deleted - I'm not going to respond to a baiter.

---

BETA = Bahhh

Way too many errors, killing both the credit & RAC.

And I still think the (New and Improved) credit system is not ready for prime time...

Stevea, here's the chart of your entire Rosetta "credits/day" contribution for the last 60 days. The last 4 days might show a decline.

Note: this is a total, not individual pc's. Those might vary more.



This picture acquired by typing "Stevea", in the search box at www.boincstats.com, and picking the Stevea from team "killer frogs".

I don't know how many times I can type this out without banging my head against the wall any harder.

...I'm just searching for an answer... Does this coincide with the release of 5.43?

Steve, can you help me to answer your question? How much has my RAC dropped since 12/6? ...and how much of that drop is due to the BOINC disconnect problem several people have been having? (or, put another way, over that period of time, how many hours of crunching have I missed?)

(...and thanks for letting the bait float)

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

Hey Thanks Feet1st. I've got to read that FAQ about 3 more times~! Per your suggestion, I've checked the settings for my machines. I belive I have them set for 24/7.

---

@stevea,

We are trying to give explanations to why it may be dropping. Since it is based on averages it is just that and average. Rosetta@home said it should be even over time. You have had a period of hig, low and bit inbetween. This will happen as different computers and different crunching in the client is better or worse for different processors. Maybe Athlon's just do not do well with the Docking part or the CAPRI targets or somthing. We do not know, you still do not unhide your computers so we can have a clearer picture of the setups, what tasks seem to be running or what sort of variation to the claimed you have (that is is anyone can be bothered to do it).


Your post asks for how much people's RAC has dropped and some people have posted.
Though it has an implied why is this happening and like all good R@H post it'll drift from the original question, though it did only take one post to do that ;-)
Where is your proof that this 'has to be the lowest PPD project' Since you don't run other projects.

We are all in the same boat since we all use the same credit system.
Looking at your overall RAC for R@H at kwsn stats you do have a very slight decline (but RAC is then an average of an averaging credits system). Without your clients unhidden it cannot be seen if the drop is identical from out pov across all your clients. We have to take two approximate 'range' numbers.

---

Team mauisun.org

Feet1st,

Your Message 33647 post raised an issue ... Are failed, overdue, or other "not nominal" WU results contributing to the rolling average credit per model that is computed for a given type of work unit?

Without sitting down to work through the math, this just seems troubling if so. While I certainly agree with Rosetta's philosophy of granting people credit for failures and the like, it seems like the per-model metric should be computed from successful returns, possibly even a subset of successful returns where the number of models crunched is substantially more than one.

If failures are used, what did this WU contribute to the average? It has a claimed and granted credit, but I don't know how to determine how many models it crunched.

---

Alan, as some already know, I have speculated in the past that such an averaging in of the invalid or otherwise odd results might be a problem. I've never seen confirmation from the project team either way on the subject. It's really not possible to know for sure, from the outside, how the numbers are handled. So, it sticks in my brain as being possible. So, don't let it trouble you, I really doubt they are averaging in zeros or such problem results. I'm just trying to bring up ideas on possible ways to explain what is being reported, and if we can find other supporting evidence then perhaps we pursue a specific answer to the direct question.

The WU you pointed to does not show the completed models. I've noticed the same seems to occur when the watchdog ends a WU, even if it had indeed completed models prior to the watchdog kicking in. I plan to pursue the whole topic in more detail with the project team in the new year. But I hope they fix the screensaver issues first.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

OK, so I found these two charts (I'm looking at the credit per day last 60 days chart). I crunch only Rosetta and Ralph
Rosetta
Ralph

I've been having problems with BOINC losing contact with the crunching threads, which results in a blank BOINC screen and no crunching. To avoid losing work over Xmas, I pulled down extra work, then suspended communications, so there's quite a bump after the holiday.

I really can't tell if there is any trend here or not. If anything, I'd bet you'll find a trend that I lose work over the weekends (due to the above problem), and yet this effect sometimes is shown as a low credit on Sunday and a higher credit on Monday (when I get to restart BOINC, and thus report Saturday's work).

