TBH, I'm not a credit chaser and can't be bothered with the science of optimization, so I'll always be open to correction, but herewith my take.

It is my understanding that an Optimized Client does only 1 thing, that is to inflate the Benchmarks for that Particular Computer.

Yes, and in some instances for a reasonable reason.
The standard windows client is compiled with some level of optimization, whereas the standard linux client isn't. With this in mind someone can determine that on their dual boot system (i.e. identical hardware), in windows they get 10 credits, but in linux only 5. IMO it's reasonable to expect the same reward for the same work on the same hardware, irrespective of client preference.

An optimized Client bundled with an Optimized Application does 2 things, inflate the Benchmarks plus speed up the processing of the WU.

Agreed, and the standard client has already some optimization, so introducing an optimized science app ought to do no more than speed up the throughput.

With that understanding I don't really see how it is any different than Blatant Cheating is. If you use an Optimized Client your getting more Credits than you were intended to & if you use the Optimized Client with the Optimized Application now all your doing is getting more Credit and getting it faster ...

Like everything other tool, optimization has it's uses (i.e. linux vs. windows levelling), but it's also open to abuse so things start to get suspect when people optimize the already optimized, in effect using the tool for the suspicious purpose of solely inflating benchmarks to pump up credit. I know that that sounds judgemental, but I can't think of another way to phrase it. Please believe me when I say that I'm not judging anyone. :-)

Unless everybody runs the same Client & Application @ a Project then somebody's Cheating...IMO

I think that yours is a point of view that many share. If the clients all returned a result which earned a reward that scored within a fraction of every other reward for the same result, it would be immaterial which client was used. And IMO, that's how it should be. But with the exception of CPDN, it aint so. And it aint looking like getting so any time soon. ;-)

Jeesh, we already get a new Client from Berkeley every 2-4 days that's hard enough to keep on top of, then we are supposed to keep up with all the Optimization thats going on too ...

Precisely why I pick a std version that suits my needs, and remain blissfully ignorant of all the optimizations.

I also keep my mouth shut anymore when I run across something that seems quit fishy because I got in in trouble once before for saying something. Everybody sided with the person that was cheating or at the least fiddling with the Credits & I was made out to be the villain.

Yep, discussion surrounding credit is always a minefield. ;-)

Right now I know of a serious & provable case of cheating going on at one of the Projects, but I'll be darn if I'm going to say anything about it. Let the chips fall where they may is the Motto I go by anymore ...

No matter how tight the system, there will be someone that can work a way around it. Usually not for long though, and if anyone of substance is concerned enough, the hole gets closed eventually.

There's a hole in this project (not due to the project's making) and the participants are concerned that it will be and/or is already being exploited. Unfortunately the easiest fix (quorum) carries implications, so it hasn't yet been implemented while other possible options are being debated. The hole will have to be closed, otherwise people may refrain from participating and that wouldn't benefit the project.

---

Ad 2) on ideal but what one can do when some projects have optimalized apps, some only claim to release the source code, some are not intending to do so (for example due to copyright restrictions).

Valid point.
However, in such projects it would be expected that fair players would use the standard client. :)

I know, this will always be the area of greatest debate, since people with only one system and many projects would fear prosecution as 'cheaters' if they use an optimized client + standard app. That's why I termed it "grey". But let's not forget that for someone to fall into this category it is an active action, rather than passive. They had to go looking for the optimized app, and they know what they're doing.

---

Any feedback from the admins? Because at the end of day, we may all have our opinions, yet we can't change a single thing. What do the Davids think of the situation / what are the plans?

I'm not going to single out anyone, but just have a look at the boincstats page I've mentionned above and see for yourself the mobile pentiums 1.5ghz that bench infinitely higher than the opterons... or the pentium 4 3ghz that return RAC in the thousands...

---

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

I do not understand these speeches. Everyone makes what it wants in house its, its PC to satisfied way its and of pocket gestice its. As far as the Boinc plans, contests to who are plans of Science do not arrive first or second, and second my every customer who puts to disposition a PC also for a single WU deserves great respect. And for the second time that for information gone to see this link ufficiale:http://boinc.berkeley.edu/download_other.php.

Sorry my english write for translator

---

... The solution to this problem that is easiest for the project to implement is a quorum system on returned results. ...

However, as far as the project is concerned, this does more harm than good. Right now, they are incredibly CPU poor.

