Glitch pictures anyone? back
(L) [2007/06/15] [toxie] [Glitch pictures anyone?] Wayback!I hereby declare that anyone should post their coolest glitch pictures in this thread.  :)
By glitch pictures i mean the -really- strange stuff (kD bugs that resulted in bizarre, but cool artifacts, QMC that resulted in horrible correlations, coding mistakes that produced "abstract art" ;) and the like), not just a badly chosen parameter or a noisy picture.
I start off with a mediocre picture (simply because it's the -only- one i still have on my HDD :( , thus this thread that hopefully should keep such art alive)
[IMG #1 Image]
[IMG #1]:
(L) [2007/06/15] [toxie] [Glitch pictures anyone?] Wayback!I hereby declare that anyone should post their coolest glitch pictures in this thread.  :)
By glitch pictures i mean the -really- strange stuff (kD bugs that resulted in bizarre, but cool artifacts, QMC that resulted in horrible correlations, coding mistakes that produced "abstract art" ;) and the like), not just a badly chosen parameter or a noisy picture.
I start off with a mediocre picture (simply because it's the -only- one i still have on my HDD :( , thus this thread that hopefully should keep such art alive)
[IMG #1 ]
_________________
The box. You opened it. We came.
[IMG #1]:
(L) [2007/06/15] [beason] [Glitch pictures anyone?] Wayback!Halton sequence correllation produces interesting pattern:
[IMG #1 Image]
Floating point error in isosurface intersection:
[IMG #2 Image]
[IMG #1]:
![[IMG:#0]](images/6ad76d6de784cb0383dc09c4d94a62e84cb509b32f4b42f6b43ec21647b2dcae.jpg)
[IMG #2]:
(L) [2007/06/15] [beason] [Glitch pictures anyone?] Wayback!Halton sequence correllation produces interesting pattern:
[IMG #1 ]
Floating point error in isosurface intersection:
[IMG #2 ]
[IMG #1]:
![[IMG:#0]](images/6ad76d6de784cb0383dc09c4d94a62e84cb509b32f4b42f6b43ec21647b2dcae.jpg)
[IMG #2]:
(L) [2007/06/15] [MooZ] [Glitch pictures anyone?] Wayback!The glitches were caused by a quite stupid error in bvh traversal code.
[IMG #1 Image]
[IMG #1]:
(L) [2007/06/15] [MooZ] [Glitch pictures anyone?] Wayback!The glitches were caused by a quite stupid error in bvh traversal code.
[IMG #1 ]
[IMG #1]:
(L) [2007/06/16] [toxie] [Glitch pictures anyone?] Wayback!great stuff, guys! keep it coming!!!
(L) [2007/06/16] [tbp] [Glitch pictures anyone?] Wayback!I'll spare you countless floating point bits as texture for that LSD coated view of interpenetrating geometry with bogus reflexion/refraction.
[IMG #1 Image]
{ thread already enlarged, show no mercy }
[IMG #1]:
(L) [2007/06/16] [tbp] [Glitch pictures anyone?] Wayback!I'll spare you countless floating point bits as texture for that LSD coated view of interpenetrating geometry with bogus reflexion/refraction.
[IMG #1 ]
{ thread already enlarged, show no mercy }
_________________
May you live in interesting times.
