-

Jump to content



Sign up for a free account!

Signing up for an account is fast and free. As a member you can join in the conversation, enter contests and you won't get the popup ads that guests get. Click here to create your free account.

Photo
- - - - -

THX needs some feedback


This topic has been archived. This means that you cannot reply to this topic.
60 replies to this topic

#41 of 61 Mark Techer

Mark Techer

    Agent

  • 48 posts
  • Join Date: Sep 21 2005

Posted September 24 2005 - 12:47 AM

"Put the subtitles on the top portion of the bottom black bar!!! ...for 2.35:1 movies of course."

3 weeks ago I would agree, but since upgrading to a constant height setup, I think that subtitles should be in the picture area as they are on the actual cinematic film release...

Mark

#42 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted September 24 2005 - 05:32 AM

say guys: subtitle location can be adjusted in the source! As long as the subtitles are coded as subtitles and not part of the actual image itself (thus not adjustable in location), I'm happy.

#43 of 61 Mark Techer

Mark Techer

    Agent

  • 48 posts
  • Join Date: Sep 21 2005

Posted September 27 2005 - 12:44 AM

Some question about the optimiser...Reading this thread suggests (and so do the instructions on the disc for that matter) that each optimiser is different. Why would this be done? Isn't all video mastered at D6500K? And if so, why, once calibrated, would you need to adjust anything?

I use JPK DVE and have cross checked with all of the optimisers that I have. The only discrepancy I found is colour on PAL. There is a difference between the PAL and the NTSC versions where I can not get the the PAL version to look like 6 bars through the THX Blue Filter Glasses, but the words Color and Tint do match on R1.

With JPK DVE, the Grey back ground and Blue, Cyan and Magenta patches match when looking throught the Blue Filter, as does the Red, Magenta and Yellow against the grey background when looking through the red filter. I do have a Green error that I can not fix. There is a clear differece between Green, Cyan and Yello but the Green and Grey are close.

But of course I am doing this by eye...

Mark

#44 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted September 27 2005 - 04:33 AM

Quote:
And if so, why, once calibrated, would you need to adjust anything?

Which is precisely why the original idea was a terrible one.

#45 of 61 Mark Techer

Mark Techer

    Agent

  • 48 posts
  • Join Date: Sep 21 2005

Posted September 28 2005 - 09:40 AM

So quickly back to the THX optimizer colour test pattern. The inconsistency is the same for all of the optimizers I have. Are they really different?

Mark

#46 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted September 28 2005 - 10:17 AM

Some are off, and some are correct. The ones that are off are probably correctly implemented as per THX's original idea behind optimode, which is a very bizarre one. The ones that are accurate patterns aren't implemented as per THX's original idea, but to tell whether the patterns are accurate you need to check the levels on them.

I don't know what patterns on what movies you mean, and I haven't looked at many, but you'll need to look at the pattern's encoded digital levels to see whether they are correct or not.

#47 of 61 Mark Techer

Mark Techer

    Agent

  • 48 posts
  • Join Date: Sep 21 2005

Posted September 29 2005 - 02:40 AM

Sorry I don't have access to that kind test equipment...

Think I'll stick with DVE...

Mark

#48 of 61 Mark Techer

Mark Techer

    Agent

  • 48 posts
  • Join Date: Sep 21 2005

Posted October 02 2005 - 01:42 AM

Flagging of EX/ES.

Most of the original DD EX discs like SW1 and Fight Club were not flagged.

Today many, but not all are, and what I have found is that sometimes, if the decoder is set to "on", the info that should go to the back surrounds gets lost on flagged discs. It does not just go to the LS and RS, but is not heard at all. I have found this on several DVDs and now have reached the conclusion that if the disc is flagged, then the decoder must me set to "auto" and not to "on".

The problem is that if other decoders do this, and their owners have not worked it out, then they might very well be missing out on some important audio.

Personally I think that given the fact SW1 was not flagged, that no discs should have been flagged. I have also experienced this with DTS ES, both 5.1 matrix and 6.1 discrete...

