My "Fifth Element Superbit" review (a bunch of thumbs up, IMHO)

Discussion in 'Archived Threads 2001-2004' started by Bjoern Roy, Oct 6, 2001.

  1. Bjoern Roy

    Bjoern Roy Second Unit

    Joined:
    Oct 15, 1998
    Messages:
    315
    Likes Received:
    0
    Trophy Points:
    0
    Hi folks,
    i just got my first SB and i now know why i don't trust any reviews but my own.
    You know i am a videophile first, so i checked out the picture but not the sound yet.
    Comparisson against the old R1 DVD. (I also have the new R2 SE DVD, which has several issues and is far subpar IMO to both R1 versions.)
    Detail:
    The SB version has a fair bit more horizontal detail in ALL scenes, but in SOME the difference is HUGE. There is no increase in vertical detail at all. The increase in horizontal resolution is NOT DIRECTLY due to the higher bitrate, but rater INDIRECTLY. Let me explain:
    DVDs are usually slightly filtered to remove the highest frequencies, thus limiting the finest detail.
    1) In the vertical dimension this is done to minimize interlace flicker artefacts on regular TVs. This wouldn't be necessary if DVDs where optimized for progressive output.
    2) In the horizontal dimension its done to minimize aliasing artefacts in the DVD players output stage. Not all DVD players can fully resolve 6.75Mhz patterns, some show heavy aliasing if such frequencies are present.
    But there is another, more important reason why filtering high frequency information (in both dimensions) is used: it reduce the ENTROPY (real information, detail) of the image. And the lower the entropy of the encoded material, the lower the MPEG bitrate can be, without yielding a considerable amount of compression artefacts.
    Think of it like this. There is a consensus about a certain threshold of compression artefacts that you don't want to surpass. On the other hand, going far beyond this threshold isn't necessary, because there is a point of diminishing returns.
    Now, if you encode your transfer and choose the bitrate in each scene in a way, that you are always reasonably above this artefact threshold, the result will be certain amount of space that you need (average bitrate * running time). If you simply don't have the space for this on the DVD (due to extras, single layer or whatever), you need to make a decision:
    a) Either you save bitrate in several scenes, thus surpass the artefact threshold which results in visible MPEG blocking (esp. in darker scenes) and mosquito noise artefacts.This is obviously a bad choice, but if used carefully isn't too obvious, especially if you have your black-level calibrated properly, becaues blocking is most noticable in the lowest IRE levels, which are hardly distinguishable once calibrated.
    b) Or you reduce the overall entropy of the image by filtering away the highest frequencies. Now, as mentioned above, a lower entropy means that you need less bitrate to achieve that same artefact threshold! So you can now lower the bitrate slightly in all scenes without introducing any blocking or mosquito noise, but get a slightly softer picture. Like so many things, its a tradeoff.
    But, if you have the space available, like on the extra-less superbit titles, you can use the opposite reasoning. You can choose NOT to filter the content at all, leaving the original entropy intact. If you do so, you will need to raise the bitrate in most scenes to a higher level to achieve an artefact free image. If you have a full dual-layer disc to spare for a movie with not-too-long running time, you can even run the bitrate close to its possible maximum most of the time.
    Col/Tri's transfers from day one used route b) to dodge compression artefacts. Their transfers are always slightly soft, never quite reaching the highest amount of achievable detail as Criterion's 'The Rock' , Warner's 'The Pledge' or Fox's 'X-Men'. On top of that, their encoder is said to be the most efficient, thus yielding the least amount of artefacts for a given bitrate. Combine these two issues and you will understand why Col/Tri transfers over the years constanty yielded among the least amount of compression artefacts. Compare early (1997-98) Warner transfers to Col/Tri and you will see what i mean.
    Now, given the above, why in gods name would Columbia introduce 'Superbit' titles (which have higher bitrates) if their transfers are among the least compression artefact plagued off all? Easy: the higher bitrate allows them to open the throttle on detail while remaining their anal standards for compression artefacts. As easy as that.
    At least thats what i hoped for in theory. But thats exactly what happened on Fifth Element SB. The horizontal resolution, that was filtered on the old transfer is now completely unfiltered, yielding the highest amount of detail possible on DVD, and indeed, it matches the magnificent transfer of 'The Rock' in this regard.