[LINK https://gna.org/projects/radius/ radius] | [LINK http://ompf.org/ ompf] | [LINK http://ompf.org/wiki/ WompfKi]
[IMG #1]:
(L) [2007/06/16] [Wussie] [Glitch pictures anyone?] Wayback!that looks real nice indeed =] priceless glitches [SMILEY :D]
(L) [2007/06/16] [Wussie] [Glitch pictures anyone?] Wayback!that looks real nice indeed =] priceless glitches [SMILEY Very Happy]
(L) [2007/06/16] [spalter] [Glitch pictures anyone?] Wayback!Hi there, my first post..and my glitch contribution: a bug in the intel compiler screwed up ray generation, resulting in small versions of the image being tiled across the screen [SMILEY ;)]
[IMG #1 Image]
[IMG #1]:
(L) [2007/06/16] [spalter] [Glitch pictures anyone?] Wayback!Hi there, my first post..and my glitch contribution: a bug in the intel compiler screwed up ray generation, resulting in small versions of the image being tiled across the screen [SMILEY Wink]
[IMG #1 ]
[IMG #1]:
(L) [2007/06/16] [lycium] [Glitch pictures anyone?] Wayback!recent mistake in generating uv for some cubes: [LINK http://www.fractographer.com/propaganda/vfx1.exe]
awesome thread! :D
(L) [2007/06/17] [Tecla] [Glitch pictures anyone?] Wayback!As talked about in a post of mine at: [LINK http://renderspud.blogspot.com/2007/03/bih-trees-much-progress.html]
I got this sandblasted looking Ajax due to not checking my rays' param values against a minimum properly.  Whoops.  [SMILEY :D]
[IMG #1 Image]
[IMG #1]:
(L) [2007/06/17] [Tecla] [Glitch pictures anyone?] Wayback!As talked about in a post of mine at: [LINK http://renderspud.blogspot.com/2007/03/bih-trees-much-progress.html]
I got this sandblasted looking Ajax due to not checking my rays' param values against a minimum properly.  Whoops.  [SMILEY Very Happy]
[IMG #1 ]
[IMG #1]:
(L) [2007/06/17] [Alan] [Glitch pictures anyone?] Wayback![img width=730 height=445]http://snowreverie.com/arclight/wp-content/uploads/2007/04/messup1.png[/img]
I can't remember what caused this one anymore, something to do with tracing shadow rays for area light sources. The little cross/star shapes on the ground are fun. I like the strange look of the shadows from the lower horizontal struts on the torii behind them. Obviously the image is not very realistic, but there's something surviving the artifacts which the mind seems to pick up on in an almost hyper realistic way. I get the same thing looking at Toxie's first picture, not to claim that my pic is in the same league though [SMILEY ;)]
[img width=825 height=512]http://snowreverie.com/arclight/wp-content/uploads/2007/04/verystrange.png[/img]
This is a "direct visualisation" of a photon map for a simple 144 polygon model I used for debugging. The spiral pattern you see goes a full 360 revolution "around the surface of a sphere", from pole to pole. I never worked out what caused this, though I suspect it may have had something to do with dodgy NR refinement of the reciprocal estimates I was getting from Altivec.
Edit: un-embedded pictures because they're too large and they distort the layout of the whole page { tbp: that's what the [img width=??? height=???] tag is for [SMILEY ;)] }
(L) [2007/06/17] [Alan] [Glitch pictures anyone?] Wayback![IMG #1 miniature]
I can't remember what caused this one anymore, something to do with tracing shadow rays for area light sources. The little cross/star shapes on the ground are fun. I like the strange look of the shadows from the lower horizontal struts on the torii behind them. Obviously the image is not very realistic, but there's something surviving the artifacts which the mind seems to pick up on in an almost hyper realistic way. I get the same thing looking at Toxie's first picture, not to claim that my pic is in the same league though [SMILEY Wink]
[IMG #2 miniature]
This is a "direct visualisation" of a photon map for a simple 144 polygon model I used for debugging. The spiral pattern you see goes a full 360 revolution "around the surface of a sphere", from pole to pole. I never worked out what caused this, though I suspect it may have had something to do with dodgy NR refinement of the reciprocal estimates I was getting from Altivec.
Edit: un-embedded pictures because they're too large and they distort the layout of the whole page { tbp: that's what the [img width=??? height=???] tag is for [SMILEY Wink] }
[IMG #1]:Not scraped:
https://web.archive.org/web/20071028083944im_/http://snowreverie.com/arclight/wp-content/uploads/2007/04/messup1.png
[IMG #2]:Not scraped:
https://web.archive.org/web/20071028083944im_/http://snowreverie.com/arclight/wp-content/uploads/2007/04/verystrange.png
(L) [2007/06/24] [necro] [Glitch pictures anyone?] Wayback!ok, something from my archives...
1. sophisticated light tracing, who cares about visibility?
[IMG #1 Image]
2. the dimension hole - reflection is just not easy enough for me to do it right...
[IMG #2 Image]
3. 1st iteration of quasi Monte Calo BDPT, keep book of your dimension [SMILEY ;-)]
[IMG #3 Image]
[IMG #1]:
![[IMG:#0]](images/a6b4984f9ac09cf6d1fb56b470b3b3ebfc15a146c963e8c0938dcda8d57465d0.png)
[IMG #2]:
![[IMG:#1]](images/71765296f4f2acd508aefce6b30ab450651c0419e3f869c0ba6a0fbf0045b9ae.png)
[IMG #3]:
(L) [2007/06/24] [necro] [Glitch pictures anyone?] Wayback!ok, something from my archives...
