Blu-Ray mkv to .m4v - How to see the perceptible difference?

Discussion in 'Video Streaming Boxes & Services' started by Norman, Feb 3, 2014.

  1. Norman

    Norman
    Well-known Member

    Joined:
    Jul 14, 2000
    Messages:
    1,748
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    116
    Location:
    Herne Bay
    Ratings:
    +597
    Until now I've always accepted that transcoding my Blu-Ray mkv rips to .m4v (mp4) will cause video quality loss (and of course the loss of DTS-HD), but I finally decided to try and find out how much and do a proper comparison to see if this loss is very perceptible or not.

    Lets use Star Wars Episode III as an example. I ripped the Blu-Ray using MakeMKV and ended up with an mkv file that contained three tracks, video track, DTS-HD (lossless) track and the PGS Subtitle track. This file is 35.45GB in size.
    Using Handbrake (High Profile, x264 Tune:Film, x264 Preset:Slow, CQ:21) the transcoded .m4v file came out at 5.65GB.

    Each file was played in VLC, stopped at a precise moment (using the goto seconds counter) and a snapshot was taken.
    I opened each snapshot side-by-side on my iMac and scanned for differences between the two images (including zooming in) but I couldn't see any.
    Thinking my 27" iMac screen might be too small, I also viewed the images on my Samsung UE55F8000 TV, again I can't see any differences. Colours are exactly the same, detail looks exactly the same, colour graduation looks exactly the same (no banding or steps).
    Not content, I done the same test using Lord of The Rings FOTR, this time choosing a meadow scene, and again, I can't see any differences. Every blade of grass, meadow flower etc, was just as detailed in the .m4v as it was in the original mkv.

    I now want to ask the community if my testing method is flawed? There has to be a difference, I know there does, but why doesn't my snapshot comparison show it?

    I want to justify to myself why it's worth keeping the original mkv's, and I'm looking forward to your replies and perhaps offering some guidance on where I'm going wrong?
     
  2. next010

    next010
    Distinguished Member

    Joined:
    May 29, 2005
    Messages:
    10,843
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    163
    Ratings:
    +2,128
    I've seen comparisons done (Doom9 used to do them but it long stopped being interesting as everyone knew x264 smoked everything) but the differences are often so tiny that keeping the original is really no longer necessary if your willing to take the time and convert. Most x264 encoders default to high quality encodes unlike many other H.264 encoders and Blu-ray's of course are excellent source material.

    I remember a few years back one of the x264 devs saying Blu-ray wouldn't even have been necessary, x264 HD video on DVD would have delivered satisfactory results.

    This guy did some comparisons of commercial rentals, stuff that comes from iTunes and Netflix, he also gives some tips on how he went about it that you could probably try for your self.
     
  3. MadScientist

    MadScientist
    Active Member

    Joined:
    Apr 30, 2006
    Messages:
    765
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    46
    Ratings:
    +103
    Make sure you assess some material that was originally encoded in the VC1 format. (I think the Bourne trilogy is an example; Animal House, another). Some Dr Who BRs are encoded as VC1 1080i and are a real pain! If these don't show any artefacts nothing will
     
  4. Norman

    Norman
    Well-known Member

    Joined:
    Jul 14, 2000
    Messages:
    1,748
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    116
    Location:
    Herne Bay
    Ratings:
    +597
    I'm starting to feel a bit better about the test I done, and having read the linked article (above) it seems to verify my findings (although he could see slight detail differences between the iTunes v Blu-Ray versions whereas I can't (might be my handbrake settings are doing a better job of encoding)).
    I don't have any VC1 titles but I'll keep an eye out for one to test.

    It takes my iMac about 3 Hours to transcode a typical 2hr Blu-Ray, so I'm starting to lean towards using Handbrake to re-encode them to mkv (rather than .m4v) so I can Passthrough (keep) the DTS and PGS tracks.
     
  5. HeadBanger

    HeadBanger
    Well-known Member

    Joined:
    Sep 22, 2000
    Messages:
    2,311
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    116
    Location:
    Ashingdon
    Ratings:
    +538
    I agree with you completely and I have done the very same tests as you. I have posted my same findings on AVForums and there have been a few people who just will not accept that it can be so.
     
  6. HugoFJH

    HugoFJH
    Active Member

    Joined:
    Jan 3, 2007
    Messages:
    2,346
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    63
    Location:
    London
    Ratings:
    +133
    Several things still need to be considered;

    1) a still image doesnt really prove anything, it also depends on how movement is handled (especially alot of movement in the same scene), differing colours and shadows just to name a few.

    2) DTS-HD/True-HD usually accounts for a large proportion of any BR rip Ripping the same movie once with True Hd and a second time with plain stereo reduced file size in my experience by about 60-70% (I did it years ago but I dont see why it would be any different now)

    3) Everyone is different. The op may be very happy with the results but someone else may see things when watching the m4v file that they personally cant ignore.

    4) Most importantly to me - if you have spent a few thousand ££ on a very good tv, a few more on amps, speakers etc - is there any point in scrimping a few hundred on smaller hard disks for the media server!!
     
  7. Norman

    Norman
    Well-known Member

    Joined:
    Jul 14, 2000
    Messages:
    1,748
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    116
    Location:
    Herne Bay
    Ratings:
    +597
    Thank you for your input and you raise some valid points, but in response...

    1) I'd have thought still images are relevant. I would expect to see differing colours and missing or blurred detail in poorly encoded frames.

    2) I can test that easy enough, I'll demux the DTS Lossless track and see how big it is on its own, but I suspect it'll only account for a few GB.

    3) That was my main point - I can't see a difference. Granted, my TV is only 55 inches and I don't pretend to be a video expert, but I easily spot the shortcomings of broadcast HD, Streamed HD (i.e Love Film) and low bit-rate encoded films, so I think I'd see it here.

    4) That depends. I'm happy to spend "a few hundred quid" on fitting more 4TB HDD's to my QNAP NAS, but that on its own isn't safe enough for me. I'm not happy to leave all my media content on one device in one location (house fire, theft) so off-site backups (for me at least) are a must, and setting up an identical off-site storage solution isn't just "a few hundred quid".
     
  8. HeadBanger

    HeadBanger
    Well-known Member

    Joined:
    Sep 22, 2000
    Messages:
    2,311
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    116
    Location:
    Ashingdon
    Ratings:
    +538
    I use the MKV container and keep the DTS/DD tracks (sometimes lossless) and do not convert audio to stereo AAC (or equiv).
     

Share This Page

Loading...