One player's perspective

Case number:699969-996889
Topic:General
Opened by:auntdeen
Status:Open
Type:Question
Opened on:Tuesday, February 11, 2014 - 00:55
Last modified:Thursday, February 13, 2014 - 16:37

I thank beta helix for his apologies in the feedback thread about rolling back - vmulligan for his excellent front page post - Dr. Baker for his video, and all the other devs who have posted thoughtful and informative comments.

The problem, IMHO, is that anything fixed will be an improvement of some sort - because almost everything is broken to some extent right now.

Some improvements will not change the fact that this has turned into a completely different, and highly buggy game, within the shell of what we know as Foldit.

And that is the real heart of the problem here. This is a completely different game. Right now there is no way to distinguish a feature from a bug, we simply do not have the tools to be able to tell the difference.

From my perspective:

Game engine issues:

-- Rebuild is more or less broken both for hand folding and scripts (this despite the recent devprev upgrade - more on that below), having been more or less replaced by cuts & idealize

-- Most scripts are not functional - even the ones that our wonderful scriptwriters have adapted for the new client - banders, rebuilders - all types of scripts

-- Wiggle appears to lock us into "refinement only" after a certain point in folding

-- Many are experiencing constant hangs and crashes

-- In the past month, just about every major & minor bug that has ever plagued Foldit has reared its ugly head at one point or another (almost 5 pages worth of feedbacks)

Community issues:

-- There is very little, if any, enjoyment in folding for most players right now - and a lot of aggravation

-- The result of that is that many players are rage quitting - some are drifting away or just checking in once a day, hoping to see if anything has changed - many are playing with many less clients running

-- Protein improvements that once took 1-3 steps now take 12-15 steps… very tedious and boring, making the experience feel more like work than play

-- Much of our wiki is now obsolete

My own experience has been very frustrating, and I see no end in sight if this continues. My usual method for denovos is to bring the protein in the first puzzle as far as I can in low CI, save that, then step it up to 1.0, do a little more work (I don't have the patience to be my own DRW), then walk away. For the second puzzle, I go back to the early save, run a couple scripts on that save, then proceed to step it up to 1.0 CI with scripts. The result is usually that I rank higher on the second puzzle.

I tried that in the updated devprev, assuming that rebuild was working better. On the plus side, a rebuild script did find points on that save. On the minus side, that one cemented a very few points above my high (loaded into the second puzzle and some work done, the high was already cemented).

I just don't have the personal resources to do the tedious work that I did the first week of rollout, where I did everything I could think of, spending many many hours trying to find techniques that work with this (and placing in top 20 for the ones that I did really work on).

Yes, I can do it, but very honestly, there is no enjoyment left in it. I still wake up with that "run to the computer and see what happened overnight"… then I remember that I only have one client open, and little hope of much improvement at all.

So I am now running just one client instead of 3 or 4, and only poking around with it. I really miss my game, but have finally run through the emotions of denial, anger, and I guess, acceptance. I've started playing Eyewire again - they are having celebrations and events in honor of their first year.

I am not trying to "make a statement" by saying what I have above - I am simply stating what my personal experience is now. I haven't stopped playing, but will likely just do one client / one puzzle at a time until something changes radically.

There is obviously an underlying issue to all these problems, and I suspect that the game engine simply will not handle the wiggle change, and / or score function. Since that is the basis of the client change, it will probably take a lot more than a few days to fix that.

The early cementing of proteins is reminiscent of the SA / WA slider major bug. To my recollection, after the client was rolled back for debugging, there was never a fix or anything else said about them. That makes me surmise that there was a deep incompatibility with the game engine. Sadly, with the massive change in the current client, we have no tools for comparison or investigation, so if that is the underlying cause of many of our issues, as players, we can't help determine that.

