@giku, something I’ve noticed which would be enormously useful for Visual would be for users to be able to easily reorder the layers of a mixer without having to remove all channels and start again so that the correct images are on top of each other. This could be by long pressing and sliding the channels to move them or via ‘Move up/Move down’ text options in their double tap menus.
Also, a one tap way to select all code could be worth considering. When working with AI generated code and the constant back and forth that requires, a quick way to select all four deletion would really help workflow.
Hopefully this is just an oversight, but why is there a 4:3 aspect ratio option for the projector but no 4:3 ratio option for export? You set things up in 4:3 and then can’t export in that format?
I’ve been experimenting with how possible it is to apply predictable rhythmic motion to hand drawn images using Drambo Visual. Here’s an experiment from yesterday.
It’s frustrating, you have to double tap quickly and hold the second tap until you get the option “select”. Tapping that will open a menu with select all
don’t tap on text but on blank space to the right of text
The process I’ve been attempting to use is quite simple really. After feeding it the rules from the Drambo Docs, I asked Gemini to write me a processor shader of a dancing circle shape with various parameters.
I then applied this processor several times to different parts of the cat image (just one image, not broken up or layered), with a different gate/velocity sequencer or LFO driving each one. Mixer for background and drum on different layers.
The important parameter the processor shader needs is blur/falloff so that the dancing circle isn’t just a hard edged distortion on the incoming image texture. I don’t know whether there’s a more elegant way of doing this, it is the main flaw in the system (the distortions even when blurred are like lens distortions) but I know nothing about the coding intricacies and possibilities.
I’ve been experimenting a bit today with making an adjustable dancing line instead of a circle which is probably a better, more flexible approach ultimately?
Essentially, I’m trying to rig the image texture using a processor shader skeleton underneath.
If this concept/approach seems interesting or viable to @giku, it’d be great if some processor shaders could be included specifically for this purpose? Cheers.
@Robin How do you see it? Which params, how should it move etc... you would probably end up in dozen of control and behaviour dimensions.
Thats why it's better to divide it into smaller modular parts: modulation texture generator -> transformer -> target mask transform. and that's how it works now :)
You can start from a basic Spectrum analyzer in angular mode to modulate radius of glowing circle.. so we have a basic "dancing circle" responding to audio. Then you can modify it's behaviour, appearance and nuances by other modules in the chain.
@giku Yeah, the ‘problem’ as I see it is that that very quickly sounds very technical. No doubt perfect for the people who know about such things but I’m guessing people who come from a non modular animation background are likely to think of hacking it in the way I’m describing - in terms of rigging.
Having said that, I’ve only recently bought Visual less than a week ago so you somewhat have the advantage over me! I’ve not even attempted to build a shader from scratch using a chain of modules so maybe it is all possible and really easy to do? I’ll certainly try.
To play devil’s advocate though, if the current tools are the best to do it with, why haven’t I seen any examples of other people trying to do what I’m experimenting with? Either it’s because what I’m trying to do is shit and no one else wants to do it (which in fairness is possible!) or the tools don’t naturally lend themselves to it? I mean, how is a newcomer supposed to realize that the Spectrum analyzer can be used to create a dancing circle? Isn’t it more helpful just to provide them a dancing circle or line?
You suggest that providing such a processor shader in app would result in it having lots of parameters. Does this mean you fear users would see it as too complicated? Again, I’d suggest a lot of users will see the prospect of chaining together lots of modules as being a lot more complicated than a ready made processor with many parameters. The one I asked Gemini to code has eight exposed parameters.
I was only suggesting it might be useful so If it doesn’t appeal, that’s cool. The great thing is that Visual doesn’t stop you from doing it anyway! Cheers.
By the way, small possible bug to report: I keep finding the system qwerty keyboard (in floating mode) keeps getting stuck on screen (can’t be dismissed) when in Code mode for the processor shader.
Speaking of the code editor, there is no way to maximize it or make it wider, right? I have some long lines with multiple indentation, sometimes it becomes harder to follow the logic when the line gets broken.
Would it be good if the editor could be made floating like the XY pad?
Are people mostly coding outside of Drambo and doing the select all copy-paste routine all the time? Or perhaps using a Mac with Drambo when coding.
