1.20: 6:10min
1:21: 5:30min
1:22: 6:09min

Windows XP SP2, C2D

Well... to get "backwards" compatibilty we had to sacrifice some speed on 1.22...
That's the way life goes but we won't exclude older OS just for having a faster app ;)

If all settles down without any computation errors we'll have a look at speed improvements again ;)

Dropping g++/cl is one of them ... switching to icc/icl will help a great deal but
we're not there yet.

Hope ya'll uderstand that we have to iron out the bugs first.

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=554767

This unit caused one of those MS popup errors. After I submitted the error report the little BOINC icon in the system tray added a little red-x to it. Looking under my projects tab everything had disappeared. Exiting the manager and starting up a new one restored everything.

41,697 credit is the amount you need to credit back to me just in case you don't know.

You were another person with extremely high avg credit for a single computer (over 600) and this just wasn't in line with what everyone else was getting. This was similar to all the other machines that had received erroneous credit for work units.

Looking in the database now, i only reduced the work on your Xeon, and left the others alone (because they were more realistic), however, it seems i didn't leave their credit for you, so i'll update that.

Maybe leave your xeon crunching for a day so i can see what kind of credit it's getting? It really did seem very off from the other machines out there (that hadn't gotten extremely high credit for a couple work units), having at least 2x the avg credit of anything else.

unfortunately the exact workunits that gave the bad credit weren't still in the database so i just had to go by removing credit from computers that looked really off.

Travis, Xeons and Conroe Quads or similar to this are exceptions, you can reach a RAC of 1000 if you crunch a while :) so take a look into it. I can also have a RAC of 600 because my CPU is overclocked!

Rebirther-Crystallize had said in a thread getting 4-8 cs per unit which IS 2-4 times higher than everyone else....I don't think Travis changed many quad-dualquads..most got left alone.

41,697 credit is the amount you need to credit back to me just in case you don't know.

You were another person with extremely high avg credit for a single computer (over 600) and this just wasn't in line with what everyone else was getting. This was similar to all the other machines that had received erroneous credit for work units.

Looking in the database now, i only reduced the work on your Xeon, and left the others alone (because they were more realistic), however, it seems i didn't leave their credit for you, so i'll update that.

Maybe leave your xeon crunching for a day so i can see what kind of credit it's getting? It really did seem very off from the other machines out there (that hadn't gotten extremely high credit for a couple work units), having at least 2x the avg credit of anything else.

unfortunately the exact workunits that gave the bad credit weren't still in the database so i just had to go by removing credit from computers that looked really off.

Seems the worst ones got away :)

How does that work?

Computers belonging to RAMSES RHA 254 credits for 700 seconds of work?

Another cheater? Owner: AF>HFR>RR Benson007. More than 27 credits for 300 seconds of work?

This person is also very questionable: Account data for Jenik How can he have a RAC of more that 40,000 credits when there is a work unit limit of 2,000 work units/day?

Now here is a weird result: Result ID: 491638 Granted credit: 787537.651434724 for 272 seconds of work?

Travis, can you reduce the credits of two hosts got 2x250k and 700k credits / WU? These are the 2 ones on top computer list, we must fair enough but there are many other too but not so extremely.

after we get the binaries updated i'll be doing this :)

How does that work?

Computers belonging to RAMSES RHA 254 credits for 700 seconds of work?

Another cheater? Owner: AF>HFR>RR Benson007. More than 27 credits for 300 seconds of work?

This person is also very questionable: Account data for Jenik How can he have a RAC of more that 40,000 credits when there is a work unit limit of 2,000 work units/day?

Now here is a weird result: Result ID: 491638 Granted credit: 787537.651434724 for 272 seconds of work?

I suggest you screen just the Top-20 users for manipulated credits and manually reset them to 1. All the rest would probably be much too much work and will most probably upset all users.

Now hang on...

I was the participant leader here for a bit, and am still in the top 20. I have never done anything but used standard BOINC clients, and fiddling with benchmarks. Check with my BOINCstats history. I have done no wrong. And resetting my 33k hard earned credits (especially with all the pop-ups!) is unfair. However, perhaps you mean to reset each result to 1, rather than set the total credits to 1? That I can agree with if it is applied to everyone, not just the top 20 or whatever.

since we've gone to a new credit granting scheme and have some problems with the previous one, i'm thinking about starting all the credit over from scratch (if i can even figure out how to do this). I've been looking for a way to remove credit for individual workunits but so far haven't been too successful. if anyone has some knowledge about this let me know :)

Ah, the user referred to in my post was the same as the last one you mentioned, Martin. Looking at all of these, I've got to say, it doesn't look like cheating as much as a miscalculation in whatever process asks for and/or grants credit. I wonder if the machines we're looking at are seriously overclocked. And if so, why it doesn't generate an error on the WU. Questions, questions.

Somebody needs to be looking at these, right soonish.

i hope to have the validator updated sometime today, and it should be generating fixed credit per amount of work done, calculated by how many stars are in the work unit and the size of the volume. this should fix these abnormalities.

The High Benchmarks could be causing a problem if that is how the Core client figures out scores. That is an Old version of BOINC and was known to cause weird results. They did not cheat but should update their Boinc client to a newer version.

Version 5.10.20 had no problems with correctly benchmarking a CPU.
The BOINC versions that possibly gave weird outcomes due to benchmark inconsistencies were 5.8.17, 5.10.1 and 5.10.6

None of the above versions gave integer benchmark claims in the way the two clients in this thread have them. So either someone took the core client code and adjusted how the benchmarks should be done before compiling it as a 5.10.20 client (easily done by changing the version number before compiling), or they've changed their values in client_state.xml

BINGO! client_state.xml
But "changing the version number" is OT.
Try edit (make higher) benchmark in this file.
For example <p_iops>500002814793331.347700</p_iops> instead of <p_iops>2814793331.347700</p_iops>
and "credit jumped from 1.xx to 1xxx.xx per result with no increase in crunch time."
Version number of the Boinc client is irrelevant.
To admin: ...Houston, ve have problem...
Best and simply is FIXED credit.

Thats probably a good idea. The work units are (for the most part) fixed size, so fixed credit might be the way to go. Currently, the amount of work done is based off two things: 1. the size of the volume, and 2. the number of stars.

Between a quorum of 2 and a way of calculating credit not based off boinc's benchmarks, maybe that will fix the problem?

I'll try and update our validator tonight to get a quorum of 2 going, so hopefully this will fix any cheating.

best way to get rid of them is to have a quorom.

I disagree. That also has the side effect of reducing the total crunching by 50% or more. Just slows down the project. Increasing the quorum should be done only if the science requires it.

A better way to deal with this kind of thing is fixed credits, or some sort of step counting.

Hm, yes, these is as far as I can see an obvious cheat attempt ????

or is it a error of some kind ???

I never get more than 4-8 CS per WU, and it shouldn't be more either with such a short WU completion time and certainly not over 200CS for every WU !