Sorry this might be a "technically" silly question, but I was just thinking, since exporting in an uncompressed AVI format means you already have 100% quality (at least before you run it through another application), would Super Sampling do anything? It'd be a waste right? I assume it's for situations where you're exporting in a compressed format, but still want high quality.

--------------------------------------------------------------------------------------

My iClone short films: https://www.youtube.com/iClone316

That's a good question, I'd love to know the answer to that too, as super sampling and high shadowing crash the hell outta my poor ol  computer.

  

Uncompress video and supersampling are two different things. You can have uncompressed video with "preview"quality and you can have "mp4"compressed video with 3x3 supersampling with High quality shadow.
You have "render"quality , the best is 3x3 supersampling with High quality shadow. And you have "compression"" video quality wich the best is uncompressed.
But generating high quality render with uncompress video codec needs tremendous hardware ressources. I suggest to render in sequence images with high setting , then make an uncompressed video in an another app.
Or you can edit you image sequence in an editor app. The benefit of the sequence image is that you don't loose the tasks even if your pc chrash. You restart the render at the frame when your pc chrashed.

Thanks for the response, I have one more question.

Based on what you said, which is better quality: an uncompressed AVI file (on highest settings), or a "normal" MP4 file with 3x3 super sampling?

Sorry if your post already implied the answer, I'm a noob at this lol.


--------------------------------------------------------------------------------------

My iClone short films: https://www.youtube.com/iClone316

which is better quality: an uncompressed AVI file (on highest settings), or a "normal" MP4 file with 3x3 super sampling?

Good question!!   The answer is : "Try it" !! 
Because I d'ont know how far is the mp4 compression used by Reallusion (Data rate, multipass, B-frame,etc...) , I can't tell you. Make the test on 2 second video and see the result.
For my case , I use mp4 only on the final product for sharing, never to edit
or for post-effect.

Besides quality considerations, there is the difference between editing and delivery formats, as explained here: http://telestreamblog.telestream.net/2012/04/save-yourself-frustration-use-editing-formats-when-editing-2/.

Editing works best if individual frames are directly available with best quality. In the case of compressed formats, the editor needs to reconstruct the frame by applying the relevant decoder (the same that needs to happen when viewing the video). This is an extra step. That is why some of us prefer to work with image sequences. Using PNG you have lossless compression, so the sum total of the files for all frames is less than that of an uncompressed AVI file.

I've found that the PNG files from iClone are not the smallest possible. If you load your sequence in Virtual Dub and then save it out as sequence again, the files will be smaller (this was true the last time I tried this, which is a while ago).

The advantage is that you now have pristine quality after which you can choose any delivery format you want: MP4 for YouTube, Vimeo, and other online outlets, MPEG-2 for Blu-ray, etc.

The disadvantage is that you need more disk space, but you have worked hard on your movie, so I think it deserves it.

I hope this helps. Time for me to make breakfast...

---

There is no choice to make between compressed/uncompressed video and multisampling.
Whether you use compressed, uncompressed video or images sequence, you should always render with the highest multi-sampling setting available, here 3x3, or else your images will suffer from aliasing whatever the image or video format you use, compressed or not.

--
guy rabiller | GOETZIWOOD STUDIOS
"N.O.E." (Nations Of Earth) Sci-Fi TV Show, Showrunner.

I appreciate everyone's reply. I will keep all these tips in mind next time. 

Unfortunately I didn't have that option this time (apparently) as it was a 4 minute long VR video, with about 40 or so lights, and about 1.4 million polygons.

Super sampling even on 2x2 crashed halfway though the render. Video is in the link. 


 

--------------------------------------------------------------------------------------

My iClone short films: https://www.youtube.com/iClone316

Your option is to render images sequence *then* create the movie file. If your rendering crashes half-way then you can resume at that point.

Aliased images are of poor quality, they fill flicker and will look unfinished to your audience, whether it is for game, movie, tv or VR.
You should always seek for properly anti-aliased images, whatever you do.

Sorry to be a bit 'hard' on this, but it is a shame to sometimes watch videos with great ideas, assets, textures, camera work, performances, narration, etc.. but unpleasant to watch because ruined by devastating aliasing giving images that flicker everywhere.

--
guy rabiller | GOETZIWOOD STUDIOS
"N.O.E." (Nations Of Earth) Sci-Fi TV Show, Showrunner.

On average, does rendering as a set of images take longer than a video? 

Also, how do I sequence them as a movie after? Windows Movie maker? I assume it will still appear exactly as a video to the audience right? 

I didn't realize that it would retain the VR properties if I turn it in to an image. 

--------------------------------------------------------------------------------------

My iClone short films: https://www.youtube.com/iClone316