I try to make something that looks good (or at least doesn’t look like random static) by running pictures I’ve taken through audio editing software. There are some extra steps that go into it to “trick” the program into importing the picture as if it were a sound file, making sure the header (information that tells your computer that this is a picture) doesn’t get fucked with, and then exporting the data in a way that it will be saved as a picture and not an mp3 or something else.
On the rare occasion I do bring it up, I can literally watch people’s eyes glaze over. Until I show them a picture
Edit: internet is really bad right now, will reply with an image when I can
Edit2: picture was too big at 7MB. Hopefully a screenshot of the picture doesn’t look too bad
I do a lot of photography and I’ve been trying to find something that I could do with some of my more experimental shots that makes them more… more‡. If that makes sense?
You wouldn’t happen to have more details on how to do it would you?
Edit: ‡ My more experimental shots are more done as like experimenting with how a shot is taken for like evoking a specific feeling or doing something strange in camera or really any number of reasons. Hell some of my experimental shots were accidentally taken pictures that are disorienting or confusing. I don’t share them often, because IDK it just seems like really personal sometimes. Those experimental shots feel less like photography and more like painting with photos.
I have a bookmark saved on my computer at home to an old forum with the instructions I followed when I started doing this, and I can send that link later.
There are two programs that I use, and both are free.
GIMP - image editing software
Audacity - audio editing software
Here is the basic process from that bookmarked forum post that I can remember off the top of my head. If something is wrong (especially the Audacity import settings, since I don’t ever change them), I will fix it later.
In GIMP (or other software of your choice) convert the image to a bitmap (.bmp). This step is very important!
Use the option to import raw data as A-law with “little endian” (I have no idea what those setting do, but I assume it’s for keeping the header intact)
Change the timeline in Audacity from time to samples and select everything after the 34th sample to edit and add effects (samples 1-34 are the information that tells your computer that this is a picture CHANGING ANYTHING IN THE HEADER WILL STOP YOU FROM OPENING THE IMAGE AFTER THE EDIT)
Export the audio using the raw data option, selecting A-law again. This should re-save the “audio” as a bitmap image as it will not add an audio file header to the data.
I believe the blue parking garage image uses reverb, or maybe a phasor… possible both to get that effect? But there are a lot of setting to mess with for each audio effect that can dramatically change the outcome. The trees picture was made by putting the original picture in the left audio channel, and putting a horizontally flipped copy of the image in the right audio channel. Delete the header from the flipped copy, and exporting the data smashes them together in this really strange mirror effect. Afterward, I would use GIMP for any color correcting, changing saturation/hue, simple stuff
Fun fact: Lightroom has no idea how to process the image results in any semblance of consistent way, so literally which sliders you mess with first changes how the image as a whole is effected. And sometimes even moving the slider vs typing a number in does different things. Which leads to more and stranger distortion.
I just wanna know how people even came up with data moshing. It blows my mind that there’s a thought process that incorporates these steps, and ends with awesome (sometimes) results!
That’s pretty cool and definitely falls under the category of a hobby that you do because you can, although I’m sure there are people in the world who would pay for art like that (not suggesting that you do).
I have had the thought about trying to make money off of this, and some friends have joked that I should be making album art or something like that.
However, I don’t want to feel like I have to meet a deadline or feel the pressure of making something that someone else wants/likes. I just want to make something I like
you just post the link to image hosted on imgur.com or catbox.moe, don’t really need to upload to lemmy itself. Like this - 
Most of the time I don’t have a plan of what I want out of something. Sometimes I can get something that looks interesting or cool right away, other times I have hundreds of files trashed over a couple of days and there’s not a single one that I personally liked
Considering it’s audio-software, I guess the changes are related to frequency changes. You should look up Fourier transform (the function that allows to see the sound frequencies of music, for example) applied to images and play with it. If you are not afraid to do a little bit of Python coding, you should be able to have much more control on the parameters responsible for the visual effects you’re looking for.
The image equivalent of bass frequencies (long wavelength) are big details (ex: the trees) and high frequencies (short wavelength) are small details (ex: the leaves).
I haven’t had to think about Fouier since college, so thanks for bringing up that trauma lmao. I do realize that there are ways to reliably get certain effects or even learn how to do this in Photoshop or GIMP, but I like the shotgunning, spray-and-pray of not knowing what the outcome will look like
I guess I find the process of going back to make small changes to the settings and then seeing how that affects the image more satisfying. Getting something that looks good is just a bonus
I did try turning a song into an image last night, and it looked like TV static. Kinda disappointing, but I’ll try some other songs and see if I can mess with some settings
I used to do a similar thing where I saved photos as .txt files, but they were an awful lot glitchier, and usually split the RGB channels to look as if you had three eyes and were crossing them.
I try to make something that looks good (or at least doesn’t look like random static) by running pictures I’ve taken through audio editing software. There are some extra steps that go into it to “trick” the program into importing the picture as if it were a sound file, making sure the header (information that tells your computer that this is a picture) doesn’t get fucked with, and then exporting the data in a way that it will be saved as a picture and not an mp3 or something else.
On the rare occasion I do bring it up, I can literally watch people’s eyes glaze over. Until I show them a picture
Edit: internet is really bad right now, will reply with an image when I can
Edit2: picture was too big at 7MB. Hopefully a screenshot of the picture doesn’t look too bad
This one looks amazing
Thanks!
This belongs on the better call Saul intro
Ok now that’s really cool
I do a lot of photography and I’ve been trying to find something that I could do with some of my more experimental shots that makes them more… more‡. If that makes sense?
You wouldn’t happen to have more details on how to do it would you?
