BD-Rebuilder Recommendations

[Sounddude]

I'm new to BD-Rebuilder and am trying my first backup of a full disc to BR-25. (DVD-RB Pro-user, new to BR)

The disc is Van Helsing size ripped is 41.18GB
What settings do you recommend using? the main movie with DTS-MA is just over 35GB.
4.5+hrs in total time for whole disc.

Speeds not a problem as I have a quad-core intel @3GHZ and 4GB ram.

And how much quality difference between this and movie-only to DVD9 or 5 with Dolby instead of DTS-MA? :thanks:

[GaPony]

For what its worth, I copy most (90%) of my movies as Movie Only w/AC3 to DVD9. The video quality is comparible to the original and my ears seem to prefer AC3 over DTS (or DTS-MA). Van Helsing is among those I've copied this way. (I use the AVCHD setting with 640kbs AC3 and custom size to 8250).

With the price of BD-25 media dropping so much, I may do more to BD-25... but I also copy everything to my server for streaming around to all my TVs, so maybe not. :)

I think if you want to use DVD9 media, you pretty much need to use AC3 audio. Sometimes the DTS tracks need more room than is available on a DVD9, all by themselves.

[Sounddude]

I just got my BR burner, Just had a BD-ROM before and I bought some BD-R 25's to try out both ways. On the 25GB I plan to try to keep the DTS audio - full disc and then I think I'll try the DVD-9 method with movie only and w/AC3 and compare the two. Only paid 21.00 for 10-4X BD-R 25 discs so I'm anxious to try them out.

Thanks,

[jdobbs]

I just got my BR burner, Just had a BD-ROM before and I bought some BD-R 25's to try out both ways. On the 25GB I plan to try to keep the DTS audio - full disc and then I think I'll try the DVD-9 method with movie only and w/AC3 and compare the two. Only paid 21.00 for 10-4X BD-R 25 discs so I'm anxious to try them out.

Thanks, It's always a good idea to keep a BD-RE around too. They carry them at Target if you're in the US.

[Winenut]

I know this thread is old, but just a note on BD Rebuilder "Quality" settings. Since building my new pc (I7 930), I've been able to play around, re-encoding lots of movies using all the different quality settings...and quite frankly, the only quality option that does a good job consistently (at least for BD25) is the High-Speed option. All the other options add some level of DNR, which softens the picture (which reducing the grain), shifts the color (to the red) while reducing it, hardens the contrast, and worst of all...introdues motion artifacts (common to DNR). The higher "so-called" qualkty setting you choose, the worse it gets. The High-Speed option appears to do almost no DNR and the only side effect I can see, is a slight increase in the color level (whichout shifting it at all). This is most noticeable with films that are already over-saturated (such The Music Man). For BD9 and BD25, the GOOD option appears best, but unlike the High-Speed option does introduce some DNR. Like with the BD25, the fastest option does the least clean up (and introduces the fewest DNR related artifacts). Just my opinion of course, but it's based on quite bit of experimentation.

[jdobbs]

You have a point and conclusion holds up when the bitrate is high such as when encoding for BD25 -- but you'll also find that when you lower the bitrate and introduce distortion due to more compression, the other options are essential to a good picture.

[hautem]

I know this thread is old, but just a note on BD Rebuilder "Quality" settings. Since building my new pc (I7 930), I've been able to play around, re-encoding lots of movies using all the different quality settings...and quite frankly, the only quality option that does a good job consistently (at least for BD25) is the High-Speed option. All the other options add some level of DNR, which softens the picture (which reducing the grain), shifts the color (to the red) while reducing it, hardens the contrast, and worst of all...introdues motion artifacts (common to DNR). The higher "so-called" qualkty setting you choose, the worse it gets. The High-Speed option appears to do almost no DNR and the only side effect I can see, is a slight increase in the color level (whichout shifting it at all). This is most noticeable with films that are already over-saturated (such The Music Man). For BD9 and BD25, the GOOD option appears best, but unlike the High-Speed option does introduce some DNR. Like with the BD25, the fastest option does the least clean up (and introduces the fewest DNR related artifacts). Just my opinion of course, but it's based on quite bit of experimentation.

@Winenut

I just graduated to a BD burner - all of my previous backups have been full disc to BD9 using the High Quality (Default), 2-Pass setting and have been more than satisfied with the results. I just tried my first full disc BD25 backup using the settings determined by the Automatic Quality Settings - Good (Very Fast), One Pass (ABR) Encoding. While largely negligible, I did notice some softness in identical frame captures between the 1-Pass BD25 and 2-pass BD9 and still experimenting/feeling my way around the BD25 format. Are you referencing using the High-Speed Option (BD-25) with One Pass (CRF) Encoding or One Pass (ABR) Encoding and is this full disc or movie only?