1. sophisticated light tracing, who cares about visibility?
[IMG #1 ]
2. the dimension hole - reflection is just not easy enough for me to do it right...
[IMG #2 ]
3. 1st iteration of quasi Monte Calo BDPT, keep book of your dimension [SMILEY Wink]
[IMG #3 ]
[IMG #1]:
![[IMG:#0]](images/a6b4984f9ac09cf6d1fb56b470b3b3ebfc15a146c963e8c0938dcda8d57465d0.png)
[IMG #2]:
![[IMG:#1]](images/71765296f4f2acd508aefce6b30ab450651c0419e3f869c0ba6a0fbf0045b9ae.png)
[IMG #3]:
(L) [2007/06/24] [phresnel] [Glitch pictures anyone?] Wayback!inspired by necro (grats on those great shots)
doh, my very first try on "global illumination" (not really GI, nowadays I'd say reversed-canonical-ray-tracing (forget about the terms "backward"/"forward"/whatsoever-ray-tracing(too much discussion on the correct usage of those terms in teh intahrweb))).
I shot rays in the light->scene order, for each intersection i verified the visibility with the camera (shooting a "shadow ray" to the camera). Only "primary hits" traced:
[IMG #1 Image]
edit: linkage updated
[IMG #1]:Not scraped:
https://web.archive.org/web/20110720174139im_/http://greenhybrid.net/images/glass_log_0028.jpg
(L) [2007/06/24] [greenhybrid] [Glitch pictures anyone?] Wayback!inspired by necro (grats on those great shots)
doh, my very first try on "global illumination" (not really GI, nowadays I'd say reversed-canonical-ray-tracing (forget about the terms "backward"/"forward"/whatsoever-ray-tracing(too much discussion on the correct usage of those terms in teh intahrweb))).
I shot rays in the light->scene order, for each intersection i verified the visibility with the camera (shooting a "shadow ray" to the camera). Only "primary hits" traced:
[IMG #1 ]
_________________
[LINK http://greenhybrid.net/ greenhybrid.net]
Real Men code Software Graphics.
[IMG #1]:Not scraped:
https://web.archive.org/web/20071028083944im_/http://www.root-engine.net/greenhybrid/images/glass_log_0028.jpg
(L) [2007/06/25] [fpsunflower] [Glitch pictures anyone?] Wayback!>> necro wrote:1. sophisticated light tracing, who cares about visibility?
Funny but seeing this image makes me wonder if there's a way to combine that with this paper:
[LINK http://www.ee.technion.ac.il/~ayellet/Ps/KatzTalBasri.pdf]
(Great thread by the way - I need to start saving my glitch images)
(L) [2007/06/28] [Wussie] [Glitch pictures anyone?] Wayback!this (to me at least [SMILEY :P]) is a rather old image, think it was taken roughly 20 weeks ago, a short period before started actual production with Phantom's engine. Something must've been messed up in the UV's (twas a quicky export of a Quake III Level to a useable .obj format).
[IMG #1 Image]
[IMG #1]:
(L) [2007/06/28] [Wussie] [Glitch pictures anyone?] Wayback!this (to me at least [SMILEY Razz]) is a rather old image, think it was taken roughly 20 weeks ago, a short period before started actual production with Phantom's engine. Something must've been messed up in the UV's (twas a quicky export of a Quake III Level to a useable .obj format).
[IMG #1 ]
[IMG #1]:
(L) [2007/06/29] [Phantom] [Glitch pictures anyone?] Wayback!>> Wussie wrote:this (to me at least ) is a rather old image, think it was taken roughly 20 weeks ago, a short period before started actual production with Phantom's engine. Something must've been messed up in the UV's (twas a quicky export of a Quake III Level to a useable .obj format).
Haha, you couldn't resist. [SMILEY :)]
(L) [2007/07/12] [toxie] [Glitch pictures anyone?] Wayback![IMG #1 Image]
[IMG #2 Image]
[IMG #3 Image]
camera ray precision falling apart in the powerplant scene..
