Always have high reduction level

[neo squidward]

Curious as to why I always have a high "reduction level"?

Most recent:
- VTS_01: 2,787,717 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 219,869 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 87.8%
- Overall Bitrate : 4,462/3,569Kbs
- Space for Video : 3,995,618KB
- HIGH/LOW/TYPICAL Bitrates: 4,765/1,890/3,569 Kbs

Even on shorter movies with lower bitrates, the reduction is still 70%+. I'm ripping "main movie" with decrypter, editing the start/stop with shrink (no compression), then preping with rb. What am i doin' wrong?

tia

[Harrysmiith]

why do you think something is wrong? My understanding is that Reduction level figures mean that the movie needs to be reduced to 87.8% of its true size to fit on a DVD-5 ?
the higher this figure the better I would have said.

[jdobbs]

Higher is better. You're probably doing movie-only encodes if you are getting that high consistently.

[neo squidward]

"Higher is better." I guess I completely misread another thread, and yes, I am doin' movie-only.

I guess I thought something was wrong since on my test encodes I've been getting rather blurry results. CCE Basic, VBR_bias:25, VBR_Passes:2, Quality Prec:16, default encoder matrix. Screens available if curious.

Any ideas or recommended settings?

[jdobbs]

Two things to look at:

1. Make sure you haven't added any filters to the filter editor and forgotten about them.

2. Check the original for comparison -- see if it is "blurred" as well. At the compression levels you are using you shouldn't normally be able to distinguish between the original and the backup. There's always exceptions, but that's usually only for originals that have extremely high numbers of frames to be encoded. Yours doesn't.

[jdobbs]

One other thing... try doing the movie with no Shrink step. I've been told that running a job through Shrink, even though you choose no-compression, can result in blurring. I can't confirm that, though, as I don't use Shrink.

[neo squidward]

Here's an example of the blur with stats from rb and cce:

original - http://upload.hoogetech.com/encode/o1.jpg

encode (directly from original with no filters) - http://upload.hoogetech.com/encode/rb1.jpg

VTS_01: 2,466,980 sectors.
-- Scanning and writing .D2V & .AVS files
-- Processed 211,767 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 92.4%
- Overall Bitrate : 4,743/3,795Kbs
- Space for Video : 4,091,480KB
- HIGH/LOW/TYPICAL Bitrates: 5,043/2,726/3,795 Kbs

Other info from 5 mins encoded (other settings in previous post)
vbr: 4507 (500-9000) kbps

[jdobbs]

Was this sent through Shrink first? If so, try one without it.

[steptoe]

As I understand and read the figures, its only reducing the original by 12.2%

How I see it, 100% is the original source, so you're getting 87.8% of the original size, not its being reduced by 87.8%

If you have a movie that will easily fit on a single DVD, and run it through DVD-RB ignoring the warning that it will fit on a single DVD already, you'll end up with a reduction size higher than 100%

[neo squidward]

Was this sent through Shrink first? If so, try one without it.
Yeah, it was sent through shrink first, but I got the same blurry results from the original.

[Boulder]

What is the default matrix? It could be a bit too aggressive for your bitrate.

[jptheripper]

dont send it through shrink

[FredThompson]

Try HC. It gives a sharper output than CCE for quite a lot of source.

[Msc_Alex]

dont send it through shrink
Wait a minute, you mean to tell that shrink messes with the picture even when you have selected no compression? I have used it a lot too remove the audio and sub flags I have striped? :eek:

[jdobbs]

I don't know that for a fact... and my initial feeling was that it couldn't be so. But I have seen enough reports to that effect that I'm just not sure... and they all reported blurring. I guess someone could run it through some time and compare the demuxed sources. But as I've said before -- I can't chase problems that may be related to preprocessing with 3rd party sources... there'd be no end to it.

[FredThompson]

That doesn't make sense. DVDShrink does do some goofy stuff to the IFO structure which PGCEdit will detect. It doesn't modify I-frames at all, just B and P. Those are only modified if the stream is to be shrunk. AfterDawn posts are screwed.

If anything, loading PGCEdit and allowing its sanity check should be helpful. If there are broken or corrupt nav structures, PGCEdit will fix them. I've seen this a lot on TV episode discs.

[jdobbs]

Like I said, I don't know. The only way to tell is to compare the MPEG streams before and after.

[jdobbs]

Ok... since I don't want to be the kind of person who spreads unsubtantiated rumours I did a comparison myself... and the MPV files created by Shrink in "No Compression" mode match the originals exactly (at least on the ones I tested). One thing to watch, though, is ensuring you select the entire disc before making the "No Compression" setting (I accidentally didn't the first time I tested it and got a little concerned at first).

So the "blurring by Shrink" reports are urban legend and nothing more.

As for this thread...

@neo squidward

What settings to you have for the Matrices in DVD-RB? A matrix could have that kind of effect...

[Jeffster]

He said he used CCE basic. On it's own, basic doesn't have the option of changing matrices... is DVD-RB Pro able to use custom Matrices with basic?

I think neo squidward should post the avs script (copy & paste) for the segment that he obtained the screenshots from.

If CCE was starved of bitrate (unlikely at a reduction level of 92.4%), it may show other artifacts but I highly doubt you would get blurring, at least not on it's own, I wouldn't think. As jdobbs said, at the compression levels he is using you shouldn't normally be able to distinguish between the original and the backup.

[jdobbs]

Hmmm.... your right about CCE Basic. I'm stumped. The picture looks almost as if a light smoothing filter was used.

[Jeffster]

Hmmm.... your right about CCE Basic. I'm stumped. The picture looks almost as if a light smoothing filter was used.
Yes, to me also... that's why I wondered if he could post the avs script, since he hasn't ruled out yet that no filters were used.

[jdobbs]

@new squidward

Could you send me the following files? ( dvd-rb@dvd-rb.com ): REBUILDER.INI from the DVD Rebuilder installation directory, and REBUILDER.INF, REBUILDER.ECL, and REBUILDER.LOG from the working directory.