Jose, I just wish people wouldn't jump on someone that actually tries to compile the program themselves. It is open source abnd people are allowed to do it. If they wanted to stop it they could build in restictions (certificates blahdey blah blah etc..). But don't have a go a someone giving things a try.
Yes we know the credit is a mess, yes we know rosetta are bringing something new. And I at least know the 8th is very sson (I ahve my oral presentation/viva for my MSc then :-s ).

Now you say BoincManager runs better :-) I'll give that a go, the current one is sluggish and BoincViews to much for a single computer.
Is that just a compile time optimisation on wxWidgets?

As for getting on with CASP7, what has that got to do with this ? This in know way slows CASP down. It's just a boinc client.
(and P.S. I am, two computers just went full time for a week, one of them a new one)

Fluffy: what bothers me is thre resurgence of the "öptimizers" vs "non optimizers". Lets say: I have seen this fight re emerge too many times . And lets say I am kind of getting tired of it, specially since it is diverting the attention of what should be the primary focus for this week.

It is the venue I question. It bothers me a lot that the "öptimizers" vs "non optimizers "bring their fights into specific projects. Want to talk and discusss plain BOINC do it in a BOINC only MB.

I am saying nothing re Boinc... What the heck!!! Want to hear my opinion of BOINC? I wish Rosetta were Boinc Free. Okies start roasting me.... :)

Ps

Congrats on the MSc... :)

[quote]Jose, I just wish people wouldn't jump on someone that actually tries to compile the program themselves. It is open source abnd people are allowed to do it. If they wanted to stop it they could build in restictions (certificates blahdey blah blah etc..). But don't have a go a someone giving things a try.
Yes we know the credit is a mess, yes we know rosetta are bringing something new. And I at least know the 8th is very sson (I ahve my oral presentation/viva for my MSc then :-s ).

Now you say BoincManager runs better :-) I'll give that a go, the current one is sluggish and BoincViews to much for a single computer.
Is that just a compile time optimisation on wxWidgets?

As for getting on with CASP7, what has that got to do with this ? This in know way slows CASP down. It's just a boinc client.
(and P.S. I am, two computers just went full time for a week, one of them a new one)

Fluffy: what bothers me is thre resurgence of the "öptimizers" vs "non optimizers". Lets say: I have seen this fight re emerge too many times . And lets say I am kind of getting tired of it, specially since it is diverting the attention of what should be the primary focus for this week.

It is the venue I question. It bothers me a lot that the "öptimizers" vs "non optimizers "bring their fights into specific projects. Want to talk and discusss plain BOINC do it in a BOINC only MB.

I am saying nothing re Boinc... What the heck!!! Want to hear my opinion of BOINC? I wish Rosetta were Boinc Free. Okies start roasting me.... :)[quote]

Have to agree with your there (especially the last bit :-)

What's needed is for the mods to step in! (as mods) Direct it into one post (I actually put the suggestion of having a link to the BOINC forum specifically designed for removing most the boinc talk, but making it look like a normal forum section link) I do hope after things settle down after CASP that they can eventually bring the Ralph style sectioning over here. Most orignal regulars have left the rosetta forums since it's hard to have a good chat and everything gets muddled up (to many stickied threads, no designated area's for anything really).

To me if people really don't like it and it bothers them that much then bugger off, yes create a feedback thread saying why you buggered off then maybe something will get done.

Anyways I hope we've not scared the poor person off, that at least is having a go at improving things.

---

Team mauisun.org

So you suggest that nothing Boinc be talked about here? Hmmm, interesting. Let's look at a forum conversation without using any part of Boinc:

New user> "How can I get more wus on my machine, I'm planning a trip"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "I just started, How come I don't see any credits yet?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "I can't seem to download any work?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "What are credits for?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "How do I attach to Rosetta?"
Answer> "Sorry we can't tell you, it involves Boinc"

Boinc is involved in EVERY management function of the project, from gettting work, to returning work, to reporting work, to communication, and "yes" even the credit process is Boinc. Heck, even this forum is provided by Boinc.

What you really should have said is "I'm one of those overclaimers and am tired of hearing about it. Let's move on to something else, Please, I feel guilty everytime I am reminded of the fact"

Let's look at this thread alone. All the voices that don't agree with an optimized client voiced their opinion ONCE (except this post makes 2 for me, except I didn't choose a side in my first post). The "Pro" overclaimers have posted multiple times. Hoping if they say it enough, It will become fact. All the while hiding their actions behind Pro-Project rhetoric. As if not everyone wants the best for Rosetta.

