Will any credits be granted for clients that only run WUProp? As nothing useful will be done on them, I sincerely hope not. Otherwise the credit jerks will set up tons of VMs with BOINC running only WUProp and recieve credits for nothing, in other words: they will severely cheat.

If you will only grant Credits for systems with projects really running in parallel, this could be avoided.

Edit:
Of couse this other projects have to be CPU-intensive ones, not something like FreeHAL, otherwise they will cheat nevertheless.

---

Grüße vom Sänger

Dont give the cheaters any ideas!!! LOL

---

Will any credits be granted for clients that only run WUProp?

so far it gives full 50 kredits/day on WUProp-only machine. Actually, on MacOS it is not 50/day, but 50 per 100000 sec

Of couse this other projects have to be CPU-intensive ones

what about QCN which is not intensive but eats around 5% cpu anyway ?
also, freehal eats some traffic, and this is useful WU stat (i said WU stat, not project, of course :D)

---

I collect WU properties for Ukraine!

There are of course useful non-CPU-intensive projects, and for some it may even be useful to have 10 VM running in parallel with only that project in it. Afaik FreeHAL is such a project, although I recently don't crunch it.

WUProp and QCN are no such ones. As well as there's no use in running more quake sensor reading projects for the same sensor it's not useful to run a project that's only purpose is looking at other WUs.

The credits jerks of course don't give a damn about usefulness, as long as they can climb in the ranking without any efford they will do next to anything. This should be prevented.

---

Grüße vom Sänger

maybe it would be useful to give credits depending on number of CPU-intensive results per WU (like it is done in dynaping where it gives kredits for successful pings)

this will make useless running wuprop on 100500 virtual machines

---

I collect WU properties for Ukraine!

maybe it would be useful to give credits depending on number of CPU-intensive results per WU (like it is done in dynaping where it gives kredits for successful pings)

this will make useless running wuprop on 100500 virtual machines

Good idea!

---

Grüße vom Sänger

they will severely cheat.

Do you *have* to infect every BOINC project forum with your childish, petty spew?

Go away you whining retard.

KTHXBYE

B.

Without taking sides or getting in the middle of a big "thing" here , I must say that cheating is an issue at some projects. Milky way for instance just had to redo apps or their credit system or something because people were running scripts to abort lower paying workunits. Myself, I have never understood this. Of course its fun to watch yourself climb up the credit leaderboards, but c'mon people---ITS NOT REAL MONEY!!!! The "credit" is worthless. Edit: Please allow me to add-- I, myself, have asked on the forums about the possibility of running 25 concurrent workunits on this project because I thought at the time that it analyzed data BOINC wide. My understanding now is that it only analyzes data from the hosts upon which it is run. Somebody please clarify this for me if I am wrong. The only thing which I could possibly disagree with Saenger about is that the data/project is worthless (see my post in science section). Perhaps there are many people with the skills to compile the data provided by this project independently, but I for one am not one of them!!

---

The only thing which I could possibly disagree with Saenger about is that the data/project is worthless (see my post in science section).

I never said that this project is worthless, au contraire, I like it very much and think it's really useful.

What's useless is running more than one WU per machine, unless you have a VM that does real crunching running in parallel to the ordinary BOINC, perhaps because it's projects run better under another OS or why-ever. Running only WUProp in a BOINC gives no useful data at all, as no data to collect is generated, that's what I called worthless.

Edit:
And that's what I like about rillians idea: Credits according to projects in parallel, perhaps with ever decreasing amount per next project, starting with 25 or 30 for only one and ending at 50 for more than 60 or something.

---

Grüße vom Sänger

One credit per user per day, no matter how many project he crunshes. :)

Sorry Saenger for misquoting you about the utility of this project. I misremembered what you said in an earlier post. BTW: dont ever "go away" as someone suggested in an earlier post. Reading your posts in the past has helped me ( and I am sure many others ) tremendously!

---

Will any credits be granted for clients that only run WUProp? As nothing useful will be done on them, I sincerely hope not. Otherwise the credit jerks will set up tons of VMs with BOINC running only WUProp and recieve credits for nothing, in other words: they will severely cheat.