    Edge Enhancement:
    Probably to compensate for their slight filtered softness, Col/Tri transfers always show an unpleasing amount of edge enhancement. I won't go into this again, search posts under my name and read my EE guide (in signature) if you want more info.
    While i was quite certain (or at least hopefull) that Col/Tri would open the throttle on detail on their SB titles, the first thing that popped into my mind was "maybe they reduce the EE while they are at it!". Note that i tried to clarify in some recent threads that by simply increasing the bitrate, you don't affect the amount of EE artefacts at all. You decrease a different edge artefact called mosquito noise, but the trace-contoured halos that us videophiles hate so much, stay where they are.
    But i am ULTRA happy to report that at least on the Fifth Element Superbit DVD, they did indeed reduce at least the horizontal EE a considerable amoung. Fifth Element never was one of the very strong EE contenders anyway, it was just a bit too much for the most picky souls like me.
    A concrete analysis of the improvement is that the amplitude of the ringing stays the same, but the frequency increased by a fair amount, which means that the halos are less disturbing, because they are thinner.
    Hoooray!
    Everything else:
    The Fifth Element transfer has always been reference quality in every other aspect: Shadow detail, contrast delineation, black level, color tone and saturation. Perfect. And the new SB transfer does indeed inherent all these characteristics. In all these aspects they look IDENTICAL.
    I have read reviews where people found a difference in shadow detail, better colors (esp. in the faces) etc... Must be a psychological thing, because its not true. Apart from the higher resolution and less compression artefacts in darker areas (remember Fifth Element was a single layer disc), the discs look completely alike. Print defects etc. are the same as well, of course the same master was used.
    Conclusion:
    I have done over 50 screenshots each to compare all the difference (yes, i was excited to get my hands on this) and will post some later, that will demonstrate each and every point i made above.
    How do i rate the improvement? On my setup with my eyes, the improvement isn't subtle. The picture is so much more detailed, so much clearer. The difference, at times, is as big as going from a non-anamorphic transfer to an anamorphic one, only here in the horizontal and not in the vertical dimension. In other scenes, the improvement is still noticable but not quite as dramatic. Although the term is abused, the image looks a lot more '3-dimensional'. Thats what added resolution does.
    Few people have a system as revealing as mine. Nor do most people have as critical an eye as i have. On my second, lesser, system, i only see the difference slightly because i know its there. So i am not surprised that most people will have a hard time seeing the difference and calling the new DVDs a scam.
    But for videophiles this is great news. I really hope the other Superbit DVDs show similar improvements. Although Dan at DVDFile already posted proove that on the Air Force One SB, the EE halos are NOT minimized. Grrr! I won't comment on that, before i have more Superbit DVDs in my hand (should have all next week). Read his review here .
    This is at least a step in the right direction for Col/Tri, whose transfers just never reached the best of the other studios, detail wise:
    (not supposed to be comprehensive! No animation listed on purpose)
    Warner: The Pledge
    New Line: Blade, Seven
    Criterion: The Rock
    Paramount: Braveheart, Rules of Engagment, Ferris Buller
    Fox: X-Men, White Men Can't Jump
    Universal: U-571, Pitch Black
    MGM: Hannibal, Tomorrow Never Dies
    Dreamworks: Gladiator, Saving Private Ryan
    Disney: Remember the Titans, The Insider
    Starship Troopers from Col/Tri being their best effort yet (no EE, high detail). The rest just look alike, the 1.85:1 transfers (Hollow Man, Vertical Limit) being better than the 2.35:1 ones, which i wouldn't choose a single one from as being reference.
    But this new Fifth Element SB transfer is up there with the best of the other studios. So while its a shame that Col/Tri needs a 'Superbit' label to achieve what others have been doing for some time now, its still a good thing!
    Best regards
    Bjoern 'Anal and proud' Roy
    P.S: screenshot comparissons coming up soon if folks are interested
    ------------------
    "Never underestimate the predictability of stupidity" (Bullet Tooth Tony in 'Snatch')
    My HT in action (Screenshot Page) | My Ultimate 'Edge Enhancement' Guide | My DVD/LD SPL page
    [Edited last by Bjoern Roy on October 06, 2001 at 06:43 PM]
     