Perhaps you can give us a time frame for getting this back to a reasonably enjoyable game again. Maybe a week? And then roll it back if the issues are still huge (in the players opinions)? Whether you run a concurrent game - give extra points to those who choose to muddle through NC - whatever you do, if this situation doesn't improve, in a major way and rapidly, the game is dying for lack of enjoyment.

It's an obvious thing, but please remember that players are here because they want to be - we aren't paid, our careers are not on the line. We all want the game to be as productive as possible - we all have as motivation the desire to contribute to science - but in the end, if we can't derive some pleasure from it, if it isn't a game with enjoyable elements, then we cease to be "players" in one respect or the other.

Please remember, that from the player's perspective, this is a game with a scientific component (a wonderful, enjoyable way of wasting time and computer resources with a potential benefit) - not science with a little game thrown in. The dividing line is thin, but it is a major difference between the devs perspective and the player's perspective.

One last cautionary note… many players, and I think all the devs, are too young to have lived through the debacle of New Coke. If unfamiliar with the story, check it out on wikipedia. The takeaway quote: "New Coke had the spotlight for only three months but casts a long shadow, in both the business world and popular culture, that can be seen today. It is most frequently mentioned as a cautionary tale among businesses against tampering too extensively with a well-established and successful brand."

(Tue, 02/11/2014 - 00:55  |  21 comments)


alcor29's picture
User offline. Last seen 2 days 6 hours ago. Offline
Joined: 11/16/2012

I agree that all the explanations made by the developers and Dr. Baker make good sense, nevertheless, auntdeen's comments above once again focus light on the inescapable. The application as it is constituted today just seems to require a different kind of participant. Computations are too intense for the resources some of us have. The time required to work a problem now takes too long for some of us, the constant overheating is too threatening to the health of our weaker processors, the constant crashing gets really tiresome. As auntdeen states the inability to run more than one client decreases the pleasure of the game. I'm not much of a folder, but running only one client forces me to commit early to a particular track, and more often that not it is the wrong one so I get frustrated and just quit on the puzzle.

Just to throw out a couple of suggestions:

1) Since you admit that we are in a difficult period similar to what the scientists went through in adjusting to the Rosetta changes, perhaps there is a way you can lend foldit much more support than is being given currently to clear up some of the mysteries which some of the most experienced folders and scripters seem to be addressing in feedback and in chat. Clearing up how some of these processes are now operating will help them devise new strategies and scripts to do the tedious work.

2) This is perhaps a bit far out and possibly incompatible with the competitive nature of the game, you may want to deal with the limitations some of us are dealing with in trying to play, by changing the criteria for considering a puzzle to be 'finished' or submittable. You may consider finding a way to lower the requirement for refining our folds, for example, thereby, sparing us the endless hours of cranking scripts for the purpose of getting a few more energy points; this may also help save our hardware, and possibly even help us run more clients simultaneously.

Joined: 04/15/2012
Groups: Beta Folders

Also, many people don't have the time to do the heavy manual labor needed. I know school has already greatly reduced my play time, and this makes it very hard for me to be able to really do much. I'm sure it's frustrating for new players as well.

Angus's picture
User offline. Last seen 3 days 4 hours ago. Offline
Joined: 06/04/2008
Groups: Beta Folders

I applaud auntdeen for taking the time and effort to so eloquently explain the situation many of us find ourselves in.

The recent client brings us back to the days when all we did was handfold, laboriously working every segment, hoping rebuild would do something, and eventually spending days manually walking the backbone to get those final points.

Then GUI scripting came along, and we could wait longer before turning it loose to get those last few points, giving more time to work on the other parts. Lua scripting really freed us up - we could spend quality time getting the starting pose to look pretty or getting all our "oranges in", then turn the scripts loose to do all the amazing stuff to really improve our protein. A by-product of that is that we didn't need to spend 8 hours a night for days on end, we could spend one hour, and let the scripts do the rest, and still get better results than all that hand folding time could produce.

I spent days on one of the first NC puzzles, and produced a result that was embarrassingly bad, and I know the old client would have done way better.

