Thank you, Tralala! That would change things a bit.

I composed my previous post and then the internet went out, so I wasnt privy to that info prior.

Mea culpa.

-SS

---

On the plus side, people who have been using the stock clients all along will get the excitement of clawing our way up through the ranks of people we were never able to catch, before. I think that would be more fun, myself. Especially this fall/winter, when I can really "unleash the hounds"!

Realistically, though, it would be very difficult, if not impossible, to fully correct total credit. Some people have never used inflated benchmarks. Others have always used them. Still others have used them sometimes, but not always. This is what would make it difficult to fix total credit, as I see it. You would have to look at each individual work unit, to see what credit was granted for it, what computer ran it, etc. I'm not sure they still keep all of that information (considering you would need to know what those values were, as they were at that time).

The part that concerns me is that I'm sure there will be some people who, seeing that they will soon lose the ability to overclaim, will try to get as much as they can before the new credit system comes out. I'm afraid we might see some unbelievable overclaims, during these next couple days/weeks.

---

"There are obviously many things which we do not understand, and may never be able to."
Leela (From the Mac game "Marathon", released 1995)

We should also see lots more posts on how to increase actual work done. Tips on when to add memory - how to streamline a system (kill off a few unneccessary memory hog apps), etc.

When I started DF, I had one machine.. and a particular person had around 300,000,000 points. I caught up to them by the time the project ended. Slow and steady (with a buildup of systems) can match those that are incredibly fast and burn out quickly. Early leads just mean you have to work harder, and encourage more to join your team or add their machines to your account, or increase the size of your own pharm.

---

What about starting it right from zero all over again?
If we get a fair app that puts all on a even playing field and gives credit based on work done, no matter what the points are, why not go from zero?
I think that would make a lot of people happy, both the power users and the guy with his one machine.
Here are what I see as the major points to making this work:
1)Must be "fair" to all major brands of cpu's.Intel,AMD and Mac
2)Must be "fair" to all major OS.XP,Linux and since I'm a mac know-nothing, whatever OS the mac systems use..Last I heard ot was OSX BUT I could be way off!
3)MUST be work based.The faster the machine, the more it crunches in a given time frame, the more "points" it gets..
I'm not a software engineer, just a hardware guy with some software knowledge, so I'll leave the designing to the people with the know-how.
Thanks for your time,
Movieman

---

What about starting it right from zero all over again?

I've had the same thought, but it would certainly be a DE-motivator for many in the nearterm. I think the way to "SELL" that idea would be to roll your existing credits over in to "Rosetta Classic" credits, and make all credits going forward be the new and improved "Rosetta Opti-Express" credits. Have fun with the names, but you get the idea.

---

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/

What about starting it right from zero all over again?

I've had the same thought, but it would certainly be a DE-motivator for many in the nearterm. I think the way to "SELL" that idea would be to roll your existing credits over in to "Rosetta Classic" credits, and make all credits going forward be the new and improved "Rosetta Opti-Express" credits. Have fun with the names, but you get the idea.

Agreed 100%..Yes..
Movieman

---

What about starting it right from zero all over again?

I would support this for sure. Separate the old credits from the new ones, don't mix them together. No worrying about converting any old credits too. Announce a rollover date. Do it.

Why should the users who have been using stock managers, clients and stock-clocked machines have to start at 0? It's not wise to punish the innocent.

---

Why should the users who have been using stock managers, clients and stock-clocked machines have to start at 0? It's not wise to punish the innocent.

I agree! I am proud of the honest effort I have put into obtaining my score, and to lose it through no fault of my own would be very frustrating. These many months for naught! I would probably go to another project where I would not be "cancelled". :)

---

Regards,
Bob P.

Why should the users who have been using stock managers, clients and stock-clocked machines have to start at 0? It's not wise to punish the innocent.

I agree! I am proud of the honest effort I have put into obtaining my score, and to lose it through no fault of my own would be very frustrating. These many months for naught! I would probably go to another project where I would not be "cancelled". :)

As long as they separate the old from the new I don't see how it would "punish" those who've ran stock, you would still be recognized for your previous work. It would actually be beneficial as it would put everyone on a level playing field and ranking would be as it should, based on the actual work done. Integrating the new credit system with the current would in my opinion be impossible, as things may be dramatically different in terms of actual work being done vs the synthetic estimate currently being used, that and the numerical range of credits.


Inter-BOINC credit shouldn't even be thought about as it has a different basis, just consider it another DC project not based on the boinc platform. Consider it like comparing the results of two separate benchmark applications(like the different boinc projects), for example the time it takes to encode a certain movie with two different applications, the time it takes to complete with one program might be faster on one machine, but the other program may end up being slightly slower than the other machines compared to...

