View Full Version : Rendering an Image Sequence - What format is best?

04-04-2007, 07:42 AM
Hi everyone, i'm new here and i hope some of you guys can help me out with this.

I'm doing a project now for uni that's eventually going to put onto DV PAL. At the moment i'm rendering at 768 x 576 (sq. pixels), no fields. I've been trying to render .mov files at 25fps but no matter what i do in lightwave, my .mov files always end up as 30fps no matter what.

I read through the forums here and came across another thread where someone had the same problem, and they were told to render an image sequence instead and compile the image sequence in AfterEffects/Quicktime later.

So my question is, what format should i be saving as? I'm not doing .jpg because i plan to do some post-production later in After Effects, and the image will deteriorate the more i play around in AE. So i was thinking i might try saving as .pct files or .tiff files, but i'm not sure. I figured you guys would know best.

What format should i be saving as? Should i save as 24-bit or 32-bit?

And taking into consideration that i'm dropping this to dv pal, could there be any problems between 24-bit and 32-bit?


04-04-2007, 05:25 PM
The main difference between the 24 bit and the 32 bit is an embedded 8 bit alpha channel. This you will only need for compositing and you would be setting up the scene specifically for that anyway. So 24 bit is fine unless you are compositing. But it depends, you can also save the alpha separate depending on what program you are using to composite with.

Just make sure and save as the standard for pal. That is, the image size, the FPS and the pixel aspect ratio. I would leave fields to none. You can always go interlace if you need to after the fact in your NLE, but this gives you progressive all the way should that be your deciding output.

To get the standard for pal just open up any NLE and set the project to PAL then look at the properties and use that as your guide.

04-04-2007, 07:44 PM
Personally I render sequences to .TGA if I'm using them straight, or I render to .HDR or .RPF if I'm going to process it. Rendering to a HDR image format isn't necessary but I think it's really nice to have the extra bit depth when your processing.

And depending on what kind of compositing you're doing you might find the extra buffers that RPF supports handy too (surface ID, Zbuff, etc)