Robert Harris on The Bits - 8/3/04 column - OFFICIAL THREAD (1 Viewer)

[Mike Maloney]

Eric,
In general (as used in the entertainment industry) the term "post" means everything that happens after a performance is recorded.

In the movie industry, post is everything that happens after the film is exposed. In the music industry, post is everything that happens after the recording session whether it was live before an audience or in the studio.

A post- (fill-in-the-blank) is any facility that does something to that original recording. Hence the ubiquitous (in Nashville as well as Hollywood) phrase: "We'll fix it in post."

Some post facilities specialize, others offer a variety of services. Normally, projects use the services of several post-houses, so that the best provider of all the specialities can be used.

Hope that helps,
Mike

(edited on 8-13 to fix the second paragraph. What was I thinking?)

 

[Eric Stewart]

Thanks, Mike ... do I understand you correctly that the "original recording" in this case is the video master transferred from film via telecine, and that the "post" is everything that happens after that, en route to the finished DVD?

 

[Cassy_w]

Mike...

Too much Edge Enhancement:
Patton
Lethal Weapon 4
Devil's Advocate
Cold Mountain
Pirates of the Caribbean (this one is horrendous)
Every Tarantino movie from Disney/BV

Low Edge Enhancement:
Matrix Reloaded (and maybe Revolutions)
Pitch Black
Raiders of the Lost Ark (unless I am remebering the HD version???)
LOTR Extended editions (still a bit much, but far less than any recent Disney title)

Here's a pisser: Shanghai Noon has only an average amount of EE. But Shanghai Knights has way too much EE. This will show you, very clearly, how older Disney titles were OK, with more recent releases being ruined by excessive EE.

Mike, I really do hope you can get to the bottom of all this. I think I speak for everyone here when I say this: NOT ONE PERSON anywhere in the world will complain if BV releases are stripped of edge enhancement/ringing/halos (or whatever you want to call it). In fact, your sales will only be improved, because people like me, who have stopped buying all Disney/BV releases, will start buying again.

 

[Eric Stewart]

David Grove wrote:



Amen to that. Mr. Harris, would you consider pitching the idea to whoever you know in the industry that might green-light such a documentary? Perhaps it could be like a super-extended version of the "Obsessed with Vertigo" program you made with Mr. Katz. In structure and tone it could be like Mr. Scorsese's "A Personal Journey ... through American Movies" but with the emphasis on film restoration/preservation techniques, problems, and results. In fact, since Mr. Scorsese is big on preserving our film heritage, maybe he would like to take a hand in some way.

 

[Robert Harris]

While I appreciate the kind thoughts, my extended discussions in regard to the technicalities of film reconstruction and restoration have been known not only to clear rooms, but may become a new form of relief for insomniacs.

For those who joined this thread after the intermission, it should be noted that Mr. Maloney is not an employee of either Disney or Miramax. He works with a post house involved in the various DVD compressions and authorings.

RAH

 

[Eric Stewart]

Re: the debate about inadequacies of recent Miramax DVDs.

Mr. Maloney, is the following correct?

At the compression/authoring facility where you work, which I’ll nickname “CAFAC,” you typically receive a compression master in the form of a digital tape. It’s standard-definition with (say) a 270 Mb/s data rate, compressed at a ratio of (say) 5.5:1 from an original high-definition video master with (say) a 1.483 Gb/s data rate. The latter master is the original output of the video transfer process. There has been a video downconversion from the original master to the compression master, but the downconversion may not have been (or definitely hasn’t been?) done at “CAFAC” itself. It may have been done (or definitely was done?) elsewhere, it may have had its color timing changed with the approval of the director, etc. Unless there are “obvious technical problems” with the compression master, “CAFAC” simply accepts it as is.

Moreover, “CAFAC” doesn’t do noise/grain reduction or edge enhancement to it.

It looks to me, accordingly, as if some of Mr. Harris’s complaints about recent Miramax DVDs are with the part of the process that turns the original video master into the compression master. Call that the “compression master generation” process (CMG).

And then, a separate group of his complaints are about the compression and authoring itself: in particular, the bit budget allocated to the film, in view of how many extras are slated to appear alongside it on the DVD. Call that the “compression/authoring” process (CA).

Both CMG and CA depend on the quality of the “original video transfer” (OVT), which Mr. Harris and all others concerned seem to agree is probably fine.

(Quote from Mr. Harris: “If [the compression/authoring facility doesn't do them], then where do the grain reduction and EE come from. I'm aware that some can be added during the transfer, but if the transfers or files do not have the problem....?”)

(As a further aside, I’d like to know if the OVT is typically done more than once -- say, once for high-def TV and again for home video. That is, does it even make sense to compare My Fair Lady on Showtime HD with My Fair Lady on DVD? If they’re different transfers right from the get-go, probably not. Maybe the former transfer has no EE and the latter has plenty of EE, for example. It all depends on the definition of “some can be added during the transfer.”)

