[tig] RE: I'm worried and angry

sklein54 at earthlink.net sklein54
Sun Oct 19 22:17:09 BST 2003


I wish to wager $40 American on mhafner. 

-----Original Message-----
From: mhafner at imdb.com
Sent: Oct 19, 2003 3:56 AM
To: tig at tig.colorist.org
Subject: [tig] RE: I'm worried and angry

thanks to Thomas Tomlinson for supporting
the TIG.
---

Good morning,

#The amount of grain reduction that is applied either with hardware or =
#software is strictly the creative's choice and not Technique as a =
#company doing Digital Intermediate films.

I agree that the final decision is surely the client's since he pays.

#I resent Mr. Hafner's bashing =
#of Technique with not obviously knowing the decision process during the =
#session.

I do not intend to bash Technique. I am bashing the look of the film and
I'm questioning the creative decision behind it.
Also, I'm not willing to think of DI post houses as mere execution factories
of all demands coming in from all sorts of clients. I see the role of such
houses more like that of a doctor, aware of illnesses and the cures available
for them, but also aware of what the basics of a healthy life style are and
how it helps to avoid getting into trouble in the first place.
If a patient comes in and says: Doctor, doctor, I have been smoking too much!
I can clearly feel that the flow of blood is not what it should be in my right
leg. Doctor, amputate my right leg! 
And the doctor says: Stop smoking! I have looked at your leg. It's not in
good condition. But if you stop smoking now and apply this cream twice a
day it will improve.
Patient: No Doctor. I will not stop smoking. And the leg has to come off!
I'm sure it will be my doom if it's not cut off right now! Do it!
...
Should the doctor cut it off? Should he send the patient to another doctor?
If he does cut off the leg, does he use this knife or that saw, this
sterilization procedure or that one?
It's not 100% the same situation, but neither is it very different. The
expertise of both parties for different aspects of the whole issue is uneven
and if one does not listen to the other or does not make a good case for what
is right and what is wrong both parties may very much regret what they did
later on. And the actual technical details of the implementation of the
client's wishes are very much the post houses responsability. DNR comes in
all kinds of quality levels, and it's not just the amount of reduction
that is asked for which determines the level of artifacts. All of this
is not meant to propose anything specific about what happened at Technique.
I merely have to say that the leg was cut off and I see no need for it.

#Mr. Hafner should respectfully meet with =
#Mr. Tarantino and Mr. Richardson to discuss their creative decisions

I would not object to that at all. The chances they are interested are
very slim, though, I'm afraid.
It's kind of funny and fitting that another thread here is about audio.
There are clear parallels between the development of digital audio and
the development of digital film (processing). Audio is about 10 years 
ahead of the learning curve and had the same teething troubles. There
was a time (maybe still is) when analogue tape hiss and cracks and pops
from records were deemed unacceptable and agressive digital filtering
was used to get rid of it. Unfortunately they also got rid of the natural
timbre and higher frequency parts of the original signal representing
instruments and human voices. The result no longer sounded like real
music. The baby had been thrown out with the bathwater. As a result
criticism rose and the filters were improved, the noise better separated
from the useful signal and noise reduced while keeping the music sounding
like music. Are we doomed to have the same cycle with film?

#Look, all non-trivial work goes through a fairly rigorous process of
#viewings and approvals.  I don't think I've ever seen a client simply drop
#off the film and offline list and simply walk away.  Returning several weeks
#later to pickup his finished 100 million dollar motion picture.  The process
#is supervised.  Highly supervised.

I believe you. All the more puzzling what I see. It makes no sense to me.
Zero, nada. There are some very unpleasant explanations I can think of,
though. 

#You have a myopic view of the process and choose to place all the blame on
#the shoulders of the post-production facility.

