26 replies [Last post]
Joined: 04/24/2014
Groups: None

This thread idea started here: http://fold.it/portal/node/2000856#comment-30609 (with some of http://fold.it/portal/node/2000842#comment-30611 thrown in.)


With the onset of summer (at least for some of us in the world), it's a good time to talk about puzzles and balance*. As always, your input is important to us, and we try to maintain a strong balance between "hard core science goals" from the lab, and your ideas of "what makes Foldit fun". We heard the recent feedback from 1102/1103, too.

I am happy to report that we've been talking more about a variety of puzzles as a result of these recent comments, and the community should look forward to seeing some progress in that direction going forward. :)

We can't promise that you'll enjoy every single puzzle (and we would be silly to think that's the case) - that's totally fine. Nor can we guarantee that the most enjoyable puzzles will have significant science merit or incredible publishable results (because science). And of course you all know, it takes time to construct our puzzles in a way that will provide results we can use. But what we can find out in the meantime is what type of puzzles are enjoyable (or not), what you enjoy about them, and work at steering more of those type of things into the mix as much as possible**. I thought about doing a poll, but that really doesn't get into the type of particulars we're looking for.

This thread is meant to be constructive (this is not a bug report thread, we have the specific puzzle threads and feedback for that), so non-constructive feedback is likely to be culled out (that's your warning). Specifics always are great and appreciated.

Help me help the rest of the team, and by default all of you, and chime in! Thanks, everyone!

* it's always a good time for this conversation and it really should be ongoing, but there's never a better time to actually start doing it than now :)
** you might find yourself in the minority opinion on a particular puzzle type, but you'll never know until you post, just saying.

Joined: 04/11/2013
Contact Puzzles

Thanks for the opportunity to give an opinion.
These seem to have faded out. They seem to be enjoyable.

jeff101's picture
User offline. Last seen 8 hours 28 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
We'll need more practice on them before the next CASP

There were a number of Contact Puzzles during 2014's CASP competition.
We will need more practice on such puzzles before 2016's CASP competition.

Joined: 04/24/2014
Groups: None
More good news!

Some contact puzzles should be on the way in the near future. We won't always have a steady stream of those, but if we keep mixing them in with everything else, it's good to know they're enjoyed.

jeff101's picture
User offline. Last seen 8 hours 28 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
More feedback about our results:

I like the Science and Developer Chats and the Blog Posts that give us feedback about our results (see posts http://fold.it/portal/node/997313 and https://fold.it/portal/node/2000356 and https://fold.it/portal/node/2000276 and http://fold.it/portal/node/1998900). I was hoping the next Chat would be sooner.

Have any of the H-Bond Network designs held together well in the lab? Also, when we do revisit puzzles, can you tell if we are doing better now than in the past (see posts http://fold.it/portal/node/1998460 and https://fold.it/portal/node/988112#comment-28798)? Also, in the E Bola puzzle 1000 (see https://fold.it/portal/node/1998572 and http://fold.it/portal/node/2000209), what has become of our designs? Are they helping?

Enzyme's picture
User offline. Last seen 51 years 40 weeks ago. Offline
Joined: 07/10/2008
Groups: None
feedback results

More information on our scientific progress would be appreciated. How much progress actually occurs. How often does progress occur(I fear it's far less than most players believe)? Is there something we can do as players which can help generate faster results? We get 3 types of puzzles: baseline puzzles to refine rules for future puzzles, scientific puzzles which might save a life someday, and busywork puzzles only given to us because there is nothing better available. Knowing which puzzles help you the most can help players allocate their folding time more appropriately.

We all joined for the science but stayed for the game. While I enjoy a good rank up as much as the next person, there are plenty of games with the same level of instant gratification, the long term gratification is why we continue to chose Foldit over other entertainment.

Enzyme's picture
User offline. Last seen 51 years 40 weeks ago. Offline
Joined: 07/10/2008
Groups: None

As a side note. Have you considered recruiting some of the long term players to assist in preparing puzzles or generating results. I'm sure there are some menial tasks which can be accomplished by volunteers, allowing your paid employees to handle more difficult/restricted tasks. I would say anyone who has folded continuously for 4 years or more most likely has your best interest at heart and could be trusted to complete repetitive tasks on non-confidential items.

jeff101's picture
User offline. Last seen 8 hours 28 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
More Quick Frozen Cutpoint Puzzles:

I enjoyed puzzle 630. It helped me see the importance of sheets.
http://fold.it/portal/node/993498 630: Quick Frozen Cutpoint Puzzle

You could do bigger puzzles like this since it has less
degrees of freedom per segment than a normal puzzle.
Should go faster than a normal puzzle of the same size.

jeff101's picture
User offline. Last seen 8 hours 28 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
More Puzzles with Outside Collaborators:

I thought the catalyst puzzles posted by DrLemming were interesting
(see http://fold.it/portal/node/993778 and http://fold.it/portal/node/994493).
Did any of our designs prove useful? Can you post more puzzles like these?

egran48's picture
User offline. Last seen 4 years 51 weeks ago. Offline
Joined: 03/31/2014
Groups: Go Science
Thanks to the Foldit team.

