Open letter to PG [with response]

[SnW]

Great and all but i am just so close to leave all the FAH alone
only 6701 WU's yet i see others getting BIGadv WU's and non 67xx WU's

611GPu WU's all over the place 9xx WU's nice but cripple the SMP performance by o whathever !!!!!!!!!!!!!
i get GPU clients hanghing all over the place (great software) yes stanford you can go and have a change but you had to DO this a longggggggggg time ago

Ati cards ow NVM ha

Yes i am a so sick of this Folding and nothing is getting better just a few words that solve about NOTHING

[7im]

Yes, we will see if a new v7 client solves anything or not.

[AtwaterFS]

I for one am dying to use my evergreen gpu plus its little brother Cayman (6870) is coming home.

Cayman is 69x0 - Barts (which just "dropped") is 68x0... I'm intrigued by Cayman but most likely I'll wait til 28nm generation.


How nice would it be to actually use my ATI card for FAH again? Its been dormant in protest for a while now

[VijayPande]

Mr. Pande, I can't understand why you are reluctant to release v7 to beta testers I Think everyone who are participating in beta program will agree, that we are quite understanding bunch in there. We worked with Joe when testing a4 core(if I remember correctly) and me personally I liked his enthusiasm during that period. Unless we were pain in the arse for him
I for one am dying to use my evergreen gpu plus its little brother Cayman (6870) is coming home.
So if there is any showstoppers in a way during the beta testing it wouldn't be any problems for us.
And as I understand base code is almost ready, just new features are added.

There are a lot of rough edges that Joe already knows about. Beta testers would naturally point those out, but that would be a waste of your time, since we already know about it. I think it makes the most sense for Joe to get those issues finished and smooth it out a bit more before putting it out for beta.

[muziqaz]

AtwaterFS, yes, I was writing early in the morning on my phone half asleep. Was thinking about 9 clicked 8


Mr. Pande, fair enough.

[codysluder]

There are a lot of rough edges that Joe already knows about. Beta testers would naturally point those out, but that would be a waste of your time, since we already know about it. I think it makes the most sense for Joe to get those issues finished and smooth it out a bit more before putting it out for beta.

Like others, I hope to be able to beta test v7 soon but I do understand what you're saying. Creating a unified client that can do everything that the latest smp and the latest gpu clients can do (when both continue to be moving targets, albeit not moving rapidly) has to be a major challenge, particularly since you've said it's a total rewrite, thereby introducing a whole family of new bugs. I guess it all boils down to how many issues you already know about and how important they are.

Beta testing is only useful if it finds bugs you don't know about yet. As long as Joe's team has their hands full with critical known bugs there's no point in finding more. I trust you to distinguish between serious and minor bugs and to decide when it's ready for beta testing, even though I'm just as impatient as the next guy.

[7im]

I agree with codysluder. There is no point in wasting the time or duplicating the efforts in beta testing until all the big known issues from internal testing are fixed, and the minimum feature set is ready to go. Besides, finding new bugs is more fun, so fix all the old ones first.

[road-runner]

Yes please try and get the big bugs out before releasing it, if not some people will be complaining why was it released without testing...

[muziqaz]

Road-runner, are you a betatester? If not then you don't need to worry about bugs; if yes - you are moaning too much for a beta tester.

[road-runner]

Road-runner, are you a betatester? If not then you don't need to worry about bugs; if yes - you are moaning too much for a beta tester.

No not a beta tester, but ever time something new comes out there are bugs, problems it seems. Then people complain why was it released and not tested is what I was saying. I hope your right nothing to worry about, but I have seen plenty of stuff released that has bugs...

[VijayPande]

We do a great deal of testing, but sometimes new issues crop up when we go to a larger scale. For example, v7 is going through alpha testing with a group of people outside of Stanford who are very familiar with FAH. However, testing can't guarantee zero bugs, since going to a larger scale often finds new problems. This is true of all software: I don't know any company that releases software without bugs.

The alpha team has done a great job to flesh out several issues and Joe has been working on them. When we do a beta release, there will likely still be some issues, but hopefully fewer and less severe.

[theo343]

Really looking forward to test the v7 beta When it releases after all testing is done it can be a true icebreaker in bringing even more people into FAH.

[VijayPande]

Really looking forward to test the v7 beta When it releases after all testing is done it can be a true icebreaker in bringing even more people into FAH.

Yea, that was the plan. Also, all the backend stuff should help too with sever reliability, which is a big deal for donors of course. In the end, we would have upgraded almost all of the FAH software by 2011, which is actually a lot of work. That should also set us up nicely for the future, as the new code base is clean, more functional, easier to modify, and much higher performance (esp the server code). There will be (and has been) some growing pains in switching over, but the future is especially exciting I think.