Glitch Art

ED̀͘Į̛T :̢ As it has been close to a year since beginning to experiment with Glitch Art, I would like to update the links here for anyone else interested in Glitch art or for further research. I will aim to continue posting my experiments and I also encourage anyone to post their results too!

ED̀͘Į̛T 2 (2018) : I have added a link/resource page to my website which lists everything from various software, tutorials (including my own) and further reading here:

https://vaeprism.com/resources/

original post:

Re̷s̴o͘u͢r̀ces:

http://facebook.com/groups/GACToolTime

http://glitchtheory.tumblr.com/resources

http://www.animalswithinanimals.com/bent/

http://nickbriz.com/thoughtsonglitchart/

https://www.reddit.com/r/brokenanimegifs/wiki/tutorial

https://prezi.com/gow_u3gtew1w/glitch-workshop/

http://datamoshing.com/

http://freestockfootagearchive.com/free-free-glitch-art-tools/

http://www.glitchet.com/resources

Recently I have been experimenting with glitch art. This was a progression from glitching images to produce sound. I often import images into audio software (Mostly audacity, however as renoise can open any file, i do use it too), sometimes apply effects or just save/export the audio file. This can produce interesting results which can vary depending on file format and other factors. As I am moving more towards creating my own samples, I have begun to create more glitched sounds rather than rely on others. I also find satisfaction in creating or capturing a “real glitch” as opposed to something controlled. There is art in the mistakes created when pushing technology beyond it’s intended use too and I am really enjoying exploring this.

As I have also been searching for art to complement my sound, it seemed natural to seek methods of visual glitching. I found a few helpful resources and started experimenting with data bending using a hex editor (Hexedit) and Wordpad. Results varied from different file formats (raw,jpg,bmp,tiff,psd) and processes. The possibility of manipulating image layers in a psd (phtoshop file) seems really interesting to me too.

For those unfamiliar with glitch art, here is a brief description on Wikipedia:

" Glitch art is the aestheticization of digital or analog errors, such as artifacts and other “bugs”, by either corrupting digital code/data or by physically manipulating electronic devices (for example by circuit bending)." (read full article here).

I have only been experimenting with databending for a week, though I am enjoying the process and will continue. There is much to explore.

Some of you may already be quite familiar with glitch art, using hex editors etc. Please share your art

Some examples of my images:

vaeprismclouddodglitchdsml.jpg694×400

stdiogltch.jpg800×450

Is there a way to turn sound files into images too?

The times i imported non audible files into renoise i rarely get very interesting sounds, can you recommend a file format or something that makes more interesting results?

Cool stuff. But I imagine it a tedious process of manually “glitching” images over and over in random fashion until it produces something interesting. I imagine compressed formats giving the most interesting results.

Bellows: in contrary to glitching compressed images, I found uncompressed data the most interesting for generating audio. Plain ascii text files for example can sound similiar to old modems. I think raw uncompressed images (like .bmp) are also good candidates for making audio of them, the audio kind of reflects the image still. Compressed stuff - will most often sound very noiselike, but with a little blip in the beginning that is the file header. You can also experiments with the audio import settings, bit depth, stereo or not, sample rate.

here’s something I made a few years ago using hexeditors & kyma;

Is there a way to turn sound files into images too?

The times i imported non audible files into renoise i rarely get very interesting sounds, can you recommend a file format or something that makes more interesting results?

I do plan to explore this further. I have only been able to produce basic patterned images so far. there are many ways to go about it. As mentioned by OOpslFly uncompressed files such as bmp. Also try .raw.

I have gotten better results from audacity so far, though renoise has worked. Sometimes files just don’t sound good, I just experiment a lot. I think it’s too early though for me to give any solid tips. i found the link to the data bending and glitch art primer has given me some kind of understanding, and a kind of starting point, though i feel there is much to explore.

Cool stuff. But I imagine it a tedious process of manually “glitching” images over and over in random fashion until it produces something interesting. I imagine compressed formats giving the most interesting results.