I code outside, using notepad++ and then copy the results. Minor edits I do occasionally directly on ipad or iphone but big edits would be painful. The font is too small, and even with a BT keyboard you can’t find words, find and replace, etc. Even a simple editor like notepad++ can highlight all instances of a keyword, which is very convenient
True, that is. For work I use VS Code, but Notepad++ is also good.
Benefit of doing it on iPad is easier transition to Drambo (I don’t have D on a pc). I’ll try using GitHub Codespaces, it gives you a VS Code inside a Safari tab on iPad. Hopefully with syntax coloring and highlighting all instances etc.
Btw still on the topic of select all, no, merely press hold doesn’t work in many places, including notes (where it brings up the magnifier). Double-tap hold works everywhere
Thanks @pedro - I'll stick with Textastic on the Mac, but your post inspired me to get Notepad++ up and running on my Mac with Wine. It's good to see my old Windows days friend Notepad++ again. 😎
Just in case you weren't aware (I only noticed it the other day) - there's a shared clipboard between iOS devices and Macs, I'm not sure if they have to be on the same network + signed into the same Apple Account or whatever, but it's definitely working for me and is a huge timesaver.
Trying to get my head around Scenes. I’ve read and reread the pertinent sections in the Docs but none of the descriptions have quite explained it in plain enough language for me to fully get it. Anyone care to try to give a definition that’s really dumbed down? What are their unique abilities that Shaders aren’t capable of?
On a technical level, is it possible to code a scene that accepts in more than one texture? In the example Scenes, you can have an incoming texture on a rotating cube - would it be technically possible to have a scene which accepted multiple textures so that each face of the rotating cube had a different image/texture on it?
Comments
@giku, something I’ve noticed which would be enormously useful for Visual would be for users to be able to easily reorder the layers of a mixer without having to remove all channels and start again so that the correct images are on top of each other. This could be by long pressing and sliding the channels to move them or via ‘Move up/Move down’ text options in their double tap menus.
Also, a one tap way to select all code could be worth considering. When working with AI generated code and the constant back and forth that requires, a quick way to select all four deletion would really help workflow.
Hopefully this is just an oversight, but why is there a 4:3 aspect ratio option for the projector but no 4:3 ratio option for export? You set things up in 4:3 and then can’t export in that format?
I use duble-tap-hold on the blank space to the right of a line to select all. Works for me
+1
Do you mean the options that pop up? I don’t get a Select all option - no idea why not?
I’ve been experimenting with how possible it is to apply predictable rhythmic motion to hand drawn images using Drambo Visual. Here’s an experiment from yesterday.
It’s frustrating, you have to double tap quickly and hold the second tap until you get the option “select”. Tapping that will open a menu with select all
don’t tap on text but on blank space to the right of text
I’m on ios 26 btw
Okay got it now, I wasn’t holding the second tap. Works, thanks.
@Robin love it! Care to share your process? Did you break the cat into layers, or something else?
@Robin @pedro
Long press on an empty space in the Scene Code module text input area brings up the Paste/Select/Select All/AutoFill menu.
That's on an Air 3 iOS 17.6.1 and an Air 4 iOS 26.4.2.
You’re not the 1st person to tell me that but I can’t get that to work, only double tap hold. Unless aI’m supposed to hold for 5min or something
wait: it works… wtf I could swear it never worked all the times I tried
iOS is like that. ;)
I'm here, gathering the issues
Many thanks @bcrichards.
The process I’ve been attempting to use is quite simple really. After feeding it the rules from the Drambo Docs, I asked Gemini to write me a processor shader of a dancing circle shape with various parameters.
I then applied this processor several times to different parts of the cat image (just one image, not broken up or layered), with a different gate/velocity sequencer or LFO driving each one. Mixer for background and drum on different layers.
The important parameter the processor shader needs is blur/falloff so that the dancing circle isn’t just a hard edged distortion on the incoming image texture. I don’t know whether there’s a more elegant way of doing this, it is the main flaw in the system (the distortions even when blurred are like lens distortions) but I know nothing about the coding intricacies and possibilities.
I’ve been experimenting a bit today with making an adjustable dancing line instead of a circle which is probably a better, more flexible approach ultimately?
Essentially, I’m trying to rig the image texture using a processor shader skeleton underneath.
If this concept/approach seems interesting or viable to @giku, it’d be great if some processor shaders could be included specifically for this purpose? Cheers.
@Robin How do you see it? Which params, how should it move etc... you would probably end up in dozen of control and behaviour dimensions.
