Hmm.

A very strange thing is happening on my host.

I´m using BOINC 6.3.10 with default settings (no messing around with ncpu+1 or something like that)

After I uploaded a new WU for GPU Grid today I found my PC crunching happily two CPU Seti units alongside one GPUgrid WU....
-- Can´t say what impact that has on the GPU-crunching time...
I´ll have to wait until the WU is finished to take a look on the time per step.



Cu KyleFL

Same thing on my computers!

I will see this tomorrow morning!

Jim PROFIT

Don't want to be rude, but this related how to "New applications coming up with 260,280 optimizations"? ;)

MrS

Thats something I want to figure out.
I didn´t upgrade my Boinc to 6.3.11 or 6.3.12 nor changed anything on the configuration so I thought, maybe it could have something to do with the new applications, because that was the only thing that could have changed something...

The new downloaded Wu has the name:
UQ27568-GPUTEST3-0-10.acemd (Full-atom molecular dynamics 6.45 (cuda)).

Is it possible that this WU has changed the behavior of my BOINC or is the error somewhere else?


If it´s not related to the new applications coming up than I´m really sorry for misusing this threat and I will be very glad to drink the next beer on everyones health! :)


Happy Crunchin, KyleFL

If it´s not related to the new applications coming up than I´m really sorry for misusing this threat and I will be very glad to drink the next beer on everyones health! :)


Happy Crunchin, KyleFL

Well, I'll join you with that beer too, because I now have this same (welcome) behaviour too!

I manually updated a few of boxes using BoincView and each one that downloaded new WU's now shows this feature. To clarify, no boxes run with the ncpus+1 setting, all run Boinc 6.3.10 and they are a mix of Windows XP32 and Linux.

Two of my machines did not download new WU and have not started up extra CPU WU's.

---

After I uploaded a new WU for GPU Grid today I found my PC crunching happily two CPU Seti units alongside one GPUgrid WU....

It's a failure of the SETI application I've seen in the past, too. Sometimes SETI runs further even though the BOINC manager has stopped the application.

---

My apologies for hijacking this thread: perhaps the last several posts could be moved off into a thread of their own?

Anyway, the two boxes that did not download new GPU WU's last night (because they didn't need any more work at that time), have downloaded new WU's overnight. Those machines are now crunching an extra CPU WU too.

It's not the science app failure that Kokomiko refers to because this is too systematic across more than one user and in anycase it is not a WU that is running on past 100%. In my case a new CPU WU was started on each machine.

Just to confirm again I am not using ncpus+1: I am running stock boinc 6.3.10 it happens across both Windows and Linux platforms.

Perhaps GDF will tell us what is happening?

---

The only thing I have done yesterday night it is a server update. This should NOT have changed the scheduler policy. Please create a new thread on this.

GDF

The only thing I have done yesterday night it is a server update. This should NOT have changed the scheduler policy. Please create a new thread on this.

GDF

Ok, I´ll do so. I´ll create a new threat on that problem and - as promised - will drink a beer later on everyones health here for hijacking this threat! :)



Cu KyleFL

How can I check "Time per step" ?

How can I check "Time per step" ?

You have to take a look on you´re finished tasks:

For example
http://www.ps3grid.net/result.php?resultid=70401

Your Acount -> Computers (choose one) -> Task ID
click for details -> The Time per step is hidden in the big text field.


Cu KyleFL

and - as promised - will drink a beer later on everyones health here for hijacking this threat! :)

Cheers mate :)

Actually your question was not as unrelated as I first thought - it could have been related to new apps. I just assumed the new GTX 260/280 apps would get a version number update, so without further thinking I inferred that this new BOINC manager behaviour would be unrelated to new clients.

MrS

---

Scanning for our furry friends since Jan 2002

so when is the new app going to be released?

This coming week.

gdf

That is really good news.

Here the promised drink on your health:





Cheers to you all!!!


And always happy crunchin!



Cu KyleFL

hrhr.. :) nice picture...

:push:

..looking fwd to that new release... \o/


cheers! ;-)

This coming week.

gdf

Have they already been released?
There is a way to know if a task is a one of the new ones?

They'll surely get a version number update (current one is 6.45).

MrS

---

Scanning for our furry friends since Jan 2002

As soon as the situation with boinc 6.3.14 is stable, we will come out with new applications.

gdf

As soon as the situation with boinc 6.3.14 is stable, we will come out with new applications.

gdf

Is it necessary to use the 6.3.14 to benefit from the performance incerease, or do the new apps run with the older 6.3.10 as well?



Cu KyleFL

It might be better.

gdf

GDF... :)

..r u doing well bringing up the new app ? :) I'm so curious about it... :P