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

I am not sure yet, but i think they use the fastest computer available on Rosetta to define the RAC of all the other PCs.
I started running a machine that is dramatically much faster than anything else, if they use a bi or tri lineary interpolation using Zero and to top machine, that may explain your drop. My Dual Xeon machine is 75% faster than the previous machine on top (Was a 4 sockets Opteron 875, who decided to leave, because ridiculized ... hummm hummm ), if they really use the TOP 1 machine to calculate the RAC, this 75% delta can explain the drops. relatively to my machine, all the other machine need to drop (If you interpolate the scores, i am not saying i agree, i am saying that it is a possibility)
on this kind of interpolation, the Fastest is on the demoninator side, bigger means small interpolation result.

To test my theory, i ll be shutting down my 2 computers for a week in January, and we will see if the RAC goes up for the rest of the people. I ll go and process some seti for a week , with my new SSSE3 code :) hehehhe

Who?

I am not sure yet, but i think they use the fastest computer available on Rosetta to define the RAC of all the other PCs.
I started running a machine that is dramatically much faster than anything else, if they use a bi or tri lineary interpolation using Zero and to top machine, that may explain your drop. My Dual Xeon machine is 75% faster than the previous machine on top (Was a 4 sockets Opteron 875, who decided to leave, because ridiculized ... hummm hummm ), if they really use the TOP 1 machine to calculate the RAC, this 75% delta can explain the drops. relatively to my machine, all the other machine need to drop (If you interpolate the scores, i am not saying i agree, i am saying that it is a possibility)
on this kind of interpolation, the Fastest is on the demoninator side, bigger means small interpolation result.

To test my theory, i ll be shutting down my 2 computers for a week in January, and we will see if the RAC goes up for the rest of the people. I ll go and process some seti for a week , with my new SSSE3 code :) hehehhe

Who?

RAC is just calculated using the same methods as all project and is only based on your own credit.

Awarded credit is based on a rolling average for that particular type of target.
The first person get claimed, second get an average of the 2 probably, third average of the 3 etc... The probably is from memory, either that or it uses the average of all thoose before excluding itself, though I doubt this is used.

You computers have no effect other than adding you claimed credit into the averages, oh and crunch a lot of tasks :-)


So if a load of Linux computers join up using the current default client, these get a lower benchmark than the windows computer so would drag the average credit score down. It would be noticable if there was a substantial amount of these linux computers.

---

Team mauisun.org

Listen ta da chicken!

There is only ONE prediction of what your machine is capable of that enters into Rosetta credit anywhere, and that is based on the BOINC benchmarks of your machine. And THIS only enters in as the "claimed credit". It is the number that is averaged in with everyone else's so far reported for a given type of task to determine credit per model.

And from there, RAC is just computed from credit history over time. "Recent Average Credit"... for each user account and each host.

So, what might be new with your machine would be how quickly it crunches the BOINC benchmark as compared to the Rosetta work units. You would get a feel for that by reviewing your completed tasks and looking at claimed vs granted credit. The claim is based on the BOINC benchmark, and the granted is based on the Rosetta running average.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

Here is a chart of my RAC from Boinc Manager.



As you can see, there is a steady decline from about the 6th, and it does not seem to be stopping yet. Another strange thing is that my overall RAC is remaining at just over 300th place.

See here.http://www.boincsynergy.com/stats/teams.php?project=rah&team=1138

Which would only happen if everyone else's RAC is also dropping. So you would think that if it's only me then other people should be passing me in the overall world rankings, right?

This is just another of the things that has me baffled?

I just don't get it...

As far as the points allowed being the lowest in Boinc. I just remember the chart that Astro had comparing the projects, and that Rosetta was on the low side to begin with, so it must be lower now. (As far as my points per project would be)

---

BETA = Bahhh

Way too many errors, killing both the credit & RAC.

And I still think the (New and Improved) credit system is not ready for prime time...

...

To test my theory, i ll be shutting down my 2 computers for a week in January, and we will see if the RAC goes up for the rest of the people. I ll go and process some seti for a week , with my new SSSE3 code :) hehehhe

No need to shut down your PCs, the credit/RAC doesn't work like that.

If your claimed credit is roughly in line with your received credit, then you won't be affecting the overall credit calculations at all (since the benchmark would be in line with the actual work done - granted credit is the cumulative average of claimed credit versus actual work done).

If your claimed credit were way below the granted credit (i.e., very poor benchmarks but good progress on the Rosetta code), then you could theoretically be affecting the overall average of granted credit, but it'd be a very small effect unless you happened to grab one of the first few of a particular strain of work unit.