  2. Matt_Stevens

    Matt_Stevens Supporting Actor

    Joined:
    Dec 3, 2000
    Messages:
    747
    Likes Received:
    0
    Trophy Points:
    0
    I feel like I was just in school after reading all that! [​IMG] Good work. I hope your screenshots are BIG.
    ------------------
    www.deceptions.net/superman
     
  3. RobertR

    RobertR Lead Actor

    Joined:
    Dec 19, 1998
    Messages:
    9,951
    Likes Received:
    340
    Trophy Points:
    9,110
    Now Bjoern, why'd you have to go and give me a reason to want the Superbit 5E? [​IMG]
    ------------------
     
  4. Reginald Trent

    Reginald Trent Screenwriter

    Joined:
    Feb 18, 2000
    Messages:
    1,313
    Likes Received:
    0
    Trophy Points:
    0
    Bjoern, just curious about what your system consist of?
     
  5. Mark Turetsky

    Mark Turetsky Supporting Actor

    Joined:
    Aug 7, 1998
    Messages:
    832
    Likes Received:
    0
    Trophy Points:
    0
    I'd love to see some full res uncompressed screenshots to show the difference. (I mean, anything less would miss the point, n'est ce pas?).
    ------------------
    [​IMG]
    "Sex? We're men! We wrestled!"
    -Barton Fink
     
  6. Bjoern Roy

    Bjoern Roy Second Unit

    Joined:
    Oct 15, 1998
    Messages:
    315
    Likes Received:
    0
    Trophy Points:
    0
  7. PatrickM

    PatrickM Screenwriter

    Joined:
    Aug 10, 2000
    Messages:
    1,138
    Likes Received:
    0
    Trophy Points:
    0
    Bjorn,
    Another very informative post. I'm still on the fence about getting SB titles though since I don't have my reference video setup yet but maybe I will just to have it for when I do get upgraded.
    Thanks again,
    Patrick
    ------------------
    My DVD Collection
    Patrick The 69th most popular name for boys according to the Social Security Administration.
     
  8. DanR

    DanR Supporting Actor

    Joined:
    Dec 27, 1998
    Messages:
    676
    Likes Received:
    0
    Trophy Points:
    0
    Bjeorn,
    Nice review. Most of the SuperBit reviews on the internet, in my opinion, are cases of "psychological confirmation bias". If you read most of the DVD forums, you can easily see how many people out there want this SuperBit idea to fail. My feeling is that people cannot psychologically deal with the fact that a new DVD can look better than the first version they have already purchased. It's human nature to think "you have the best there is." Thus, most people have immediately passed SuperBit off as a marketing gimmick or CTHV trying to rip everybody off.
    I planned to pick a few of these new discs up. I cannot wait to see Bram Stoker's Dracula on December 11.
    Regards,
    Dan
     
  9. Matt_Stevens

    Matt_Stevens Supporting Actor

    Joined:
    Dec 3, 2000
    Messages:
    747
    Likes Received:
    0
    Trophy Points:
    0
    A lot will also have to do with the size of your screen. Anyone viewing these on a 32 inch TV will have a hard time seeing a difference. But if you have a RP-HDTV, then I think you have a much better chance of seeing a difference.
    Ever since I moved from a 36" Sony Wega to a 47" Panasonic HDTV, I can say that my eyes were opened to the world of artifacts, which I just could not see on a smaller TV. DVD's that I had thought to be perfect, were clearly flawed by artifacting that is visible on a large display. Especially in darker scenes or fade outs. [​IMG]
    DVD's with a very high bit rate tend not to have these problems. [​IMG]
    ------------------
    www.deceptions.net/superman
     