Assume that 25% of the current user base decide to up and leave because of the cheating issue. That means they're running at 75% efficiency.

If they implement a 3 way quorum, they have to process each WU 3 times, which reduces them to 33% efficiency.

Which way would you go if you were running the project? Quorum for 33% efficiency, or "lose the people that can't stand the cheating" for 75% efficiency. I sure know what I'd do in their shoes.

---

The quorum enters nothing not there is a credit on the job that carries out a PC. In order to resolve the problem 1 WU 1 credit as it was the SetiClassic. The aim of the Boinc plans is not the quorum but the search. Every PC that joins to be a project towards the solution of 1 problem.

---

Assume that 25% of the current user base decide to up and leave because of the cheating issue. That means they're running at 75% efficiency.

You are assuming that both the participants who care about cheating and the participants that don't return a proportionnaly equivalent number of WU... I'm not sure about this. If you look at the stats, the top 20 teams are all stats and/or competition oriented. They also happen to return most of the results.

Anyway, if the quorum is not the solution, then maybe we could come up with another way to stop the cheaters? Secure benchmarks maybe?

---

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

Secure benchmarks maybe?

flops-counting...

---

However, as far as the project is concerned, this does more harm than good. Right now, they are incredibly CPU poor.

Yes, that is the implication I referred to earlier.

Assume that 25% of the current user base decide to up and leave because of the cheating issue. That means they're running at 75% efficiency.

There is a problem with basing statistics on assumptions. The way I read this, you're implying that there are more people willing to do evil (75%) than good (25%). Maybe I'm just naive, but I'd see it the other way around. :) If the goal of the project was evil in nature, it would likely attract more evildoers, but it isn't.

Now, if we assumed that 75% of the user base decided to leave and followed the rest of your logic, we'd be at 33% vs. 25%. Which option looks more attractive to the project now?

Without clear facts, I don't expect that it would be possible to reach any conclusion in exploring this line of thinking.

---

Maybe I'm just naive, but I'd see it the other way around. :) If the goal of the project was evil in nature, it would likely attract more evildoers, but it isn't.

Now, if we assumed that 75% of the user base decided to leave and followed the rest of your logic, we'd be at 33% vs. 25%. Which option looks more attractive to the project now?

How many people have left because they can't bear competing with people who are getting more credit than they should? I'm not going to make any more assumptions but all I see is growth in users and hosts.

I for one like what this project is doing (the science) and I like the fact I get credit as soon as the work is done. Whether others get more credit for the same amount of work (which is likely still useful to the project!) is for me not a reason to leave.

---

*** Join BOINC@Australia today ***

How many people have left because they can't bear competing with people who are getting more credit than they should?

Well, I have nothing against someone getting say, up to 10 or 15% percent credit on a workunit as long as it averages out in the long run. But if I continue seeing cheaters who get 50000% more credit than me for the same CPU, I would quit, absolutely.

Now since no one since to be taking this seriously, I'm going to start naming names. Let's start with this guy

He looks like someone who gave it a shot to see if cheating would work, and it does. 48000 credit in one day for a pentium 4 CPU 2.40GHz. Now ranked host #6 wordwide.

---

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

Anyway, if the quorum is not the solution, then maybe we could come up with another way to stop the cheaters? Secure benchmarks maybe?

These types of ideas have been hashed and re-hashed.
The proper solution lies in tamper-proof and accurate benchmarks across all projects. Unfortunately the jury is still out on how to achieve that and this project cannot achieve it alone. It is an issue that requires a collaborative solution between projects and the BOINC developers. I harbour no doubts that it will eventually happen, but it's not likely in the short-term.

For most projects, this type of problem is addressed via a quorum.
This project requires an alternate solution since neither the normal solution (quorum), nor trying to make the square peg (client) fit the round hole (project) is it.

If indeed there are people cheating, there must be a reason to do it.
How about removing the reason? Nullify the gain.

Can the project issue 0 credits?
Yes, but they'd face having 0 participants.

Can the project issue fixed credit per result until they have sufficient resources to implement a quorum system?
With some project work, I would expect so.

Could the project stop exporting statistics, thereby constraining any false statistics to the project alone and preventing "BOINC World" stats contaimination?
One possible option.
Another could be for sufficient participants that felt strongly enough to lobby the stats collaters to segregate the stats from this project.

