E@H not using the whole processor

knight4761
knight4761
Joined: 18 Jan 05
Posts: 3
Credit: 26283
RAC: 0
Topic 188132

I tried running the E@H and the old S@H programs at the same time. Through windows manager it showed that the processor was being split 50/50 between them. But E@H only took about 5 to 10 minuents longer to complete the wu's. I have an Athlon 3000+ and it takes about 6h5m with only E@H running and about 6h10m running both E@H and S@H. I was wondering if anyone could tell me why I don't see a large change in time when I do this. I do see a difference in the time for S@H it increases from about 3h to about 8h.

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5874
Credit: 118343065693
RAC: 25446824

E@H not using the whole processor

> I have an Athlon 3000+ and it takes about 6h5m with only E@H running and about
> 6h10m running both E@H and S@H.

That is very interesting. Obviously EAH has a higher priority than SAH and SAH is just using the leftover CPU cycles. Please report again after you have accumulated several EAH results to make sure you didn't have some special case WU. It sort of implies that EAH is leaving plenty of available CPU cycles. Strange indeed!!

Cheers,
Gary.

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4331
Credit: 251619338
RAC: 36045

Which client version are you

Which client version are you running?

BM

BM

knight4761
knight4761
Joined: 18 Jan 05
Posts: 3
Credit: 26283
RAC: 0

I have ran between 10 and 20

I have ran between 10 and 20 WU's like this now. I thought it was a special case at first too. I am running BONIC 4.19 with einstein 4.79 running and SETI version 3.08.

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5874
Credit: 118343065693
RAC: 25446824

> I have ran between 10 and

Message 6182 in response to message 6181

> I have ran between 10 and 20 WU's like this now. I thought it was a special
> case at first too. I am running BONIC 4.19 with einstein 4.79 running and
> SETI version 3.08.

Wow!! I've got a number of boxes running Seti classic and had turned one completely over to Boinc/EAH. If the EAH performance stays about constant and the SAH performance is about 40% of what it used to be, I'll just add EAH to all those running SAH. I've just now restarted SAH on the EAH box to see if I can reproduce your findings. I'll post back later after some work is done.

Thanks for the tip!!

Cheers,
Gary.

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5874
Credit: 118343065693
RAC: 25446824

> I have ran between 10 and

Message 6183 in response to message 6181

> I have ran between 10 and 20 WU's like this now. I thought it was a special
> case at first too. I am running BONIC 4.19 with einstein 4.79 running and
> SETI version 3.08.

OK, I've had Seti classic running with EAH for a little while now. According to task manager, EAH is getting 99% of the CPU and SAH is getting 00%. Setispy has gone idle as the state file is not being updated. Obviously SAH is dead in the water, as one would expect. EAH is using every available cpu cycle and SAH is getting zilch. I could up the priority of SAH but I'm sure that would just slow down EAH. I have an Athlon XP2700+ which does EAH WUs in just under 6 hours, ie similar performance to yours.

What magic incantation do you use to get both to produce?? :).

Cheers,
Gary.

knight4761
knight4761
Joined: 18 Jan 05
Posts: 3
Credit: 26283
RAC: 0

Both S@H and E@H are set to

Both S@H and E@H are set to the same priority. Could it be because I have a 64 bit processor? That's the only thing I can think of that would be different now.

Comment viewing options

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