  10. RobertR

    RobertR Lead Actor

    Joined:
    Dec 19, 1998
    Messages:
    9,951
    Likes Received:
    340
    Trophy Points:
    9,110
    My projector is essentially the same as Bjoern's (NEC XG135LC), as is my player (Radeon based HTPC), so I should see similar results [​IMG]
    ------------------
     
  11. Bjoern Roy

    Bjoern Roy Second Unit

    Joined:
    Oct 15, 1998
    Messages:
    315
    Likes Received:
    0
    Trophy Points:
    0
    Here we go.
    Screenshot 1-0-1
    Showing a downsampled or highly recompressed alteration of a screenshot would obviously completely miss the point. But even showing the raw screenshots would be a compomise. Why?
    Well, DVD frames consist of only 720x480 pixels for NTSC. People often argue that showing these frames upscaled to any resolution above that (1024x576, 1280x720 or 1920x1080) is a waste of time, because "there isn't more information than 720x480 anyway" or "you can't create detail from nothing".
    While both statements are true, they COMPLETELY miss the point!
    Upscaling 720x480 to a higher resolution is not about creating 'detail' that isn't there in the first place. Its about presenting those 720x480 pixels on a display device with the least possible amount of high frequency aliasing noise.
    Huh?
    Ok, only because you have 720x480 SAMPLES of an analog medium (like film in this case), doesn't mean that displaying them AS SQUARE PIXELS WITH SHARP BOUNDARIES on a display device is the right form of output to reapproximate the smooth analog waveforms it originally captured! By upscaling digital samples to a higher resolution, you interpolate a smooth waveform between the discrete samples. The more interpolation steps the better. The ideal would be to use a sinc filter that yields a sine wave, but thats currently too demanding processing power wise. A typical interpolation that is used is bicubic filtering, which is already better than bilinear filtering, but not quite spline or sinc quality.
    If you are a bit familiar with digital information processing in the audio domain, you will see that this is exactly the same concept that is used in DACs in CD players for years. Its called oversampling. Think of the same thing in 2 dimensions and you are there.
    So, again. Displaying 720x480 samples as square pixels means that the difference between the square waveform and the ideal sine waveform is noise that masks the actual detail in picture. This is what we perceive as 'aliasing'. The harsh edge of the square yielding very high noise in the frequency domain (fourier).
    That is why watching a DVD on a high resolution device (e.g. 9" CRT) at a high resolution like 1920x1080 does actually reveal MORE detail (of that original 720x480) than a 800x600 digital device, because actual detail is masked by less aliasing noise that distracts the eye.
    Thats why i use the following strategy when doing screenshots analysis. I scaled the 720x480 grabs to 1920x1080 also with a bicubic filter to mimic what a Faroudja 3000/5000 or HTPC would do. To make my own comparissons, i watch these images on my monitor (which is capable of resolving 1920x1080) with a program that is able to switch between 2 images almost instantly (
     
  12. Robert George

    Robert George Screenwriter

    Joined:
    Jul 3, 1997
    Messages:
    1,102
    Likes Received:
    88
    Trophy Points:
    1,610
    I see it too, Bjoern.
    Now tell me about the DTS, preferably with graphs.
     
  13. Rob Gillespie

    Rob Gillespie Producer

    Joined:
    Aug 17, 1998
    Messages:
    3,632
    Likes Received:
    5
    Trophy Points:
    0
    Impressive. Most impressive.
     