[IMG #1]:
![[IMG:#0]](images/7c68999d85ee581093fae00478dd9a9f3027ed3dd5322f5f08776b15a0f7f9c1.png)
[IMG #2]:
![[IMG:#1]](images/6951786ab347f6176c31eaa7de03c772958262c8644b223d145774520b723ba9.png)
[IMG #3]:
(L) [2007/07/12] [toxie] [Glitch pictures anyone?] Wayback![IMG #1 ]
[IMG #2 ]
[IMG #3 ]
camera ray precision falling apart in the powerplant scene..
_________________
The box. You opened it. We came.
[IMG #1]:
![[IMG:#0]](images/7c68999d85ee581093fae00478dd9a9f3027ed3dd5322f5f08776b15a0f7f9c1.png)
[IMG #2]:
![[IMG:#1]](images/6951786ab347f6176c31eaa7de03c772958262c8644b223d145774520b723ba9.png)
[IMG #3]:
(L) [2007/07/18] [toxie] [Glitch pictures anyone?] Wayback!QMC si teh r0xx0r!
[IMG #1 Image]
and no, this is actually NOT photon-mapping
[IMG #1]:
(L) [2007/07/18] [toxie] [Glitch pictures anyone?] Wayback!QMC si teh r0xx0r!
[IMG #1 ]
and no, this is actually NOT photon-mapping
_________________
The box. You opened it. We came.
[IMG #1]:
(L) [2007/07/20] [fpsunflower] [Glitch pictures anyone?] Wayback!A failed attempt at optimizing ray/triangle intersection:
[IMG #1 Image]
[IMG #1]:
(L) [2007/07/20] [fpsunflower] [Glitch pictures anyone?] Wayback!A failed attempt at optimizing ray/triangle intersection:
[IMG #1 ]
[IMG #1]:
(L) [2007/07/20] [slartybartfast] [Glitch pictures anyone?] Wayback!WHOA!!!!   Cool vertex shader program ..... can you post the code ?   [SMILEY :wink:]
LOL - just kidding.
I had something similar when trying to code up my BIH tree traversal / tri intersection code. Except mine looked like little tris pushed out from the surface, rather than little cubes.
(L) [2007/07/20] [slartybartfast] [Glitch pictures anyone?] Wayback!WHOA!!!!   Cool vertex shader program ..... can you post the code ?   [SMILEY Wink]
LOL - just kidding.
I had something similar when trying to code up my BIH tree traversal / tri intersection code. Except mine looked like little tris pushed out from the surface, rather than little cubes.
_________________
S. Hurry, or you'll be late.
A. Late ? Late for what ?
S. Late. As in "The Late Dent-Arthur-Dent"
(L) [2007/08/25] [Alan] [Glitch pictures anyone?] Wayback![IMG #1 ]
[IMG #1]:
(L) [2007/08/29] [coldcity] [Glitch pictures anyone?] Wayback!Some lovely glitches in this thread!!
I forget what I was trying to do, but when it broke it broke pretty:
[IMG #1 Image]
[IMG #1]:
(L) [2007/08/29] [coldcity] [Glitch pictures anyone?] Wayback!Some lovely glitches in this thread!!
I forget what I was trying to do, but when it broke it broke pretty:
[IMG #1 ]
[IMG #1]:
(L) [2008/01/19] [Phantom] [Glitch pictures anyone?] Wayback!Photon mapper gone mad:
[IMG #1 Image]
During an attempt to bring down memory requirements...
[IMG #1]:
(L) [2008/01/19] [davepermen] [Glitch pictures anyone?] Wayback!>> Phantom wrote:Photon mapper gone mad:
[...]
During an attempt to bring down memory requirements...
is this the matrix? [SMILEY :)] (make it green, make it green!!!)
(L) [2008/01/19] [ypoissant] [Glitch pictures anyone?] Wayback!When I implemented Photon Mapping for Animation:Master, I made a blooper page I called "[LINK http://www.ypoart.com/tutorials/photon/bloopers.php Different ways to render a Cornell Box]"
Here is one example:
[IMG #1 Image]
[IMG #1]:
(L) [2008/02/09] [auld] [Glitch pictures anyone?] Wayback!Hi, my first post at the forum. Naturally its a "glitch". This glitch resulted because I had reflected rays going back into an object *and* I didn't move the ray off the surface by a small delta. I'm still unable to explain the result but I put it down to floating point accuracy of graphics cards.
(L) [2008/02/26] [slartybartfast] [Glitch pictures anyone?] Wayback!Hey Lycium. That's not a glitch picture !!  It's impressionistic art ...  [SMILEY :wink:]
(L) [2008/02/27] [toxie] [Glitch pictures anyone?] Wayback!this is most definetly the most beautiful glitch art ever!