It's extreme I know, but is it a big enough issue to warrant such action?

It all boils down to how the majority of participants feel.

---

One practical, but not perfect solution would be to implement some sort of sanity checking on credits granted. The fastest CPU on the planet is known - and the number of credits that CPU could return given a work unit that was returned in a given time after download. If someone goes significantly over that, a variety of approaches could be taken:
1) Their credit could be recalculated based on assumptions about
a) how long they have had a work unit
b) the probable performance of their declared CPU

2) First offenders lose the credits for that work unit and are warned. Second offenders lose the last 5 work unit credits returned. Third offenders are reset to zero. If this were to be done, it should be publicised to discourage cheating up front

Additionally - given enough results returned, it should be possible to see whether some people seem to be getting, on average, more credit for their given CPU than is normal. Rather than send duplicates of all work units - a sufficient percentage of work units to be statistically significant could be sent to duplicate machines to confirm whether the credit claimed is plausible.

I'm not saying a sufficiently motivated person couldn't find ways around some of these - but it would hopefully get rid of the most blatant attempts and, by having a declared punishment system, you have a deterrent.

Regards
Chris

---

I'm not saying a sufficiently motivated person couldn't find ways around some of these - but it would hopefully get rid of the most blatant attempts and, by having a declared punishment system, you have a deterrent.

Good idea. I'm going to compile a list of cheaters and post it publicly. Maybe that will be a good place to start cleaning up for the admins.

---

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

How many people have left because they can't bear competing with people who are getting more credit than they should?

Speaking for myself, I'll be sticking around to see if the situation improves or worsens.

I'm not going to make any more assumptions but all I see is growth in users and hosts.

Aren't these stats total values, or active values?
I'm not in the habit of deleting my hosts. I have 6 registered with this project, but presently only 3 of them still participate. If I were to stop, I expect that neither the host nor user count would be reduced.

When a beer truck overturns and news of the event spreads, do people not come running? ;-)

---

I'm going to compile a list of cheaters and post it publicly. Maybe that will be a good place to start cleaning up for the admins.

Before you accuse anyone in particular of cheating... Do your research and gather the facts. Accusing someone of cheating because it LOOKS like they are doesn't mean they actually ARE cheating.

---

*** Join BOINC@Australia today ***

Good idea. I'm going to compile a list of cheaters and post it publicly. Maybe that will be a good place to start cleaning up for the admins.

While your intentions may be worthwhile, I'd strongly encourage you to exercise extreme caution.
Without direct access to the data, you are at the mercy of third party stats publishers. Should they make a mistake....
Labelling someone as a 'cheater' could backfire on you. Just scroll down and re-read the second last paragraph of PoorBoy's post, no 4247.

---

Good idea. I'm going to compile a list of cheaters and post it publicly.
==========

I would be very careful about accusing somebody of cheating if I were you Stephan. Right now you have no Positive Proof that anybody is Cheating other than in your assumption that they are.

That host that you pointed out in Message ID 4335 doesn't mean anything other than the guy probably has a bunch of the same OS & CPU Equipped Computers and he simply Merged them all together to get the big increase for 1 day in that Host. That sort of thing goes on at all the Projects and isn't by any means an indication of Cheating.

It may be a little UnEthical to do it if doing it was to get your Host to or near the Top of the Host List but it isn't Cheating since you have the Option to do that.

With your constant Harping about the Credits your going to do more to Drive people away from the Project than any of the supposed Cheaters will do...

Labelling someone as a 'cheater' could backfire on you. Just scroll down and re-read the second last paragraph of PoorBoy's post, no 4247.
==========

Exactly, take my advice and Live and let Live Stephan ... In that instance I mentioned that I was made out to be the Villain everybody agreed the person Cheated big time from the Administrators of the Project on down.

Even the person that did it admitted to it, yet I was made out to be the bad guy in the whole affair ... :/

Live and let Live Stephan

cheat and let cheat, then?

With your constant Harping about the Credits your going to do more to Drive people away from the Project than any of the supposed Cheaters will do...

*constant harping*: wow, didn't expect that sort of comment from this project. Very, very disapointed here.

... In that instance I mentioned that I was made out to be the Villain everybody agreed the person Cheated big time from the Administrators of the Project on down.

Sorry to hear that, Poorboy. But if things happens that way around here, I'm going to have to reconsider my participation completely. I'm considering this thread closed.

---

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