-- Edit:

For a 6 hour work unit your claimed credit is typically in the 70s, your granted credit is typically in the 130s (=21c/h per core, which is reasonable for a core2. My X2 gets 13-14c/h per core which is OK given it's now a generation behind). If you happened to pick up the very first one of a new batch of work units, and returned it before anyone else, then the next few people may be slightly affected, but after a few more PCs have crunched it, there'd be no detectable alteration in granted credit.

---

...Which would only happen if everyone else's RAC is also dropping. So you would think that if it's only me then other people should be passing me in the overall world rankings.
...

Not a great drop, less than 10%?

If the mix of computers (Linux vs. Windows vs. Apple, and Intel vs. AMD) were changing, then there could be a drop or rise in granted credit as a result. Now, a lot of people are home at christmas from University, and simultaneously people are taking time off from work at Christmas and so forth. Not suprising the mix of PCs has changed.

Secondly, if the mix of Rosetta jobs were changing (i.e., bigger or smaller memory requirements due to the number of amino acids being modelled, different working set size due to algorithm changes), then it'd have a similar effect.

---

my claimed credits are about half the granded credits ... heheheh I like that ;)

i was wondering if somebody can check the calculation on the global RAC.

who?

It's just a cumulative average, given the large number of PCs running each type of work unit, it won't make any difference in the long run.

Using a very simple example, the first is your PC, the subsequent ones are all claiming the average

PC1 - 70 Work claimed, 130 work done - 70/130 granted/done cumulative, you get granted 70 credit
PC2 - 30 work claimed, 30 work done - 100/160, gets granted 18.5
PC3 - 30 work claimed, 30 work done - 130/190, granted 20.5 credit
PC4 - 30 work claimed, 30 work done - 160/210, granted 22.8 credit
PC5 - 30 work claimed, 30 work done - 190/240, granted 23.8
PC6 - 30 work claimed, 30 work done - 220/270, granted 24.4
PC7 - 30 work claimed, 30 work done - 250/300
PC8 - 30 work claimed, 30 work done - 280/330
PC9 - 30 work claimed, 30 work done - 310/360
PC10 - 30 work claimed, 30 work done - 340/390, granted 26.2

Now the opposite example, your PC is the 10th to receive the work unit.

PC1 - 30 Work claimed, 30 work done - 30/30 granted/claimed cumulative, granted 30
PC2 - 30 work claimed, 30 work done - 60/60, granted 30
PC3 - 30 work claimed, 30 work done - 90/90, granted 30
PC4 - 30 work claimed, 30 work done - 220/220, granted 30
PC5 - 30 work claimed, 30 work done - 250/250, granted 30
PC6 - 30 work claimed, 30 work done - 270/270, granted 30
PC7 - 30 work claimed, 30 work done - 300/300, granted 30
PC8 - 30 work claimed, 30 work done - 330/330, granted 30
PC9 - 30 work claimed, 30 work done - 360/360, granted 30
PC10 - 70 work claimed, 130 work done- 430/490, granted 114

You see the effect of the first PC is fairly quickly drowned out. Now keep in mind that there would be hundreds or thousands of PCs running each work unit type.

Something like that anyway. With a 6 hour work unit, you'll never be the first, or even the 10th, to return a work unit, most likely the 100th.

---

So much for my RAC. A power transformer right at the end of my block just went up in a very nice display of smoke an fire, now, Mr. Rock Steady P3, may it rest in peace, boots no more,...Oy, it's always something.

---

It's just a cumulative average, given the large number of PCs running each type of work unit, it won't make any difference in the long run.

Using a very simple example, the first is your PC, the subsequent ones are all claiming the average

PC1 - 70 Work claimed, 130 work done - 70/130 granted/done cumulative, you get granted 70 credit
PC2 - 30 work claimed, 30 work done - 100/160, gets granted 18.5
PC3 - 30 work claimed, 30 work done - 130/190, granted 20.5 credit
PC4 - 30 work claimed, 30 work done - 160/210, granted 22.8 credit
PC5 - 30 work claimed, 30 work done - 190/240, granted 23.8
PC6 - 30 work claimed, 30 work done - 220/270, granted 24.4
PC7 - 30 work claimed, 30 work done - 250/300
PC8 - 30 work claimed, 30 work done - 280/330
PC9 - 30 work claimed, 30 work done - 310/360
PC10 - 30 work claimed, 30 work done - 340/390, granted 26.2

Now the opposite example, your PC is the 10th to receive the work unit.