[A.Fenderson]

@jdobbs and/or winenut: So, if the goal is for the BD25 to come out with the picture being as close to the original as possible (no DNR introduced, no color shifting/boosting/suppressing, no alteration to the contrast, no introduction of "motion artifacts"), then throwing 2-pass and "Highest" quality settings at it is not necessarily the best idea? I've been doing a lot of back-to-back encodes this way since having gotten my BD burner, and not testing many of them extensively--I've done almost no A/B testing between the BD-RB output and the original, and none between the BD25 output and a BD9 output with high-quality and 2-pass. I assumed BD25 @ Highest/2-pass would be at worst overkill and hoped it might be the ideal settings to maximize similarity of the reencode's picture to that of the original BD--is this not really the case?

@hautem: you mentioned softness in frame caps between 1-pass BD25 and 2-pass BD9, but specifically do you mean the BD25 was softer?

[hautem]

@hautem: you mentioned softness in frame caps between 1-pass BD25 and 2-pass BD9, but specifically do you mean the BD25 was softer?

Yes, that the Good (Very Fast), One Pass (ABR) Encoding to BD25 appeared a bit softer in areas than the High Quality (Default), 2-Pass to BD9. But that was based on looking at just a few frame captures and I've looked at a number more since (using MPC). I've been comparing:

High Quality (Default), 2-Pass to BD9 (before getting the BD Burner)
Good (Very Fast), One Pass (ABR) (from the Automatic Quality Settings for BD25)
High-Speed Option, 1-Pass ABR to BD25
High-Speed Option, 1-Pass CRF to BD25

After looking at several dozen frame captures (mostly under 200-300% magnification), it kind of boils down to this (at least for me). There's really no free lunch here - the re-encoded copy will never look identical to the original (on a pixel-for-pixel basis), but you can get pretty close - i.e. virtually identical. In high contrast areas/bright light areas, they all look about the same to me. Not exactly the same as the original, but sufficient to capture a near identical rendering when viewed at a distance. It's in low contrast/low light areas where you can sometimes see a difference - the resolution can become mottled a bit depending on which setting was used. And I can only see this stuff when I overlay the frame captures in Irfanview at 200-300% and toggle back and forth. FWIW, the settings determined by Automatic Quality Setting (Good (Very Fast), One Pass (ABR)) is what wound up looking truer to the original in those low contrast areas - not identical, but less mottled than the others. The differences here are so subtle/minor, I'm not sure my conclusion would be the same if I compared 100's of frame captures vs. just dozens.

Quite honestly, I couldn't see any difference re: resolving the issues that Winenut was speaking to by using the High-Speed option (reduction of DNR artifacts, color shift etc.), maybe because I'm comparing those High-Speed settings to High Quality (Default), 2-Pass (BD9) and Good (Very Fast), One Pass (ABR) (BD25) to begin with. Also, the High-Speed option, whether using 1-Pass ABR or CRF only saved about 15 minutes encoding time, so I'm staying with dobbsy's recommended settings for now. I know it's been described as overkill, but maybe for s&g's, I just may try the High Quality (Default), 2-Pass to BD25 just to see what shakes out - but it's going to be difficult giving up the new found 3+ hour encode time - my High Quality (Default) 2-Pass to BD9's were taking 10-12 hours.

[setarip_old]

@hautem

Hi! Please clarify - What settings are you using when you say: it's going to be difficult giving up the new found 3+ hour encode time

[hautem]

@hautem

Hi! Please clarify - What settings are you using when you say:

Before getting the BD burner, all my backups were full disc to BD9 using an LG BD reader drive with the following settings:

Full Disc
Quick Encode for Extras (unchecked)
High Quality (Default), 2-Pass
Custom Target Size 8210
English only for audio and subtitles (enabled additional audio tracks manually as needed)
All other settings default
Encode times were typically 10-12 hours, sometimes longer on a few occasions

Settings now with the LG BD burner going to BD25:
Full Disc
Quick Encode for Extras (unchecked)
Good (Very Fast), 1-Pass (ABR)
Custom Target Size 23450
English only for audio and subtitles
All other settings default
Total encode time for the first one I tried was 3:37:06. Using High-Speed, 1-Pass ABR, it was 3:17:55. Using High-Speed, 1-Pass CRF, it was 3:23:11

This was all for the same disc that I happened to have on my hard drive at the time ("The Mask of Zorro"). Total encode time for this going full disc to BD9 using High Quality (Default), 2-Pass was 10:06:22

Running on a Dell - Intel Core 2 Quad CPU Q6600 @2.4Ghz 3GB SDRAM

[setarip_old]

@hautem

Thank you ;>}

[hautem]

@hautem: you mentioned softness in frame caps between 1-pass BD25 and 2-pass BD9, but specifically do you mean the BD25 was softer?

@A.Fenderson

Not sure how scientific this is and/or if it's an accurate indication of quality level of some of the various BD-Rebuilder settings, but was fooling around with some of the frame captures that I collected, did some crops and stuck it on a web page.

http://hautem.dyndns.org:81/html/Pictures/BD-Rebuilder/BD-Rebuilder.html

Just my opinion, but if you're looking for near identical going to BD25, it's there - you just have to pay the encoding time price.

P.S. - the link is on my PC that I use as a server and may not/won't be available when I need to shut the PC down etc. I'll most likely remove it after some time.