Just curious. Would Rosetta's total project points have to be separated as well? How would this display on Rosetta's Serverstats "Total Credits" and how would Boinc Stats be affected? I assume that "Boinc Stats" reporting would still show only total credits.

---

Just curious. Would Rosetta's total project points have to be separated as well? How would this display on Rosetta's Serverstats "Total Credits" and how would Boinc Stats be affected? I assume that "Boinc Stats" reporting would still show only total credits.

I am probably getting too complicated, but given the difference of opinions here maybe it is worth at least floating this idea:

Have two playing fields: One would be to leave the existing credits as is and the new credit system would build on that (Call this Option A); and the second option (Option B) would be an option for the user to switch to the level playing field while keeping the old credits as an historical record, much as rcofell suggests below.

I realize then you would really have two credit systems for the same project, but whichever one felt more comfortable with is the one that one could choose.

Just a thought...

---

Regards,
Bob P.

Just curious. Would Rosetta's total project points have to be separated as well? How would this display on Rosetta's Serverstats "Total Credits" and how would Boinc Stats be affected? I assume that "Boinc Stats" reporting would still show only total credits.

I'm sure they could just add a new listing for the new credit system and rename the old to reflect the differences.


Regarding my last post, I'm not saying that there shouldn't be any attempts to normalize the new credit system to the old, but it should just be something like finding a global multiplier/ratio that affects all results the same regardless of cpu architecture or OS, as I thought that the new credit system will be based on actual models crunched, which is the actual work done.

Time for the SETI people to go home and mind your own business...

Love

Haltech

---

SETI...Grown Men Looking For ET's That Dont Exist... Do You Dress Up Like Starwars Characters Too?

I'm curious for the new credit system, but I'm afraid the whole thing is sadly going to cost a lot of members...

(Warning: Percentage used below are just a wild estimation of my own, and are used as example)
* Only 1% of the members is real cheating. These can be found very easely.
* Other people are using optimized clients, to give the diffent cpu's a more fair creditsystem like the normal of BOINC.
* Others who aren't interested in credits use a normal standard client.

I want to warn you before changing the credit system:
- Personally I don't like cheating in credits. That's for sure! The discussion of using optimized clients is being held several times. I don't want to start it over again. Is this a so big prolem?
- As learnt from the past, changes in credits systems leaded often to losses in memberbase. See Seti, see TSC, see D2OL, ...
- Resetting the credits will cause more users to retire. I can guarantee that I'm sure that lots of the current members will leave the project if that's the case!!
- If there is a new credit system, fine. But see to it, that no-one will have less credits than the current system.
- Last point: As said, I don't like real cheaters, but the discussions here are also to much, and will new people away!


My 2 cents


As a representative of team Du

---

Member of Dutch Power Cows

It would be nice if the old credit got fixed, but getting the new credit fixed first is the more important thing! After a month or so the RACs will be nearly 100% reflecting the new system.

Indeed it will, however the credit race by "total credit" will be flawed forever. lol Anyone one ever wonder what the leader board would be like without optimized apps?

No not really as XS would still be on top.

Do you ever stop your bleating ? Your like a broken record.

---

It would be nice if the old credit got fixed, but getting the new credit fixed first is the more important thing! After a month or so the RACs will be nearly 100% reflecting the new system.

Indeed it will, however the credit race by "total credit" will be flawed forever. lol Anyone one ever wonder what the leader board would be like without optimized apps?

That is an easy one, something that you should have figured out before even posting this absurdity....

The leading teams would still be leading, in the same positions, albeit with lower total score. In terms of sheer tera-flops the top 3 teams kick the remaining teams in the promordial soup.

Just because SETI fell to pieces does not give the ET-lovers the right to come to this project and attempt to undermine the REAL science being achieved with this project. Go home ET....

---

This discussion is totally going in the wrong direction. Just because we have a few whiners who spend more cpu cycles complaining about the credit system than crunching science, we have to have a new credit system, and then wipe out the old credits?

In a true democracy, this issue would be resolved by a vote, with voting rights earned by amount of science done. I have no doubt in my mind about the outcome.

DDTUNG

---

Why should the users who have been using stock managers, clients and stock-clocked machines have to start at 0? It's not wise to punish the innocent.

As one of those "innocent" users myself, I would support this idea if the total credit score cannot be adjusted to be realistic under the new system. Fix total credits too so we don't have to consider this and everyone is happy.

Is any information which shows claimed credits are modified attached to submitted results eg client version?

---