8 replies [Last post]
tzarl's picture
User offline. Last seen 4 hours 49 min ago. Offline
Joined: 10/16/2017
Groups: None

Hi, my name is Matthew Hantsbarger. I'm a programmer new to Foldit and I'm excited to start working on improving the game! I'm a recent biomedical engineering graduate with a particular passion for coding. I also love playing all sorts of games and I can't stop thinking about and discussing the social and collaborative aspects of those games.

I have to warn you, if you are ever interested in starting discussions about games design, including mechanics reflecting message, hidden role games, the role of interactivity in storytelling, modularity of card games, or social contracts in role playing games, you better be prepared for some impassioned discussion! I always enjoy listening to other's thoughts and feelings on those topics and sharing my own.

With regards to Foldit, I'd love to hear from the community what features and bug fixes the community are really looking for. There are quite a few fixes and features being worked on in the wings from our other game developers, but I'd love to know from our players if there are any frequent ideas (older or recent) that we haven't been able to address yet.

I can't wait to hear from all of you, and I hope I can be helpful!

jeff101's picture
User offline. Last seen 3 days 57 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
Welcome to Foldit!

If you watch Global & Veteran Chat now & then,
you'll get a taste for what players want.
If you haven't already looked, there are many
Feedbacks with ideas about how to improve
Foldit (http://fold.it/portal/feedback).
If you pick one at random & implement it,
you will make at least one player happy!
Another idea is to watch some Twitch streams
of Foldit (http://fold.it/portal/node/2003849)
and see how players actually play the game.

Below are more links to help you:
https://www.twitch.tv/s0ckrates/videos/all
https://www.twitch.tv/susume2/videos/all

Sockrates' YouTube Channel:
https://www.youtube.com/channel/UCPXuGlA4-kiGVxQOFh0B3Jw

Susume's YouTube channel:
https://www.youtube.com/channel/UC2XTlyviglWWiK_3bCcKn1w

https://www.youtube.com/user/UWfoldit
(it has some fairly recent videos showing how to play Foldit)

Joined: 06/24/2008
Groups: Void Crushers
Welcome

Welcome, I hope to you enjoy yourself.

Joined: 06/06/2013
Groups: Gargleblasters
Welcome!

Welcome Matthew
Just yesterday we discussing how using remix on a puzzle can lead to odd results -- a hydrophobe sticking out, ideal loops turning non-ideal. Tuck it away and play with it for a bit -- maybe even borrow a puzzle someone else can build for you and see what it does. I think it looks for strings, and ignores SS. I could be wrong, but so be it.
I'd recommend reading the excellent forum input from sOcrates about Tetris vs. Foldit. I do think we need a progression path and better rating of puzzles to convert more tryers to regular folders. Sock took a good deal of time thinking about what would help
Mostly, I recommend you open an IRC chat account that you can leave logged on. That way you can save forums and read back input from each day even if you are not around at the time.
I'm really happy you are here, and hope you try a puzzle or two along the way!
Skippy (female Skippy from Philadelphia)

Joined: 05/19/2017
Groups: None
Welcome welcome!

Hi Matthew! Sock here, been playing the game since March of last year. I run a weekly livestream of my folds over on Twitch (https://www.twitch.tv/s0ckrates), and I'm among the newer experienced folders here, and one of a considerably younger demographic! You can rely on me for a perspective of someone who's a video game enthusiast and scientifically somewhat familiar with the subject matter Foldit works with as a Pharmaceutical Chemistry major at UC Davis.

I could talk at length about this, even longer on a livestream of mine with discussion in chat, but currently my gripes with the game revolve around UI intuitiveness, the core gameplay loop, and player progression. As it stands, the UI is a decent scientific array of tools, but one that has a super low quality of life rating for newer players. The core gameplay loop can get stale if you don't know what you're doing or can't appreciate what's being done on screen, and recipes add a sort of "set it and forget it" idle metagame after the hand-folding early and mid-games are done on a puzzle. Lastly, player progression has a huge difficulty spike; the tutorial is present but a little nebulous, the beginner puzzles are not completely representative of the current ranked rotation, and it's easy to get discouraged from competing if it feels like the effort is for little achievement. A single player campaign in the vein of the tutorial but with real life example folds is a possible solution I've thought about that could prepare players to feel competent enough to hop on the weekly grind and know what they're doing.

I've got a laundry list and I've said my piece on other things elsewhere on this forum and on my stream archives so if you get the spare time feel free to peruse those and get in contact with me for further comment here, on Twitch, or on Twitter (@MrSockrates). Lookin' forward to see what you bring to the table!

tzarl's picture
User offline. Last seen 4 hours 49 min ago. Offline
Joined: 10/16/2017
Groups: None
Thanks everyone for the warm

Thanks everyone for the warm welcome! I'm still getting familiar with the job but I'll be sure to keep an eye on the feedback section and see what I can do about some of the topics in there. I'll also make sure to check out a livestream or two when I can!

LociOiling's picture
User offline. Last seen 59 min 51 sec ago. Offline
Joined: 12/27/2012
Groups: Beta Folders
Welcome aboard!

There are certainly plenty of challenges available, but a couple came to mind immediately as starting points.

One issue is the Rama Map. The Rama Map is available on most puzzles, and normally, it looks fine, with clearly defined regions. On design puzzles, the nice crisp regions become huge colored blotches with undulating edges. See the feedback Rama map not showing accurate color areas since recent update. It seems likely that the conditions on most design puzzles are involved.

(The current condition-free design puzzle 1473: Primitive 80 Residue Monomer Design doesn't offer the Rama Map, so a missed testing opportunity there.)

Another issue also affects design puzzles. There's a "recentbest" pose that's supposed to get updated automatically as the protein improves. The function recentbest.Restore can be used to load the recent best pose. There's also recentbest.GetEnergyScore to get the recent best score without loading the pose.

Most recent design puzzles allow players to add segments. A condition applies a small penalty for each added segment. The problem is that the score for the recentbest pose doesn't include the penalty. This makes using recentbest confusing and error-prone. See the feedback recentbest ignores residue count penalty. The same issue seems to affect the absolutebest pose and possibly creditbest as well, but these auto-saved poses aren't often used in recipes.

Having a nice crisp Rama Map and a reliable recentbest would be big plus. It seems like these issues would be relatively easy to track down by comparing a design puzzle to a non-design puzzle.

Joined: 05/19/2009
Groups: Contenders
Welcome and good luck

Hi Matthew, welcome aboard and best of luck with the assignment.
A small initial modification, but a big plus for the processing would be the expansion of the number of quicksaves. I would love to see 999 quicksave slots, possibly numbered in this format: 001 .. 999.

Regards,
BP.

Joined: 09/24/2012
Groups: Go Science
Welcome

It sounds like some fundamental "memory" or CPU usage strategy gets its limit now (causing a variety of crashes).

See abstract and discussion here:

http://fold.it/portal/node/990207

=============

Suggestions for Lua functions are here:
http://fold.it/portal/node/2004288

Sitemap

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, Microsoft, Adobe, RosettaCommons