So - why should I continue with this?

Joined: 09/24/2012
Groups: Go Science

It's strange. I don't share this bad feeling. On the countrary, I feel excited by:
-the fact that we'll get more scientific valid results
-the "new rules" that challenge us to find new solutions instead of relying on our experience

It all remembers me my first moments with foldit. I came in Foldit end 2012 when there were already powerfull Lua scripts. I just had to test a lot of scripts in order to find the good ones. Then I went to a team and we shared the best practices and scripts. Because of this, I did not develop high hand folding capacity. Now, I'm obliged to hand fold a little bit more, that's great ! I get some more satisfaction when I found something myself (hand fold, or using one of my scripts) or when I apply what a Human (of my group, or from the chat or wiki) told me. Now we have a lot of possibilities for satisfaction: we have to reinvent a lot of thinks, re-write scripts, sharing with team members etc.

There are a lot of "research" for us to solve in the game: try to understand how Wiggle and Wiggle power work with different recipes and tools. Just try this discovery: the Idealize 2.0 [NC] from spvincent with wiggle power "High" gets points, not at all with Low or Medium. When you find a solution, you are happy, you share it with the others, and you want to play again ...

Concerning the clients, I noticed 2 opposite effects (or is it me?) and I learned something. Before, I used to limit to 3 clients on a quite powerful computer because I had more crashes when I tried with 4 clients. Now with big puzzle 841, I had to limit to 2. I learned to carefully check the tracks in order to avoid crashes, and to look at the Task Manager to look at the CPU usage %. And by the way, I noticed that with small puzzles 842 & 843 .. I can go till 5 clients !

So yes, after one year this all renew my interest in the game. There is a new continent for discovery ...

Joined: 04/11/2013

I want to qualify bumping Bruno's post up.
Firstly I think its a bad idea to take the attitude, "If they don't like it, they can leave, no one is forcing them to play."
This kind of attitude leaves an undesirable track record.
There will be a select few, like Bruno, who take up the challenge with the attitude, "Bring it on.This is good for the science and to develop my skills."
Sadly, the rest of us don't have the resources, including time, to make meaningful contributions any more.

My recommendations:

1) Give the Bruno's their challenges, but please give the rest of us our toys back. We can still do something good with them.

2) I don't mind running one or two NC's along with a few of the old style.

3) "Perhaps the evo'ers can do the NC, as an optional on all the solutions.

4) Perhaps one or two server models from our toys can be used by the NC players.

We will find our fun somewhere, but we prefer rather to stay than to leave with a bad taste in our mouths.

Please daddy Dave, give us our toys back. Please! Pleeeeeease!
We love you daddy, Please!

frood66's picture
User offline. Last seen 8 hours 56 min ago. Offline
Joined: 09/20/2011
Groups: Marvin's bunch

Many thanks to Auntdeen for putting the time and effort into such an excellent and accurate critique. I can safely say that there is not one point or personal observation that I can take issue with. Here is my personal experience.

I’ve been playing foldit for about 2 ½ years now, every day bar a handful. during that time I’ve moved from script heavy play to handfold based play (which took me rather a long time since I’m not that good at this). It suits both me (not having a huge machine) and my group – I can try to find a basic solution – those in the group with less time can go to work on it – and they do…way better than I can solo it usually.

So the vehicle that is foldit has been modified, added to and tinkered with. The community scripters have worked long and hard to make it look sleek, get the best out of the engine and balance the handling. Now and then a wheel has fallen off (these things happen) though changing the front steering was a disaster (wiggle slider). But, with work from the designers and the mechanics, the wheels have been bolted back on, the old steering rack reutilized - and tweaking continues. Not a bad vehicle. So I’ve been looking forward to the next race – CASP. ☺

‘We really need an upgraded car if we’re to kick butt’ say the designers ’So we’ve been working on something’

“Fantastic!” say the mechanics, “We had no idea…when can we test drive?”

