Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
Warp wrote:
micro500 wrote:
Nintendo related stuff
So it's completely ok to show their games, but not their artwork (even though the games themselves show a thousand times more of the game artwork than some still images that are shown for a couple of seconds)?
The rationale behind this baffles the mind. I don't even.
I totally agree, it is silly. But even if it really is fair use Nintendo could still make a copyright claim against GDQ and threaten a lawsuit, which would likely end up being a giant headache even if we are in the right. While this does seem like fair use, I'd much rather avoid the whole issue and make a copyright safe run.
In other news, I've picked an RNG seed I plan to use for the first 20 questions. Here is the WIP which starts with that seed. The WIP should sync in DeSMuME 0.9.9. I've put the question set into the script so we can start doing the art for the questions that we needed to know the answers to (1-6).
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
After a discussion with Raelcun, I've decided to take a much more conservative approach to this project and do our best to avoid any possible copyright issues. I want to do my best to prevent GDQ from getting into any legal trouble, so here is my plan going forward:
We need to avoid any copyrighted material for the art we show on stream at AGDQ*. This includes the Futurama images, Nintendo related stuff, and anything of the sort. References to indie games should be ok however (Minecraft, Super Meat Boy, Shovel Knight, etc.).
We should put together a list of over 200 possible ideas that we think are legally safe for a GDQ. We should include specifics about what we plan to draw (or the image if it is finished already) and what the reference is for someone who doesn't understand it. We will then have the GDQ PR team look over the list and pick out the ones they think are the safest, and they can give us approval for those. From there we can get to work making the art for those ideas.
When we finish the art I will put together a finalized video of what we plan to show on stream and get that to the PR guys for another approval.
So what I need from you guys is a lot of image ideas. Just start throwing out ideas, and remember that we need 200+. I will assemble a list and we will go from there.
*My goal was to make this TAS in such a way that with some small modifications it could also be published here. The modifications would include things like taking out anything GDQ related, and just generally making the run targeted at a wider audience. I really love all the images that have been made so far and I would hate to see them not be used, but with the copyright concerns we won't be showing them on stream. I'd really like to see a run with these images, and we might still be able to make a publishable run that includes them. Whether or not copyrighted material is safe for a published run is off topic for this thread, so I'll start a discussion about that when the time comes. For now I think it would be best to only post GDQ safe images in this thread. Any other images you make can be posted in the Brain Age thread where we can discuss making a TAS with them, and I'll post WIPs as I work on them.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
samurai goroh wrote:
I made one drawing of "Shut up and take my money". Maybe it can be added after the donation reader as the 20th one?
I love this! I'm actually probably going to move the donation reader one out of the first 20 since we probably won't know who will be hosting us for a while and I'd like to get those finished soon. But this one is definitely going in the queue.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
Thanks to Masterjun I now understand how the game draws arbitrary lines. I added that functionality to my path drawing tool along with a few new hotkeys, and it now looks very usable!
Here's the first example with the upgraded tool:
Link to video
That drawing now takes ~5 seconds less compared to the old method, and I think it looks smoother.
To move this project forward I am soon going to finalize the first 20 questions of the script and start working on the TAS for those. If you have any comments/suggestions for the first 20 questions let me know soon. And if you are looking for images to work on, pick from the first 20 so we can get them out of the way.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
Raelcun wrote:
Warp wrote:
Question: How does Brain Age interpret those images? In other words, what numbers do they correspond to? Can this be chosen somehow (other than, you know, using that misleading trick of drawing off-screen)?
On the first page xy2_ linked to a large series of posts going into that here
That page describes how we think the game's OCR works, but that is entirely based on observations.
After the discussion about what we should allow for this TAS, the consensus seemed to be that as along as the images we draw are actually being scanned by the OCR, we can add extra input to trick the game. The published TAS does this by adding enough extra taps to the image (hidden in things he has already drawn or off screen) to trick the game into accepting the answer. After we finish getting the images into brain age we will add some extra input to the end of each, possibly with a bot. If you are wondering, the example videos I have been posting do not have this extra input, so the game doesn't accept the answer yet.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
Raelcun wrote:
Does this translate to just making the lines a bit thicker than the minimum of 4 pixels when doing slanted/curved lines? Because that's not a hard fix if so.
Yes, the end result is that lines become slightly thicker, but the more important part is the points where the pen needs to be touched.
Here's a zoomed in example from Zoidberg's neck:
The blue pixels are where you need to touch the pen. You can see that they are separated, meaning that you would need to lift the pen to go to the next point.
