I’m working on my first project with vvvv, in just over a week and a half we go onsite. The staging company has been kind enough to loan me the computer that will run the patches so I can tweak and finalize them before we get there – Windows7 64-bit, 12GB RAM, 3.47GHz i7 CPU, Radeon HD 6900 series. I’ve used the Eyefinity to created a 4x1 series of monitors, final size is expected to be 4725x1050.
Since I’m a newbie, I’m sure that this question has been answered before, but I can’t find it on the site.
The project requires a spread of photos (that will be taken morning of show) to appear at a certain point over a video that will run full screen behind it. So I’ve got a Translate Z to zoom the spread up to fill the frame/fly through them. I also have a fade on them because otherwise, they show up as the video is playing. The problem is that if I use D24X8 on the renderer, the images are correctly placed (all others, there is weird overlapping where images in front appear much larger behind smaller ones that should be behind them) but, with the fader that I jury rigged up, it causes the quads to flicker when they come in (doesn’t do that on the other settings, smooth transition). I’ve probably done it wrong or there is a better way to do it that I don’t know about.
Except for the flickering, it works pretty much as I want it to. I am probably going to have to rework it somewhat this week because now the client wants the photos to move in very fast (they haven’t seen anything yet) and I don’t think what I’ve got will be fast enough for them, so I think I’m going to have to learn about DeNiro tomorrow.
If someone would be so kind as to take a look at what I’ve done and give me some pointers where I’ve gone wrong, I would be very grateful. I’ve been picking away at this for almost a month now in between my other projects and don’t know what else to try.
People Mosaic5 with video.v4p (46.8 kB)
It’s just a patch I found on the site that sets the renderer size and position – very useful. Since all modules will be the same size, made sense to do it this way.
Another problem is that when I try to make the renderer full size, it knocks out the Eyefinity group and compresses everything onto one monitor. Not sure if that’s normal or not.
TD_screen size.v4p (23.7 kB)
Set the near plane of you camera to 10 and see how that looks
I tried that and it does fix the flickering. Even setting it as low as .01 works. When you’re working in XYZ, do you have to set the near plane to something? Or is that general “best practice”?
But, I’m going to have to redo it. The photos snap on and given the language the client has been using, I have a feeling that’s not what they want to see.
Generally, yeah I find that there has to be a little bit of near clipping to avoid the flickering unless a depth buffer isn’t an issue.
The smooth route for avoiding the pop is some distance-to-alpha patching, but the lazier (easier…probably) route is to use a fog render state the same colour as your background. Linear mode is the quickest way to set it up.
“The smooth route for avoiding the pop is some distance-to-alpha patching”
I’ve spent the better part of the day trying to figure this out, and failing. Is there somewhere I could see an example to get an idea how it works?
Sorry, I don’t mean to be a pain, I’m just finding 4V on the frustrating side. If I could do this in AfterEffects, things like this wouldn’t be too much of a problem. 4V is a completely different way of thinking things through.
I see what you mean. Hadn’t noticed how tight your camera was, so the quads are covering a big distance in very little time. Set the far plane of you camera to 1200 or summink and you should be alright.
Actually, depth buffer wasn’t involved the second time
Try applying this and reducing the the deviation between the ohotos so they emereg from the dark at similar times.
just some general remarks:
- from looking at your patch you’re using unusual value scales. vvvv typically operates in a space of -1 to 1. you seem to be a pixel guy and like to use high values. while this is possible in vvvv this typically just complicates matters.
- to draw a fullscreen background video you’d typically use WithinProjection (Transform) on the quad to make it stay there no matter what and then use a ZWriteEnable (DX9.RenderState) with its ZWrite Enable set to false to make it not interfere with the scenes depth.
- also in your case you probably need no depth/camera at all just rely on the fact that things get drawn one after the other…(but i might be missing somethign from your description)
- when using the Camera (Transform Softimage) you’d typically not mess with Near Plane and Far Plane. only in rare situations when you know what you’re doing you should need to change those. same goes for the FOV. a fov of 0 like in your patch cannot be recommended.
Thanks for the pointers Joreg
@joreg – No, I’m not a high pixel guy, just a newb who doesn’t know any better! The screen size was given to me by the technician at the staging company 4725 x 1050, a massive pan 20’ x 90’.
But I thank you for your tips and I will look into applying them. Since this program is kind of a play-as-you-go learning structure, I don’t really know what the best practices or other usual stuff is yet.
And thanks too Hadasi, I’ll try out that fog and see how it works for me.