So, anyway, the sequence is: OVT -> CMG -> CA. Mr. Maloney, you say you can definitely vouch for the “compression/authoring” part of it (at least, given the compression master and bit budget that “CAFAC” has to work with), but can you vouch for the “compression master generation” part? Maybe that’s where most of the trouble is.

Who exactly does the CMG part of the process, anyway? That’s what I’d mainly like to know. Is it done by the facility that does the original video transfer -- perhaps even as part of, and during, the transfer, or perhaps after the original transfer? Or, on the other hand, is it done by some third-party CMG facility that is interposed between the OVT and CA stages?

Can anybody provide some authoritative answers to these questions?

 

[Eric Stewart]

Mr. Harris,

While we're waiting for Mr. Maloney to rejoin this thread with further reports concerning edge enhancement and lost high-frequency information on the Cold Mountain DVD ...

... perhaps you'd like to tell us more about the Sony XBR910 video monitor you mentioned earlier in the thread.

Do you find that this direct-view CRT with its Super Fine Pitch tube lets DVDs look their most film-like because (not to put words in your mouth) it's not a hard-edged, fixed-pixel display?

Did you have yours professionally calibrated? Did it improve the picture?

Did you defeat scan velocity modulation for a more film-like rendering?

What size is it? 34"? 30"?

How close do you sit to it?

When you review a DVD for your column, do you find the XBR910 gives you a more accurate impression of it than your other, larger displays?

Cheers,

 

[Robert Harris]

Do you find that this direct-view CRT with its Super Fine Pitch tube lets DVDs look their most film-like because (not to put words in your mouth) it's not a hard-edged, fixed-pixel display?

****
I find that it offers the finest representation of DVD as well as HiDef broadcasts, with an extremely "film-like" look to films, with superb black levels, saturation and shadow detail.

I believe that this Sony is the finest CRT device made for consumer consumption. They've created a superb product.
****

Did you have yours professionally calibrated? Did it improve the picture?

*****
I do my own calibration.
*****


Did you defeat scan velocity modulation for a more film-like rendering?

****
no. Haven't yet experimented to that point.
****

What size is it? 34"? 30"?

***
34XBR960, sorry if I earlier referred to 910. The 960 is an updated unit.
***

How close do you sit to it?

***

8 feet or so from box
***


When you review a DVD for your column, do you find the XBR910 gives you a more accurate impression of it than your other, larger displays?

***
More accurate in many respects, while hiding certain flaws because of the image size. I find that overall it gives the best image of any device which I own, while the Runco gives the most impressive as a "theatrical" experience.

IMHO, the Sony 34" 960 is bar none, the best monitor produced today. I heartily recommend it to anyone with the space to put its 200 pounds.

RAH

 

[Ed St. Clair]

RAH,

Would you like too comment on a 2001 SE DVD?

Thanks for you last post.

 

[Michel_Hafner]

Interesting thread. Some remarks for Mike Maloney (thanks

for coming to the forum and discuss this):



- EE, level of detail: Please study in detail

this excellent source:

http://www.videophile.info/Guide_EE/Page_01.htm

and this to see how standard and feasible are two very

different things:

http://www.videophile.info/Review/FOTR/FOTR_01.htm

- If a DVD has excessive EE artifacts and ringing there are

basically no excuses acceptable. Especially when the

source is a DI that delivers high quality digital images

to begin with. Even when there are many supplements and

the bit budget is quite limited there must be no EE and

ringing. Lack of high res detail, yes, EE and ringing,

no.

Some MPEG artifacts are unavoidable, but they don't look

like what we have on the Cold Mountain DVD.

- If the EE and ringing is on the compression master you

got whoever made that master did a horrible job and needs

to be told about it. This is not about supplements and

bits. It's about lousy mastering of the NTSC compression

master. Such a master makes no technical sense whatever

you are required to do with it afterwards. EE and ringing

are hard to compress and ugly to look at whether well

compressed by you or not.

- If the EE and ringing is not on the compression master

you added it with your MPEG encoding. This is avoidable

and bad encoding. Even with low bit rates it's neither

necessary nor desirable. Some scene dependent high pass

filtering (that adds no EE or ringing) is enough. If

grain is a problem some grain reduction might be needed.

It adds no EE or haloes but other problems.

- Another issue that ruins many DVD (and hampers even the

quality of DIs!) is noise and grain reduction. More

information about this here:

http://www.cinedrome.ch/hometheater/dvd/dnr/text.html

- Once you know if the EE and haloes are your doing or not

you can help stopping such bad DVDs. If it's on the

compression master tell the mastering house about it.