I added some extra points to fix the problem:
This results in a slightly thicker line (expanded to the left), but doesn't require the pen to be lifted numerous times. Another way to fix the problem would be like this:
This time the new points are on the other side, making the line slightly thicker to the right. Picking which way to add the points depends on the picture, and you may need to mix.
I hope that explains it a little better!
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
I am loving the images you guys are coming up with! Nice work everyone!
One thing that will help me out a lot is if you try to reduce the number of times lifting the pen is required. Take a look at these pictures:
The left is the unedited picture, and on the right is where you would touch the pen to draw the image. If you zoom in you can see there are many pixels that are not connected to any others, meaning that lifting the pen is required to start each segment. Lifting the pen adds 1-2 frames per extra segment required, so reducing the number of segments is a good idea.
You can see that a lot of those pixels form a line in the resulting image. A lot of times you can add a few extra pen taps and connect the points that were previously separated without altering the resulting image too much.
I did a quick edit to the picture above as an example:
I left some places alone where small important details would be lost and the result isn't dramatically different from the starting image, but will be much quicker to draw.
If you guys could try to reduce/eliminate those small segments before submitting the pictures, that would be really handy. I can/will do it if necessary later on, but I prefer to let you guys touch up your images to look best. I've updated the tools to allow you to generate the blue images above. The tools are now hosted on github: http://pjgat09.github.io/BrainAgeTools/. I will also update the link in the previous post.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
I've been working on a tool to convert these images to paths/coordinates for brain age. Here's the first example:
Link to videoHere is what the lower half of that looks like in my tool
Green and red points are the beginning and end of each path I drew, and yellow are the intermediate control points. The game will automatically add the blue points when the touch input jumps between the given coordinates.
You'll see a lot of control points on slanted lines. Right now my tool only works with vertical, horizontal, or 45 degree diagonal lines. I need to add support for other line angles. I will also likely need to tweak some drawings to fit the game better, which would make drawing them quicker.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
I did more work with the hardware today. I spent the majority of the time trying to debug why my formula to hit precise pixels wasn't working. I finally figured that out, recalibrated the screen, and I now have pixel perfect input that matches the emulator!
I also tested trying to get a static RNG seed on the hardware. I changed the DS date/clock to 1/1/2000 00:00, started a stop watch, rebooted into Brain Age, and started the game 1 minute after I had set the clock. I repeated the process and got the same question set!
I then tried the same steps in the emulator. I started a movie at 1/1/2000 00:00 and entered the game 3600 frames later, and I got the same question set as on the hardware!
This news means that we can pick an RNG seed and know what the answers will be. We can then cater our image set to work for those answers. We don't need to make each image work for all answers! This will save a large amount of work, and we can now focus on the art.
I love this idea too! *added to the script*
Spikestuff wrote:
Created 2 both have 20 bad (useless) pixels.
I'm ignoring the rule about PMing this.
Edit:
The fossils look great, and the bad pixels look pretty easy to clean up. If someone wants to fix them I'll add them to the script. The flareon looks great too, but I agree that it might look better smoothed out. What do other people think?
AdituV wrote:
Where do people stand on references to past events? I recreated one of the most tense moments (imo) of this year's SGDQ tetris block for Brain Age, but wasn't sure whether that sort of thing would be suitable, preferring last-minute-rush references to the coming AGDQ.
Picture in question: http://i.imgur.com/OchvvAP.png (semi-WIP)
That is another great picture. I think we're going to try to stick to references from the AGDQ we're at so that they are fresh in people's heads, but if nothing else we can use this drawing as just a tetris picture.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
feos wrote:
I still ask to add this to the script (maybe by the cost of one correct answer):
Draw the wrong answer, scratch it out, draw another wrong answer, scratch it out, draw "Pffffff whatever" and have that accepted.
Added to the list of ideas. I'm going to change around the script in the near future, so I'll slot that in when I do that.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
We're now ready for people to start working on the artwork! Here are the guidelines that every picture must conform to:
180x196
Black on white (no grayscale)
Drawable with a 4x4 pen
You can use any drawing program you want (Photoshop, GIMP, MS Paint, etc), but make sure your resulting image is in a lossless format like PNG.
I've made a utility to check if an image can actually be drawn in Brain Age
http://pjgat09.github.io/BrainAgeTools/
When you finish an image in your image editing software check it with this tool. This tool will crop your image to 180x196 and will remove any pixels that are not black. It will then check that the image can be drawn with a 4x4 pen. Any pixels that can not be recreated with a 4x4 pen will be highlighted red. You will need to fix those sections.
