7
Replies
Hi Joe.
In all honesty, you can get a much lower bitrate if you use a good H264/AVC encoding rather than the MPEG-2.
AVC will get you the same quality as an MPEG-2 at a much, much lower data rate & file size simply because the compression is much more efficient.
I would also HIGHLY recommend the use of 3rd party encoders rather than the stock AME version (which is based on the MainConcept engine) and personally I use one of the following 2 encoders:
x264 PRO | Adobe Creative Suite H.264 Encoder
Both of these are built on the same encoder engine - the x264 one - but they are different implementations and as such the same footage encoded with both encoders will look slightly different - and the way we handle this is to pick a section that is particularly "difficult" to encode and give it the old College Try, passing the snips to the client for approval - some discs it is A, on others we use the B. Technically, the B version is a little less comprehensive in it's tweakability (and if you want to tweak it then you really do need to understand not only the encoder parameters but also exactly what the Blu-ray specifications are) and the only downside I can find to it is that there is no "quicktime player compatibility" checkbox as there is in A.
As you are already well aware, your biggest problem is the combined bitrate - you do not want to exceed a total rate of more than 40Mbps otherwise you may struggle on Profile 1.0 machines (and sadly there are a lot of these still in use so all replicated titles should be authored to work in these machines). So let's have a closer look at your assets as well - can I run some questions past you please?
1 - What is the DTS-HD stream, exactly? (1.5 Mbps implies a DTS Core Audio stream, not DTS-HD Master Audio)
2 - What did you use to create it?
3 - What is the 5.1 track at 448 Kbps? It smells like AC3 to me......and if you already use a DTS stream then the Dolby one is pointless unless your client is insisting on it - AC3 is the lowest possible quality you can actually put on a Blu-ray, although the whole area of audio on Blu-ray is kinda complex & we should probably have a discussion about this at one point in these forums (if anyone else reading this is interested in seeing such a thing please let me know and I will make it so)
Carrying on with audio, I am assuming (yes, I know what assume means - make an ASS out of U and ME, but we have to start somewhere 
) that you have a stereo mix and a 5.1 mix yes? If this is the case then I would set it up as follows:
Audio Stream 1 - PCM Stereo or DTS-HD Master Audio stereo (this can be either 16 or 24-bit and should usually be at 48kHz unless you have a good reason to use 96kHz)
Audio Stream 2 - DTS-HD Master Audio 5.1, setting a downmix to RPGA standards (see pages 58-59)
That is all you really need and using DTS-HD MA for your stereo stream gets you a reduced file size of around 40% due to the lossless data compression. As I said, I can give you chapter & verse on audio if anyone is interested, but for the moment suffice it to say that on all Blu-ray players the support of DTS-HD Master Audio is absolutely mandatory (although with "gotchas" in 5.1, just like Dolby True HD) and is guaranteed to work on all machines.
Your freezing problem could be caused by several factors:
1 - Peak bitrate is just too high for P1 players (The Adobe Media Encoder lies to you about peak bitrate in MPEG-2 encoding as well as AVC encoding using the stock codecs - for MPEG-2 DVD it is around 1 Mbps above what you set and for Blu-ray I am not sure how far off it is, but again the peak rate will be higher than the one you set - the only MPEG-2 encoder I know of that keeps to the peak bitrate you set is CinemaCraft CCE-SP and CCE-HD)
2 - The total stream bitrate is too high for the player being used as a test bed
3 - The disc was burned too fast (always use the slowest possible speed you can achieve with your media/burner combination - IMGBurn will tell you this when you set a burn up, and believe me we always use IMGburn for Blu-ray tests as it is by far the best burn engine out there bar none)
4 - the quality of the blank media is not what it could be (sad but true - not all blanks are equal)
5 - the burner is emptying the buffer during the process (set this higher & see what happens if using IMGBurn)
6 - general written disc weirdness - this happens much more than you might think (although it is rarer with single layer media) as written disc support is sporadic & varies from player to player -- remember a written disc is using a thin layer of vegetable dye whereas a replica is using a thin sheet of metal between the plastic substrates.
All we can do is go through the options & gradually rule stuff out.
Sticking in the same place every time implies to me (and I stress this is a quick guess based on the limited info we have) that either the image is faulty or the disc/burner combo is struggling - is the "stick" near the end of the disc by any chance?
Replication & test moulds
When replicating, use the best you can - a cheap plant is cheap for a reason and that reason is usually that they are cutting corners somewhere, either in press time or in staff quality. In the former case all you can do is go somewhere else (some run their machines 24/7 using the minimum time per cycle and carrying out maintenance when things break yet the better ones run slower (fewer discs but better quality) and do preventative maintenance on the production lines) and in the latter case you need to test your disc as widely as possible as they will be employing people who do not understand the Eclipse ImageAnalysis reports & either pass discs they should not or fail discs they should not (both have happened to me and we ended up buying the analysis software so we could run these checks at our end)
Test discs are often charged extra (especially by the cheaper plants) unless you go to one of the big players who tend to offer this from the start - again, the cheaper ones will not even mention test discs unless you raise the subject first.
In the end, a test disc is really the only way to go for 2 very good reasons:
1 - Blu-ray carries a mandatory copy protection system called AACS. Before you can replicate, either you as rights holder or the plant must have an AACS license or the disc should not be replicated. Personally I think this is daft but there you go (I do not have an authoring house AACS license because it is too expensive, as you need to buy several BD "books" plus the license and this runs into tens of thousands of dollars, so all our masters use Type A BDCMF replication masters and our clients have the AACS license as rights holder - this is much cheaper. You probably already know all this though, but in case you do not again let me know & I will write an article & sticky it here)
Because of this encryption, which is not possible to do on a written disc (written discs use blocks of 2048 where copy protected discs use 2054 - just like in DVD), a test disc is your only chance to actually check a replica disc with the AACS encryption for any faults (and factories do make mistakes - they are only human after all) before you press hundreds if not thousands of copies. It is madness not to do this.
2 - The way a replica disc is handled by players is not the same as the way a written disc is handled. Written discs have much lower rates of reflectivity and are therefore much more likely to "stick" or generally play up (pun unintended) so just because a burned disc is flawed does not necessarily mean the replica will be -- it could just be that your player does not like written discs much, or that particular brand of discs very much.
I hope this helps a little - please please get back if you need more & I will keep an eye on this thread daily.
Summary:
1 - Dump the MPEG-2 and use AVC/H.264 instead.
2 - For the H.264, try the plugins referred to above (both work beautifully in Premiere Pro CC 2018) and use a bitrate of 20 Mbps maximum and 15 Mbps target. This will be plenty enough.
3 - Use DTS-HD Master Audio for your stereo and your 5.1. Do not add the commentary as a secondary audio stream but instead create a separate stream for this. DTS-HD MA gets you a bitrate of 2475 Kbps (peak) in stereo (this sums both the core & the lossless sections) and 6204 Kbps (peak) & 5727 (average) in 5.1
I look forward to hearing from you with reports of success!!
AdChoices