25%
cmdock 6.2% max and around 5.7 averag
hsgama (Einstein GPU with CPU control) 6.2-6.3 x2
FAH as usual uses 6.3 x 2
RAH 6.2 (peak) to 4.8 (minimum) x6
WCG 6.2 to 4.7% x 2

By my reckoning that's around 113%- well overcommitted.
By BOINC uses 17%- what exactly is using that? On my system the BOINC client & BOINC System tray for Windows and BOINC Manager use nothing.



Edit- OK.
Make sure that the BOINC Manager is closed (not just minimised) when checking out what is using the CPU. If it is open, then it shows it as using heaps of CPU (when in actual fact it's the Applications running under it. And for some reason not all of them show as being under BOINC, some still show up as Background processes (at least on my system they do)).


So taking that 17% off of the 113% gives less than 100% which is what we want.
So what ever else it is that's using the CPU time, wasn't doing it the time you were checking Task manager.

No BOINC 16-17% of which 14-15% is FAH CPU control of the 2 GPU's

FAH just kicked in hard and took the CPU up to 72% for a few seconds.
Note: FAH is set at Medium. If I set at low then nothing gets done. If I set it on high it takes up to much of the system.

Ok this is interesting...looking at Resource Monitor and it claims that I am using 110% of the CPU.
The highest values jump between WCG (CPU only) and Einstein (GPU only)