*TaDaaa*

“Right now! Here it is!” say the designers.

“ Ahhh…Hmmmm” ponder the mechanics ”It doesn’t look very nice”

“ Well it needs some work on the aerodynamic, a paint job and polish up, yes. But u guys are good at that”

“ And twin steering axles?”

“ Looks good huh?” say the designers

So the mechanics push the car out of the garage.

“ Do u think I could have the steering wheel clipped in before I go any further?” asks the driver.

“Ahh” say the designers “small snag with that just now – the clip to the column appears to be incompatible. But don’t worry, we’ll have that fixed in the next little while. Here, this spanner sort of fits – just don’t steer like you are used to”

So the test drive takes place and the mechanics start to get somewhat worried. ‘Should have been back by now” and “do u think he’s alright?” they mutter.

Eventually the car limps back into the pits.

“How did it go?” beam the designers. “Any feedback?”

“ Not terribly well” says the driver “ can I ask a few questions?”

“Sure” the designers reply in unison.

“ The engine appears to be running on only half the cylinders it’s supplied with “ the driver intones “ and the new gear shift gets stuck in high power”

“ Ahh” respond the designers “ we believe those are both problems you mechanics will be able to help us with – there is a chance they are interface glitches of course. We’ll fix those if that’s the case – won’t take long”

“ And the front axles turn in opposite directions” the driver continues

“ We weren’t sure if it would work better going forwards or backwards” the designers smile “ We know u guys can fix that in a trice. Look, we know there are few issues we need to work on – but the faster u guys get to it, the quicker we’ll be race ready ☺”

“Can we have a copy of the drawings?” asks a head mechanic.

“ Ahh – the drawings can’t leave the design room – sorry about that – company policy” they reply – looking at the floor.

“ Can we come in and just sneak a look? Get just a few questions answered?”

“Sorry – no mechanics in the design room – company policy”

A pregnant silence enveloped the garage – eventually one mechanic could stand it no longer and piped up “And the questions we will have?”

“ Of course, of course – just submit them and we’ll take a look”

The weeks passed. The mechanics worked furiously at first – keen to make progress. Then an upgrade was announced from the design room – how excited the garage crew were! But it was deferred and found short of the mark. Despite the disappointment the mechanics continued to try different solutions. But progress was slow and no further upgrades were forthcoming.

The crew had asked if they couldn’t run the existing car and work on the new one separately. This idea was brushed aside and the serviceable old car locked away out of sight. Enthusiasm began to wane. By the time 4 weeks were up some of the crew were taking sickies whilst others took holiday (some notified, some not)

One day the driver walks into the pits and, with a grim look on his face, calls the designers (he’s awoken that morning to remember that it’s unlikely any progress has been made overnight)

“Okay – work in progress” says driver “ Tell me. It now takes 3hrs to complete a circuit that normally took 90 seconds in our old car. Can anyone tell me when we might expect to get this baby race ready?”

Silence was the loud reply.

Finally a designer reiterated the need for car improvements – something the crew were well behind anyway. He repeated the mantra..” I know this is causing some frustration right now. But the benefits will be outstanding”

Many of the crew looked at the heap of car in the middle of the garage – then wistfully glance across the pits to the locked garage containing the last 5 years work. The driver decides that the term ‘some frustration’ can only be a euphemism – similar to that one so employed by doctors – “u may feel some discomfort”. A crew in the corner muttered “ad astra per alas porci”

And so it was that an all-pervading despondency settled across the crew.

But the driver is a hoopy frood who always knows where his towel is. He also has a plentiful supply of new ones. So, despite having grown very attached to this particular one, he throws it on the floor of the garage, collects his electronic thumb, can of beer and packet of peanuts, walks out of his beloved garage - then proceeds across the circuit to the motorbike pits and offers his free services there.

