comparing p2p to scene high def releases

[ipa]

I've been downloading hd from usenet over a year now and found that a lot of good releases come from various p2p torrent sites. Assuming it was always the case, I sometimes wait for the p2p release, or replace my scene release when it arrives.

But when I look at the nfos, I notice that a lot of releases are almost the same exact file size and bitrate. Sure, plenty are monsters, with 20GB files for some 1080p movies, but especially 720p movies tend to track the scene releases. My limited understanding of encoding leads me to believe that there are basic settings to get the most transparent encode to source.

So what is it that makes them better quality? How much of a factor is bitrate? How much depends on the source quality? I realize these are basically encoding questions, but I hoped the most active section oriented towards torrents, where these files originate, might have some folks who could enlighten me.

[TONiC]

Firstly, The Scene have rules/standards which Scene groups must abide by. However, many P2P groups, abide by these rules - including the packaging ones about RAR compression, even though it's not necessary! The reason why many P2P groups stick to these rules (even though they don't have to) is that these rules give a decent length/size ratio for video files, be it XViD releases or X264. It also ensures that, at least for XViD, releases are compatible on all XViD capable hardware.

BitRate is a high factor when encoding, but this needs to be combined with resolution and movie/tv show length considerations to ensure a good level of quality. All these factors affect the Bits/(frame X pixel) measurement <- the true sign of quality.

[ipa]

Could you explain more about the Bits/frame x pixel measurement please?

[ca_aok]

There are several factors that determine the quality of an encode. Bitrate does not tell all, especially since it's a VBR algorithm.

To name a few standard x264 parameters: Various filters, SAR, deblocking, quantization methods, motion smoothing, ref frames, bframes, etc.

The type and quality of audio used also has a large effect on the filesize.

The reason P2P encodes are generally better is because they can take all the time they need to polish settings, check interlacing, retry several encoding parameters, and can optimize for quality rather than speed. Additionally, they don't need to conform to scene filesize regulations that limit video quality. P2P encoders can also wait for a better source to come along, as quality is also source dependent.

[ipa]

Thanks for the explanation. I guess only some groups provide the information about those settings, like Framestor's media-info file. I guess I could always pull that info after downloading, but it would be great if more groups provided all the necessary facts.

I'm just glad that I was correct in "seeing" the difference between releases even when the limited nfo stats seemed comparable.

Thanks for the info.