We have been through this before a couple of times. But here it is again.
If that was the impression I'm sorry. I'm not blaming someone specific
for the time being except the 'system' as a whole that produces a print like
"Seabiscuit". The system includes the filmmakers and all the technical
facilities that did work for the film, without getting any more specific
for the time being.
I'm also more than willing to see "Seabiscuit" as just a failed experiment,
which nobody is really happy with as far as the look of release prints
is concerned, although the people in charge do not say so in public for
various reasons. If it is not a glimpse of things to come more and more
often it might be a mosquito turned into an elephant. It's way too early
to tell. But the total silence about the look of the film concerning its
problems, from the public, the film critics and the industry alike was
enough of a bad omen for me to raise the issue. Maybe I'm simply not on
the right internal lists, don't speak to the gurus of the trade, don't
know the buzz, and it's all cold coffee, snow of yesterday. You tell me.
It's the first time I have seen something like it on a high budget feature
film shot on 35mm negative.

#It's OK and perfectly acceptable to discuss the issues at professional
#forums such as this one.  No problem there.  But dont' come of as if you
#undestand the topic, 'cause you don't.  There are very few people who do,
#even within the professional community.  Why?  Because, generally speaking,
#the nature of specialization is such that each of us is exposed to one
#sliver of the overall game.

I certainly do not understand the topic well beyond the core issue of
DNR and its artifacts. I hope this thread helps all of us understanding it
better.

#During that stint I colored a set of music videos for Michael Hall (yes, the
#actor).  He wanted DNR applied to a level that was simply ridiculous.  I
#tried as hard as I could to dissuade him from taking that approach.  He
#wouldn't budge.  If you ever saw those music videos you might puke (for more
#than one reason), there's nothing subtle about the DNR problems they have.
#But, ultimately, it's what the client wanted.

Which brings me back to the unpleasant explanations I can think of.
Gary Ross and John Schwartzman are no actors. They are professionals
who made many films and have a trained eye, at least as far as traditional
film image quality issues are concerned. Some people are color blind.
Are there DNR blind DOPs and directors? Are there grain haters that
go to any extremes to get rid of grain? If the answer is two times no,
what explanation is there for the look of "Seabiscuit" that is compatible
with the logic of the whole technical process?

#I shouldn't put my coins into a box office to find DNR artifacts onto the
#screen.

Indeed. Actually if I knew in advance I would not go see the film.
It's probably a severe case of 'd?formation professionelle' but I was 
unable to follow "Seabiscuit" at all on the plot, acting and directing
level. All I could see were artifacts. I'm just not used to this in cinemas.
I can watch bad DVDs and forget about the technical issues and just enjoy
the film. So far I can not do this at all in a cinema when it comes to DNR
issues.

#Anyway, with the Q.T Film "Kill Bob" do we know if the
#Artifacts are a mistake or maybe "artistic value?"

I would not recommend to kill Bob Richardson over the look of "Kill Bill".
:-) 
But, yeah, are some people thinking of DNR as a cool sfx tool that does
'funny' things to textures in motion and reduces grain somewhat at the
same time, which they don't mind, after all. The 'actor' that was mentioned
seemed to think so. Is it something like an embossing filter or these median
filters that turn the picture into an oil painting?

>The vast majority of HD transfers have EE (or aperture correction) too.
#Yes, they do. Because they basically have to. But I don't expect that you'd
#understand that right now.

Try me.

#Have you ever visited a high end transfer facility and sat in on a telecine
#session? Or gone to a facility (maybe even Technique) that does digital
#intermediate work and seen a colorist in his working environment? 

I have been to PO houses and I have seen professionals at work.
I do know about digital image processing, have written my own DNR
software, interpolation filters, motion estimators, statistically
robust estimation procedures. I have a degree in computer science.
Is that enough for starters?
Please spare me the usual tantrum of the offended professional wolf
slightly annoyed by an amateur flee. 
I have to learn about things like the rest of the world. But I'm not
a clueless ignoramus either that postures for posturing's sake. 
I'm all ears to hear what you have to say beyond personal attacks and
things are they way they are because that's the way they are and there
is an explanation for everything. I guess there is.
cheers
						Michel Hafner


--
Help wanted ads are welcome here on the main TIG; blatant advertising
or marketing is not but can be sent to tig-announce at tig.colorist.org.
For contributions see http://tig.colorist.org the "make a donation" button at the bottom, thank you!.






More information about the Tig mailing list