(L) [2008/02/27] [lycium] [Glitch pictures anyone?] Wayback!thx [SMILEY :)] you can blame the halton-zaremba sequence for this mess, it seems to be interacting with the metropolis-hastings sampling in a weird way... quite fun to watch it render, check the exes if you have a 64bit box handy!
now to hook it up to some autechre/brothomstates music via fft... [SMILEY ;)]
(L) [2008/02/27] [fpsunflower] [Glitch pictures anyone?] Wayback![SMILEY :thumbsup:] You should definitely consider making a real "glitch renderer" out of this one! Its a new art form! [SMILEY :D]
(L) [2008/02/27] [lycium] [Glitch pictures anyone?] Wayback!the "art" of glitch rendering:
1. back up your rendering system
2. go out, find a nearby bar
3. get completely hammered, crawl home by your eyelids (if you can still walk, goto step 2)
4. fire up your most complex sampling code, and "improve it" as much as possible
5. start a render to cook while you recover from hangover [SMILEY ;)]
this patent-pending algorithm can even be adapted for non-alcohol-drinking people (like me) with minor modifications...
(L) [2008/02/27] [derethor] [Glitch pictures anyone?] Wayback!the problem is that the second point, normally, is an unexpected exception, and the point one can not be predicted in advance...
(L) [2008/02/27] [lycium] [Glitch pictures anyone?] Wayback!>> Currently working on Quasi Montecarlo Sampling
to be a little more honest, that's where this effect comes from [SMILEY ;)]
as for periodically backing up code, that's always a good idea even without drinking!
(L) [2008/02/28] [toxie] [Glitch pictures anyone?] Wayback!I like this easy 5-step approach!
Be an artist within hours! Try our new and easy 5 step program to impress your friends and other nerds!    [SMILEY ;)] [SMILEY ;)]
(L) [2008/02/28] [phresnel] [Glitch pictures anyone?] Wayback!>> lycium wrote:as for periodically backing up code, that's always a good idea even without drinking!
/one shalt backup the backup ( of the backup)*, plus see if the backup ( of the backup)* is okay\. store it external\./;
once i backed up my whole "/multimedia" partition, and a backup inside the backup of the rock-partition did not restore properly ...
(L) [2008/02/28] [derethor] [Glitch pictures anyone?] Wayback!so, the solution is a bored, but deterministic, quasi montecarlo drinking.... like saturday sex with your wife: fast, efficient, optimized, and using a small amount of system resources. In that case you dont need a backup (and never talk about external backups, of course)
but if you want to test new sexy "algorithms".... you should backup, at least, your credit card [SMILEY :)]
(L) [2008/03/05] [movax] [Glitch pictures anyone?] Wayback!Hi, my first post on the forum [SMILEY :)]
I have working raytracer, but it's not anything special. But some glitch pictures can be...
[IMG #1 Image]
This triangle should have solid color, but reflectance vector is wrongly calculated for some pixels
[IMG #2 Image]
texel avaraing for bilinear interpolation broken
[IMG #3 Image]
error in SSE code for rgb color operations
[IMG #4 Image]
u/v coordinated for normal interpolation flipped
I have also few other pictures with errors, but not really interesting (like rounding errors - lots of black pixels in random places)
[IMG #1]:
![[IMG:#0]](images/641ae7a6a54952bba8c0b35b7de1c8a4dca8f945c413ef78b360c7b4d25efc21.png)
[IMG #2]:
![[IMG:#1]](images/96bc29151e5eadaef3f62a67840b8d096ef838827242cf63dc635a0d0a239faa.png)
[IMG #3]:
![[IMG:#2]](images/83bc4cfe916cb50d5d402cc8c7c489ecf447b79468215991d4718318b7a3ad12.png)
[IMG #4]:
(L) [2008/03/06] [toxie] [Glitch pictures anyone?] Wayback!i like the SSE-mixup pic..
(and welcome to the forum.. [SMILEY ;)])
(L) [2008/04/22] [toxie] [Glitch pictures anyone?] Wayback![IMG #1 Image]
[IMG #2 Image]
images (C) funky
[IMG #1]:
![[IMG:#0]](images/b56e8f30e4ec26a416cae969beaf97f36dea82a315de7b23a6fe32eea872524f.png)
[IMG #2]:
back