PC1 - 30 Work claimed, 30 work done - 30/30 granted/claimed cumulative, granted 30
PC2 - 30 work claimed, 30 work done - 60/60, granted 30
PC3 - 30 work claimed, 30 work done - 90/90, granted 30
PC4 - 30 work claimed, 30 work done - 220/220, granted 30
PC5 - 30 work claimed, 30 work done - 250/250, granted 30
PC6 - 30 work claimed, 30 work done - 270/270, granted 30
PC7 - 30 work claimed, 30 work done - 300/300, granted 30
PC8 - 30 work claimed, 30 work done - 330/330, granted 30
PC9 - 30 work claimed, 30 work done - 360/360, granted 30
PC10 - 70 work claimed, 130 work done- 430/490, granted 114

You see the effect of the first PC is fairly quickly drowned out. Now keep in mind that there would be hundreds or thousands of PCs running each work unit type.

Something like that anyway. With a 6 hour work unit, you'll never be the first, or even the 10th, to return a work unit, most likely the 100th.

I guess, the thousands of Core 2 Computers can be responsable for other lower performance RAC drops ... if i understood well.

who?

I guess, the thousands of Core 2 Computers can be responsable for other lower performance RAC drops ... if i understood well.

who?

Well there we have it, it all Who?'s fault with his superdooper-prefetchers and poly-unsaturated FSB's

Bring on Hyper'RAC'sport to increase our credit.

:-D


Stevea
There is another Stevea running on rosetta, not linked to you account? Is that you or someone else (just in case you attached badly at one point)
https://boinc.bakerlab.org/rosetta/show_user.php?userid=125547 A P4 running linux ?

Since your boincmanager stats are about to loose the high RAC points, you can increase the number of days stats are kept for by the manager using the
<save_stats_days>N</save_stats_days>, read more http://boinc.berkeley.edu/client_msgs.php

---

Team mauisun.org

I guess, the thousands of Core 2 Computers can be responsable for other lower performance RAC drops ... if i understood well.

who?

Well there we have it, it all Who?'s fault with his superdooper-prefetchers and poly-unsaturated FSB's

Bring on Hyper'RAC'sport to increase our credit.

:-D


Stevea
There is another Stevea running on rosetta, not linked to you account? Is that you or someone else (just in case you attached badly at one point)
https://boinc.bakerlab.org/rosetta/show_user.php?userid=125547 A P4 running linux ?

Since your boincmanager stats are about to loose the high RAC points, you can increase the number of days stats are kept for by the manager using the
<save_stats_days>N</save_stats_days>, read more http://boinc.berkeley.edu/client_msgs.php

Nope thats not me I only have 4 XPM2600's running right XP now. And never had a Linux box crunching because they were granted lower credit. So I went and spent more money on Windows. I had another WU error out last night with a video error, but after only 1 hr so thats not going to bring down the RAC any further, but it's still going down?

Now I want to explain that I am not a credit hore, if I was I'd be crunching QMC right now. I had a family member taken from cancer. So I will only crunch disease related projects.

But if it continues to fall, at some point I will leave and go folding. I just want a fair amount of credit awarded for the work I do. I mostly use it to make sure the boxes are running fine. But at this point the boxes are running fine but the RAC continues to drop like a rock. It was steady for 8 months, at over 1325 PPD now it's not?

I think something is broken with the new Boinc credit system, I was never a fan of the cross platform uniformity thing. I never did think it would be possible. When the cry to change things went up with all the spreadsheets and constant bickering, I never said a thing. I only started to complain when the credit I was used to seeing started to drop as a by product, of the new credit system. I was not happy that I was getting 125 PPD less here than at Predictor with there 3 WU average, and am less happy now that I'm receiving 100 PPD less than what I was receiving for the last 8 months.

The tipping point for me will be around 1200 PPD, since these same 4 boxes running Predictor were averaging over 1450 PPD with there 3 WU average, and I was getting over 1325 PPD here. I just can't in my mind accept that these Water Cooled, Highly Over-clocked boxes are getting such a low amount of credit when compared to what I am used to getting. Be it here or at Predictor.

Let's see 125 x 365 = 45,625 PPY less here, and 250 x 365 = 91,250 PPY less than Predictor. That 10% looks a lot different looking at it this way, does it not?

Like I said I'm just trying to figure this out, before it's too late...and late looks like it's coming.

OR we can all blame it on who?

---

BETA = Bahhh

Way too many errors, killing both the credit & RAC.

And I still think the (New and Improved) credit system is not ready for prime time...