[A.Fenderson]

@hautem: Thanks for the link! Those comparisons are very cool. The high-quality 2-pass BD25 encodes seem to capture much more of the detail of the original than any of the other captures. Have you tried any "highest" quality 2-pass to BD25? That's what I'm using for now--the increase in encode time from the "high" quality was relatively minor in comparison to the overall encode time, so I thought I'd try squeezing every last bit of quality out of the reencodes that I could.

[jdobbs]

Unfortunately when you use JPG as your output format you never know if any differences seen are from the AVC or the JPG compression...

[hautem]

Unfortunately when you use JPG as your output format you never know if any differences seen are from the AVC or the JPG compression...

Good point. The original frame captures in MPC-HC were saved as .bmp. The crops in PS were saved as .jpg mainly because I initially couldn't seem to get the grid .css to work with with the .bmp crops from which the .jpg versions were derived and I thought it might take too long for the .bmp crops to toggle etc. (the .bmp crops are about 3MB each).

I just replaced all the .jpg crops with the .bmp crops and the visual relationship appears the same (at least to me). There's always the issue of what may have been added in the bicubic upsizing interpolation or just the further processing, but it was applied consistent across all the samples. The visible differences may not prove anything.

@A.Fenderson

I haven't ventured past Good (Very Fast), 1-Pass ABR for BD25 at this point. I just got the BD burner last week and found Winenut's comments re: the High-Speed Option in the above post interesting, so I was curious and still feeling my way around. On an encoding time price/quality basis, I'm pretty satisfied with what Good (Very Fast), 1-Pass ABR is delivering on the few backups to BD25 that I've made so far. After seeing these 3 1/2 hour encoding times, I just don't have the patience to go back to the 10-12 hours (or now longer) that it was taking for High Quality (default), 2 Pass to BD9. When viewing the screen captures full frame, the differences that are apparent at 200% are to me, barely discernible. The crops just seem to reinforce what's already a known - if you're looking for the highest (or higher) quality, 2-Pass appears to be the way to go.

[A.Fenderson]

I haven't ventured past Good (Very Fast), 1-Pass ABR for BD25 at this point.

Hm, maybe I'm not understanding what you mean here, but everything on the linked page you included seemed to indicate that (unless otherwise stated), the encodes you were doing were to BD25. So do you mean you just haven't done any BD25 compressions besides this one movie as a test, or were the "High-Quality (default), 2-Pass" and "Good (Very Fast) 1-Pass ABR" screenshots on your page actually from BD9 encodes or something other than to BD25 target size?

[hautem]

Hm, maybe I'm not understanding what you mean here, but everything on the linked page you included seemed to indicate that (unless otherwise stated), the encodes you were doing were to BD25. So do you mean you just haven't done any BD25 compressions besides this one movie as a test, or were the "High-Quality (default), 2-Pass" and "Good (Very Fast) 1-Pass ABR" screenshots on your page actually from BD9 encodes or something other than to BD25 target size?

@A.Fenderson

Sorry if I wasn't clear. All of the screen captures at the various settings were to BD25 except the last one referenced (as indicated in the grid) - that one was to BD9 using High-Quality (default), 2-Pass. I just threw that in comparing my "old" BD9 format to the new BD25 format options.

I was just responding to your question re: "Have you tried any "highest" quality 2-pass to BD25?", i.e. that no, I haven't gotten past exploring Good (Very Fast) 1-Pass ABR to BD25 (which I'm happy with right now).

FWIW, you can be your own judge with this stuff. Just find a frame using MPC to open up the main movie .m2ts file (I prefer using the Media Player Classic - Home Cinema version) - usually a frame right after an edit is easy to find again. Use the right arrow keyboard key to frame advance (the left arrow key back frame advance is less cooperative) - "Alt-I" to save the frame either to .bmp or .jpg (.bmp preferred). Do this for the original file, then your "highest" quality 2-pass, open both up in a quick viewer like Irfanview, magnify to 200% and toggle back and forth between the two instances. If there are differences, you should be able to readily see them as in the crops.

[A.Fenderson]

@hautem: got it, sorry if I was obtuse. Thanks for the advice. :-)

[hautem]

@hautem: got it, sorry if I was obtuse. Thanks for the advice. :-)

@A.Fenderson

Hey, no biggie - I tend to get a bit diffuse, so it's understandable.

Personally, if you've been doing "highest" quality 2-pass to BD25 for all your encodes, you've probably maxed out what H264 can deliver via BD-Rebuilder anyway (I saw the new AVS filter editor options in the new release, but that's nature's way of telling me, "Do not touch"). My suspicion is that if you did side-by-sides, they still wouldn't be identical, but you probably wouldn't get any closer with any other (current) setting. The trick I think, is using one that gets "close enough", but taking substantially less than 12+ hours to do it in - at least for me, anyway.

[Winenut]

ABR seems to be a better job than CRF, particularly in action scenes requiring a higher bit rate...and I've not seen much different (or benefit in using multi-pass). So I'm down to always using Single Pass ABR with the High-Speed option. I mostly do Movie-Only encodes, as I'm after the least degradation possible. But I have a done some full disc enclode, when the compression is minimal. And I do like that BD Rebuilder does do more compression on the extras and tries to leave the main movie if possible.