Joined: 09/24/2012
Groups: Go Science

Ha ha ha !! Excellent ! Thanks for this post !

v_mulligan's picture
User offline. Last seen 1 year 6 weeks ago. Offline
Joined: 03/04/2009
Groups: None

I just wanted to let you know that the devs and scientists have read this, and we are working to address what we can (hangs and crashes, problems with rebuild, and increased tediousness of hand-folding). I think that some of these might have a common underlying cause that has now been identified, and there's a possible solution that we'll be talking about at this week's developer's meeting. Sorry that our replies are a bit sporadic. Until we get a new community liaison, the scientists and developers are having to take on community relations in addition to the actual work of making the game playable. We are listening, though -- fear not!

In the mean time, the best thing that the community can do is to try to play the game, and to continue to give feedback about the specific ways in which the game is harder to play -- particularly gameplay issues that might be difficult for someone who isn't an expert Foldit player to see. (Remember that we can usually recognize a bug, but since we're not experts in Foldit strategy, it's harder to anticipate how a change will affect ease of use or playability, so we rely on you for that.)

v_mulligan's picture
User offline. Last seen 1 year 6 weeks ago. Offline
Joined: 03/04/2009
Groups: None

One little note: wiggle is supposed to only subtly refine a structure after a certain point. Mathematically, wiggle is performing downhill gradient marching to find the nearest local minimum in the energy landscape. It isn't intended to hop up and out of local minima in order to search for the global minimum -- that's what rebuild and hand moves are for. This means that as a structure gets closer and closer to its locally optimal conformation, wiggle can only make smaller and smaller improvements. This is particularly true if you've been using high-powered wiggle. This isn't a bug -- it's just a part of gameplay strategy that comes out of the nature of the different tools and how they interact with the scoring function.

Joined: 10/30/2012
Groups: Beta Folders

I pretty much agree with Bruno, it may take a bit of time and effort to adjust our play and scripts etc. but if at the end of the day we do more accurate folds. Then it will have been worth it.

Having said that, I spent the first year on fold-it hand folding, I did not have a powerful enough computer to use scripts effectively. Now I do, and size matters.

So perhaps it has not hit me as hard.

So as long as they don't hit us with and 'electron density' puzzle before this is mostly straighten out I am content.

beta_helix's picture
User offline. Last seen 6 hours 12 min ago. Offline
Joined: 05/09/2008
Groups: None

It is for this exact reason that we haven't posted a new Electron Density puzzle!

Thanks to all of you for posting your opinions and thoughts in this thread, and thank you again for your patience and understanding.

Joined: 06/24/2008
Groups: Void Crushers

Sorry, I play Foldit for the science, the science is not something on the side. The only reason I joined years ago was to help do science. If I wanted to play a free game I would play Mafia Wars or Farmville, both of which I have played. I think people forget that the primary reason for the game is science; not your ego.

As for players quitting, I have been heard this forever. It comes up whenever there is any change. Yet I have to see this major exit. I heard that many many players left because of scripting and hand puzzles would bring them back; I am still waiting for their return. What was forgottenwas that the players asked for scripting; we were bored doing the same procedure over and over and over. SCRIPTING WAS NOT IMPOSED ON US; WE ASKED FOR IT. I accepting grumbling about the game; I actually think it is part gestalt of the game.

I know there are problems with NC; there were many bugs, many freezes, many crashes when the game started. They were cleared up. The problems will be cleared up in NC. As has been pointed out, the bugs will be cleared up faster if we, the players, help by playing the game and pointing out the bugs; and the difficulties we are finding.

Personally I would be dismayed for Foldit to turn back to the old game and not keep pushing the science forward.

Joined: 04/11/2013

Understood.
They are probably under fire to get results or be closed down.

Angus's picture
User offline. Last seen 3 days 4 hours ago. Offline
Joined: 06/04/2008
Groups: Beta Folders

I think you have an issue with the concept of a "game". A game is a competitive event, played for pleasure, following a set of rules to determine how the game is played, and having a way to determine the winners and losers.

