Discussion: what is holding F@h back?

[iceman1992]

The amount of processing power in each game console has NOT grown very rapidly -- although the number of units keeps growing. The processing power in home computers that are used for gaming does continue to grow. Today, the PS3 is just about as useful to FAH as a relative cheap PC with a low to mid-range GPU (though the were mid-to high-range GPUs at the time).

Why not make a client for the XBOX360? Rumors say that the next console generation will pack quite a number of processing cores, that might be good for FAH

[Jesse_V]

The amount of processing power in each game console has NOT grown very rapidly -- although the number of units keeps growing. The processing power in home computers that are used for gaming does continue to grow. Today, the PS3 is just about as useful to FAH as a relative cheap PC with a low to mid-range GPU (though the were mid-to high-range GPUs at the time).

Why not make a client for the XBOX360? Rumors say that the next console generation will pack quite a number of processing cores, that might be good for FAH

Sorry, probably not going to happen.
http://folding.stanford.edu/English/FAQ-Press:

Is there anything about the PlayStation3 that makes it especially well-suited to the task, or could the project feasibly run on the Xbox 360 too? Do you have any plans to further expand the project?

There are several aspects which makes the PS3 well suited for this task. First, it is powerful: its main processor – the Cell Processor – is very powerful. In fact, we get a 20x speed increase over PC's. That's not 20%, but 20x, i.e. a 2,000% increase over a typical PC. That sort of speed can't be found on PC's or on the xbox's central processor.

[iceman1992]

Sorry, probably not going to happen.
http://folding.stanford.edu/English/FAQ-Press:

Is there anything about the PlayStation3 that makes it especially well-suited to the task, or could the project feasibly run on the Xbox 360 too? Do you have any plans to further expand the project?

There are several aspects which makes the PS3 well suited for this task. First, it is powerful: its main processor – the Cell Processor – is very powerful. In fact, we get a 20x speed increase over PC's. That's not 20%, but 20x, i.e. a 2,000% increase over a typical PC. That sort of speed can't be found on PC's or on the xbox's central processor.

Okay then let's just wait for the PS4 and XBOX720.

One of the things that can be annoying for folders is the points system. For example, my i5 folds a 7022 in about 3,5 hours and gets around 4800 points, but then folds a 7200 in 5,5 hours and gets 3000 points, or a 8024 in 2,5 hours for 1300 points, which seems really inconsistent. For folders who really care about the points more than the folding itself, this may turn them off.

And yes, the visualization does help when I try to show FAH to people.

[SASinUtah]

When we make a direct cash contribution to Stanford for Folding, we can take a tax write off. Since we, via folding, make a direct contribution to Stanford is there a way to assign a dollar value per work unit? These work units become intellectual property of the respective lab of the respective university (Stanford, Temple, Hong Kong) which can be used for monetary gain via medicinal patents, medical procedure development, medical equipment development; this is a common practice in the university arena to help continue to pay for research (see http://www.yale.edu/ocr/pfg/policies/patents.html, http://www.montanakaimin.com/news/um-wa ... -1.2750497, http://www.sfgate.com/cgi-bin/article.c ... 1FO7R8.DTL).

Although there may be some unintended consequences (i.e., if the owner/user of the pc is not the tax filer), this may be a way to:
A) encourage corporations to donate their formidable resources
B) encourage individuals not familiar with Folding to join
C) create an explanation for hesitant spouses for those who have caught the bug and want to expand
D) provide the headline to those who have suffered personal tragedy due to the diseases F@H targets (I suspect Jesse_V understands this aspect) to encourage participation.

Our purpose, it seems, is to encourage more legal cpu/gpu resources, and to do so we need to: 1) get the word out in the sound bite/four second attention span headlines so that the word makes an retained impression, and 2) put the work done into a language which makes sense to the layman during that sound bite so that we can create a mutually beneficial hook. We need to get the word out in a way which shows an immediate return (tax write off), return for those for whom pc use/upgrading/optimizing is a favored past time (points, I suspect the majority of us now), and help retain those who have the Folding bug yet are becoming discouraged for one reason or another.