If you will only grant Credits for systems with projects really running in parallel, this could be avoided.

Edit:
Of couse this other projects have to be CPU-intensive ones, not something like FreeHAL, otherwise they will cheat nevertheless.

I monitor the results returned.
If I detect cheating, I will remove credits

---

Hi guys. Wow I am obviously very niave about all this BOINC business, I did not ever contemplate that people would feel the need to cheat on any project and it is a bit of a shock to read these posts and realise that the original concept(as I understood it) has been thrown to the wayside. I joined BOINC because I wanted to be able to help real people (virtual strangers of course) by finding cures, medicines, and of course life on mars!!! It saddens me to think that what is a truly giving concept has been turned around to make it something of a free for all competition. Is cheating a big problem on projects and if so which ones? I see you mentioned milkyway already, do people actually cheat on the biology based projects also? Does the cheating nullify the results the rest of us are sending in? I hope not. Anyways happy crunching all and lets hope this problem can be erradicated.

---

...... it is a bit of a shock to read these posts and realise that the original concept(as I understood it) has been thrown to the wayside.

Its still alive and well - the overwhelming majority of crunchers dont bother posting anything anywhere - they just Crunch. In any Group activity there are always the vocal ones, no different at BOINC, dont judge a whole enterprise on an unrepresentative few. At the end of the day, the vast majority just Crunch and could care less about league tables yaddie yadda. Those that are into league tables, thats great no issue there, thats the beauty of BOINC, play it as you see it - just dont expect anyone to be somehow impressed that an individual is somehow a "vertuous cruncher" or is the "high scorer" - two different genres that can never be directly compared. At the end of the day, just enjoy it and take from it want you want to take from it, never mind others motivations, just have fun.

Is cheating a big problem on projects and if so which ones?

It can be if Project Admins dont keep on top of it by putting in checks and balances, validations etc. As to which ones, it will vary hugely along the lines of which way the wind blows, whether there is an R in the month, and more usually the amount of credit a Project gives out - the higher the credit rate, the more it attracts the less than desirable numbskull who seems to think mythical credits gained by cheating methods is somehow worthy. Usually the latter are easy to spot, so they come and go, and are usually treated with contempt by other Crunchers. It really is a pointless activity, but thats human nature I guess.

.... Does the cheating nullify the results the rest of us are sending in?

By in large no. Occasionally it will have an effect, however Projects Admins are alive to the pointless stupidity, and put in place measures to nullify it in the long run.

..... lets hope this problem can be erradicated.

Never happen - just like there are always people who will steal from a Church Alter, so there will always be people who attempt to cheat in BOINC - all part of life's sweet pattern. However, dont let it put you off BOINC, just ignore the idiots like the overwhelming majority of Crunchers do.


The cheaters are just a boring itch on the backside of progress, quick scratch and the irrelevance goes away. Just enjoy BOINC and have fun:)

Regards
Zy

i have latest BOINC 6.10.43 on MacOS 10.6.3 with WUprop attached, beside some projects with Mac apps that get rare work.

I monitor the results returned.
If I detect cheating, I will remove credits

These WUs run 100000 sec each, and all get full 50 points, even "empty" ones. You need to fix validator so it will not validate useless results.

http://94.23.23.123/wuproj/results.php?hostid=6

---

I collect WU properties for Ukraine!

Seems like this is fixed! Next WU from empty BOINC got 0 kredits

---

I collect WU properties for Ukraine!

Now, results with only WUProp or returned before 24 hours will be invalidated.

---

just like there are always people who will steal from a Church Alter, so there will always be people who attempt to cheat in BOINC - Zy

Thanks for the detailed reply, it is nice to know our work is not affected by such things.

---

Sebastian, I have a wu that has managed 10% in 2 hours which means it is very likely to report in under 24 hours, what shall I do , suspend it?

---

Sebastian, I have a wu that has managed 10% in 2 hours which means it is very likely to report in under 24 hours, what shall I do , suspend it?

Its ok, seems the first ten percent ran quickly then it seemed to readjust and finished in 24 hours and 2 mins , phew.

---