I assume then we'll be free to discuss the credit situation on August 8th? Unless of course the project follows Joses' desire and drops Boinc by then. Then there'd be no point in discussing it.

So you suggest that nothing Boinc be talked about here? Hmmm, interesting. Let's look at a forum conversation without using any part of Boinc:

New user> "How can I get more wus on my machine, I'm planning a trip"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "I just started, How come I don't see any credits yet?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "I can't seem to download any work?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "What are credits for?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "How do I attach to Rosetta?"
Answer> "Sorry we can't tell you, it involves Boinc"

Boinc is involved in EVERY management function of the project, from gettting work, to returning work, to reporting work, to communication, and "yes" even the credit process is Boinc. Heck, even this forum is provided by Boinc.

What you really should have said is "I'm one of those overclaimers and am tired of hearing about it. Let's move on to something else, Please, I feel guilty everytime I am reminded of the fact"

Let's look at this thread alone. All the voices that don't agree with an optimized client voiced their opinion ONCE (except this post makes 2 for me, except I didn't choose a side in my first post). The "Pro" overclaimers have posted multiple times. Hoping if they say it enough, It will become fact. All the while hiding their actions behind Pro-Project rhetoric. As if not everyone wants the best for Rosetta.

I assume then we'll be free to discuss the credit situation on August 8th? Unless of course the project follows Joses' desire and drops Boinc by then. Then there'd be no point in discussing it.

Tony: dont waste your time with me. I am a little cog in the process. There are too many things that are more important in my life now than your provocation and bad mouthing people in different projects. This includes a possible visitor to my country

Your interference created a lot of problems in SETI. I take it Rosetta is the next project you want to disrupt as you keep boudmouthing the top crunchers in the project. So my last question for you: after you finnish your "work" here, what will be the next project you will attempt to disrupt?

As to my opinion of BOINC: It is the open source nature of BOINC that allows any modification and the outright cheating that has been documented. So BOINC, yes that BOINC you defend is the source for the problems. So if BOINC is the problem, it follows that BOINC should be disposed of.

I assume then we'll be free to discuss the credit situation on August 8th? Unless of course the project follows Joses' desire and drops Boinc by then. Then there'd be no point in discussing it.

That sounds good to me. So the season for ranting about "overclaimers" starts August 7, 00:00:00 GMT. The season ends when Rosetta switches to the new fair credit system.

During the season -- have at it!

But not until then.

:)

I assume then we'll be free to discuss the credit situation on August 8th? Unless of course the project follows Joses' desire and drops Boinc by then. Then there'd be no point in discussing it.

That sounds good to me. So the season for ranting about "overclaimers" starts August 7, 00:00:00 GMT. The season ends when Rosetta switches to the new fair credit system.

During the season -- have at it!

But not until then.

:)

Ranting season begins August 8 :)
10-1 Once the fair credit system is implemented , they will rant about the fair credit system. :) Or they will go to another project to rant.

So you suggest that nothing Boinc be talked about here? Hmmm, interesting. Let's look at a forum conversation without using any part of Boinc:

New user> "How can I get more wus on my machine, I'm planning a trip"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "I just started, How come I don't see any credits yet?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "I can't seem to download any work?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "What are credits for?"
Answer> "Sorry we can't tell you, it involves Boinc"

New user> "How do I attach to Rosetta?"
Answer> "Sorry we can't tell you, it involves Boinc"

Boinc is involved in EVERY management function of the project, from gettting work, to returning work, to reporting work, to communication, and "yes" even the credit process is Boinc. Heck, even this forum is provided by Boinc.

What you really should have said is "I'm one of those overclaimers and am tired of hearing about it. Let's move on to something else, Please, I feel guilty everytime I am reminded of the fact"

Let's look at this thread alone. All the voices that don't agree with an optimized client voiced their opinion ONCE (except this post makes 2 for me, except I didn't choose a side in my first post). The "Pro" overclaimers have posted multiple times. Hoping if they say it enough, It will become fact. All the while hiding their actions behind Pro-Project rhetoric. As if not everyone wants the best for Rosetta.

I assume then we'll be free to discuss the credit situation on August 8th? Unless of course the project follows Joses' desire and drops Boinc by then. Then there'd be no point in discussing it.

