{Slaps self} Boinc does do auto reporting. It doesn't do reporting immediately. Which is what you probably meant and which is what I meant in my post above. ;-)

---

Thanks for all the help and advice in this thread guys, much appreciated. The new chip is crunching away for dear life and is getting some quite inpressive results compared to the old P4 640. Now if only I could afford a Conroe E6700 that would really motor!!!!! Problem is it means a new m/c or motherboard as the Core 2 Duo's need the latest chipset.

---

Yes I am using Crunch3r's 5.5.0 but ONLY because it auto reports finished units which I don't think the current 5.4.11 does.

All of the BOINC clients report work automatically -- you really can just ignore it and it will work.

What it doesn't do is report a work unit the moment it uploads -- to avoid loading the scheduler unnecessarily.

---

Yes I am using Crunch3r's 5.5.0 but ONLY because it auto reports finished units which I don't think the current 5.4.11 does.

Auto report was a bug in some of the late 4.xx versions of BOINC. It was also used as a test option in some of the early 5.x.x versions. It's never been in any of the public releases as it adds a lot of overhead on the server.

Reporting 1 result to the database takes as much time as reporting 10 results at the same time. So for each result you report immediately, you take up CPU time, RAM, disk time etc. that you could just as well do when reporting multiple results at the same time.

It won't speed up you getting credit anyway.

---

Yes I am using Crunch3r's 5.5.0 but ONLY because it auto reports finished units which I don't think the current 5.4.11 does.

Anyway, updating preferences isn't working.....

October 12, 2006 - 19:00 UTC
A new scheduler was installed the other day and there was a small bug regarding user preferences which has since been found and fixed. The fixed version will be employed shortly.

From technical news.

---

So, if Boinc Master 2 is using 5.5.0 now, he's already "testing" and might as well move to the current version.

5.5.0 was never a Berkeley released version of BOINC. It was released by Crunch3r. Any version released outside the fray will up the ante for an alpha release from Berkeley. So at that time the Alpha release was 5.5.1

I'd hope that people releasing their own versions would be careful not to collide with the official numbering, and not the other way 'round.

---

So, if Boinc Master 2 is using 5.5.0 now, he's already "testing" and might as well move to the current version.

5.5.0 was never a Berkeley released version of BOINC. It was released by Crunch3r. Any version released outside the fray will up the ante for an alpha release from Berkeley. So at that time the Alpha release was 5.5.1

---

PSST, Ned 5.5.0 was only a boinc build for LINUX. An official Boinc 5.5.0 for win was not made.

You're correct. Boinc 5.2.x, 5.4.x and 5.5.0 (that Boinc Master 2 is using) will not use this setting. 5.6.x is the minimum version to use that option.

That checks out with my findings. Altering the settings as in my earlier posts has made no difference. I'll try it later when 5.6.x becomes the standard download.

If I remember the numbering scheme correctly, anything with an odd-second-digit is a development (or alpha) version.

So, if Boinc Master 2 is using 5.5.0 now, he's already "testing" and might as well move to the current version.

---

You're correct. Boinc 5.2.x, 5.4.x and 5.5.0 (that Boinc Master 2 is using) will not use this setting. 5.6.x is the minimum version to use that option.

That checks out with my findings. Altering the settings as in my earlier posts has made no difference. I'll try it later when 5.6.x becomes the standard download.

5.6.x is never going to become a standard download - see various threads, esp. posts by JM7. You'll have to wait for (at least) 5.8.x

You're correct. Boinc 5.2.x, 5.4.x and 5.5.0 (that Boinc Master 2 is using) will not use this setting. 5.6.x is the minimum version to use that option.

That checks out with my findings. Altering the settings as in my earlier posts has made no difference. I'll try it later when 5.6.x becomes the standard download.

---

Anyone may correct me, if i'm wrong...

Regards
Andy

You're correct. Boinc 5.2.x, 5.4.x and 5.5.0 (that Boinc Master 2 is using) will not use this setting. 5.6.x is the minimum version to use that option.

---

I would set it at 25% that way it uses 25% across all four cpus so that all 4 projects running get 100% usage all the time

emm... sorry guys, but AFAIK this setting ("use at most... CPU time") doesn't work yet (at least with BOINC version 5.4.11, rsp. SETI). AFAIK it is planned to involve this in one of the next versions of BOINC (maybe 5.6.xx or 5.8.xx). So, no throttle at this time...

You may proof it and set it to 1% (and hit the update button in BOINC manager), you will still have 100% CPU usage...

Anyone may correct me, if i'm wrong...

Regards
Andy

John Clark you have to many lower end pcs it would effect all of them but if you took off the others and just ran the one with 4 cpus I would set it at 25% that way it uses 25% across all four cpus so that all 4 projects running get 100% usage all the time

---

the universe is to big to just have life on one planet

exactly and you should still get 100% cpu usage it'll put 50% across both processer so in essence both projects will still get a 100% cpu usage

How does this affect those running several single core machines?

What I mean is single CPU, dual CPU and Xeon (with HT) machines!

I have set my old dual CPU Xeon system to use 3 processors, and this does no affect my two single CPU P3 systems nor my dual P3 CPU system.

However, as I run only one project (S@H) I have traditionally left the procesor usage at 100%, for all the machines.

If I set the processor use for 33.3%, for the Xeon system, how might this impact on the other machines?

---

It's good to be back amongst friends and colleagues

exactly and you should still get 100% cpu usage it'll put 50% across both processer so in essence both projects will still get a 100% cpu usage

Ok, I've added the new settings and we'll see what increase we'll get in the next 24 hours. Be interesting to see what difference it makes.

Chris

---

exactly and you should still get 100% cpu usage it'll put 50% across both processer so in essence both projects will still get a 100% cpu usage

---

the universe is to big to just have life on one planet

go to your acount preferences for seti at home and set it to use 50% cpu that splits the workup and still uses 100% cpu speeding things up a little bit and if it's a dualcore processer make sure you set it for as many cores as you have if you have four like me set cpu usage to 25% it spreads the work accross all cpu cores litghening up the load so it can go faster

I presume you mean like this?

Processor usage

Do work while computer is running on batteries?
(matters only for portable computers) no
Do work while computer is in use? yes
Do work only between the hours of (no restriction)
Leave applications in memory while suspended?
(suspended applications will consume swap space if 'yes') yes
Switch between applications every
(recommended: 60 minutes) 60 minutes
On multiprocessors, use at most 2 processors
Use at most 50 percent of CPU time

---

go to your acount preferences for seti at home and set it to use 50% cpu that splits the workup and still uses 100% cpu speeding things up a little bit and if it's a dualcore processer make sure you set it for as many cores as you have if you have four like me set cpu usage to 25% it spreads the work accross all cpu cores litghening up the load so it can go faster

---

the universe is to big to just have life on one planet

Well, all that work seems to have paid off. It crunching away at a vast rate of knots now. The RAC is rising and should top out at about 1000 from its previous 700 with the older chip.

CPU Time for a credit of ~62 has gone down from 13,800 secs to 8,850 secs, and the RCF from 1 to 0.63. Is that about right for a Pentium D 940 running an enhanced app?????

That sounds pretty good. My P-IV D (3.0 GHz - slower than yours) takes around 11,000 seconds for 62 credits with Simon's app. - you seem to be ahead of the curve there.

Expect your DCF to drop a bit further - perhaps around 0.35. Mine's about 0.40 at the moment, but I've had a run of short results: and it went back almost to 0.50 when we had that run of VLARs a couple of weeks ago.