It can be, though I do like a bit of tedium and especially if it means I am able to capture something that I would otherwise struggle to recreate manually. I like the process in that sense. I am also strangely attracted to cut and paste, shortcut keys and repetition. Combine that with find and replace in a hex editor, where i can create a cycle to produce creative results and i can lose hours

here’s something I made a few years ago using hexeditors & kyma;

…

Enjoyed that cheers. Nice sound design. Had not heard of kyma, Is this what you mean? http://kyma.symbolicsound.com/kyma-sound-design-environment/

If so, looks interesting.

executables (.com, .exe, .dll, …) also have a “special sound” unless they are packed, which would let them sound noisy again. remember to dc correct/hpf your glitch audio results at the right stage, or else the offset can blow up your mix.

as for audio glitches -> importing raw uncompressed audio with the wrong sample format settings is the devil’s bitcrusher-distortion…

Big fan of glitch. Made a series of glitched GIF animations and used them for a music video:

Recently made some simpler ones to have some background noise for the music:

I’ve a few ,more I need to wrap up and upload.

I wrote some Ruby code to disrupt image files, creating a sequence of new images, which are then combined using Image Magick. There’s a lot of time spent trying different parameters to get good results.

It’sinterestingto see how different file formats handle disruptions bmp files are pretty robust but tend not to end up looking terribly distressed. png and jpgs can get that WTF look but there’s a higher chance of anunreadablefile.

And glitch makes for nice Vine content : )

https://vine.co/Neurogami

Really into the whole glitch art/databend aesthetic. Recently been using effects in audacity to manipulate bitmaps loaded in as raw data, though I wish I could manually manipulate header data to make images larger or smaller. There’s nothing special about audacity, any DAW that can load raw will work, but audacity does include some nice import and export options. I’ve been using audacity and gimp to edit images. But using gimp to edit audio results in nothing but noise. I’d be really interested in figuring out a way to use image processing filters as audio effects but loading in wav files as bitmaps definitely doesn’t provide any useful features other than turning an audio file into noise, but I haven’t dug too deep yet. Definitely exciting territory. I think scripting could work wonders in this domain.

^^ For sound I use different file types too, psd files can be interesting (maybe it’s the layers?). Mostly it is just noise, though i just cut small samples to use as loops or one shots in renoise.

Using image effects on audio would be interesting.

Following this method: http://www.hellocatfood.com/databending-using-audacity/ and saving copy of an image as a non interleaved .raw file (instead of using the suggested bmp to import into audacity etc) from photoshop, I have been getting some interesting results.

I am also getting variations when i vary the import settings once i have exported from audacity and open the .raw file in photoshop. Finding the right balance between dimensions/channel count/bit depth seems key. (the first image is the raw file unprocessed for comparison):

vaeprism_audgltchdatabendzzzzzz.jpg1243×1240

Following this method: http://www.hellocatfood.com/databending-using-audacity/

I’m listening to Xanopticon’s Liminal Space while doing this right now.

I’m listening to Xanopticon’s Liminal Space while doing this right now.

Zang. Been listening to Richard Devine and Vaetxh whilst bending.

I finaly published a simple video for Swarm Intelligence, done with with 3D sources reacting to audio and transformed in a glitchy fashion in jitter.

The glitchy parts are around in the middle…

here’s something I made a few years ago using hexeditors & kyma;

https://www.youtube.com/watch?v=Uiuy0EiYijo

Kyma!?!?Did you buy it?thats mega bucks $$$$

databending some of my art in hexedit…

jmorbus_glitchd_bw.jpg?w=10001000×318

And this photo…

CnurUQqVIAAa12t.jpg750×750

recent experiments…

Short Audiovisual glitch experiment here: https://www.instagram.com/p/BIkEN9fAgwb/

I’ve always enjoyed glitch art - saved loads of pics over the years where the computer has messed up. Would love to know how to glitch videos

I’ve always enjoyed glitch art - saved loads of pics over the years where the computer has messed up. Would love to know how to glitch videos

Feel free to share. There is a section on datamoshing available on this page: http://www.glitchet.com/resources

I decided to try bending a gif file and later a mp4 video.
I got results with both only using ‘find and replace’ with a hex editor (i use hexedit:http://hexedit.com/ ).

Here is the result from bending the gif file:

Original:

After processing:

This was achieved relatively quickly. I did also have immediate results with the video file (though it also glitched the audio a little).

Edit: Tried datamoshing video using avidemux (Must use 2.56 or earlier. available here: https://sourceforge.net/projects/avidemux/files/avidemux/2.5.6/) and it worked.

If you are interested, this tutorial helped: http://datamoshing.com/2016/06/26/how-to-datamosh-videos/