I think what Jose is refering to is that in every thread where any talk of any use of the optimised files comes up, there are certain people, yourself included, that go off on this rant about their use. If you don't want to use them, don't, plain and simple.
You line following the "Pro" overclaimers comment. Is your memory that short?
It was the comments by you and your small group of "holier than thou's" with your constant banter about the optimised files that brought forth a response from what you call the "Pro's". I'd much rather spend my time getting new people into the project than over here writing these posts but to sit and ignore your comments is a mistake also.
Correct me if I'm wrong, but I was told yesterday that it was you who started the big argument at the SETI forum that resulted in Crunch3r quiting the seti project? Is that correct? My understanding is that this accusation of yours caused many to leave the seti project.
It appears to me that your trying to do the same thing here. Causing a disruption of what used to be a project that focused it's resources on the Rosetta issue alone.
As to BOINC, it's a mess. It's benchmarks even when done back to back to back are all over the map. It doesn't give fair awards to Intel based machines compared to the AMD based machines at least up to version 5.2.13 and that is a provable fact not speculation.The funny thing on this point is that according to the BOINC stats page a few days ago, 65.8% of the machines running Rosetta were Intel based.
My personal feeling is that I'd love to see Baker Labs dump BOINC all together but I suspect that won't be the case due to cost and implementation issues. I stick with what they have for one simple reason, of all the DC apps out there, Rosetta seems to me to be the most important and the one that has the greatest chance of making the medical "breakthroughs".
So lets at least agree that BOINC is very flawed and stop beating this issue to death in every other thread on this forum.
Both sides have voiced their opinion ad nauseum.
To continue it day after day after day just announces to the world that your prime motives aren't a better credit system, but to cripple the project by driving away any new person that might happen to look at these forums for information about what to run on their own machines.
Last point, I sincerly doubt that on August 8 your going to see a new system magically appear. It will take time for Baker Labs to design and impliment a new system whatever it is.
What say we all just cut them some slack and let them do what they will do.
If your that unhappy with their decisions, you always have the right to leave and go elsewhere.
They damn well know how everyone feels by now so what is to be gained by harping on it daily?
Thanks for your time,
Movieman

---

I am kind of annoyed at some of the people here that have decided to attack me.

So lets see if I make it clear:

1- If there are optimized BOINC clients for Windows, Linux and yes even for Macs,it is because the BOINC developers have kept the source code open. One cannot optimize what is not accesible. So dont blame the users, blame the developers that have not secured their code.

2- There is only one way to prevent BOINC optimized clients: close the source code and encript it in such a way that it will take an explicit act : the hacking of the code to produce the optimizers.

3- The use of optimizers was known and was encouraged and worst it has never been prohibited in the specific projects. So no project rules have been broken by the use of the optimized clients.

4- The use of multiple quorums, while pausible for some projects is not a solution for others. Trying to force all projects in the same mold of runing is not practical. BTW there are ways of getting arround multiple quorums, so they are not safe either.

5- It is also a known fact that a group of people have made it their crusade to attack any individual or teams that use the optimized clients and that in their crusade they have been very free in their use of the verb and name "Cheat". It is also know known that those people were responsible for a serious fight that affected the civility in the oldest DC project there is : SETI., It is also known that thay have moved their crusade here.

6- It is also a known fact that those people that have come here to create the fighting have targeted the members of the top teams in the project. Once they attack, then they cry foul when those team members defend themselves.

Those are facts. You may not agree with me but your disagreing with me doesnt make those facts less valid.


If a person doesn't want to use the optimized clients : that is their personal call and I respect it. But, as long as optimized clients are not explicitely prohibited , I expect the same respect for those of us that have decided to use them. Call me a cheat, insinuate that I am a cheat if you want but then do not expect me to allow you to insult me and not challenge you. Simply stated A LIE repeated and unchallenged becomes a truth. And calling a person that uses an optimized client a cheat , is a lie.


So all I am saying is : want to stop the manipulation of the BOINC client; want to stop the optimized clients ( for windows, for linux and for Macs) join me in asking the BOINC developers to stop having their source code accesible and to heavily encript the credit granting/parameters code of the client. For as long as the source code for BOINC is open, NOTHING can stop tampering with the code, regardless of the project. Nothing can prevent the optimization of the clients.

The only other alternative to prevent tampering with the credits is removing the project from BOINC, in a more secure platform whose code is not open for every one to tinker with it. That is a fact. If this fact gets the "Boinc Purists" angry at me so be it.

