Newbie questions

vvvv beta general
1

Hi everyone,

Very new here to vvvv but excited about finding a new tool. I’ve got some experience with Jitter and Gem so not totally new to graphic type programs. I’m having some trouble wrapping my head around vvvv though and it’s very intriguing.

In Jitter I’ve built a vj type patch for experimental music that takes in 3 live feeds from firewire cams, and mixes them into a three panel projection. Each panel has a selection of composite settings and 3 or 4 shader fx (thanks to vade for those) that can all be triggered by various audio input settings or random parameters using a matrix (not a jitter matrix, but an input/output matrix). I can also mix in two video files on top of all of this.

In anycase, that’s where I come from to vvvv experience-wise and I’m wanting to know if this is possible with vvvv. Not that I want to build the whole thing again, but I’m just trying to see what the strength is of vvvv as opposed to jitter. I also thought that trying to build something I’m already familiar with might be a good way to learn a new program.

I’m WELL impressed with some of the examples I’ve seen but the documentation is obviously still in development and I can’t imagine how to get to a more advanced level based on the beginning tutorials. Funny thing about max/msp/jitter is that so much of the value of that package is actually the tutorials that come with the program itself (this isn’t a gripe at all, just a reflection on the struggle that often faces non-commercial software.)

Am I right in understanding that vvvv seems more involved in the ‘generative’ aspect of creating visuals, algorithmic etc? And some of what I’ve read has said that vvvv had audio concerns built into it from the beginning, but I’m not sure how I understand that? Also, I know that this has been discussed before both on the max/msp list and here, but can someone who’s used both jitter and vvvv discuss their differences a bit more in detail in terms of what each does well. I’m sure to an extent that you could accomplish very similar things with both, but I’m also sure that some things are easier in vvvv (spreads) and others in Jitter etc, but I’d like to hear from someone with more experience as to what those might be. I’m still not sure what the significance of spreads is, and I’m feeling like there’s likely a fundamental conceptual difference between jitter and vvvv that I’m not getting yet. Also, so much of my jitter programming involved designing a useful gui, I’d be interested if anyone has experience with the various gui soloutions on the site)?

Thanks again and sorry for the long post. I’d like to explore vvvv more and if anyone has any other tutorials or workshop materials that you wouldn’t mind sharing I’d greatly appreciate it.

looking forward to digging in more,

Bill

2

no time right now to explain, but the graph calculation is different in vvvv than max/pd. vvvv is designed for video processing, it ‘thinks’ in frames instead of message forwarding and the graph is a bit more strict (only one connection to inputs, order of nodes is not important), which simplifies some things. please look for the other discussions about that with the search function…

3

I moved from jitter to vvvv around 1 year ago.
The main differences i found were:

  1. vvvv is much more efficient (can deal with lots more data)
  2. vvvv is more elegant (data can be spreaded, you’ll learn about that as time goes on)
  3. vvvv is easier to use (the interface is much better designed)

The two advantages that I can see to using jitter are:

  1. You can tie in visuals with a max/msp composition without needing to osc variables around
  2. jitter has a JS node, which allows you to write your own nodes very easily using javascript (which is heaven sent when you need to make things like gamestate engines or anything you cant get your head around how to write in nodes).

spreads allow many layers of data to be travelling around your patch simultaneously. If you go through the beginner tutorials you should get a grasp of how spreads work. Most nodes are optimised to use spreads, i.e. 1 node accepting 3 textures will be more efficient than 3 nodes each accepting 1 texture, and as time goes on you’ll find spreads more natural to work with than unnecessarily making redundancies in your patch.

For your purposes, you’ll need 3xVidoIn nodes at the top of your patch, feeding data in from your cameras, then pass those into VideoTexture nodes which will convert the videos into textures.

You’ll then need a set of shaders, each accepting the videotexture as an input. these shaders only work on 3d objects, so you’ll need to attach an object to each of them. to do this add a “Grid Ex9” and attach it to the mesh pin on each of the shaders.