Mark

#49 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted October 02 2005 - 04:47 AM

Mark: not sure what you mean about it being lost, that sounds like a decoder problem, not an encoding issue. I agree though that discs should be more accurately flagged for both cadence and soundtrack.

#50 of 61 Mark Techer

Mark Techer

    Agent

  • 48 posts
  • Join Date: Sep 21 2005

Posted October 02 2005 - 04:45 PM

I'll give two examples "lost" -
1. T2UE chapter 3(?) future war. A terminator aims two guns and fires right at camera. In the original 5.1 mix, this sound effect was from centre, but after sweetening for EX, is now heard in the back surrounds. If the decoder is set to "auto" all is OK, but if set to "on" is not heard. The sound is not steered to LS/RS it just is not there as if the back speaker's cables had removed.
2. JKP DVE (can't recall the exact chapter) where the narrator talks about what should be heard in the surrounds. The narrators voice comes from back surrounds, but again, if the decoder is not set to "auto", there is no narration, just the sound effects in the LS/RS.

I'm not sure if it just my decoder, or if anyone else has experienced and identified these problems. It just becomes annoying if I forget to turn the decoder to "auto" for flagged discs and even more annoying if I leave it in "auto" as I miss out on back surrounds from all of the unflagged discs that are out there.

THX stated in designing Ultra 2, that they wanted one system, one speaker layout that would work for everything, and the system would take care of the processing. It appears that this is not the case if I have to reset the ASA for each soundtrack. This problem could be rectified by simply not flagging the discs...

Mark

#51 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted October 02 2005 - 07:49 PM

I've never run into this problem, I do think something is wrong in your decoder or your setup. Are you sure your rear speaker is connected and functioning? That it does not on DVE leads me to believe something is amiss in your system, not anything related to disc mastering.

#52 of 61 Mark Techer

Mark Techer

    Agent

  • 48 posts
  • Join Date: Sep 21 2005

Posted October 02 2005 - 11:38 PM

I may be new to this site, but I have been in HT since 1990.

I calibrate my system at least once a week (don't ask why, I just do)and know very well that both back surrounds are working and giving me a reading of +75dB/c/slow on my SPL meter.

What I can not explain is why I am losing this information when the incorrect decode mode is chosen. As I have stated, this only happens if the decoder is set to "on" for a flagged disc instead of "auto".

Another example is Finding Nemo. Bruce the shark swims over head during the shark's meeting when Dori is allowed to speak, and again, if set to "on" you get sound pan from screen channels to the surrounds, if set to "auto" the pan extends to the back surrounds and is much more enveloping. This sounds different to if the decoder was switched to "off", where the back surround info is played through the LS and RS...

Mark

#53 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted October 03 2005 - 08:40 AM

This is an obvious Q, but something unclear to me because I don't know your system: do you have a 6th channel in your array? If you force your receiver to do EX/ES, you're sending that sound to a non-existent speaker. Other than this, something is amiss in the design of your receiver it seems.

#54 of 61 Mark Techer

Mark Techer

    Agent

  • 48 posts
  • Join Date: Sep 21 2005

Posted October 03 2005 - 02:35 PM

Hi Chris,

Sorry if I missed that point, yes I have four surrounds - 2 at the sides and 2 at the back of the room connected to a 7.1 THX receiver.

I am using four identical bipoles (not dipoles as is specified by THX) that are mounted 90 degrees to the listening position - that is they fire front and back, not at the listening position. The pairs are located at +/-90 degrees for LS and RS and +/-150degrees for the LBS and RBS.

I moved away from dipoles because I prefer the slightly more directional sound field I get now. I also used the +/-150 degrees positions for the backs surrounds as opposed to two centred speakers as described on the website for ASA for two reasons -

1. cinema surround arrays are supposed to cover from +/-60 to +/150 degrees, and
2. when seated off the centre axis, the sounds from the back surround tend to pull if using one back surround or if a pair is too close. To achieve a wider sweet spot requires the back surrounds to be spaced at least the distance of the width of the seating area.

Even with four directional surrounds (+THX processing Adaptive De-correlation etc), the ambience is restored because of the many arrival times of sound to my ears, including a rear presence.