And calling a person that uses an optimized client a cheat , is a lie.

No, it's the truth.....

Getting more credits for the same work by using "optimized clients" than others using the normal clients is cheating. That's a fact and not a lie....

---

Teddies....

Regardless of our personal feelings on the matter - the project has stated that it's okay to use optimized boinc clients with the non optimized Rosetta client. We'll be using a Rosetta only benchmark on Ralph to get an credit score for a model/decoy for each WU by running each general WU on Ralph.

It's pointless to keep bringing this up over and over and over and over and over. And over. And over.

Those who have enough ram to run the client in ram and not in virtual ram will have an advantage. Those that Overclock reliably will have an advantage. The dynamics of the game will change - and we'll start to see WU/hr ratings or new credit/hr ratings for each cpu type. We'll be encouraged to produce more (good!) to get a higher score. Those that see flaws in the proposed new system should bring them up in a thread devoted to the new system - so they can be looked at and worked around if they're legitimate flaws. Some of the creative types here should be able to think of ways around the currently proposed new system..

But the issue needs to be put on hold until after the 8th..

---

And calling a person that uses an optimized client a cheat , is a lie.

No, it's the truth.....

Getting more credits for the same work by using "optimized clients" than others using the normal clients is cheating. That's a fact and not a lie....

The problem, is that optimized clients are allowed. When a tool is allowed it's use is not illegitimate.

If you are against optimized clients join me in asking that BOINC cease to be an open source code application. Ask them to encript the credit granting code. Be consistent with your anti optimizing stance...optimizing happens because BOINC Code being open allows it. Change the root of the problem, complaint about the root of the problems : the BOINC Application perse.

BTW: I will like to know how the members of your team that use optimized clients feel about your claim that those who use optimized clients are cheats.

BTW: I will like to know how the members of your team that use optimized clients feel about your claim that those who use optimized clients are cheats.

Not using the optimized version at the present but have in the past. Morphy is entitled to his opinion as anyone else is. My stamp of approval isn't necessary.
Sometimes it is best to agree to disagree and leave it at that. :)

---

BTW: I will like to know how the members of your team that use optimized clients feel about your claim that those who use optimized clients are cheats.

Not using the optimized version at the present but have in the past. Morphy is entitled to his opinion as anyone else is. My stamp of approval isn't necessary.
Sometimes it is best to agree to disagree and leave it at that. :)

It is not your stamp of approval of your teamate that is the issue .

So how does it feels to have been a cheat? And to be a teamate to cheats?

Your teammate has directly accused ALL that use optimized clients as cheats; no if or buts. So he is accusing ANY and ALL the members members of your team that use or have used the optimized clients of being cheats. That is the only logical conclusion that can follow from his statement:

"Getting more credits for the same work by using "optimized clients" than others using the normal clients is cheating. That's a fact and not a lie.... "

So, nice teamate.

Your teammate has directly accused ALL that use optimized clients as cheats; no if or buts. So he is accusing ANY and ALL the members members of your team that use or have used the optimized clients of being cheats. That is the only logical conclusion that can follow from his statement:

"Getting more credits for the same work by using "optimized clients" than others using the normal clients is cheating. That's a fact and not a lie.... "

So, nice teamate.

Must be some of that special kind of 'team spirit' and 'community' that they're ranting about over in the Cafe.

---

Proudly Banned from Predictator@Home and now Cosmology@home as well. Added SETI to the list today. Temporary ban only - so need to work harder :)



"You can't fix stupid" (Ron White)

I like the 'agree to disagree' comment.

In an effort change direction on this thread. . What does everyone think of the 'per wu type' credit system? (By that I mean a given WU is run by several dozen people on Ralph, and a credit per decoy/simulation is determined) Is this fairer? Since it's based on an average calculation time, is it ok if intel gets more credits per hour than amd (or vice-versa, just pointing out that one type of CPU might be inherently better at crunching the code)?

What is the best method for determining the credits per decoy/simulation? Perhaps by taking the average of all the Ralph times and base it around x credits per hour? As an example, if one simulation takes an average of 1.5 hours, and the standard reward is 10 credits per hour, each simulation on Rosetta would grant 15 credits. Regardless of how fast your computer is (or the version of boinc you're running), whenever you return one simulation from that wu type, you get 15 credits.

