?kbps = dvd quality

[paddyspub]

If I'm encoding the main title of a dvd at a constant bitrate in either Theora or x264 then what should I set the kbps at to ensure a dvd quality picture? As a comparison, when converting a cd track to mp3 or Vorbis it's generally acknowledged that 192kbps = cd quality. So what would the equivalent be for dvd quality encoding?

[guns1inger]

I don't consider 192 kbps MP3 as CD quality. It is an arbitrary measure. So is setting a specific bitrate as "DVD Quality".

The bitrate you need to use is driven by the type of video you have, the quality you want, and the size budget you have for the output. So long as you stay within the DVD specification, the rest is up to you.

For example, a simple static text menu might look very good at 2000 kbps, but a fast action motion menu may require 8000 kbps to maintain the quality.

[manono]

If I'm encoding the main title of a dvd at a constant bitrate...

There's your first mistake. Don't use CBR encoding.

...then what should I set the kbps at to ensure a dvd quality picture?

There's no way to tell. Different movies compress differently.

As a comparison, when converting a cd track to mp3 or Vorbis it's generally acknowledged that 192kbps = cd quality.

CBR again, I suppose. Nonsense.

If you want good quality, maybe do a 1-pass at Q 18-21. The lower number will give you a larger file size.

[Noahtuck]

As a comparison, when converting a cd track to mp3 or Vorbis it's generally acknowledged that 192kbps = cd quality.

I always get a kick out of it when people say 192kbps MP3 is cd quality

Cd quality is 1411kbps, 192kbps MP3's are crap!!
Unless you are listening to them on crap equipment then you can't tell the diff.

[edDV]

X264 is not DVD. DVD is MPeg2. See "What is DVD?"
http://www.videohelp.com/dvd

Some DVD players will decode h.264 outside the DVD spec. Ideal h.264 bit rate depends on type of input and the bit rate limits of the particular machine. You need to test maximum bit rate for each player model.

Consumer camcorder video requires far more bit rate vs. pro shot progressive. Films are shot and processed for ideal MPeg encoding. Ideal source would be uncompressed with high signal to noise. If you are working off a compressed source, more recode bit rate is required.

A typical VBR range (not the best quality) h.264 bitrate for SD video is 2-6Mb/s. Experiment with higher and lower and make your choice.

[jagabo]

Use constant quality encoding if you're interested in quality. You'll always get the quality you specify.

[lordsmurf]

192kpbs is *usually transparent compression to the audio CD. By the time you start to use 256kbps and 320kbps, you're just pissing away bits to bloat the file size. Any added quality is perceived by psychology only, not from reality.

So many audio CDs are cooked anyway, that the term "CD quality" is almost a joke in itself. Don't give me that "vinyl is better" BS either, it's not.

There are downsides to all audio storage formats.

The "better quality" ones are boondoggled by expense and lack of popularity.


* I say "usually" in that first sentence because because there are some CDs that need more bits. Although most of the times, when somebody says their CD needs it, they're still wrong.

[guns1inger]

A lot of bitrate now goes to encoding auto-tune warble. It is so bad now that many artists seem to include it as an effect, rather than just learn how to actually sing in key.

But that is getting way off topic.

[t0nee1]

As noted above, when conducting proper blind ABX tests, the results have shown most "192kpbs is *usually transparent compression to the audio CD"..
Cracks me up when someone shouts, 192kbps MP3 is "crap" and then fails a blind ABX test..
Crap/quality is in the EARS of the beholder..

And yes, have to agree..Getting way Off-Topic.. 8)

[hech54]

Why make commercials for your "superior" television with "superior to all others" picture quality when it is obvious that everyone is watching the commercial on their "inferior" television sets?

[paddyspub]

In case anyone missed the point of my original post, I'm attempting to rip the main title of a dvd to either Theora or x264. Considering that this thread was posted in the DVD Ripping forum I'm not sure why this was so confusing. Secondly, I do consider 192kbps to be cd quality and can't hear any difference in audio quality between the mp3 and cd track. In fact in most instances I can't hear any difference between 128kbps vorbis files and cd tracks.

So based on the posts above there doesn't appear to be a de facto cbr setting that is considered to be dvd quality, right? The program I'm using offers either a cbr setting or a constant quantizer setting. Is the quantizer similar to a variable bitrate setting?

[edDV]

In case anyone missed the point of my original post, I'm attempting to rip the main title of a dvd to either Theora or x264. Considering that this thread was posted in the DVD Ripping forum I'm not sure why this was so confusing. Secondly, I do consider 192kbps to be cd quality and can't hear any difference in audio quality between the mp3 and cd track. In fact in most instances I can't hear any difference between 128kbps vorbis files and cd tracks.

So based on the posts above there doesn't appear to be a de facto cbr setting that is considered to be dvd quality, right? The program I'm using offers either a cbr setting or a constant quantizer setting. Is the quantizer similar to a variable bitrate setting?