If YOU added it take your MPEG encoder apart and find

out where exactly it happens and how to turn it off.

If it can't be turned off the encoder belongs in the

garbage can. Unfortunately some decoders have internal

sharpening and even noise reduction that are not (well)

documented and hard or impossible to defeat. :-(



Finally



- Not everybody posting and complaining on public fora is

a bloody amateur. And not every statement and denial

of professionals, studios and facilities are the gospel.

Trust your eyes first, know the limitations of your

equipment and go to the bottom of things. Crap is out

there in large quantities despite all the naysayers, but

high quality work is there as well. It's not so difficult

to tell one from the other on decent display hardware.

- I have complained about EE and DNR issues since DVDs

were born. I see little progress in all these years.

There were great discs back then and there are now.

There was crap than and there is now. It's more a lottery

than anything else.

 

[Marty Christion]

Too much EE:



Tombstone! (Vista series)



No EE:



Remember the Titans





And the examples here of course.



http://www.videophile.info/Guide_EE/Page_01.htm

 

[Robert Harris]

Michel...



Thank you for posting.



As usual, your expertise is a great help in defining the reality of the situation.



RAH

 

[Nils Luehrmann]

I'll gladly second Robert's appreciation for Michel's contributions.

Related to DVD PQ, if you have not already I advise you to read Michel's explanation of Digital Noise Reduction on DVD.

 

[Ed St. Clair]

quote:By the way, allow me to add that your experimentation in attempting to find a solution to this problem is appreciated by all at HTF and sets a most positive tone for your facility.

GREATLY appreciated!!!

Thank you,

and your flame proof suit!



I wore one of those babies, tough too see out of though.

Butt, it beats getting BURNED!!!

 

[Ed St. Clair]

Fantastic screen captures, Ted.

Nice web site as well.

A TRUEly Great HT!

 

[Ed St. Clair]

Michel_Hafner,

I would be interested to see your list of five DVD's with & five w/o EE.

Thank you for your insights & questions.

 

[Ed St. Clair]

Yeah, Marty Christion,

Bjoern is BOSS!!!

 

[Ed St. Clair]

Who authorized Mickey's The Three Musketeers?

Both reviews that are up gave the transfer high marks.

 

[Michel_Hafner]

which in the case of Cold Mountain, was not exceptionally high. Please keep in mind that film frames had already be resolved to 2k as either a printing negative or an IP. Precisely how much information is lost generationally going from the film Oneg to a 2k file and then out to another 2k file, and then finally to film is a most interesting question when compared to an Oneg to an IP to a dupe to a print.

------------------------------------------------------------

I'm not aware of any studies that tried to explore these

chains in detail and compare end results. The current

DCI files and prints seem to be a good starting point.

As far as I know a pristine 2K version from the ON digitally projected matches a print from the ON in apparent detail, but will not be as smooth looking (on a 2K projector).

It will be more stable as well. Color and contrast will

depend on the projector. The dynamic range of a top film

print is not available from digital projectors for the time

being. Only a top CRT projector will give you that, including blacker blacks than I have seen from film so far. But the CRT can not match the digitalos in brightness and sharpness. There are no projectors to fully resolve what's

on a digital intermediate that captures the essence of

a 35mm ON. In this regard putting it out on IMAX comes

closest to see what's on the negative.

The terms 2K or 4K themselves mean not much concerning

the quality. They define only upper limits of possible

spatial resolution. If that resolution is actually used

is another issue. The recent Spiderman 2 was 4K (4K

scanned, downres to 2K, 2K SFX, upres to 4K) but was it

sharp as hell? No. Was it free of aliasing? No. Actually

the end credits had the worst aliasing I have seen in

a long time on a digital intermediate. These credits were '4K' rubbish.

To see good 4K watch one of

the IMAX jobs on Matrix. They have issues too but at least

there is often 4K detail up on the screen with good contrast.

I'm rather skeptical concerning this digital intermediate

technology as long as it has not progressed beyond the

current state of infancy where film and digital artifacts

in long processing chains often add up to more than dubious end results on the local screen. The potential is there,

but with traditional printing chains it will never reach the end user, and with silly ideas like film grain needs to be reduced and dust busted automatically on principle we are heading for less image quality, not more.

 

[Michel_Hafner]

My list of 5 EE free and EE 'horror' films?

Ask me something easier.

To say a DVD has zero EE is quite daring.

Usally the HD transfers already have EE or HF boosting

of some kind.

One's own display equipment often adds some ringing.

A DVD with little to no EE is Star Wars 2 mastered from

the HD 1080p source. Some of the direct digital Pixar

titles have little to no EE.

Finding 5 DVDs with lots of EE should be easy. You

will find examples on Bjoern's site.