Thats why it's better to divide it into smaller modular parts: modulation texture generator -> transformer -> target mask transform. and that's how it works now :)
You can start from a basic Spectrum analyzer in angular mode to modulate radius of glowing circle.. so we have a basic "dancing circle" responding to audio. Then you can modify it's behaviour, appearance and nuances by other modules in the chain.
@giku Yeah, the ‘problem’ as I see it is that that very quickly sounds very technical. No doubt perfect for the people who know about such things but I’m guessing people who come from a non modular animation background are likely to think of hacking it in the way I’m describing - in terms of rigging.
Having said that, I’ve only recently bought Visual less than a week ago so you somewhat have the advantage over me! I’ve not even attempted to build a shader from scratch using a chain of modules so maybe it is all possible and really easy to do? I’ll certainly try.
To play devil’s advocate though, if the current tools are the best to do it with, why haven’t I seen any examples of other people trying to do what I’m experimenting with? Either it’s because what I’m trying to do is shit and no one else wants to do it (which in fairness is possible!) or the tools don’t naturally lend themselves to it? I mean, how is a newcomer supposed to realize that the Spectrum analyzer can be used to create a dancing circle? Isn’t it more helpful just to provide them a dancing circle or line?
You suggest that providing such a processor shader in app would result in it having lots of parameters. Does this mean you fear users would see it as too complicated? Again, I’d suggest a lot of users will see the prospect of chaining together lots of modules as being a lot more complicated than a ready made processor with many parameters. The one I asked Gemini to code has eight exposed parameters.
I was only suggesting it might be useful so If it doesn’t appeal, that’s cool. The great thing is that Visual doesn’t stop you from doing it anyway! Cheers.
By the way, small possible bug to report: I keep finding the system qwerty keyboard (in floating mode) keeps getting stuck on screen (can’t be dismissed) when in Code mode for the processor shader.
@Robin I'm happy to see you diving into the vast capabilities of DV!
What @rs2000 was suggesting, for now, is to use the full rack view to position your projector on the top of the screen, like this:
@Robin keyboard goes away if you close the shader/disable code editor
Speaking of the code editor, there is no way to maximize it or make it wider, right? I have some long lines with multiple indentation, sometimes it becomes harder to follow the logic when the line gets broken.
Would it be good if the editor could be made floating like the XY pad?
Are people mostly coding outside of Drambo and doing the select all copy-paste routine all the time? Or perhaps using a Mac with Drambo when coding.
I code outside, using notepad++ and then copy the results. Minor edits I do occasionally directly on ipad or iphone but big edits would be painful. The font is too small, and even with a BT keyboard you can’t find words, find and replace, etc. Even a simple editor like notepad++ can highlight all instances of a keyword, which is very convenient
True, that is. For work I use VS Code, but Notepad++ is also good.
Benefit of doing it on iPad is easier transition to Drambo (I don’t have D on a pc). I’ll try using GitHub Codespaces, it gives you a VS Code inside a Safari tab on iPad. Hopefully with syntax coloring and highlighting all instances etc.
@pedro , are you using Notepadd++ on a PC? If so, how are you copy/pasting?
Ah, okay, thanks @pedro. Since I’ve only been pasting in code from Gemini rather than actually writing it, I just thought it was there being awkward!
Thanks @Intrepolicious.
I just use notes in icloud.com to copy back and forth to/from my devices
Btw still on the topic of select all, no, merely press hold doesn’t work in many places, including notes (where it brings up the magnifier). Double-tap hold works everywhere
Thanks @pedro - I'll stick with Textastic on the Mac, but your post inspired me to get Notepad++ up and running on my Mac with Wine. It's good to see my old Windows days friend Notepad++ again. 😎
Just in case you weren't aware (I only noticed it the other day) - there's a shared clipboard between iOS devices and Macs, I'm not sure if they have to be on the same network + signed into the same Apple Account or whatever, but it's definitely working for me and is a huge timesaver.
Trying to get my head around Scenes. I’ve read and reread the pertinent sections in the Docs but none of the descriptions have quite explained it in plain enough language for me to fully get it. Anyone care to try to give a definition that’s really dumbed down? What are their unique abilities that Shaders aren’t capable of?
On a technical level, is it possible to code a scene that accepts in more than one texture? In the example Scenes, you can have an incoming texture on a rotating cube - would it be technically possible to have a scene which accepted multiple textures so that each face of the rotating cube had a different image/texture on it?