Pick an image from the script that hasn't been claimed/finished yet, and let me know you're working on it via PM. I don't have a fancy submission method ready yet, so when you've finished an image PM me with a link to it on imgur or similar. Feel free to PM me with any questions.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
dwangoAC wrote:
I guess where I differ is I could care less how fooling the OCR takes place as long as the result fools the OCR into thinking the answer was correct but others here care a lot how it takes place.
The problem with using the new glitch is that the OCR is nothing being fooled at all! The OCR isn't even being run on what we draw. For all intents and purposes, the drawing doesn't exist to the OCR. To the viewer it looks like the drawing is fooling to the OCR, but in reality that is not happening, and this is what I have a problem with.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
dwangoAC wrote:
Samsara wrote:
To put my point more eloquently: It's not about the art, it's about the canvas.
I think this is more or less what I've been trying to say as well - the only thing that I've been going around and around about is that this particular combination of glitches should not be the focus of the run. The fact that we're using something that should only recognize numbers to draw full-on pictures is really cool, it's a canvas where one wouldn't expect to find one, and there is definitely value in that.
This is exactly the point of using this game. We are drawing pictures in a mathematics which was designed to only accept numbers. In addition to that, the game is also being tricked into accepting those pictures as the answer! Part of the challenge of this canvas is that the game is watching everything we do, drawings and all. If we use the glitch to have the game ignore our drawing it's as if we are avoiding the canvas that we purposely picked. The audience may have no idea, but this feels wrong to me, and I think it would feel wrong to the audience too if they knew what was really going on.
I believe this glitch should be avoided to the best of our ability*. I want at least a majority of the answers shown at AGDQ to be done without the glitch, otherwise I would not feel comfortable showing the run. If we end up with a majority of the drawings not using the glitch we will explain to the audience how difficult it is to trick the OCR and that we did our absolute best to trick it in the time we had, but in the efforts of time we used the glitch to get the run ready for AGDQ. But this explanation doesn't work if most of the pictures use the glitch. At that point the show ends up looking more like "we kind of tried, but then we cheated for the rest of them", and we end up looking bad. If time is such a problem maybe we should save this run for the next year when we've worked out a way to not use the glitch.
* We are still definitely going to use it to make a blank answer, because a blank answer is hilarious!
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
dwangoAC wrote:
the focus should be on what we're able to do with the game
Brain Age gives us a fixed pen size (4x4), and our only color choice is black. We aren't able to do much with the game. Why are we using Brain Age instead of Mario Paint, Color a Dinosaur, or Flipnote Studio?
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
dwangoAC wrote:
Having said that, I have no intention of using the Brain Age TAS to impress people with our fancy OCR, I plan to use the Brain Age TAS to tell a story as described in the script micro500 wrote as linked from the first post (spoilers). The really cool technical elements will be weaved in but are not the focus of this particular portion of the TAS block.
Invariel asked the question in IRC "What is the point of demonstrating a TAS block at *GDQ?", and the answers seemed to agree on "validation". People want to see that the cool glitches we use in the emulator actually work on the console if you were a super-human. I think second to that we should entertain the crowd, if possible. We shouldn't pick uninteresting movies, or movies that only a very small portion of the crown may have heard of. TAS Block started with DarkKobold taking a NESBot to a GDQ and playing back a run, and the crowd was entertained to see that everything worked just as it did in the emulator.
This continued in 2014 with an example of an ACE glitch. We validated the ACE glitch on hardware, and second to that we entertained the crowd with a new game of snake. 2015 again continued this trend with another example of a ACE glitch, again validating that it does work on the real hardware. While the crowd is also looking to be entertained, I think they are also looking to see the real hardware being used (and abused), and I think this is why some of the responses to this year's SGDQ were rather negative. Some people were entertained to just see the TASes being played back as videos, but it seems that a majority wanted to see runs on the real hardware.
If the point of the 2016 AGDQ TAS block will be to tell a story, then why are we using Brain Age to do it? Surely we could find another game that is more suitable for drawing pictures. Maybe a game that lets us use colors, or vary the width of our pen. If this is just about the art, I see no reason to use Brain Age when there are other superior options available that would give us more freedom.
I disagree with your idea of focusing on telling a story, and instead think we should talk about our hardware implementation and let the story tell itself.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
dwangoAC wrote:
To me, I'd error on the side of better art - after all, we've literally reduced this game to the point that it's a series of drawing pad sheets and we're effectively exhibiting art that serves a purpose in and of itself (albeit using a quite absurd method). The art itself *IS* the point because at that point it's the artwork that gets the attention, not the game!