[VijayPande]

It has been interesting to read this thread. We have had different priorities at different times. Years ago, we had a big push to get to new platforms such as GPUs and true SMP (many other d.c. projects just run multiple WUs on each core, which uses cores, but at least for us is a big loss scientifically compared to true SMP). Now, our main goal is to grow the number of clients so we can get more work done. This includes items like (in no special order):
* Make GPU computing easier. This includes install issues and heat issues, but especially usability issues (GPUs don't have process priority, so efficient GPU code can really monopolize a machine). I've been thinking about bringing the screen saver back for GPUs since that would be a natural way around this. If we could get 100,000 GPUs, we would be at the ~100petaflop level which would be an amazing feat computationally, but more importantly would be a huge resource for FAH's scientific calculations. Towards this end, we have been experimenting with a linux GPU client (no promises or an ETA, but it is something we've been looking into).
* Make running FAH in general easier for newbies/non-experts. The V7 client I think has helped with this, but there's still a lot we can do there.
* Provide more feedback in the client. We've made a push to do more with the blog and I think that's had positive reactions. I think the client could do a lot more as well. The v7 client rewrite took a lot of effort since we started from scratch. Adding in new features won't take nearly as long. I'd also like to improve its GUI to be slicker, more modern, and more visually appealing.

I should stress that this isn't to say that I don't think these are the only issues but that these are the ones on the forefront of my mind these days in terms of the donor experience (there are also tons of other things, in terms of the server backend and the science, but I think that wasn't the main point of this thread).

Also, I should add that I'm not so concerned about the long term prospects along the lines of d.c. going away. There will always be plenty of machines out there, but just not an exponential growth. If we can make FAH much easier to use and less intrusive for donors and make its benefits more clear, I think we can do very well. There are 100M GPUs out there. It would be great to get even 1% of them running FAH!

[VijayPande]

PS The title of this thread implies that FAH is lacking (being "held back"), but I think we've collectively accomplished a great deal. I'm particularly excited to see the Alzheimer's story develop over the years, starting with methods development years ago, then testing in biophysical systems, then direct simulation of Aß peptides, then testing of those predictions, and now most recently drug candidates designed from FAH simulations. For me, that's been a huge amount of progress considering we were starting from a time in 2000 when many people thought what we wanted to do was impossible.

Of course, I'd love to do even more. Perhaps a different way to think of this is more like "How can we do even more?" I think that's the spirit people have had in mind in these posts, so my comment here may be a bit redundant, but I think it's worth noting that we have collectively done so much. And I'm very excited about that and grateful to all of you for helping make this possible.

[Jesse_V]

PS The title of this thread implies that FAH is lacking (being "held back"), but I think we've collectively accomplished a great deal. I'm particularly excited to see the Alzheimer's story develop over the years, starting with methods development years ago, then testing in biophysical systems, then direct simulation of Aß peptides, then testing of those predictions, and now most recently drug candidates designed from FAH simulations. For me, that's been a huge amount of progress considering we were starting from a time in 2000 when many people thought what we wanted to do was impossible.

Of course, I'd love to do even more. Perhaps a different way to think of this is more like "How can we do even more?" I think that's the spirit people have had in mind in these posts, so my comment here may be a bit redundant, but I think it's worth noting that we have collectively done so much. And I'm very excited about that and grateful to all of you for helping make this possible.

Thank you for your reply and the one above as well. I'm sorry if the title of the thread was slightly misleading. I've been very impressed from the accomplishments that I've seen F@h do, and indeed as you say I was trying to see if we could list things that were preventing it from "doing more". This is similar to "what is holding F@h back" because one can ask "Why isn't F@h at the 100 petaFLOP level already?" There's got to be things that are limiting F@h's growth, and it's nice to see that you're taking steps to address them.

I'd like to take the opportunity to say that I would be in full support of a GPU screensaver; I really think that is a really good solution, and as long as it shuts down quickly you don't have to worry about lag. I'll bet that the upcoming GPU-CPU hybrid core could applied there as well. I'll be looking forward to seeing improvements to V7, I think your group has been doing a really great job there and I'd like to see more. Thanks for working on a Linux GPU core as well!

[Jorge1950]

@ Dr. Vijay Pande: I think you can do more direct communication with donors through the "About Project" window of V7. A good message in that window can really influence donors to stick with this project.
viewtopic.php?f=16&t=21367&start=0#p213460

[jcoffland]

I agree that the project description area could be better utilized. This is currently up to the project creators who likely have variable communication skills. Perhaps someone at PG should be responsible for helping to assure new projects are rolled out with engaging descriptions which are understandable by laymen. Eye catching pictures might help too.

Giving donors a choice about what they contribute to could be a way to get them more engaged. This is equivalent to allowing cherry-picking, something we've tried hard to eliminate. Part of the problem here is that there are many important projects which don't have names that people can identify with. Currently we ask donors to trust the scientists to allocate donated resources efficiently.

Regarding the v7 GUI. I've heard both complaints and praise so it's hard to know what to think. It's very difficult to balance both the needs of the expert users, which form the core of Folding@home, and the average user. This is the idea behind the novice vs. advanced and expert modes. Another issue is that we cannot afford to develop platform specific GUIs for FAH. We went with the Gtk GUI tool kit which is not perfect on all platforms. Especially there are problems with Gtk on OSX.

I am a nerd not a graphic designer so it's likely I need help with this. Here are a few ways others can contribute to this area in order of increasing difficulty:

• Incrementally improve the current v7 GUI design by submitting enhancement tickets.
• Create a mock up of a GUI design that you believe is better than v7 and post it to the forums.
• Create a new Gtk theme for v7.
• Write a guide on how to create Gtk themes for v7.
• Write your own, possibly platform specific, GUI from scratch. See: Client Remote Interface documentation

[VijayPande]

PS Regarding "Why isn't F@h at the 100 petaFLOP level already?" –– I think that's a good question. If we can get 100,000 newish (~2 years old or less) GPUs, we'll be at the 100PFLOP level. That's why I put GPUs at such a high priority.

[Jesse_V]

I agree that the project description area could be better utilized. This is currently up to the project creators who likely have variable communication skills. Perhaps someone at PG should be responsible for helping to assure new projects are rolled out with engaging descriptions which are understandable by laymen. Eye catching pictures might help too.

Giving donors a choice about what they contribute to could be a way to get them more engaged. This is equivalent to allowing cherry-picking, something we've tried hard to eliminate. Part of the problem here is that there are many important projects which don't have names that people can identify with. Currently we ask donors to trust the scientists to allocate donated resources efficiently.

Regarding the descriptions, the Beta Team members can help with this, since it seems to me that changing the description at that point is more likely to happen than later on. Diwakar seems to be doing exceptionally well here, just look at his descriptions for his 800[6-9] projects and the 8041/8042 projects. IMO, they should all be like that.

I don't mind trusting you guys with my computational resources. I see the problems with cherry-picking, and I'll bet that implementing that would require a lot of server-side changes on your end. Once again, check out Diwaker's 800[6-9] project descriptions in the link above. Notice that he's just doing general biochemical studies with seemingly little to do with disease research, yet the description is awesome and should make anyone feel that their efforts are worth it. I realize that it probably consumes an extra half-hour of time to write such a description, but I'd speculate that the payoff in donor retention is worth it. I'm sure there's a word-smither in the PG somewhere that could be helpful here.

PS Regarding "Why isn't F@h at the 100 petaFLOP level already?" –– I think that's a good question. If we can get 100,000 newish (~2 years old or less) GPUs, we'll be at the 100PFLOP level. That's why I put GPUs at such a high priority.

One thing that I often have to is pause the GPU slot. Folding on my laptop's Nvidia 240m will cause severe lag on about 20% of YouTube videos, and of course I pause it for games because it lags them out too. I've no idea how it fares for others, but I think a GPU screensaver will really help this problem. Along with better project descriptions, I think it's likely that this will significantly increase the size of the donor base and the number of GPUs that are folding. At least for me, that would really reduce my reserves about spreading the word. Might be a bit ambitious, but I'd like to see something that works well in Windows by September, since that's when school starts up again here.

[7im]

A new class of client types for -cancer, -alzheimer, -parkinsons, etc. would be a nice addition.

Like any other client types, example -advmethods, using that option would state a preference for that class of work unit, but it would never be a guarantee to get those WUs.

And in the end, Stanford can still control the weighting of the server assignments, so if too much of one class is getting cherry-picked, they can rebalance with weights. Sure, it makes Assignment Server logic much more complicated, but hopefully some recent or future upgrades would have better support for something like this. I think it would raise donor participation / retention.

[VijayPande]

A new class of client types for -cancer, -alzheimer, -parkinsons, etc. would be a nice addition.

Like any other client types, example -advmethods, using that option would state a preference for that class of work unit, but it would never be a guarantee to get those WUs.

And in the end, Stanford can still control the weighting of the server assignments, so if too much of one class is getting cherry-picked, they can rebalance with weights. Sure, it makes Assignment Server logic much more complicated, but hopefully some recent or future upgrades would have better support for something like this. I think it would raise donor participation / retention.

This sounds reasonable to me. With the v7 client in full release. I think it's a good time to consider something like this.

[orion]

A new class of client types for -cancer, -alzheimer, -parkinsons, etc. would be a nice addition.

Like any other client types, example -advmethods, using that option would state a preference for that class of work unit, but it would never be a guarantee to get those WUs.

And in the end, Stanford can still control the weighting of the server assignments, so if too much of one class is getting cherry-picked, they can rebalance with weights. Sure, it makes Assignment Server logic much more complicated, but hopefully some recent or future upgrades would have better support for something like this. I think it would raise donor participation / retention.

I couldn't agree with you more with this idea 7im.

I've always thought, with the v7 client, that it would be nice for people to be able to click a button or add a flag for the type of research that they wanted their system to be doing. Allot, if not most, start to fold because they have a family member or a friend with a disease and they want to help out in those areas of research.

Your idea would allow these people to know that they are helping, yet give Stanford control of assignments.

[Jesse_V]

A new class of client types for -cancer, -alzheimer, -parkinsons, etc. would be a nice addition.

Like any other client types, example -advmethods, using that option would state a preference for that class of work unit, but it would never be a guarantee to get those WUs.

And in the end, Stanford can still control the weighting of the server assignments, so if too much of one class is getting cherry-picked, they can rebalance with weights. Sure, it makes Assignment Server logic much more complicated, but hopefully some recent or future upgrades would have better support for something like this. I think it would raise donor participation / retention.

I couldn't agree with you more with this idea 7im.

I've always thought, with the v7 client, that it would be nice for people to be able to click a button or add a flag for the type of research that they wanted their system to be doing. Allot, if not most, start to fold because they have a family member or a friend with a disease and they want to help out in those areas of research.

Your idea would allow these people to know that they are helping, yet give Stanford control of assignments.

Good idea. I think checkboxes are a pretty good option, rather than some flags. Adding flags to V7 isn't the most intuitive thing (pros and cons to that I guess), so I think checkboxes here are a pretty good idea since you can't screw up like you can with flags. I'm imagining something like this during the installation configuration, though of course you should be able to change it during runtime:

What would you like to work on? We will make reasonable efforts to give you the selected projects if we are actively pursuing such research.

• All of the below (recommended)
• Unsolved general problems in biochemistry
• Antibiotics (drug design)
• Alzheimer's Disease (AD)
• Cancer
• Chagas Disease
• Creutzfeldt–Jakob Disease
• Diabetes
• Huntington's Disease (HD)
• Malaria
• Osteogenesis Imperfecta (OI)
• Parkinson's Disease (PD)
• Viruses (HIV and influenza)