  14. Bjoern Roy

    Bjoern Roy Second Unit

    Joined:
    Oct 15, 1998
    Messages:
    315
    Likes Received:
    0
    Trophy Points:
    0
    Obi,
    you are sarcastic here, right? [​IMG]
    Since when do you care about the DTS diff.? Or do you want me to show that they measure identical, so that we don't hear 'has lots more bass' posts? [​IMG] Admit it!
    ------------------
    "Never underestimate the predictability of stupidity" (Bullet Tooth Tony in 'Snatch')
    My HT in action (Screenshot Page) | My Ultimate 'Edge Enhancement' Guide | My DVD/LD SPL page
    [Edited last by Bjoern Roy on October 06, 2001 at 12:14 PM]
     
  15. Jens Raethel

    Jens Raethel Second Unit

    Joined:
    Oct 27, 1998
    Messages:
    473
    Likes Received:
    1
    Trophy Points:
    0
    Real Name:
    Jens
  16. Robert George

    Robert George Screenwriter

    Joined:
    Jul 3, 1997
    Messages:
    1,102
    Likes Received:
    88
    Trophy Points:
    1,610
    Bjoern:
    No agenda on DTS, just curious. I have no idea how these two tracks will compare as I don't yet have the discs. I don't have the ability to do the graphical measurements and I'd like to see how they measure. BTW, I don't really expect they will be identical in the lower frequencies and the highest frequencies, but I'll leave the graphs to you and we'll see what we see.
    If the DTS track sounds better, then that is what I'll listen to. Really. [​IMG]
     
  17. RobertR

    RobertR Lead Actor

    Joined:
    Dec 19, 1998
    Messages:
    9,951
    Likes Received:
    340
    Trophy Points:
    9,110
    It's nice to SEE what you're talking about instead of just stating your opinion, Bjoern. Thank you. And yes, I do see the difference. [​IMG]
    ------------------
     
  18. Chris Maynard

    Chris Maynard Supporting Actor

    Joined:
    Nov 7, 1998
    Messages:
    644
    Likes Received:
    0
    Trophy Points:
    0
    Excellent pictures. There is no arguing the difference.
    I am also interested in the two audio tracks graphed.
    Thanks!
     
  19. Luis Gabriel Gerena

    Luis Gabriel Gerena Second Unit

    Joined:
    Jul 18, 2000
    Messages:
    410
    Likes Received:
    0
    Trophy Points:
    0
    WOW! Great report and as I expected WSR comments were way off. Still I've got to see with my not so cutting edge projector but I am hoping for the best and in any case is a low risk move and the probability of getting better image quality makes it worth it.
    Regards
     
  20. DanR

    DanR Supporting Actor

    Joined:
    Dec 27, 1998
    Messages:
    676
    Likes Received:
    0
    Trophy Points:
    0
    [cynicism]Tapping fingers waiting for . . .
    Stage 1: The Begninning. DVD Forums around the internet erupt with links to this very thread.
    Stage 2: Denial. "The nerds at HTF have nothing better to do." Followed shortly thereafter by "Yeah, there's no difference unless you have a $500,000 system."
    Stage 3: Anger. "Why did Columbia Tri-star screw us on the first release by purposely degrading the video on the first releases?" "Fuck Columbia, I'm not buying anymore of their DVDs."
    Stage 4: Resolution. "Well, I'm going to check this out and see if it's true." "Wow, those guys at HTF were right." (for reference check out the "nerd bass charts" at HTF regarding Jurassic Park DTS)
    Stage 5: Equilibrium. Every DVD reviewer on the internet includes in his/her reviews the now "mandatory" commentary on "horizontal resolution", video "bit space", "fine detail resolution", etc. [/cynicism]
    This only proves once again that we have lost. Read all the comments on the net about SuperBit; most "revolt" against the idea of SuperBit. The majority of people utilizing DVD and participating on the web are not "enthusiasts" as most of us are.
    Oh well, some of us will carry on as we always have, seeking the best theater experiene in our homes. Let the close minded fools miss out on the experience.
    Regards,
    Dan
     

Share This Page