Edit: ‡ My more experimental shots are more done as like experimenting with how a shot is taken for like evoking a specific feeling or doing something strange in camera or really any number of reasons. Hell some of my experimental shots were accidentally taken pictures that are disorienting or confusing. I don’t share them often, because IDK it just seems like really personal sometimes. Those experimental shots feel less like photography and more like painting with photos.
Absolutely!
I have a bookmark saved on my computer at home to an old forum with the instructions I followed when I started doing this, and I can send that link later.
There are two programs that I use, and both are free.
GIMP - image editing software
Audacity - audio editing software
Here is the basic process from that bookmarked forum post that I can remember off the top of my head. If something is wrong (especially the Audacity import settings, since I don’t ever change them), I will fix it later.
In GIMP (or other software of your choice) convert the image to a bitmap (.bmp). This step is very important!
Use the option to import raw data as A-law with “little endian” (I have no idea what those setting do, but I assume it’s for keeping the header intact)
Change the timeline in Audacity from time to samples and select everything after the 34th sample to edit and add effects (samples 1-34 are the information that tells your computer that this is a picture CHANGING ANYTHING IN THE HEADER WILL STOP YOU FROM OPENING THE IMAGE AFTER THE EDIT)
Export the audio using the raw data option, selecting A-law again. This should re-save the “audio” as a bitmap image as it will not add an audio file header to the data.
I believe the blue parking garage image uses reverb, or maybe a phasor… possible both to get that effect? But there are a lot of setting to mess with for each audio effect that can dramatically change the outcome. The trees picture was made by putting the original picture in the left audio channel, and putting a horizontally flipped copy of the image in the right audio channel. Delete the header from the flipped copy, and exporting the data smashes them together in this really strange mirror effect. Afterward, I would use GIMP for any color correcting, changing saturation/hue, simple stuff
Edit: spelling and formatting
Thank you so much, I’m going to have to give this a try when I get home from work
Here’s the link to the forum post!
https://empireminecraft.com/threads/tutorial-editing-images-with-audacity.81959/
That’s a name I haven’t seen in a long time
Thank you so much, I’m so looking forward to giving this a shot (too much OT this week so far)
Fun fact: Lightroom has no idea how to process the image results in any semblance of consistent way, so literally which sliders you mess with first changes how the image as a whole is effected. And sometimes even moving the slider vs typing a number in does different things. Which leads to more and stranger distortion.
I’m really digging this way of editing pictures
source image
edited result
Hell ya dude, that’s awesome!
I’ve done experimental image editing (not like yours) and I appreciate the process.
I just wanna know how people even came up with data moshing. It blows my mind that there’s a thought process that incorporates these steps, and ends with awesome (sometimes) results!
That’s pretty cool and definitely falls under the category of a hobby that you do because you can, although I’m sure there are people in the world who would pay for art like that (not suggesting that you do).
I have had the thought about trying to make money off of this, and some friends have joked that I should be making album art or something like that.
However, I don’t want to feel like I have to meet a deadline or feel the pressure of making something that someone else wants/likes. I just want to make something I like
I completely get that, adding a timeline or even someone else’s opinions changes things
deleted by creator
Very unpleasant, basically just high pitched static. 1/10 wouldn’t recommend
What about going the other way and turning sounds into images?
I did try that yesterday, and it looked like TV static. Kinda disappointing, but I’ll try some other songs and see if I can messing with some settings
I feel like there is a middle ground where an image looks decent and sounds decent. Probably be really hard to find though.
Autechre
Looks pretty good, like some Rorschach test. War sit a picture of trees?
It was! Taken during the winter so there were no leaves, and at night too
Do you have the original?
Here is the original. Just a picture from my backyard looking at the neighbor’s house and then i did some cropping after the audio software fuckery
Wow! What a difference!
Not at the moment, but it should be floating around somewhere on my computer at home. I’ll try to remember to find it after work
you just post the link to image hosted on imgur.com or catbox.moe, don’t really need to upload to lemmy itself. Like this -
How much tinkering do they require to make the final product look like you want?
Most of the time I don’t have a plan of what I want out of something. Sometimes I can get something that looks interesting or cool right away, other times I have hundreds of files trashed over a couple of days and there’s not a single one that I personally liked
This is so fucking cool. Where can I learn more about this??
edit, I see your other post explaining!
The search term to start you with is Glitch Art :)
oh yeah! I have heard of that before. I’ll look to add some tags to my tumblr to follow, thanks!
<3
Considering it’s audio-software, I guess the changes are related to frequency changes. You should look up Fourier transform (the function that allows to see the sound frequencies of music, for example) applied to images and play with it. If you are not afraid to do a little bit of Python coding, you should be able to have much more control on the parameters responsible for the visual effects you’re looking for.
The image equivalent of bass frequencies (long wavelength) are big details (ex: the trees) and high frequencies (short wavelength) are small details (ex: the leaves).
I haven’t had to think about Fouier since college, so thanks for bringing up that trauma lmao. I do realize that there are ways to reliably get certain effects or even learn how to do this in Photoshop or GIMP, but I like the shotgunning, spray-and-pray of not knowing what the outcome will look like
I guess I find the process of going back to make small changes to the settings and then seeing how that affects the image more satisfying. Getting something that looks good is just a bonus
This is so cool
Is there a way to convert an audio file to an image? Might be interesting for album covers.
That should be possible. Haven’t tried that yet, but now I think I might
I did try turning a song into an image last night, and it looked like TV static. Kinda disappointing, but I’ll try some other songs and see if I can mess with some settings
Try looking into “fourier transform”. That gives some interestung results.
Yes it is, but don’t expect too much.
I used to do a similar thing where I saved photos as .txt files, but they were an awful lot glitchier, and usually split the RGB channels to look as if you had three eyes and were crossing them.
Removed by mod
deleted by creator
deleted by creator