Bad day at Milky Way - BOINC Credits Halved!

[Petey]

It seems that Milky Way has gone the way of many other BOINC projects and decided to half it's credits to fall in line with Seti .

BUT, the credits are only comparable if you are running an optimised Linux client - windoze clients will actually get LOWER credits than Seti .

If the credits given out by all BOINC projects are going to be continually judged against Seti and 'fixed' if they are found to be higher then why run anything but Seti?

[alpha]

If the credits given out by all BOINC projects are going to be continually judged against Seti and 'fixed' if they are found to be higher then why run anything but Seti?

Maybe if project X is a good cause? Not everyone wants to search for ETI and not everyone cares about stats.

Really, this topic just needs needs nailing, and for me it comes down to one of two options:

• Stop comparing the value of a stat unit between BOINC projects
• Normalise the value of a stat unit against something

Obviously the first option is easy and the second option is difficult. It seems we're in the situation where some projects are normalising against SETI (or others) and some are just doing their own thing. Either way, I couldn't really care less. As far as I'm concerned, each project is entitled to score however they like, BOINC or not.

Before BOINC came along we weren't saying things like "I've done 100,000 stats units on RC5-64 but you've only processed 500 SETI WUs! I'm better than you!". So why don't we just stop summing up our BOINC scores and start taking them as separate values?

[Bok]

Because (at least for me) the stats are so much more interesting due to the cross project combinations.

I think it's still a worthy goal to attempt to get the projects inline credit wise. It's obviously a difficult task, but one worth trying and I believe they will eventually get there. You are right though in that it's not worth even trying to compare between projects. It's one of the main reasons I find it so much more interesting to have the 50K/100K/500K etc stats races.

Of course there are people who have MUCH higher combined scores due to credit on one or two projects which gave out ridiculous credits. There was talk about putting in 'multipliers' on the stats sites for each project, which I'm against in some respects as I believe in reflecting exactly what the projects give us stats wise. But I did start to code it in partially... on the display side only....

Bok

[Brucifer]

That one issue is why I don't run boinc. Always dinking with points per work. Talk about Big Brother and the politically correct State....

[Petey]

I agree with Bok that it would be great to have a single consistent measure of contribution across all BOINC projects but we all know that the BOINC cobblestone isn't it!

Since the first beta of BOINC I saw I raised a ticket about the unfairness of the calculation of the BOINC cobblestone. It penalises folk not running Windows/Intel platforms. Sadly, nothing has ever been done about this.

I crunch projects because I want to support the goals of the project sponsors but sometimes the politics of BOINC makes it very hard work.

Oh, I did luv the defense of the admin at milkyway - he can't do much as David Anderson is on his doctorate panel

[gopher_yarrowzoo]

Oh, I did luv the defense of the admin at milkyway - he can't do much as David Anderson is on his doctorate panel

Cop out....

[Hobbes]

I agree with those who say cross-project stats make things more interesting and having a reliable set of those would be fantastic. However, we're not there and having project admins change credit on a whim is frustrating. There are always going to be performance gains/losses from project to project because the computations are different. People are going to figure out how to exploit those.

Regardless, what irks me most is when I read, "I'm not going to run Project X because it doesn't score well enough." There are some very worthwhile projects out there which don't attract much CPU time; some score poorly and some are not even Boinc. Stats are great, I love stats, but the goal of any DC project is more than piling up numbers on a webpage. (Seti excluded, of course. )

[LAURENU2]

Well I will tell you now I will not run Project X because it doesn't score well enough Never have and never will
In fact I have never even heard of Project X

[uBronan]

Well its easy on other projects they where whining that the ati cards users of MW@home made daily batches of 77.000 while a nvidia 295 GTX can hardly make 32.000 on other projects like gpugrid

Hence the nvidia dudes are jeaulous.

The weird thing is if i understood gpugrids forum they are gonna change the credits up for these nvidia cards.

[Beremat]

Well its easy on other projects they where whining that the ati cards users of MW@home made daily batches of 77.000 while a nvidia 295 GTX can hardly make 32.000 on other projects like gpugrid

Hence the nvidia dudes are jeaulous.

The weird thing is if i understood gpugrids forum they are gonna change the credits up for these nvidia cards.

The thing is that the ATIs actually do that much more work than the nvidia cards.