I disagree. While the art is an important part of this, making a robot to draw a predefined image is not very impressive. Making a robot to draw the right answers in a clear font would not be that impressive either. But making a robot that draws a funny picture that is also processed and accepted by the game as a real number is impressive. It is the fact that the silly drawing was instead seen as a number is the cool part about this, not just silly drawing alone. If the silly drawing has no effect on the answer at all, why bother drawing anything at all? If this is only about the art, why don't we just show a Mario Paint run instead?
This glitch is rather game breaking and also repetitive. It ends up feeling like the Final Fantasy stairs glitch. The idea is cool, but you get the idea after seeing the same thing several times in a row. Our glitch definitely warrants a quick demonstration, but it gets old really fast if you repeat it for every answer. This is why on this site we have two separate categories when a game-breaking glitch is found. An example is Pokemon Gold/Silver. We have a run using memory corruption and one without. Some people want to see more of the game played out, so the author artificially limits him/herself to not using a simple awesome glitch in favor of showing off how far they can push the game without it. This usually ends up showing more of the game, as well as more of the author's skill in the game. Sure, the author could do a simple glitch and save themselves a lot of time, but they wanted more of a challenge, and (some of) the audience enjoys that.
I feel like it is the same case with Brain Age. Making a run that shows off this glitch for every answer is interesting in it's own respect (and might be worthy of it's own category, but that is a different debate), but showing off a run that tricks the OCR for every question is even more interesting. The challenge there is to make our drawings fool the OCR. If we are planning to wow the crowd with a run that is both entertaining and technically impressive, why wouldn't we take the harder route?
However, I propose a compromise. We know how much work it is getting ready for a *GDQ, and I know how much work it will (might?) be to manipulate the images, so in the interests of time I think we should convert as many drawings as we can to not use the glitch, but if it ends up being that we can't convert them all so be it. We will need to make it clear to the audience what we did, but we are also showing off the broken OCR
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
I believe the basis for the published run is the idea that the OCR in the game is so loose/broken that you can draw a picture and the game will accept it as the correct answer. You would probably think somehow the game is finding the answer in what you drew. Watching the run with that expectation leaves me entertained, because I enjoy seeing the game being broken in that way. For some of the answers that is exactly what happens, but for others I see that he added extra taps (both in and out of the drawing area), and also traced over some of his old work to trick the OCR.
This is where the line gets fuzzy for me since my expectations were a little off. I assumed he was only drawing the picture, but there is more going on. If I had known that up front I would have still been impressed, but a little less so. I still find it neat that the game somehow finds the answer in everything you drew, even when you factor in the extra taps.
The out of bounds drawing glitch allows us to draw anything we want, and the game does not use any of that input for the OCR. Your dots and lines do show up on screen, but they are completely ignored by the OCR. No answer is detected by the game, not even "?". This lets us draw anything we want without having to do the technical feat of tweaking our drawing to trick the buggy OCR. We then could draw right on top of our image with the answer and the game would accept it. With the invisible models glitch we can draw the answer with nothing being shown on screen, but in this case it is fed to the OCR. Combining these two glitches means we can draw anything we want, lift the pen, and then just draw the right answer on top (invisibly).
If that combination of glitches had been used by the published run it definitely would have crossed a line for me. The OCR is not being abused in any way. You are drawing the real answer, but you just happen to be drawing a picture before it. As others have said, this turns into a "Color a Dinosaur" type run where the drawings serve no purpose. Removing them would not prevent you from achieving your goal.
When you reach this point, "your goal" becomes the important thing to distinguish. Is there a goal to Color a Dinosaur? There is no end game as far as I know in that game. Is there a goal to Brain Age? You can make up a goal like trying to finish the Calculations x100 as fast as possible, or trying to reduce your "Brain Age" to be as low as possible, etc. The current run aimed for a playaround completion of a round of Calculations x100. This kind of goal (playaround completion of a game/minigame) shows off some glitches or fun parts in a game and makes a sacrifice of speed. It can be done in a number of other games, so the appeal of this run is that while wasting time drawing pictures he is entertaining the audience, but the drawings are also being used to complete his goal. When watching the run you can assume the pictures serve some purpose, even if they aren't done optimally like they would be in a run meant for speed. If he had used the above glitches to make the drawing completely ignored, the drawings are truly just a waste of time and serve no purpose. It would be just like showing a few frames in a slideshow in the middle of a sort of speed-oriented TAS.
