Skip to content
This repository has been archived by the owner on Apr 13, 2022. It is now read-only.

Recommended procedure for audio recording? #14

Open
Breakthrough opened this issue Nov 1, 2017 · 1 comment
Open

Recommended procedure for audio recording? #14

Breakthrough opened this issue Nov 1, 2017 · 1 comment

Comments

@Breakthrough
Copy link

Breakthrough commented Nov 1, 2017

Hello;

Great project, seems to be working just fine for me thus far. Loving the ability to pick the shutter angle, really makes the footage more pleasant to watch back with the appropriate shutter angle. I also hope to test it using a Ramdisk via symlink instead of just recording the TGAs to my SSD and then back to RAM (will report if I find any useful performance gains).

That being said, I was just wondering, how do you deal with matching up the output audio file to the video? Is there a recommended procedure?

For some reason, I cannot seem to make the audio stream match up with the video stream.

Thanks,
Breakthrough

Edit: Have tried using a Ramdisk, no notable performance improvements surprisingly. However, it would be wise to use one if you only have an SSD, as this would avoid unnecessary writes to the drive, prolonging the drive's longevity.

@AronParker
Copy link
Owner

Hey,

Thanks for testing! I actually did these tests before I had the same results. If you're interested in the details on what I found out is essentially the Windows IO subsystem is really intelligent with caching, unless it needs to (implying that it's not forced to) doesn't flush out data to disk. You can observe these effects by opening a raw video file, the first time it'll be slow and stutter but once you play it again it's smooth even at insane rates like 360 MB/s, this is because Windows keeps the data in memory (as you can observe in the Task Manager (Standby & cached) or using tools like Process Monitor (FASTIO calls are cached, you have to enable them though as by default they're filtered). The same mechanic takes place when writing the frames.

I trust the user on setting the correct framerate, whether the game lags behind writing the audio or is ahead of it does not matter, the tool always interleaves the audio properly. Now the problem is that the Source engine doesn't always interleave properly, I've been observing these effects for years (I started making videos and recording tools in 2008+). Sadly there was never a fix for this, if there was one it would have to come from valve. I had some examples where the audio produced didn't match up with the video, but that wasn't due to avi recorder, I redid it again using VirtualDub this time and sure enough it still didn't line up. Yes it does happen and no it's not the tools fault.

You can convince yourself by keeping the TGA frames and producing another movie with a different tool like VirtualDub to remake the same movie, the effects should be equal, if they're not let me know.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants