The World's Largest (read, Only) Animated GIF Distribution Network...Hi b3tans, old user here (misteralfie way back when)but posting for High Art and B3tans coming together as Bus-Tops. Some of you may have seen the link Eight posted here: www.b3ta.com/links/741152
Basically we've made a network of big red LED screens that are scattered around London (20 boroughs) and live on the roofs of bus shelters. Tomorrow is the first day that work created by the public is sent to them (they have run through January with new work from Mark Titchner).
The screens themselves are super good at displaying GIFs though their fundamental constraints are:
* Not too much variable brightness per LED (so don't represent lots of tonal ranges brilliantly)
* 256x80 aspect ratio (the aspect ratio of a bus shelter!)
* Animations longer than 10 seconds not great due to passenger time at shelter
* Best works are those of high contrast and good separation between objects - think Frank Miller's work for best results.
Another constraint is that cocks (glass, crude or other) would need an incredibly convincing pretext to make it from the web to a screen :)
So if you're interested by all means have a look at the site and please, not too many cocks? bus-tops.com
ta!
(,
Tue 31 Jan 2012, 18:02,
archived)
Basically we've made a network of big red LED screens that are scattered around London (20 boroughs) and live on the roofs of bus shelters. Tomorrow is the first day that work created by the public is sent to them (they have run through January with new work from Mark Titchner).
The screens themselves are super good at displaying GIFs though their fundamental constraints are:
* Not too much variable brightness per LED (so don't represent lots of tonal ranges brilliantly)
* 256x80 aspect ratio (the aspect ratio of a bus shelter!)
* Animations longer than 10 seconds not great due to passenger time at shelter
* Best works are those of high contrast and good separation between objects - think Frank Miller's work for best results.
Another constraint is that cocks (glass, crude or other) would need an incredibly convincing pretext to make it from the web to a screen :)
So if you're interested by all means have a look at the site and please, not too many cocks? bus-tops.com
ta!
Seems to be a very interesting idea
(
maiden is filmed before a live studio audience,
Tue 31 Jan 2012, 18:07,
archived)
Not see him/her, not for a long time.You could try /Talk
( maiden is filmed before a live studio audience,
Tue 31 Jan 2012, 18:30,
archived)
Wow, you're so old.Is 256x80 the pixel resolution of the signs? If an animation loops, how many loops get played? How high can the frame rate go?
(,
Tue 31 Jan 2012, 18:22,
archived)
no reminders please256x80 is the absolute resolution, and anything uploaded via the tool at /create/ will be resized to this. Any multiple is of course fine. All animations are looped infinitely, but the number of times any file is played is determined by playlist programming on the server so doesn't really factor in.Frame rate is maxed out at 12.5 as we've multiplexed so many LED panels together but we suggest 10FPS to be safe.
(,
Tue 31 Jan 2012, 18:36,
archived)
However nice this is, your 260 frame 2,826.34 kB animation is perhaps a little large?
^thisand slow, I got bored after 1,2,3 what came next?
(
hekim 66 ÉŹĘš ŹÉnā² uooW,
Tue 31 Jan 2012, 18:34,
archived)
I liked it, it's just big giffs start to freeze and fuck my PC, and they're getting posted more frequently..Someone posted a 3.5Mb one yesterday, that requiered task manager to escape and restart to get things going again...
(,
Tue 31 Jan 2012, 18:43,
archived)
Looks perfectly ok and I agree about the huge gifs being posted lately Frame rate really could be faster though as potential punter isn't going to stand there waiting for each line
( hekim 66 ÉŹĘš ŹÉnā² uooW,
Tue 31 Jan 2012, 18:53,
archived)
indeedthough it's literally a function of the number of screens we have multiplexed to create the size of screen we have. With a smarter controller we could get a higher frame rate but that's a) what version 2's are for and b) a creative constraint!
(,
Tue 31 Jan 2012, 19:04,
archived)
damn those creative restraintsgo down the pub drink 10 pints go back and insist they do it your way (used to work for me) :D
( hekim 66 ÉŹĘš ŹÉnā² uooW,
Tue 31 Jan 2012, 19:22,
archived)
Frame rate is always a sod to get right with people using so many different browsers...What flys in one browser, lags in another one....
(,
Tue 31 Jan 2012, 19:07,
archived)
0.02s per frame is fine for eye melting gifs in firefox.
(
Mu Dinofiddler,
Tue 31 Jan 2012, 19:12,
archived)
I made this when all I used was IE6.. it was timed perfectly as I thoughtNow I use Firefox, the first time I reposted it an saw it in Firwfox, JEEEEEZZZE!
