coupla more questions: one practical, the other just on curiosity. :-)


1) how exactly do the different preferences work?

so far, i defined my default preference and i've defined a work preference. i'll probably also be creating a home preference.

how do i point each of my installs to the preference that i want?


2) what is the time interval between benchmark timings?

on our old project, FAD would sample the cpu every 15 minutes. this yielded a more accurate representation of true cpu load on the machine between the dc client and other active processes...thus resulting in a fairer calculation of points returned for WU's.

how often does BOINC sample the cpu?

---

- team XPC - 'Where merry times and good crunching meet head-on!'

1) in your account page is a link to your computers, there you can asign each one what "profile" it should use. you just need to consider that boinc (not only one project) uses the general settings from the project you last modified, so your best approach is to select a "master" project where you edit your preferences.

2) i don't have the exact interval in my head, i could look it up though. generally, the wu captures the cpu time it took, not the time that goes by from download to report, so having just 50% cpu available for rosetta should not affect credits rewarded.

---

--
Florian
www.domplatz1.de

so having just 50% cpu available for rosetta should not affect credits rewarded.

beg your pardon...but how would having only 50% of cpu allocation to the rosetta client NOT affect the calculation of 'points'.

consider the following example:

PC #1: 3ghz cpu with 100% allocation of cpu time to rosetta

PC #2: 3ghz cpu with 50% allocation of cpu time to rosetta (remaining 50% of cpu time allocated to process X)

if both PC's crunch the same WU and PC #2 takes roughly twice as long as PC #1 to complete the WU, wouldn't PC #1 get awarded more pts than PC #2? PC #1 may not get double the points, but it should be significantly more, right?

------

under the find-a-drug system, pts for completed WU's were awarded by a formula:

cpu_score * completion_time_for_WU == pts

the cpu_score is calculated when crunching commences on the WU and approx every 15 minutes thereafter, another sampling benchmark occurs. if the cpu was busy running other tasks during this sampling period, then the cpu_score will decrease; this accurately reflects the real world state of cpu load/resources that was made available to the dc client. and thus the pts awarded decreases as well, given the formula used (completion time would naturally rise also because less cpu time was allocated to the client).

even factoring in the averaging system that rosetta does, wouldn't PC #1 get awarded more points in the example i provided?

which probably leads to a follow-up question: is the benchmark that the client runs uses sensitive to the cpu load on the processor at the time of the sampling?

thanks.

---

- team XPC - 'Where merry times and good crunching meet head-on!'

John,

R@H also uses cpu_score * completion_time_for_WU == pts. This is a BOINC feature, not a R@H-specific issue. Plenty to read about that can be found on the wiki.

The benchmark wills stop the current WU but won't stop other processes, so if you play Doom3 while you're benchmarking, yes, you will get different results than if your machine was idle. The accuracy of the benchmark was discussed many times, for example in the following threads:

http://setiathome.berkeley.edu/forum_thread.php?id=14987
http://setiathome.berkeley.edu/forum_thread.php?id=15512
http://setiathome.berkeley.edu/forum_thread.php?id=18318
http://setiathome.berkeley.edu/forum_thread.php?id=18770
http://setiathome.berkeley.edu/forum_thread.php?id=20116
http://setiathome.berkeley.edu/forum_thread.php?id=20506

You may also find this interesting:
http://boinc-doc.net/site-boinc/benchmark/benchmark.html

---

Team CFVault.com
http://www.cfvault.com

thanks for those links...i'll take a look thru them later. i read a little of the BOINCwiki yesterday and was aware than an averaging system was used, just wanted to understand a bit more of the benchmarking and how it fits in the overall scheme of things.

i'm a tech guy and we always like to understand how things work. :) and having those answers makes it easier to answer the usual newbie questions that our own xpc members will ask about the project.

cheers.

---

- team XPC - 'Where merry times and good crunching meet head-on!'

No problem, John, I hope I didn't come across as RTFM as it really wasn't my intention :-) The more questions the better, we all probably have a lot to learn.

---

Team CFVault.com
http://www.cfvault.com

I hope I didn't come across as RTFM as it really wasn't my intention :-)

not at all. :-)

i know you didn't just list those links off the top of your head so i definitely appreciate the effort! and if you did just pull those off the top of your head, then you're a *scary* guy. [chuckle]

---

- team XPC - 'Where merry times and good crunching meet head-on!'

so having just 50% cpu available for rosetta should not affect credits rewarded.

beg your pardon...but how would having only 50% of cpu allocation to the rosetta client NOT affect the calculation of 'points'.

consider the following example:

PC #1: 3ghz cpu with 100% allocation of cpu time to rosetta

PC #2: 3ghz cpu with 50% allocation of cpu time to rosetta (remaining 50% of cpu time allocated to process X)

...

which probably leads to a follow-up question: is the benchmark that the client runs uses sensitive to the cpu load on the processor at the time of the sampling?

thanks.

i try to rephrase and give an example (english not being my native language, not sure if i was too clear before):
in your example, lets say pc 1 takes 1h to complete the wu, as it does the wu in one chunk with all cpu power, this 1h is also the actual amount of processing power needed for the wu (simplyfied, there are still processes like e.g. the clock on windows desktop that gets refreshed and other processes, same true for linux, that steal some cpu cycles).
[1h wu processing]
with pc2, only 50% of the cpu are put into the processing of the wu, so it takes 2h until the wu is finished, but 2h * 50% is 1h, which is the actual time still needed to process the wu, the other hour was used for other processes. as the processes are constantly switched between by the os process sheduler, it looks something like this:
[some process, 1m][wu, 5m][other process, 10m][wu, 5m]... (the time slices are much smaller as in reality)
boinc is aware of how many cpu cycles went into the wu, so it still reports the wu as 1h completion time.

to your follow up question: yep, that's a problem, if the machine is busy while doing the benchmark, results can be wrong (depending what priority the other processes have, somewhere between little to way off). that's why the crunching is stopped during the benchmark, but of course boinc has no influence on other processes that might run at that time. --> if the benchmark results were wrong, the claimed credits will be wrong too.

---

--
Florian
www.domplatz1.de