Display menu items any time, even when a process is running

Case number:845833-997959
Topic:Game: Display
Opened by:O Seki To
Opened on:Tuesday, June 10, 2014 - 23:12
Last modified:Thursday, June 12, 2014 - 23:56

I think I am not alone, who is struggling with this huge, resource hungry puzzles even I don't have the weakest computer - quadcore i7, 6GB RAM. Especially when 5-6 clients are active simultaneously. If I want to check what was done on a specific puzzle to figure out how I should continue and with what recipe, every time have to stop the running process or open a new client just for that. It is OK with small puzzles, but these big ones slowing down my computer close to freezing if I open an extra client.

Allowing the "puzzle menu" and the "open/share solutions" to display in a "read only mode" when a process is running/active in the client would eliminate this problem completely.
This way one can check the work history of the puzzles any time without interrupting the active process or opening a new client.
I think this modification is relatively easy on the programmers end, wouldn't affect the rest of the client and a lot of player with limited computing power would be happy with it.

(Tue, 06/10/2014 - 23:12  |  10 comments)

spmm's picture
User offline. Last seen 8 weeks 1 day ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

A lot of folders take detailed notes of what they have done to each puzzle and model, including what they have saved and the score and wiggle status of the save.

Joined: 04/19/2011

Spmm, you are absolutely right, that is a good way, if your are a professional foldit player or have a lot of time. Unfortunately I have other obligations, playing from different places/computers and tend to loose my notes... :) And yes, my team members and I always add detailed information to the saved/shared solutions.

Even so, taking separate notes about my and other group member's activities have to be repeated time to time to refresh it - interrupting the game or opening a new client just for this...

This is exactly what can be avoided by my suggestion. You can have up to date detailed information (that from the shared solutions) anywhere, anytime and do not have to carry your notes with you all the time.

I know this is not an essential issue, but we suppose to live in an advanced time, when a slight improvement in a computer program could save time and efforts and makes our life easier not more complicated.
Surely we were able to, and still can live happily without smartphones, computers, GPS devices, and so on by taking detailed notes, writing down calculations, have a phone book, or a map. I agree - I am old school.

Maybe I should have suggested this improvement as optional - just for lazy, forgetful folders like me... ;)

Thank you for your comment!

spmm's picture
User offline. Last seen 8 weeks 1 day ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

Oh I agree it would be great if we could monitor what had been 'done' to each model and track of each puzzle without us having to write it down.
Ideally this monitoring or recording would be:

- displayed 'outside' of the actual puzzle client(s) so as not to slow things down;
- would include script versions and settings if used, including iterations and scores at each change.
- one would be able to add brief text notes;
- and dare I suggest links to screen shots which may be of interest;
- also of course the whole thing would be shareable with a group or all folders.
- It would of course gracefully reset itself after any crashes.

There seem to be thousands of files in the foldit directory so a few more probably wouldn't hurt :D

In that way there would be no need to pause or stop a puzzle unless you wanted to. It seems that all of the info is actually being stored anyway, just not in a way which is accessible or useful to folders.

My suggestion was a bit tongue in cheek, as I often take woeful notes myself. :D

Joined: 09/24/2012
Groups: Go Science

Indeed, the "open/share solutions" is the main technical communication channel for current puzzle strategy, for myself and for my team (paper notes cannot be shared).

Cookbook - "pause" function
could be an second best alternative solution.

Other windows can be opened (not only in read mode) during recipes, like menu view, chat, take picture etc. so I suppose it is feasible.

A view on possible other improvements of the game display can be foreseen:
Toggle on/off graphic update of protein...,
Meta scripts etc

Joined: 04/19/2011

I have to say Bruno's idea about the "pause" function seems even better then my own. It would open a lot more possibilities aside that you can check the technical communications about shared puzzles.

I lately have the feeling that the foldit started to loose its original novelty/advantage. The ingenuity of using the intuitive capability of "amateur" human minds to figure out protein structures, rather than running a specialized scientific prediction programs.
It worked beautifully. Than we have more and more recipes making simpler to play, more and more and bigger proteins active at the same time.
To stay in the race most people (me too) just applying mechanically the proven recipes (max in different orders) and using their mind a little only in the designer puzzles.
The whole game became something that looks like running specialized prediction programs on a diversified supercomputer, consists of hundreds of individual computers. Seems kinda loosing the human factor here...

A "pause" function, which would allow to run an arbitrary "subroutine" (manual modifications, run another recipe fully or partially) during the "pause", than continue the original recipe with the newly generated data - would increase the possibilities of the dynamic adaptation exponentially and would bring back the "human factor" to the game. Probably sometimes it may lead to a dead end, but I think mostly to a faster improvement. Like an evolving, living thing applying the trial and error process.

Right now something barely similar can be done only, if you compile a new recipe from other subroutines, but that is still hopelessly static, cannot be changed while running.

I am not an active programmer, and don't know how big task is to incorporate a function like that into the game, but on the long run I am sure it would worth the efforts...

What the folders think about it? What is the developers opinion?

Joined: 04/19/2011

Seems to be my original suggestion inspired two, slightly different(and better) independent ideas, I try to summarize here:
1. (by Bruno and me) - "active pause" function that allows optional activity in the client during the pause.
2. (by spmm) - "out of client" puzzle log with detailed information about the actual puzzle (already collected, but not accesible to folders), shareable independently from the client avoiding interruption of the game.

I assume, since the detailed log is already embedded into the puzzle file and regularly updated as the game goes, it won't add any extra files to the already crowded foldit folder. Only need a "viewer" to extract from the puzzle and display that information...

Thanks for spmm and Bruno for the excellent ideas to make the game more efficient! I am sure there are tons of possible improvements out there...

Now we only have to wait for the developers' blessing... :)

Angus's picture
User offline. Last seen 19 hours 35 min ago. Offline
Joined: 06/04/2008
Groups: Beta Folders

There is a very old feedback of mine proposing the "pause" function, it's not a new idea.

AS for the OP's comment about struggling with bloated clients on his quadcore CPU, you've created your own issue here. With 5 - 6 clients on a quadcore, you're far exceeding the resources available.

I wouldn't run more than one client for each two cores if you're "watching the action", or maybe three on a quadcore if they are minimized and not rendering. It takes all of two cores is you are looking at the client run.

Joined: 04/19/2011

Thanks Angus to pointing out there is nothing new under the sun :)
Since the idea came duringt the recent discussion, I have not checked all old feedbacks.

If people keep discovering the "idea of the wheel" again and again independently, kinda suggests that would be useful to implement it...

About the quadcore overload - I am aware of it, and you are right, but you have to cook with what you have...

The evolver competition is another story, but if you want to stay in the solo race, you have to push your boundaries and run at least 4 clients simultaneously. (Exceptions the huge mutable proteins - I can run only one at a time with reasonable speed.)
I think my solo rank confirms that a little overloading can pay off well... :D

Brick's picture
User offline. Last seen 3 years 3 weeks ago. Offline
Joined: 07/15/2008
Groups: Beta Folders

Here is a link to the old feedback on the "Pause" suggestion


Joined: 04/19/2011

Thanks for the link Brick!


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