The “Group” node allows you to send more than one layer of graphics to a single renderer, so make a group node, set it to have 3 inputs and then connect your shaders to this node.

I’m presuming that you’re using a TripleHead2Go, so you’re best off with 1 single render which spans all 3 projectors. So put a renderer at the bottom of your patch and set it to fullscreen on your triplehead. then set the fullscreen resolution to 3072x768.

Attach your Group node to your renderer. you should be getting some graphics out if you’ve played along the way.
Add “transform 2D” nodes to each of your shaders, this will enable you to move and scale them to the correct position on your output.

If you make one transform 2D node and attach it to all your shaders, but have a spread of settings on its input pins (one set of setting for each of your 3 output screens), then you can send a spread of 3 variables to any of each shaders input pins, and the 3 variables will affect the 3 screens independantly. the simplest way to fade the seperate shaders on each output is to send a spread of 3 values to the V pin on a HSV node (one HSV for each shader, so you’ve got seperate settings for each)

I think i’ve talked too much and too messily here. let me know how you get on

4

a couple of other things good about jitter are:

  1. you can copy and paste patches directly in and out of forums without needing to attach any files or use a text editor.
  2. The user interface nodes often work better than vvvv (at the moment!). currently the IOBox’s are a bit lacking in terms of performance and options compared to jitter, and using dx9 controls is a bit overkill. It’s highly recommended to use something like a midi controller when using vvvv if you want to edit variables in realitme.
5

Hi everyone,
thank you for all your responses. I’m responding from my phone at the moment but looking forward to when i get back to my machine to try these suggestions out,
bill

Hi everyone,

Very new here to vvvv but excited about finding a new tool. I’ve got some experience with Jitter and Gem so not totally new to graphic type programs. I’m having some trouble wrapping my head around vvvv though and it’s very intriguing.

In Jitter I’ve built a vj type patch for experimental music that takes in 3 live feeds from firewire cams, and mixes them into a three panel projection. Each panel has a selection of composite settings and 3 or 4 shader fx (thanks to vade for those) that can all be triggered by various audio input settings or random parameters using a matrix (not a jitter matrix, but an input/output matrix). I can also mix in two video files on top of all of this.

In anycase, that’s where I come from to vvvv experience-wise and I’m wanting to know if this is possible with vvvv. Not that I want to build the whole thing again, but I’m just trying to see what the strength is of vvvv as opposed to jitter. I also thought that trying to build something I’m already familiar with might be a good way to learn a new program.

I’m WELL impressed with some of the examples I’ve seen but the documentation is obviously still in development and I can’t imagine how to get to a more advanced level based on the beginning tutorials. Funny thing about max/msp/jitter is that so much of the value of that package is actually the tutorials that come with the program itself (this isn’t a gripe at all, just a reflection on the struggle that often faces non-commercial software.)

Am I right in understanding that vvvv seems more involved in the ‘generative’ aspect of creating visuals, algorithmic etc? And some of what I’ve read has said that vvvv had audio concerns built into it from the beginning, but I’m not sure how I understand that? Also, I know that this has been discussed before both on the max/msp list and here, but can someone who’s used both jitter and vvvv discuss their differences a bit more in detail in terms of what each does well. I’m sure to an extent that you could accomplish very similar things with both, but I’m also sure that some things are easier in vvvv (spreads) and others in Jitter etc, but I’d like to hear from someone with more experience as to what those might be. I’m still not sure what the significance of spreads is, and I’m feeling like there’s likely a fundamental conceptual difference between jitter and vvvv that I’m not getting yet. Also, so much of my jitter programming involved designing a useful gui, I’d be interested if anyone has experience with the various gui soloutions on the site)?

Thanks again and sorry for the long post. I’d like to explore vvvv more and if anyone has any other tutorials or workshop materials that you wouldn’t mind sharing I’d greatly appreciate it.

looking forward to digging in more,

Bill