I don't mind what the puzzles are, yes some are large and slow and we seem to be getting more of these lately.
As long as they have scientific value I enjoy doing them.
I started folding in 2014, personally I don't mind a run of similar puzzles as it gives time to get my head around them and to try different things.

Some are complaining about computer processing power, that is a personal problem, others of us from other countries living in rural area have to contend with slow internet speed, that is life just deal with it.

I agree fully with Jeff re the feedback from puzzles. Are the old puzzles improved by the new methods? Did the Ebola puzzles make any headway, even by a small amount? etc. These don't have to be in depth reports, just a quick outline would be great. We are adults so even negative reports are OK, that is what science is about, give it a try if it doesn't work, try something else. :)
I also agree with Jeff re any type of CASP puzzle, we need the practice.

All in all, I love Foldit. I enjoy doing all of the "PUZZLES". Puzzles are meant to challenge your brain.

Joined: 06/06/2013
Groups: Gargleblasters
thanks for mix / contact map please

Thanks for the comment about mix. My team appreciates having at least one small puzzle so out newer folders can work on them with some help. We also have some members with limited machine capacity or time off work, so they like to have some things they can do relatively well and see some progress
I would like to see contact map puzzles too. I was just starting to figure them out when they went away...
I can't do the H-bond ones. Can't see well enough to band them into place. I can happily skip those while others enjoy them :D

brow42's picture
User offline. Last seen 1 year 20 weeks ago. Offline
Joined: 09/19/2011
Groups: None

I've been wondering where the CASP ROLL puzzles went, they seemed to dry up after CASP X ended.

beta_helix's picture
User offline. Last seen 10 hours 9 min ago. Offline
Joined: 05/09/2008
Groups: None

Unfortunately there haven't been any CASP ROLL targets since 2014 :-(

Joined: 06/06/2013
Groups: Gargleblasters
Need a "Rubber Duck"

One final thought.... we need something fun for the players every so often. Either a science issue that is current, or something just goofy. We had a Tall Ship parade into Philly yesterday,with the last "ship" being a giant rubber ducky. The tall ship people were really bummed that "everyone" came to see the rubber ducky (bigger than a galleon) until they realized it increased the crowd by a multiple. How do we hook new players and get them to stay? What will keep someone from being discouraged when they do not progress as fast as they would like?

Joined: 04/24/2014
Groups: None

This is a highly clever analogy that is quite useful (and put a smile on my face). Thanks, Skip! :)

spmm's picture
User offline. Last seen 1 week 6 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers
'4 days of a disappointing duck and other issues'


Don't think we need this really, sounds like those nasty times when we spend all our time in global telling people why it doesn't work properly and then having to decamp to mibbit for hours.

No problems getting people to Sydney Harbour for the free: iconic Tug and Yacht Ballet, the Tall Ships Race, the Harbour Parade, and the Great Sydney Swim. on Australia Day :P

Joined: 04/24/2014
Groups: None

Our newest Marburg puzzle is live.

If this counts in your view as a duck, then we're making progress!

Joined: 06/26/2015
Groups: None
facebook game version of FoldIt

Hi, i had posted this in the recommendation area, but i thought i would also post it here too. i would love to see this game be brought to the Facebook game platform. it would get a lot of game play from people who want puzzle games to play while on Facebook. i think that would be awesome.

Enzyme's picture
User offline. Last seen 51 years 40 weeks ago. Offline
Joined: 07/10/2008
Groups: None

a facebook version is a great idea.
Some recommendations:
1) you don't have to mantain a rotating puzzle schedule the way the Foldit client does. Puzzles can run long term similar to our current "Contests" or simply never end. An absolute best would be almost assured and would be a great source for your database lookups
2) generally use only smaller puzzles or "beginner" puzzles at least in the beginning to keep new player anxiety low.
3) You don't need all the tools. A small subset of our current tools would be fine. Keep it simple. Even simpler, less exact calculations could be used.
4) You don't need such a robust tutorial or even any tutorial. Trial by fire on small puzzles with few tools would not be overwhelming

The goal would be to create a simple to play, simple to maintain game who's only real purpose is to draw players to the foldit client to get the real "meat" of protein folding

Joined: 06/26/2015
Groups: None
Neanderthal Genome Puzzles

hey, what about puzzles involving the use of the Neanderthal genome project? puzzles which show the way their proteins fold compared to Modern Humans?