In this case, science is a by-product of the game. For this to be primarily about the science and not be a "game", then all of the competitive aspects need to be removed - no points, no standings or rankings, no teams. You would work your puzzles, submit the results, and the scientists will analyze them and hopefully provide some feedback if something good comes of it. Your energy score would be for your own use to determine whether you are going in the right or wrong direction with your creation. I'll issue my standard challenge going back to the early days of distributed computing - if the points and standings went away, would you still continue with this?

Now, the problem with taking the "game" out of Foldit, you will lose most of the participation. The competition is what keeps most people here. In this case, going back to the definition of a game, the "rules" have changed, so it's a different game. Going back to frood66's excellent automotive analogy, the game designers have now decreed that the race cars must act as the disastrous one the designers came up with. So it is with Foldit - the rules of the game have changed so much as to make it unacceptable.

Joined: 09/21/2011
Groups: Void Crushers

Totally agree with Mike on this. I am a script writer with 2 goals, one to be high in the game and second to help the devs making beter algorithms.

Angus's picture
User offline. Last seen 3 days 4 hours ago. Offline
Joined: 06/04/2008
Groups: Beta Folders

You can't agree with Mike, and still be in it to compete for ranking. He's in it only for the science.

Joined: 09/21/2011
Groups: Void Crushers

Sorry, as a game player I love to compete for ranking. And I wouldnt play this game (or write recipes) if it wasnt for a higher goal. I am in it for both.
And on another comment of you: science is the main goal of this game and people tend to forget that.

bertro's picture
User offline. Last seen 1 day 2 hours ago. Offline
Joined: 05/02/2011
Groups: Beta Folders

For me, playing Foldit is like doing a really tough jigsaw puzzle, the tougher the better. It is not about the points (even if points are good), it is about the satisfaction of knowing that a top solution could/would contribute to better understanding and maybe something that will help others. So I applaud the dev for having the guts to switch to NC. And if it means to change my ways then it is ok.

If our solutions, as solo or evo, will now be more accurate at the price of having to relearn how to play Foldit I am all for it.

I can wait for bugs/crashes to be corrected as long as the dev are committed to improving the client and I can contribute like this feedback forum.

NC is Foldit re-born.

It is a new challenge and I welcome it.

Joined: 06/24/2008
Groups: Void Crushers

Actually Angus I never said the competitive scoring was bad; what I said is "I think people forget that the *primary* reason for the game is science; not your ego."

Joined: 04/19/2009

A small interjection here, before this devolves any further...

This is not purely a game, and it is not purely science. It is a relatively new hybrid - the gamification of science. See this link for a brief explanation, and take the course if interested. https://www.coursera.org/course/gamification

I have taken the course, and it helped define that clearly for me.

We all have different motivations for folding, and some lean more to one side or the other. It's all valid.

Joined: 09/24/2012
Groups: Go Science

I must admit that the distance between work (science) and game is sometimes small. In business (and science), we have competition, rewards of the employees (or the researchers), medals etc. Games are said to be useful for children, animals and even adults (it develops or maintain brain etc). Like dreams. Games are good to develop or maintain creativity, a useful ingredient for work (and research of course).

For my work, I used to script in order to save time (with a real satisfaction). It began when I once wrote a small script on my computer, then I said to my boss "My computer is working for me, I could do something else". I showed him the screen: a sentence was writing automatically a very long letter saying "All work and no play makes Bruno a dull boy. All work ...". Now I use to say that my computer is playing for me ...

I could make it writing "All play and no work makes Bruno a dull boy", but it's not necessary, Foldit does the same on a more sophisticated way.

Is this a work? is this a game? Is my work a game or a work? Do I loose my time on work or on game? Who knows? But both are pleasant times.

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, Amazon, Microsoft, Adobe, RosettaCommons