There is no setting to rip DVD to DVD quality. There will always be loss with any recode.

If your goal is quality not minimal file size, listen to Jagabo and use constant quality. That setting says you want quality (minimal loss) and will accept whatever bit rate is required.

If your process requires CBR, that must be set to the worse case scene (action and/or soft gradient). That setting will result in the largest file size.

[Noahtuck]

Cracks me up when someone shouts, 192kbps MP3 is "crap" and then fails a blind ABX test..
Crap/quality is in the EARS of the beholder..

Well, personally, on all my home equipment i can tell in a second the difference between an original cd and any 192kbps mp3, any mp3 at all actually.

Now when i listen to a 256 or 320 mp3 on my little $30.00 portable pocket mp3 audio & video player with the cheap little earbuds, an mp3 i have made myself, it is fine.

But if someone can't tell the difference between the two on a halfway decent system either is deaf, don't care or don't appreciate good music

I know people that have said they never noticed the difference until i sit there with them and tell them what to listen for, her how that sounded like that, and this ?
Same with video, you'd be amazed at how many flaws in different videos i notice and 5 other people will watch the video at the same time as me and never notice a thing
I'm talking even constant very noticeable things like a bad jerky PAL to NTSC conversion!

[t0nee1]

Well, personally, on all my home equipment i can tell in a second the difference between an original cd and any 192kbps mp3, any mp3 at all actually.

And how can anyone dispute?..None of us share EARS...Consider yourself one with 'Golden Ears'..
But , just knowing visually the equipment being used, and how much it costs, can have an affect on one's perception..
However, not if you're one with so-called 'Golden ears..
And of course one can be trained, what to listen for...I just want to enjoy/appreciate the music, as most of us do...

[jagabo]

I do consider 192kbps to be cd quality and can't hear any difference in audio quality between the mp3 and cd track.

I think you can train yourself to hear the difference, especially with special test tones. But I agree that, with most material, 192 kbps is difficult to tell from the original CD. Also, MP3 encoders vary in quality. So a poor encoder will make the differences much more obvious than a good encoder.

So based on the posts above there doesn't appear to be a de facto cbr setting that is considered to be dvd quality, right?

First of all, DVD itself varies in quality. Bitrate typically range from near 0 to 9000 kbps. Also, video compression is more complex than audio compression and has a lot more variability. There is no hard and fast rule that says you need X bitrate with h.264 to maintain "DVD quality".

The program I'm using offers either a cbr setting or a constant quantizer setting. Is the quantizer similar to a variable bitrate setting?

Constant quantizer encoding is a variable bitrate method. There are two basic types of VBR encoding: 1) constant quality where you select the quality (be it high or low) but you don't know what the final average bitrate (and hence the final file size) will be. 2) Bitrate based encoding where you select the average bitrate but you don't know what the final quality will be.

Generally, if you are more interested in quality you use constant quality encoding. If you are more interested in file size (like you need to fit a 2 hour movie on a 700 MB CD) you use 2-pass variable bitrate encoding.

[lordsmurf]

To avoid new artifacts in a 640x480 H.264 file, you'll need about 1 to 2Mbps, with a 2-pass VBR and a ceiling bitrate of maybe 4 to 8Mbps -- AFTER it's been pre-filtered to reduce artifact-causing noise.

[Noahtuck]

However, not if you're one with so-called 'Golden ears..
And of course one can be trained, what to listen for...I just want to enjoy/appreciate the music, as most of us do...

Well, golden ears aside, even when i have pointed out to others they could then also hear differences.

And i also just want to enjoy and appreciate the music, that's why it annoys me so bad when i can hear crap :P

If 192kbps mp3's were as good as 1141kbps uncompressed audio, all our store bought cd's would contain tons of mp3's.
Oh wait... then the fatcats would not make as much $$$

But seriously, what can i say......
Maybe it's because of my upbringing ?

Both my parents played several instruments before i was born, in bands, ect.
Mom taught me how to play piano when i was 6, started playing guitar when i was 9, played heavily from the age of 13, besides all the other crap in school, violin, trumpet, ect.
So i guess i just know what i am hearing most of the time

[lordsmurf]

even when i have pointed out to others they could then also hear differences.

In my experience, this situation of friends/family agreeing with you is just to shut you up.
I've seen this so many times through the years.

Most people cave when confronted after an A/B test, and just agree with a tester.
It's not because you're right, it's just because they don't really care.

This is a classic issue with polling. It's easy to bias a "second take" poll. So unfortunately, you've still not proved your point. You've only proved you could bias your audience. You can bias in one take too, with certain nods one way or the other, be it in description or in display.

I'd also have to disagree with a piano being a good instrument to learn tonal range from. Pianos largely sound like crap compared to wood or string. The percussive sound is not ..... something .... (I forget the term, been too many years). Guitars are fairly abrasive too, not gentle sounds with a large acoustic/tonal range.