Linux vs. Windows on HCC - a study

[Someguy1982]

Well, i do have the same problem as you in most of the work units, and i'm using Windows, not all the wu, but most, it seems to depend too much on the wingman.
As for FAAH, which doesn't use a wingman normally, sometimes it's seems an obscene robbery of points.

Also, if you check the statistics for the different projects within WCG you might note that theres a huge difference in points per hour between some of them.

Fair enough, I haven't noticed this problem on the windows machines I run though. At least not as bad as the linux machine. I understand the granted credit is typically not the same as claimed credit. However, when I'm getting 45 credits for something the machine is claiming 240 credits on, that seems extreme and I wonder if something is wrong.

yeah,
you lose a decent portion of your CPU potential.
maybe it has something to do with linux,
yet, this pointing scheme do create some problems,
by the original formula, 2 CPU's should give ~the same results, yet actually, what happens, is one is taking the other up or down.
as an add,
it is not the points that are really important, yet they give the user a good estimate of it's work, it is like watching a benchmark.

Josh,
currently, doubt it...

Agreed, after reviewing the formula it seems fair and using it should yield the same points given similar hardware. It's not though, and I am at a loss as to why that would be.

[pfm3136]

Fair enough, I haven't noticed this problem on the windows machines I run though. At least not as bad as the linux machine. I understand the granted credit is typically not the same as claimed credit. However, when I'm getting 45 credits for something the machine is claiming 240 credits on, that seems extreme and I wonder if something is wrong.

Like i said, not in all work units, but i've seen up to 30% difference from claimed to granted. It depends a lot on the wingman, i've seen some ask almost half of my claimed credit, and that obviously takes the granted credit down.
But that example really is extreme, it would seem like you have a real high benchmark and they suspect something and cut the credit in half, or more...
I've already seen from 240 to 150, but not to 45, that's too much...


Well, already installed ubuntu in my Q6600, (it doesn't seem to like hdmi cable, had to use a vga cable), it started immediatelly downloading work for HCMD before i could change that pc to the profile for HCC, so i will have to wait a little to see the difference.
On Windows it did the HCC work units in about 1 hour and 40-50 minutes.

[Firechicken]

In short: It seems that under Linux, you can compute an identically structured Work Unit (=equally important piece of science) in about half the time it takes to compute an equally big WU under Windows.
In other words, Linux processes Help Conquer Cancer WUs twice as fast as Windows does.
The different credits awarded are due to Linux clients comparing only with other Linux clients, while Windows compares only with Windows. The vast majority of users are running Windows, hence the higher claimed/granted credit (at least that's what we assume). Even though technically Linux should score way better than Windows for granted credit, IF the credit system was fair and IF my findings are correct.

Awsome! Thanks mate. i'm glad I asked because I was getting things a little twisted here. I always hated having to compare to the other machines WU's that get turned in I guess thats the "Q" factor. I never understood things clearer till now. Cheers.

[pfm3136]

For now, it's taking between 40-55 minutes per work unit in my Q6600 @ 3200MHz. So that's half the time, maybe a little less than half the time it took in Windows.
Here it seems to be claiming more credit for the same runtime than in Windows.
About 50-55 for Windows and 60+ for Ubuntu for 2 hours runtime (boinc points).
The granted credit depends heavily on the Wingman, so i was getting very low granted credit, but now seems a little better, about the same as claimed.

[pfm3136]

Just saw the statistics for the first day, did some math, and wow, it seems to really be getting more points. It is claiming more points in ubuntu than it did in Windows, it just sucks to have some wingmans that take double the time and ask half the credit...
I'll have to wait for a least a week to let it average well the numbers and go to the PPD Predictor thread share the results, because my Q6600@3200MHz in Windows 7 x64 was doing 4.8 (WCG points) PPD per MHz, now in Ubuntu, the first day, it did 5.9 PPD per MHz.

[WhiteFireDragon]

so is XP somewhere in between linux and win7 in terms of points and amount of work done?

[pfm3136]

After a week running on Ubuntu, my average PPD went from 15-16k to 18k wcg PPD.
So, although the difference in claimed vs granted points, i'm getting more points in Ubuntu than i did in Windows7 x64.