Constan Quality Vs. Avg. Bitrate

HandBrake for Windows support
Forum rules
An Activity Log is required for support requests. Please read How-to get an activity log? for details on how and why this should be provided.
Post Reply
jdm
Posts: 21
Joined: Sat Jul 25, 2009 6:24 pm

Constan Quality Vs. Avg. Bitrate

Post by jdm »

I converted all of my DVDs into MKVs using the film preset, and setting the Average bitrate to 2000 Kbps. They all look great.

I decided to do the same for the handful of Bluray and HDDVD discs I have. I read the now defunct "Post you best settings" thread, and noticed a lot of people were using Constant Quality instead.
I gave it a shot, setting the slider to 61%. The results I got looked good, but the file sizes stayed pretty big. 61% quality didn't really result in a file 61% the size of the original. I tired sing average bitrate (2000 and 4000) and the results were tiny files compared to the originals, and the quality wasn't bad, but lacked the sharpness of the of the originals and even the 61% files.

What have other people had better luck with? Lowering the % in constant quality, or using a higher average bitrate? I noticed that if I use mediainfo on the original m2ts files, the average bitrates are between 13Mbps and 25. So maybe I should be aiming for files half the size, by setting an average bitrate half the original?
nightstrm
Veteran User
Posts: 1887
Joined: Fri Mar 23, 2007 5:43 am

Re: Constan Quality Vs. Avg. Bitrate

Post by nightstrm »

For HD sources, I use something between 57-59% and have been happy with the results. Output files are typically 3-4GB each (encoded for AppleTV).
jbrjake
Veteran User
Posts: 4805
Joined: Wed Dec 13, 2006 1:38 am

Re: Constan Quality Vs. Avg. Bitrate

Post by jbrjake »

jdm wrote:The results I got looked good, but the file sizes stayed pretty big. 61% quality didn't really result in a file 61% the size of the original.
....nor should it. Just like you read in the documentation.
jdm
Posts: 21
Joined: Sat Jul 25, 2009 6:24 pm

Re: Constan Quality Vs. Avg. Bitrate

Post by jdm »

jbrjake wrote:....nor should it. Just like you read in the documentation.
True, and I wasn't really expecting a 1 for 1 reduction in size based on the quality set. I was expecting more than I got however. As the 61% setting was getting me a less than 20% reduction in size. nightstrm mentioned his files were coming out between 3 ad 4GB on 57-59%, whereas my smallest so far was 13GB at 61%. Which made me think I was doing something wrong.
User avatar
s55
HandBrake Team
Posts: 10357
Joined: Sun Dec 24, 2006 1:05 pm

Re: Constan Quality Vs. Avg. Bitrate

Post by s55 »

It's a log scale, thus, that is normal.
nightstrm
Veteran User
Posts: 1887
Joined: Fri Mar 23, 2007 5:43 am

Re: Constan Quality Vs. Avg. Bitrate

Post by nightstrm »

jdm wrote:
jbrjake wrote:....nor should it. Just like you read in the documentation.
True, and I wasn't really expecting a 1 for 1 reduction in size based on the quality set. I was expecting more than I got however. As the 61% setting was getting me a less than 20% reduction in size. nightstrm mentioned his files were coming out between 3 ad 4GB on 57-59%, whereas my smallest so far was 13GB at 61%. Which made me think I was doing something wrong.
I'm also scaling down to 720p to ensure AppleTV compatibility.
jdm
Posts: 21
Joined: Sat Jul 25, 2009 6:24 pm

Re: Constan Quality Vs. Avg. Bitrate

Post by jdm »

Ah. That may be the biggest difference. I'm keeping them at 1080p.
Deleted User 11865

Re: Constan Quality Vs. Avg. Bitrate

Post by Deleted User 11865 »

If you're not downscaling, you may be able to lower the CQ setting even more (as in slightly less than 57%). The idea is to test several percentages and decide which is the best compromise for you, in terms of quality vs. file size.
jdm
Posts: 21
Joined: Sat Jul 25, 2009 6:24 pm

Re: Constan Quality Vs. Avg. Bitrate

Post by jdm »

Thanks, I'll give that a try. I'm also going to try some by setting a average bitrate as a % of the original and see what happens.

Just didn't want to be heading in the completely wrong direction, since it takes about a day each time I encode from an HD source.
Post Reply