Folding Forum

We've been working behind the scenes on a revamp of our web site. It went live today (http://folding.stanford.edu/home). This is part of our larger plan to make FAH more friendly and easy to use, especially to non-experts.

With that said, we're now thinking about next steps to make FAH more fun and appealing to experts, such as computer enthusiasts and gamers. We're in the early stages of deciding what would be useful there. If you have ideas, please do give us some feedback. As always, we can't implement everything, but we are curious to see what people think.

http://folding.typepad.com/news/2013/07 ... -site.html

I'm sure this isn't quite what you intended, but better proof-reading on (at least some) of your web pages wouldn't do any harm!

Maybe I'm picky (actually, I know I am ), but a few examples that leapt out at me on this page:

http://folding.stanford.edu/home/guide/ ... ion-guide/

"We'll get their a lot faster ..."

Should be "We'll get there a lot faster ..."

" While SMP and GPU folding is much more scientifically productive compared to their uniprocessor counterpart, they should not be run on machines which can tolerate heavy use."

Should be "While SMP and GPU folding is much more scientifically productive compared to their uniprocessor counterpart, they should only be run on machines which can tolerate heavy use." ?

Under Next Unit Percentage:

"99% is the default."

No it isn't (not on mine anyway), the default is 98%

Seriously, I'll give it some thought.

On this page http://folding.stanford.edu/home/commun ... port/#faqs under the FAQ section these links are not working:
- Introduction to V7 (current software)
- Intermediate V7 FAQ

Please update this. Thank you.

"We are scientists. Citizens. Gamers. Thinkers. Mothers. Brothers. Friends. Family."

No fathers?

John

With that said, we're now thinking about next steps to make FAH more fun and appealing to experts, such as computer enthusiasts and gamers.

I moderate a PC hardware/enthusiast forum with just over 25,000 subscribers, and I've tried selling the idea of F@H somewhat unsuccessfully for the following reasons:
Enthusiasts love comparing benchmarks of their hardware (see:http://hwbot.org/, http://www.3dmark.com/hall-of-fame-2/) and not only that, but arguing over which parts are the best based on the results of these benchmarks . As it stands, I've come across some third-party databases of user-submitted content that shows relative performance of various parts for folding such as PPD, but the manual nature of these services leads to sparse, and sometimes incomparable, data. It would be cool if you could "opt-in" to have the FAHClient submit statistics along with the WUs containing your hardware IDs and PPD/TPF stats, which would then get entered into a query-able database on the site. Now I would imagine this would require additional resources on the back-end to implement, but it's just a suggestion.

Along those lines, it would also be neat to have a "benchmarking mode" that would simulate a fold (yes, a simulation of a simulation ) for a minute and extrapolate PPD, in a standardized way. This would further aid in hardware comparisons, because I know that some WUs/projects fold better/faster than others.

As a last thought, have you considered contacting Steam (http://store.steampowered.com/about/)? That's the mecca for PC gamers today. You would get a lot more traffic if you got them to publicize F@H. Not too long ago, when they released their Linux client, they had an offer where if you installed Steam on Linux and played one of their games, they'd give you a Tux penguin hat to wear in it. I'm not sure what the numbers are at now, but shortly after the Linux client released, Linux users made up just over 2% of the total user-base (which when you consider that there's over 3 million players in-game on Steam at a single time, is no small number!).

billford wrote: Under Next Unit Percentage:

"99% is the default."

No it isn't (not on mine anyway), the default is 98%

This is a known bug with the current 7.3.6 version of the client. It worked correctly in the prior version and hopefully will be fixed.

billford wrote:I'm sure this isn't quite what you intended, but better proof-reading on (at least some) of your web pages wouldn't do any harm!

Maybe I'm picky (actually, I know I am ), but a few examples that leapt out at me on this page:

http://folding.stanford.edu/home/guide/ ... ion-guide/

"We'll get their a lot faster ..."

Should be "We'll get there a lot faster ..."

" While SMP and GPU folding is much more scientifically productive compared to their uniprocessor counterpart, they should not be run on machines which can tolerate heavy use."

Should be "While SMP and GPU folding is much more scientifically productive compared to their uniprocessor counterpart, they should only be run on machines which can tolerate heavy use." ?

Under Next Unit Percentage:

"99% is the default."

No it isn't (not on mine anyway), the default is 98%

Seriously, I'll give it some thought.

2nd and 3rd points fixed. Didn't see the "faster" comment on that same page.

dc_coder_84 wrote:On this page http://folding.stanford.edu/home/commun ... port/#faqs under the FAQ section these links are not working:
- Introduction to V7 (current software)
- Intermediate V7 FAQ

Please update this. Thank you.

Should be fixed. Please give time for the cached version to update the live version.

Ta.

7im wrote:Didn't see the "faster" comment on that same page.

It's in the "Help us reach 1,000,000!" box at the top right.

We'll get their a lot faster if you tell your friends about our project.

http://folding.stanford.edu/home/community-support/

VijayPande wrote:we're now thinking about next steps to make FAH more fun and appealing to experts, such as computer enthusiasts and gamers.

How about completing the current step for Mac users before going on to the next one?

http://folding.stanford.edu/home/guide/#ntoc8

A mature beta version of V7 for OSX is available, and an Install Guide is in progress.

My bold

bollix47 wrote:

We'll get their a lot faster if you tell your friends about our project.

http://folding.stanford.edu/home/community-support/

Thanks, will get to this shortly. I only checked the page contents last time. That sidebar box is a separate item.

billford wrote:Ta.

7im wrote:Didn't see the "faster" comment on that same page.

It's in the "Help us reach 1,000,000!" box at the top right.

The Primary Sidebar box has been fixed. Give it 30 minutes to cache out.

@Helrich:

There is a separate folding benchmarking program called FAHBench. It reports ns/day instead of estimated PPD, and it only runs one particular protein, but otherwise I think it's what you're looking for. It's not actually GPU-only, though to benchmark a CPU you have to download additional stuff, and it's not entirely accurate for them because CPU folding uses a different code base. Perhaps it should be linked to from somewhere on the new webpage, as I didn't see it in a cursory search.

As to an automatic benchmark repository, you'd have to account for hardware that doesn't fold 24/7, as that will skew the PPD figures a great deal. That's hardly a show-stopper, but it would need to be taken into account. I imagine that's more work than they want to do at this time on something that doesn't improve the science.

Zagen30 wrote:
Perhaps it should be linked to from somewhere on the new webpage, as I didn't see it in a cursory search.

Added to the Utilities Download page, where Memtest and StressCPU are also located. http://folding.stanford.edu/home/download-utilities (Again, give it a little time to propagate out...)