I say slightly because what I am listening to is still diffuse (reflected and reverberent sound, not direct sound) but because the pairs of drivers on each speaker are in-phase, the speaker becomes more directional, even 90 degrees off axis.

I love my THX system, but I wish THX would not make their technology so secrete. No-one is asking them to publish the exact curve of Re-eq, but more public awareness is needed if more people are to adopt the technology, including that if using four identical dipoles, that you have to swap the left and right assignment for the back surrounds to eliminate a phase error that occurs between LS and LBS and RS and RBS...

Mark

#55 of 61 James D S

James D S

    Screenwriter

  • 1,002 posts
  • Join Date: Nov 14 2000

Posted October 27 2005 - 02:54 AM

To echo what others have said, no edge-enhancement and anamorphic throughout.

I'd like skippable menus and non-forced previews, as well - afterall, I already paid for the movie.

#56 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted October 27 2005 - 05:15 AM

Is THX observing this thread, are they going to say anything?

#57 of 61 DaViD Boulet

DaViD Boulet

    Lead Actor

  • 8,805 posts
  • Join Date: Feb 24 1999

Posted October 28 2005 - 02:56 AM

Hey THX guys,

I've been one of your biggest critics (due to so many AV encoding problems I find on THX-certified Disney dvds) so please bear with me and keep an open mind to read my comments.

For video it's pretty simple for me:

Do a decent 16x9 HD transfer, down-convert to DVD resolution and LEAVE IT ALONE before compressing.

Most often what screws up DVDs that I've seen is unnecessary HF filtering in a misguided effort to "help" compression...then ringing and electronic noise are added to "sharpen" the image back up and all of this creates an image that looks objectively compromised on the big-screen.

The best DVDs are those that are just left alone that keep as much of the natural image detail in tact as possible and don't add any visible ringing.

As for audio, I actually find that the same is true...leave it alone. The 5.1 AC-3/DTS tracks on laserdisc that still sound "the best" were the theatrical mixes that were just compressed and put on laserdisc...without "reprocessing" for the home-theater environment. Also, THX should not approve any DD encoding that has dialogue-normalization or any other compromising processing feature applied. Many times when I've preferred the DTS soundtrack on a DVD presentation, the audio engineer contact me to tell me that the two mixes were 100% identical except for dialogue normalization which had been applied to the DD encoding. That tells me that dialogue normalization, whatever it's intended effects may be, it NOT the friend of a high-end audio system.

THX should not be certifying that a disc will look and sound good on an inferior playback system...it should be certifying that it will look and sound good on a REFERENCE system.

Also regarding audio, avoid noise-reduction to get rid of hiss..Fox screwed up the audio on Hello Dolly by trying to get rid of the Hiss from the mag tracks...and they wiped out all the musical detail in the process (the AC-3 on laserdisc is the exact same mix without the added noise reduction and sounds great). Also, just listen to the bastardized "DEHT" audio on the new Mary Poppins DVD if anyone wants to hear what NOT to do to audio. Words cannot describe how horrible that "DEHT" remixed audio presentation turned out... (not all DEHT are bad...Lion King, Bambi, and Cinderella are great, Mary Poppins and Aladdin are bad)

If the THX guys are really interested in these issues, it might be advisable to read some of the DVD reviews I've written.

Later I'll post some links to some review threads I think might help illustrate to THX what these issues are about.


One thing I can say for sure:

Studios should critically evalutate all DVD product upscaled to a minimum of 720P on properly calibrated hi-def front projection systems at a 1.5 screen-width distance before approving any DVD authoring. Direct-view CRT monitors will not do...not matter how fine the dot-pitch. That 1.5 screen-width distance is cricital. When I see ringing on Disney's Mulan in my home-theater...I can't help but think that had someone at the studio viewed the disc in the same high-resolution, wide-angle manner, it never would have passed the quality-check test.

With new projector's like Sony's 1080P Ruby streeting at 10K, there's no reason why a studio can't simply go get one, have it calibrated, and spend a few minutes spinning the discs on it in a viewing room before letting them hit the stamper.

Please allow me to repeat a comment I made earlier:


THX should not be certifying that the disc will look and sound good on inferior playback systems (it shouldn't even be concerned with that at all)...it should be certifying that it will look and sound as close to the original source as possible on REFERENCE systems.