I'm seeing this set of glitches as a game-breaking glitch like other games have. In my opinion this set of glitches should not be used if someone wants to obsolete the current playaround run. However, I can see them being used as a separate category, distinct from the one I just mentioned, that specifically says it uses this set of glitches. This would clear up any confusion or bad expectations when people watch it as they would know what they are getting into when they hit play. I personally would enjoy seeing another playaround movie in this new proposed category, in the style of my blank input TAS posted in Weatherton's first post. It would be quick, and I personally would find it entertaining, but to me it is clearly in it's own category separate from Ryuto's published playaround. A speed-based movie may also be entertaining. This would involve manipulating the RNG to give you answers that are quick/easy to draw, and also drawing the right thing so the game accepts it as quickly as possible. Both of these ideas of course would not appeal to everyone, but I think some would find them entertaining.
As for the AGDQ aspect of this, I believe we should not abuse the OOB+invisible models glitches. I believe most people watching will just assume the OCR is broken, and that is how we are getting away with the game accepting the answers. If we use the proposed glitches, the drawings are just a misdirect to hide what is actually happening. We could not tell the crowd how we did it, but surely people will ask, and some will find these forum posts. We could lie, but that is a very bad idea. I personally would rather be able to tell the crowd that the game is indeed looking at everything we drew and finding the answers. I find that to be a much more entertaining thing to show, both visually and technically. If we did decide to use the glitches, I would feel obligated to tell the audience what is going on, but I know they would quickly feel less impressed. Many people would see it for what it is: a robot that can draw pretty (but useless) pictures, and then later answer a math question. A bot that can do either one of those, or both, is not impressive to me. The technical feat of hooking it up to a real DS is cool, and so is streaming the picture in from a computer, but I think tricking the OCR is the real technical crux of this, especially when you combine it with all the other technical feats. However, I do find the ability to draw nothing (visually) on screen and still answer the question entertaining, so I am on board with answering one question like that. It would show off the neat glitch we found without deceiving the audience too much. We can tell the audience how that worked without ruining their expectations for the rest of the run. This does mean we need to find a way to make our drawings work for all answers, but that is doable, and it would be an even more impressive thing to tell anyone who asks about the technical details.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
feos wrote:
It does need some irrelevant crap like an hieroglyph or similarly absurd drawing(s?) between weird but still correct answers and calculus.
Added
As I posted in the Brain Age thread, xy2_ and I have found it is possible to have the game accept completely blank answers! I've added that to the script now too.
Emulator Coder, Experienced Forum User, Published Author, Player
(68)
Joined: 10/4/2005
Posts: 197
xy2_ discovered that any point in the drawing area with an X coordinate of 246 does not draw a dot on screen, but is used for the number recognition. This can be combined with edge breaking to create a completely invisible model that is accepted by the game.
Example:
Link to video
xy2_ also discovered that the game will not recon anything drawn if you start on the border of the drawing area. We can combine this with edge breaking to draw a complex picture (including multiple lines) without lifting the pen, and the game will completely ignore it. We can then combine it with the invisible models strategy above to make it appear as though the game accepted the drawing as the correct answer.
The blue pixels are where you need to touch the pen. You can see that they are separated, meaning that you would need to lift the pen to go to the next point.
I added some extra points to fix the problem:
This results in a slightly thicker line (expanded to the left), but doesn't require the pen to be lifted numerous times. Another way to fix the problem would be like this:
This time the new points are on the other side, making the line slightly thicker to the right. Picking which way to add the points depends on the picture, and you may need to mix.
I hope that explains it a little better!
The left is the unedited picture, and on the right is where you would touch the pen to draw the image. If you zoom in you can see there are many pixels that are not connected to any others, meaning that lifting the pen is required to start each segment. Lifting the pen adds 1-2 frames per extra segment required, so reducing the number of segments is a good idea.
You can see that a lot of those pixels form a line in the resulting image. A lot of times you can add a few extra pen taps and connect the points that were previously separated without altering the resulting image too much.
I did a quick edit to the picture above as an example:
I left some places alone where small important details would be lost and the result isn't dramatically different from the starting image, but will be much quicker to draw.
If you guys could try to reduce/eliminate those small segments before submitting the pictures, that would be really handy. I can/will do it if necessary later on, but I prefer to let you guys touch up your images to look best. I've updated the tools to allow you to generate the blue images above. The tools are now hosted on github: http://pjgat09.github.io/BrainAgeTools/. I will also update the link in the previous post.
I'm ignoring the rule about PMing this. Edit:
