Відео

Hey, thank you. If you want to do this you'll need to play around with the DAT and text TOP portion of the network where the letters are randomized and generated. Note that the font we use here automatically changes any characters you input into kana. I put together a network where random letters convert to a sensible sentence like they have in the opening matrix sequence which I haven't released yet.

@@otodojo Hmm, I'll give it a go. This is a tricky thing to experiment with. Spent a few hours messing with the columns and rows - the toggle to keep unique values in the first column and rows kind of helped to a degree. Kind of wish there was a plugin for text that allows certain characters to change, because this exact scenerio came up in a different project I'm working on. I'm just getting started in TD. Do you offer any lessons? (paid or otherwise?).

@ word! It’s accomplishable through Python scripts - converting the text to indices then back to text - this way all of the text are essentially numbers that you can apply math functions / LFO to. I do offer lessons! You can email me at otodojo@gmail.com :)

Hey there are hints for this in the 7-8:30 min section but it’s not so flexible and requires lots of iterating. You can have an easier time with more control over columns by replicating this with my ASCII tutorial!

@@otodojo Thanks a lot! I'll try!

hey, yep! It will import as a base comp. To collect a network into a base comp you can select all the nodes, right click the network background, and select collapse selected. Just put an out top from where you want the output to be and it should show up as a purple output node in the base comp! Here's a tutorial from Matthew Ragan: forum.derivative.ca/t/where-is-collapse/150767

answering myself here after a bit of digging : "as our perception of pitch is logarithmic (we hear equal jumps in pitch when the frequency doubles), using this formula ensures that the resulting audio frequencies (that we generate from pixel values) will be perceptually meaningful (sound together)"

Hey there, that has to do with reducing the number of replicants at 2:50 and subsequently the 'Count' or cache size in tex3d to match the number of replicants, as well as the speed of change for your switch depending on your cache size. Hope this helps!

www.dafont.com/matrix-code-nfi.font

that's surprising given that your computer is faster than mine - first trying increasing your cache size in your settings - then maybe reduce feedback levels of the note echos.

thanks for following! microtones.bandcamp.com/track/plexus

I believe 24 projectors hanging from the ceiling from their "WATCHOUT system". The space is 416 sq m

nvm i got it, just needed to add an otf file

glad you figured it out @@stiffyBlicky !

hope it helps!

thanks! On the simple side you can replace the text node with a movie file in TOP.

But anyway, nice tutorial! I am confused how I could get it wrong so many times because I find your explanations super easy to follow.

So, there are two parts to this, the UV map, and the texture 3D. I'm thinking that you haven't filled your texture 3D (it will disappear over time if you don't feed it anything). The switch is used to quickly flip through the letters that are in the input to fill the texture with unique letters. You need to automate the switching. The input of the switch can be anything, I just chose alphabet because that's the classic ASCII. If you're loading only blank textures that could be a reason for not seeing anything in the Remap. Have you taken a look at the reference file I put in the description? Check that out and then see if you can find out what's wrong with yours.

@@otodojo thank you for replying! I'll try to fix it with your advice :) Also, completely forgot about the reference file lol. I'll check it out

i was having the same issue, so what i did was deleted all the replicated text tops, go in the original text top and change the backgroun alpha to 1 to give it a black background, then in the replicator comp, hit recreate all operators. then proceed as normal (i ensured i redid the steps in order, connect the switch out to the text3d first to fill it up, then deactivate and it then connect the select to it and then hit replace single pulse) and then after all the ramp stuff i was nearly there, but for some reason the final result had a few squares on the grid here and there that didnt have the black background and u could see the alpha layer underneath, to fix this i changed extend in remap to repeat, and voila! im confused though as to why i had to do these extra steps cuz i followed this video exactly

Yes it can be done with adjusting camera settings to orthographic and not having movement, changing the ordering of the boxes for the main instance source. You're right that I didn't map the listener location to the camera location, so it's pseudo-spatialization, but you can try something different! Good luck on the party.

thanks for listening!

thank you!

I'm glad you liked it!

thank you!

thank you for the support!

LET ME KNOW IF YOU FIGURE OUT HOW

@@javaroque same！plz tell me if you know

Yes, I haven't implemented this but I would assume you would change the script to instead of creating a random character from the LFO to index a phrase you have ( like "hello world"[index]) you might need to tweak the randomisation of the texture map and the LFO values to correspond to you input string but that would be it I assume.