Compression questions

[ishal]

Hello there all, I am new here and do not know much about video editing and capturing so do not get to upset if I make novice mistakes.

Well I should start out by saying I like to capture footage of me and my buddies playing video games and edit them into HD quality using Sony Vegas Pro 8.0. I have found a way to get extremal good quality for my videos by rendering them into 720p. However when I do this, my file size becomes massive (around 2 gigs for 10 minutes of footage) and was wondering about ways to compress the file size.

I have done some research about compressing video files and have learned that compressing the video sacrifices the quality. I am really unwilling to do that, and was wondering if there was a program for PC (Windows Vista 64 bit to be exact) that could compress the size of the file without losing quality.

If not then is there another way to get 720p footage but with a smaller file size?

If I made any mistakes or broke any rules in my first post, I am sorry and it is not intentional.

Edit: By the way my rendering settings are as follows.

Windows Media Video 9 (.wmv)
Audio: 192 kbps, 48,000Hz, 16 bit stereo, WMA.9,
Video: 29.97 FPS, 1280x720, WMV V9 CBR Compression, Smoothness 90.

I also capture my footage using a Dazzle Capture card, and my capture settings are I believe are MPEG2 capture at 16:9.

[guns1inger]

Compression, file size and quality are a constant trade off. There is not way around it. You want smaller then you have to give somewhere. You can take a quality hit, or take a resolution hit.

My first question is : what resolution are you capturing at ? If it is SD, why bother upscaling to 720p ?

My next question is : have you tried experimenting with the settings under Custom and reducing the bitrate in increments to see what impact it has on quality ?

My final question is : what file size do you anticipate you need to get down to ? This will tell us if your expectations are in any way realistic.

[ishal]

Compression, file size and quality are a constant trade off. There is not way around it. You want smaller then you have to give somewhere. You can take a quality hit, or take a resolution hit.

My first question is : what resolution are you capturing at ? If it is SD, why bother upscaling to 720p ?

My next question is : have you tried experimenting with the settings under Custom and reducing the bitrate in increments to see what impact it has on quality ?

My final question is : what file size do you anticipate you need to get down to ? This will tell us if your expectations are in any way realistic.

Oh wow, my first response was very helpful. Thank you very much for your help. I am recording at standard definition, is there a certain way to check it? The reason is simple, I just rather look at something that looks really good and really clear picture. The only way I see making the picture even nicer looking is improving it into 720p.

For your second question, no I have not and I have no idea what the does to quality and file size, I will look that up now.

My expectations are really, can I get a 2 gig video file into a 1 gig file while keeping good quality. If so is there a recommended program.

Thank you so much, you are really helpful.

[guns1inger]

The fundamental rule for digital video compression, no matter what resolution, what codec and what platform, is universal

File Size = Bitrate X Running Time

This beauty of this rule is that you can use any bitrate calculator to work out what you need. I like Vcalc myself, but there are online calculators as well. Most were designed for DVD work, but if they are flexible enough you can use them for anything.

You can check the resolution using mediainfo. 720 x 480/576 is SD.

If you are capturing at SD you have a couple of choices. The simplest way to keep quality and keep the filesize down is to encode at SD resolution. This puts the onus on the player to upscale as necessary. If you upscale to 720p in your application then you are dictating the quality of the upscale. This may be acceptable to you, or it may not. If you are using standard resize (which is all the Vegas offers) then you may not be getting as good a quality output as you might elsewhere.

Again, if you keep the resolution as SD (assuming that is the resolution you captured at), you will be able to create much smaller files while maintaining quality.

[ishal]

The fundamental rule for digital video compression, no matter what resolution, what codec and what platform, is universal

File Size = Bitrate X Running Time

This beauty of this rule is that you can use any bitrate calculator to work out what you need. I like Vcalc myself, but there are online calculators as well. Most were designed for DVD work, but if they are flexible enough you can use them for anything.

You can check the resolution using mediainfo. 720 x 480/576 is SD.

If you are capturing at SD you have a couple of choices. The simplest way to keep quality and keep the filesize down is to encode at SD resolution. This puts the onus on the player to upscale as necessary. If you upscale to 720p in your application then you are dictating the quality of the upscale. This may be acceptable to you, or it may not. If you are using standard resize (which is all the Vegas offers) then you may not be getting as good a quality output as you might elsewhere.

Again, if you keep the resolution as SD (assuming that is the resolution you captured at), you will be able to create much smaller files while maintaining quality.

So by lowering my bit rate, I lower my file size. Thats really interesting, but would it lower the quality as well?

[guns1inger]

Yes. That's the trade off, and the balancing act. There is no way around it.

What you are looking to find is the point where you get the best compromise between the three. Initially the reduction in quality will be small, and may not even be visible to the naked eye. The whole point of digital compression is to hide the damage as much as possible. However eventually you will reach a point where this is no longer possible, and the damage will be visible. Exactly when you reach that point varies from video to video and compression to compression. You can only find it through experimentation.