I keep meaning to retime it so it's least fit inducing, don't click if you don't like flashy.....although if you use IE you should be fine..
www.b3tards.com/u/380e5b53a37701afdd20/zapparecurdouble2.gif
(,
Tue 31 Jan 2012, 19:20,
archived)
I keep meaning to retime it so it's least fit inducing, don't click if you don't like flashy.....although if you use IE you should be fine..
www.b3tards.com/u/380e5b53a37701afdd20/zapparecurdouble2.gif
I had same problem, can't remember the technical reason whybut I remember a nice person here said Firefox was fairly accurate for speed
( hekim 66 ÉŹĘš ŹÉnā² uooW,
Tue 31 Jan 2012, 19:36,
archived)
I know the the technical reasons why or understand it inside my head would be more accurate....But me being a thicko, don't know the right tecnical or none tecnical words to discribe it !! ;-)
(,
Tue 31 Jan 2012, 19:43,
archived)
true trueActually it's bit of a fail at the compression end of our drawing tool which we're working on - basically it should output at around the rate which photoshop's 'restrictive' setting does (and will, once we've iterated it).
(,
Tue 31 Jan 2012, 18:37,
archived)
It's just b3ta recomends about a 400Kb maximum for animated images.......Good luck with your project:-)
(,
Tue 31 Jan 2012, 18:46,
archived)
Well the least I can do for you then sir is give you this to post in its place....It's 425Kb so just about scrapes through, it's not as good quality as yours but gives people the idea, if they want to click you link for your full sized splendid version :-)
please feel free to use it if you want to:-)
www.b3tards.com/u/380e5b53a37701afdd20/web111.gif
(,
Tue 31 Jan 2012, 19:03,
archived)
please feel free to use it if you want to:-)
www.b3tards.com/u/380e5b53a37701afdd20/web111.gif
That sounds like a super cool ideaHere in my hometown of The Hague, there is a very large screen (about six floors high) on the tallest building, displaying all kinds of weird stuff. A lot of the stuff on there is also created by the public.
mediocre ha ha ha, you're reading this,
Tue 31 Jan 2012, 18:28,
archived)
huhI don't know that project, any chance you could pop me a link?
(,
Tue 31 Jan 2012, 18:41,
archived)
Here are some linkswww.socialbites.nl/2011/09/01/mega-led-scherm-den-haag_strijkijzer/ (in dutch, but with pictures)
en.wikipedia.org/wiki/Het_Strijkijzer
mediocre ha ha ha, you're reading this,
Tue 31 Jan 2012, 19:42,
archived)
quiteand we're not above being chuffed by the comparison, thanks.
(,
Tue 31 Jan 2012, 18:38,
archived)
She's been one of my favs since an ex pointed me to some work by her in 1991another great reason to date slightly older, scarily intelligent women
(,
Tue 31 Jan 2012, 18:47,
archived)
hahaAs I've gotten older I am still dating scarily intelligent women but they are getting younger; like I'm hung up on a certain one or something. I love Holzer's stuff and obviously she was an influence on this project.
(,
Tue 31 Jan 2012, 18:49,
archived)
Great project; I'm going to have to check these sites out. Unfortunately my lack of any artistic skillz precludes contribution, but nicely done anyway.
(
da5id <YOUR SIG HERE>,
Tue 31 Jan 2012, 18:34,
archived)
oh come now sirI'm sure that is merely modesty working against you, and thank you.
(,
Tue 31 Jan 2012, 18:39,
archived)
Nope. Not enough magenta in the panels for me ;)
(
da5id <YOUR SIG HERE>,
Tue 31 Jan 2012, 18:48,
archived)
Maybe they could switch to magenta after the 9pm watershedand then go CDC crazy?
(
Van Da Graph is sufferng wth an intermttent 'i' key,
Tue 31 Jan 2012, 19:21,
archived)
If you wanted to do a movie festival, foldsfive's work is as much art as anything ever commited to gifwww.b3ta.com/users/profile.php?id=27426
No idea how to contact the awesomeness though.
(No idea how to contact the awesomeness though.
gaijintendo Regular Member,
Tue 31 Jan 2012, 19:01,
archived)
Get posting!That would be so good. One thing - RED= LIGHTS ON; BLACK - LIGHTS OFF. Black as the background and separating objects in the frame with as high a contrast as possible whilst doing one's best to preserve detail is the name of the game.
Frankly, 8 bit animations work the best.
(,
Tue 31 Jan 2012, 19:12,
archived)
Frankly, 8 bit animations work the best.
Why can't you have red as the background and black as the foreground?What's special about a light being on that means it has to be foreground? RIS?
(,
Tue 31 Jan 2012, 19:18,
archived)
good questionIf you think about it in terms of contrast, what you have is a 3.2m x 1m screen, ENTIRELY LIT UP. With some black bits that aren't. In terms of how negative space is used effectively it is far better to have the foreground objects light and the background black. That said, as long as the separations between the objects in frame are good (these are all excellent examples: bus-tops.com/people/Mark_Titchner/ ) then all is swell.
(,
Tue 31 Jan 2012, 19:26,
archived)
Nope, still not getting it.If you had, say, a black line drawing on a red background displayed on a luxury TV screen, that might be an even bigger screen, ENTIRELY LIT UP, with some black bits that aren't. It would look fine.
(,
Tue 31 Jan 2012, 19:30,
archived)
surebut this is a very low fidelity Monochromatic LED screen with minimal variable brightness (tonal range) across the LED's at a 1cm pitch with a viewing distance of only 2.2 metres.
(,
Tue 31 Jan 2012, 19:33,
archived)
I feel sure this is something to do with the reason it can't adequately display black on a red background.I wonder what the connection is.
(,
Tue 31 Jan 2012, 19:38,
archived)
I get youbut it can display black on a red background totally fine, as long as there are few grades of colour used and there are very clear separations throughout (mark Titchner's example are good)
(,
Tue 31 Jan 2012, 19:43,
archived)
Wow dude, you have a HAL 9000 in your bedroom! hope your name's not Dave ;-)
Whahey!It looks like the problem is actually something to do with light leaking into nearby pixels. So if you have a lot of pixels lit up, the few remaining dark ones will be half-lit-up, whereas if you have a lot of pixels dark, the few that are lit up will not be darkened (since dark isn't a thing and doesn't leak). LEAKAGE.
(,
Tue 31 Jan 2012, 19:49,
archived)
sorta?I think it's to do with the controllers attempts to interpolate colour to a brightness variable and failing horribly.
(,
Tue 31 Jan 2012, 19:53,
archived)
OohWhy would it do a thing like that? Just add the three channels and divide by the maximum brightness, how hard can it be? I mean divide by three, or something. Whatever gets you the maximum brightness when they're all 255.
(,
Tue 31 Jan 2012, 19:58,
archived)
quitebut there we are. Do you think light leakage is like, an actual thing in this context?
(,
Tue 31 Jan 2012, 20:02,
archived)
No, I think I'll abandon that theory.Why does the controller get things wrong so badly?
(,
Tue 31 Jan 2012, 20:04,
archived)
hmmn`well, in it's defense I think it's largely down to the fact that you only have 256x80 and a 1cm pitch. it then tries to render finely detailed works where 25 pixels on the image is rendered by something like 3 on the screen.
(,
Tue 31 Jan 2012, 20:08,
archived)
That doesn't account for what it did to Bob's 256x80 image.That image was already 256x80 in the first place.
(,
Tue 31 Jan 2012, 20:13,
archived)
hmmyou may be right. I think it must come down to the previously noted inability of the controllers to accurately pin down variable brightnes between 0 and 255. Coupled with viewing angle and distance.
(,
Tue 31 Jan 2012, 20:23,
archived)
Does nyan cat work fine on the real screen, or do some of the values also turn out wrong?Can I see it?
Or how about this version of the monkey animation:
I removed everything that isn't in the red channel. So now it's 8-bit, right?
(,
Tue 31 Jan 2012, 20:38,
archived)
Or how about this version of the monkey animation:
I removed everything that isn't in the red channel. So now it's 8-bit, right?
huzz and aaah!I made some changes to the contollers settings and am getting much much better results, this is how the monkey gif now renders:
(,
Thu 2 Feb 2012, 11:34,
archived)
Not that simplewww.johndcook.com/blog/2009/08/24/algorithms-convert-color-grayscale/
I prefer the luminosity method meself.
(,
Tue 31 Jan 2012, 21:24,
archived)
I prefer the luminosity method meself.
Sure, but even the basic average method produces a tolerable image, not one apparently covered in mildew spotsSee this section of the image from the big screen above?
See the dark part in the middle, that looks vaguely like a peeled banana? In the original, that has red value 173 (and zeroes in the other channels). See the light pixels surrounding it? In the original, those have red value 172 (and zeroes in the other channels). The only difference in the input is the least significant bit, but the difference in the output is really big.
(,
Tue 31 Jan 2012, 21:40,
archived)
See the dark part in the middle, that looks vaguely like a peeled banana? In the original, that has red value 173 (and zeroes in the other channels). See the light pixels surrounding it? In the original, those have red value 172 (and zeroes in the other channels). The only difference in the input is the least significant bit, but the difference in the output is really big.
HmmMaybe wraparound? Depending on how the values are handled, a value of 258 can end up as 2 (258-255). Hard to say without getting into the guts of the code.
(,
Tue 31 Jan 2012, 23:07,
archived)
Yes, I was thinking that, integer overflowexcept there's no good reason for an overflow to occur at that boundary between 172 and 173, unless there's something scaling the 172 up to some 2^N-1, and though I fiddled around with the calculator for ages I couldn't find any reasonable way for that to happen. To get the 255/256 boundary you have to scale them both up by around 1.48; so that doesn't seem to mean anything. If you multiply by three, they just miss the 511/512 boundary: 172*3 is 516. Maybe *3 and then minus 5? Why, though?
(,
Tue 31 Jan 2012, 23:53,
archived)
This is getting far too geekyThe only thing I could think of was bad handling of signed/signed bytes. Then you get wraparound/overflow at 127. Java did it to me a lot when I ported a bunch of my old c/c++ code (the bastard, *shakes fist*).
[edit] Using the luminance algorithm, 173 * 0.72 is 124(ish), which is pretty close to the boundary.
[editedit] Except 0.72 is the factor for the green channel. Fucksocks!
(,
Wed 1 Feb 2012, 0:40,
archived)
[edit] Using the luminance algorithm, 173 * 0.72 is 124(ish), which is pretty close to the boundary.
[editedit] Except 0.72 is the factor for the green channel. Fucksocks!