LociOiling's picture
User offline. Last seen 4 hours 51 min ago. Offline
Joined: 12/27/2012
Groups: Beta Folders
spacing puzzles more evenly, what's next

Currently we have a small "revisiting" puzzle (1104) ending one day, followed by a large hand-folding puzzle (1105) ending 36 hours later. Then a medium-sized design puzzle (1106) ends the next day.

I'd like the puzzles to be spaced more evenly, ending 36 to 48 hours apart. Also it would be nice if the easier "revisiting" puzzles would end between the larger and more difficult puzzles. This might meaning delaying the start of the next revisiting puzzle a couple of days, but I could live with that.

Some type of calendar or even a general statement about upcoming puzzles would also be helpful. For example, we've had a flurry of hydrogen bond network puzzles recently, which required new techniques. Will there be more of these HBond puzzles in the near future, or will they go the way of contact map puzzles? A simple statement like "expect some new ED puzzles in July" would go a long way.

Joined: 03/05/2015
Groups: Gargleblasters
I agree, spacing of the

I agree, spacing of the puzzles is sometimes ... well ... challenging. Particularly for those of us who are 40 hours at work and do this evenings and weekends. Good point about fine-tuning the releases to have small/large spaced evenly as well.

Oh! Just thought of one more possible really cool thing: How about a specific page of "Best Recipes" - recipes that are most useful for... (categorized by puzzle type, function, etc.). This would be a great way of introducing new players to the tools that can help them the most, quickly. This I think would be an improvement over the search function for scripts that currently exists.

jeff101's picture
User offline. Last seen 8 hours 28 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
Favorite Recipes:

The Foldit wiki has a number of pages where
people list their favorite recipes, for example:

Favorite recipes

To prevent the loss some of the greatest recipes,
we'd like you to record some of your favorites here!,
along with links and how you personally…

Recipes - table
A table of the recipes with more votes or
average vote > 4 sort by type of recipes.

I think I've also seen a list of recipes used
most during one of the CASP competitions.
If anyone finds such a list, please post its
link here.

jeff101's picture
User offline. Last seen 8 hours 28 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
Most-Used Recipes:

I think the all.macro file that stores each player's
Cookbook keeps track of which recipes we use the
most. I think text in all.macro like below:

\"uses\" : \"13\"\n

near the end of an entry for a recipe means the recipe
has been used 13 times. Perhaps data like this could
be sent automatically to the Foldit site now and then.
Foldit could list the most popular public recipes used
for each puzzle. Foldit could also accumulate recipe
stats for different puzzle types (perhaps over the
past few months for each puzzle type). Foldit could
list the recipe names, recipe numbers, and how often
each recipe was used for each puzzle and puzzle type.

Perhaps on each recipe's page, there could be a listing
of total uses so far. Foldit could do this for public, group,
and even personal recipes. Foldit could also list the number
of different players who have downloaded each recipe.

Enzyme's picture
User offline. Last seen 51 years 40 weeks ago. Offline
Joined: 07/10/2008
Groups: None

Number of times used was listed on the recipe page years ago and has since been removed. I too would like to see the return of this feature.

Joined: 04/24/2014
Groups: None
Steer back!

Valid suggestions but we're veering a little off here from puzzle balance to recipes. :) Let's get these perfectly great ideas into their own thread or a feedback suggestion, and not lost in the shuffle of the overall picture.


Joined: 04/24/2014
Groups: None
Good news!

We're going to endeavor to space these next several puzzle end times out a bit more. :) This may involve some odd "end times" for a few puzzles as we try to balance the spacing out, as well as have them not close on the weekend as a result of this "staggered approach" (no promises or guarantees on that, more of a 'good faith effort'). We appreciate everyone's patience as we work on this one.

I can't promise this will be a permanent schedule shift (historically, I think we have had some staggered end times but yes, recently they have been quite bunched together), but we definitely understand the need as we address the larger issues at hand.

Consider it a work in progress.


Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, Boehringer Ingelheim, RosettaCommons