Folding Forum

I've been an active supporter and advocate of this project for years upon years. But I've finally had it. Running the SMP client since its first public appearance has left me jaded. Since then, I've seen excuses for every issue that this client has, but no fixes. I'm sure that you'll have a thing or two to say about this, something about it being a 'high-performance' client, beta software and the great showstopper "you could always run the normal console client", I've heard them all.

The sorry state of this 'highly valued' part of the project leaves me in doubt of the competence of those responsible for the project, the research and the software. Why then should I trust that the work units that I (try to) crunch actually help to do relevant research? I can't really base my judgment on the papers that you produce, for I have nor the time, the inclination or the knowledge required to know the difference between a shoddy paper and a gleaming example of science done through distributed computing. So I base my opinion of this project on what I do know and see, and that is a piece of poorly written software that I run voluntarily in the hope that the research done with that software has less faults.

To whom it may concern: you should take note of this feedback, as I've had a great deal more patience with this piece of software than I (or many others) would have with any other software. Yet after all these years, my patience has worn out.

To everybody else: feel free to try and convince me otherwise, lock this thread or call me names.

'bye Firestarter,
Thanks for your contributions to this noble project.
Future generations will live better lives because of your efforts.
It's been great having you with us as a folding donor and forum contributor.
Thanks again,
Alan

Hopefully in couple of years down the road all those issues are fixed, but it will take those couple of years to get there.

Sorry your patience has worn out. But the buggy nature of the client is mostly to do with MPICH, which is a third party software over which Stanford has no control.

What would you have Stanford do instead? Poorly written software? Sure, but not the software written by Stanford. Blame the 3rd party software.

And what's wrong with the GPU or CPU clients?

I also find the SMP client hard to use, limited and not intuitive, thats probably why its still at beta and hidden from non experts.
However, I like the GPU2 client and eagerly waiting for the GPU2 Nvidia; next week right?
I am guessing the Stanford team is concentrating their limited manpower in developing, debugging and testing the GPU2
because it holds a lot of potential. I am basing my intuition from reading posts and blogs from Stanford staff and news about
upcoming Nvidia support. They all seem to display great enthusiasm about the power of the GPU.
Here's a clip from the news:

The new GeForce cards are expected to hit more than 650 nanoseconds of protein simulation in a single day, while the Radeon HD 3870 is stuck at about 170 ns. The Playstation 3 is able to produce "only" 100 ns of simulation, while a quad-core CPU creates an output of just four nanoseconds. For those who are keeping count: The GeForce GPU will be about 163 times faster than a quad-core processor in this specific application.

As for Firestarter, I will try to grant your wish to convince you to go back and fold.
First, you have more patience than me. I only lasted less than a day folding SMP. I quit and abandoned SMP as soon as I saw it.
Secondly, would you rather fold for just 4ns simulation with your quad-core CPU and struggle using the SMP ?
If I were you, get a cheap 3870 and fold for 170ns protein simulation without the SMP hassle.
As for me, I am saving my lunch money for the new GeForce GPU, I heard its 163 times faster than a quad-core and a very good
folder. You can also you use it to learn CUDA.

I mush rather use the GPU2 folding client over any other. It hasn't given me problems that would stop me from contributing.
I fold because I don't want children to ever see their parents dieing in front of them, with them not able to do a single thing about it.

Try WCG if you want to contribute to another project. I have several rigs running 64 bit Ubuntu and have been having trouble with them stalling at the end of the SMP WU and I have to use Qfix and that has gotten old. I have switched most of mine back to WCG as I do not have time to baby sit the rigs when they stop....

7im wrote:Sorry your patience has worn out. But the buggy nature of the client is mostly to do with MPICH, which is a third party software over which Stanford has no control.

What would you have Stanford do instead? Poorly written software? Sure, but not the software written by Stanford. Blame the 3rd party software.

Wrong. MPICH2 and DeinoMPI are Open Source software, Stanford can invest the developer time to fix the 3rd party code they're using. That's how it's done in the Open Source world. You're not dependant on the (single) supplier for fixes.

Whether they want and can make this develope time investment is another question, but they have the ability to fix the code their using, even the code they didn't write themselves.

oh is that how its done in the Open Source world; they write crappy codes and hire them for a million bucks to fix it so you can use it?
interesting business plan.

dittopb wrote:oh is that how its done in the Open Source world; they write crappy codes and hire them for a million bucks to fix it so you can use it?
interesting business plan.

Not to deviate too far from the topic, but yes if the commercial party is interested in some OSS component then they very often hire developers to work on that OSS component and they do pay for it. For example IBM is the largest one.

dittopb wrote:oh is that how its done in the Open Source world; they write crappy codes and hire them for a million bucks to fix it so you can use it?
interesting business plan.

Wrong too. You write code and share this with the world under a license which gives the user the four freedoms, as the FSF says:

http://www.gnu.org/philosophy/free-sw.html wrote:Free software is a matter of the users' freedom to run, copy, distribute, study, change and improve the software. More precisely, it refers to four kinds of freedom, for the users of the software:

* The freedom to run the program, for any purpose (freedom 0).
* The freedom to study how the program works, and adapt it to your needs (freedom 1). Access to the source code is a precondition for this.
* The freedom to redistribute copies so you can help your neighbor (freedom 2).
* The freedom to improve the program, and release your improvements to the public, so that the whole community benefits (freedom 3). Access to the source code is a precondition for this.

With the freedoms granted to Stanford they have the ability to use freedom 1 and freedom 3 to fix the issues with MPICH2 and DeinoMPI, and while solving their own problem with the software, solving this problem for all other users too when the fix is contributed back. That's how it's done in the Open Source community. Scratching your own itch, but by sharing your solution with the community, everyones itch is scratched.

Your ignorance does not reflect well on you, please go troll somewhere else (or not at all, preferably).

oh you forgot the freedom to write crappy codes.

chill out dude, can't you take a joke?

anger.on();

Not to mention, it's not uncommon that problem reports are ignored
(I don't mean lack of response here -- that would actually be less harmful)
and reporters are considered clueless idiots.

anger.off();


Cheers!
tear

tear wrote:anger.on();

Not to mention, it's not uncommon that problem reports are ignored
(I don't mean lack of response here -- that would actually be less harmful)
and reporters are considered clueless idiots.

anger.off();


Cheers!
tear

There is room for improvement, but I do not see the situation being that bad.
The "clueless idiots" part will need some clearer explanation.

tear wrote:anger.on();

Not to mention, it's not uncommon that problem reports are ignored
(I don't mean lack of response here -- that would actually be less harmful)
and reporters are considered clueless idiots.

anger.off();


Cheers!
tear

So you've used Fedora 9 as well.