There is NO BENEFIT TO THE HT COMMUNITY in down-grading a format's ultimate performance so that it looks or sounds better on "lesser" gear. And it doesn't matter if most consumers have inferior gear or not. I have NEVER heard a consumer with average gear complain about the image quality of a DVD that was optimized for a reference system...but I *have* heard many HT enthusiasts with carefully calibrated, high-quality playback systems complain that their AV quality was compromised. It should be obvious which end of this spectrum THX ought to serve, and the "middle" serves NO ONE at all.

I think if THX could define their goal as the above, the HT community would benefit greatly. If a DVD can't meet those standards after it's authored and pressed, then it shouldn't have the THX seal of approval, no matter what the political costs in denying it.
Be an Original Aspect Ratio Advocate

Supporter of 1080p24 video and lossless 24 bit audio.

#58 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted October 28 2005 - 05:30 AM

Quote:
Studios should critically evalutate all DVD product upscaled to a minimum of 720P on properly calibrated hi-def front projection systems at a 1.5 screen-width distance before approving any DVD authoring. Direct-view CRT monitors will not do...not matter how fine the dot-pitch.

I agree, however, I do have issue with the criticism of CRT as the reference monitor. There are times where viewing on a small monitor will not reveal flaws in the same way as on a larger screen, however the reverse is also clearly true. Both is probably the best bet, though I still remain biased towards proper mastering with a CRT-based display as the reference. In my knowledge, the only limitation that this usually provides in a final transfer is perhaps banding visibility in HD material, as can be seen on the IMAX WMV-HD titles, for instance.

#59 of 61 DaViD Boulet

DaViD Boulet

    Lead Actor

  • 8,805 posts
  • Join Date: Feb 24 1999

Posted October 28 2005 - 06:12 AM

If the shadow mask/dot pitch on the direct-view CRT monitor is fine enough, it's possible for CRT to do an acceptable job if the viewer moves close enough to the screen.

However, even many "data grade" CRT displays impart a noticable grain structure at a 1.5-screen-width viewing distance that obscures fine detail...and that should be the test viewing distance ratio regardless of display technology.

And even if care was taken to select direct-view monitors that were "transparent" at a 1.5 screen-width distance, trying to convince authoring/compression artists to move that close to the screen is often impossible. Sitting even 2 screen-widths away from the image feels "really close" when using direct-view displays, but it's still MUCH TOO FAR AWAY from the image to critically evaluate the details as they would be apparent to front-projection viewers.

The DVD medium is capable of delivering an astonishing film-like, high-quality image presentation at 1.5 screen widths when properly optimized and mastered. THX certified DVDs should verify that this level of reference-quality is maintained, and nothing less.
Be an Original Aspect Ratio Advocate

Supporter of 1080p24 video and lossless 24 bit audio.

#60 of 61 ChrisWiggles

ChrisWiggles

    Producer

  • 4,791 posts
  • Join Date: Aug 19 2002

Posted October 28 2005 - 12:06 PM

You are correct, however, this is not the only attribute of a display that is important for mastering. Things like chromaticity accuracy are very important, as well things like if you are using a digital display, the pixel response time. There's other problems with DLPs in shadow detail rendering because of PWM dithering artifacts, etc. VB on LCDs and insufficient CR makes along with SDE makes this an inappropriate choice for mastering.
If you're looking at using a digital display at 720p, I don't feel this is at all sufficient for supporting a 1.5 viewing distance without pixel smearing, which is not desireable in revealing source artifacts and problems, but is for subjective viewing. A direct-view CRT will provide larger viewing angle capabilities most likely, especially if you are considering something like a 720pLCD. These resolutions are insufficient for content mastering. Some studios have explored using 1080pLCOS machines, but there have been problems in seeing things like banding and fast noise because of slow response time on something like the JVC. So while I'm all for large-screening as well, I think we should be careful before advocating a major shift in mastering practice, especially considering that many of the other behaviors of CRT displays are very different fromo digital displays, and the standards are designed with the CRT in mind, not a digital display.





Forum Nav Content I Follow