Postives, negatives, pitfalls of this method?

-Ethan

---

One of the possible problems that I've seen mentioned so far is that people can cause their system to fail WUs and return them faster than they'd be able to generate a valid WU. So we'll have to think up an approach to deal with that possibility.

---

One of the possible problems that I've seen mentioned so far is that people can cause their system to fail WUs and return them faster than they'd be able to generate a valid WU. So we'll have to think up an approach to deal with that possibility.

That would definately be a concern :) If you abort a WU, doesn't it report the actual number of simulations you finished? My bet is people could abort until their mouse fails, but each aborted WU will have 0 completed simulations.

-E

---

I like the 'agree to disagree' comment.

In an effort change direction on this thread. . What does everyone think of the 'per wu type' credit system? (By that I mean a given WU is run by several dozen people on Ralph, and a credit per decoy/simulation is determined) Is this fairer? Since it's based on an average calculation time, is it ok if intel gets more credits per hour than amd (or vice-versa, just pointing out that one type of CPU might be inherently better at crunching the code)?

What is the best method for determining the credits per decoy/simulation? Perhaps by taking the average of all the Ralph times and base it around x credits per hour? As an example, if one simulation takes an average of 1.5 hours, and the standard reward is 10 credits per hour, each simulation on Rosetta would grant 15 credits. Regardless of how fast your computer is (or the version of boinc you're running), whenever you return one simulation from that wu type, you get 15 credits.

Postives, negatives, pitfalls of this method?

-Ethan

Any system that awards credits based on actual work done I am in favor of and I would love to discuss this with any project administrator or developer at length on my dime.
Short version:
Dump this ability to adjust the timeframes from 1 hour to 24.
Come up with a system where the WU itself has X decoys/simulations that need to be run on it to give the needed data to the project.
IE: If the WU needs 1000 simulations run, thats what it needs. Lets give back meaningful data that can be used.
If you use a method of X points per simulation, then the speed with which the donor PC can "do" those simulations will answer all the credit issues.
This is what I have suggested in the past, like piece work in a factory, you get "paid" for what you do, not what some arbitrary benchmark says your machine "can" do.
Thanks for your time,
Movieman

---

Movieman,
Thanks for your input! I might be missing one of your points, however.

Why would it be necessary to limit WU length? The length of time it takes to process a simulation/decoy is what would determine credits, not how many your computer does as part of a given WU. For those on dial-up, the ability to download a 24-hour work unit is a positive thing.

The project may need 100,000 simulations/decoys (I use the both terms since decoy is a funny term to most of us), but they don't care which WU's they come from (1 hour or 24 hour).

Let me know if that makes sense, and if I'm addressing your concern about WU length.

Thanks,
Ethan

---

Postives, negatives, pitfalls of this method?

Yes, the current system of granting credit for failed WUs would have to be reviewed.

Positives, AMD vs Intel, Windows vs Linux, there will still be variations, but at least there is some standard that everyone is held equally to.

Negatives, I've noticed that what appear to be identical WUs can vary in their time per model. I've got a dual-CPU, and it is not uncommon to see two WUs with same name (excepting last numbers on the end) to both be crunching for 12+hrs and be within 10 minutes of each others elapsed time, and one will have 50 models done, and the other has 60. I am guessing that this is due to varience in the course travelled by the random starting point... and so this randomness should work its way in to the benchmarking as well, and average out. But I did want to point out the variance. I can see that changes to the algorythms used in Rosetta over time could easily cause some models to be aborted early due to lack of prospects, and others to be crunched intensely, due to excellant prospects. And it will all depend on the random number you draw as to which you are working on.

Movieman, I for one like the WU runtime preference. It is a bit confusing to new users... if they even discover it. But I think it gives needed flexibility for people to tailor the work to how they use their PC. I mean if the PC is only on 4 hours a day, and 3 days a week... giving it a WU with a fixed number of models required, with a 7 or 14 day deadline seems to me it would lead to a lot of missed deadlines for some people that don't crunch 24/7.

But I think the ideal would be to have incremental uploads of completed models and a dialogue with the Rosetta server to determine if more models for the same protein would be desirable. I believe this can be achieved if the trikle-up method of reporting results is used. This way, you could basically take advantage of limited internet time, greatly reduce bandwidth, and always have useful work to do (because more models on a given protein is never a waste of time, so just keep it and crunch more models if unable to contact server).

---

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/