Genesis to be trialled for Superman

[Michael Most]

But the most important thing in this business is what the pictures look like on the screen.  If the pictures look great, nobody gives a s--t what the numbers say.

 

Except for people on Internet forums who think they know more than those who actually use this stuff, and some people on certain selected mailing lists who will continue to berate any electronic capture system based on theory, not reality.

[David Mullen ASC]

Hi,

 

It surprises me to hear people talk about PAL conversion as a reason for shooting film, when most of the US produced material we see here is a conversion of the NTSC video, and looks rotten.

<{POST_SNAPBACK}>

 

I'm sure that's changing as more shows post in 24P HD and then make NTSC and PAL downconversions from that.

 

Certainly for theatrical movies, PAL versions are not made from NTSC transfers.

[Michael Most]

Hi,

 

It surprises me to hear people talk about PAL conversion as a reason for shooting film, when most of the US produced material we see here is a conversion of the NTSC video, and looks rotten.

 

I don't know of any studio that is currently using NTSC masters for PAL conversions, at least not on any programs that are posted in HD (which includes nearly all current scripted shows). All conversions are done from the HD masters, although it is true that in the past (i.e., shows produced more than 2 years ago) the conversions were indeed done from NTSC, albeit with 3:2 sensitive standards converters that do a pretty darned good job.

 

You need to define what you think "looks rotten" and why. If you're talking about color correction, there are creative choices being made that are based on American visual aesthetics. This is often different than European sensibilities, particularly in terms of contrast (US producers and cameramen like a lot of it, not so in Europe) and flesh tones (European material usually exhibits much paler flesh tones than US material). If you're talking about technical issues, I would like to know what you're referring to, because almost every PAL conversion I've seen in the last 10 years, at least from 24fps material, has been almost completely transparent to the HD or NTSC original. If you don't happen to personally like the creative choices, that's your right. But that doesn't mean there's anything technically wrong with it.

[John Sprung]

in the past (i.e., shows produced more than 2 years ago) the conversions were indeed done from NTSC, albeit with 3:2 sensitive standards converters that do a pretty darned good job.

<{POST_SNAPBACK}>

The first system to "un-do" 3-2 was Doug Jaqua's "Image Translation" at Laser Pacific in 1987. They had to note the 3-2 info in telecine and trace back to it thru the EDL in those days. Later systems were able to detect the 3-2 drumbeat from the image data. This was a major improvement in PAL conversion for film originated shows. But of course neither it nor anything else has helped much with the more difficult problem of turning 483/60i into 576/50i.

 

 

 

-- J.S.

[Phil Rhodes]

Hi,

 

> I don't know of any studio that is currently using NTSC masters for PAL

> conversions

 

Not any more, no, I'm sure you're right, but as you say was happening up until the late 90s and a lot of archive material was bad. The Star Trek stuff is the example I always use - and that was all Promisted as well, so the more recent Enterprise stuff is not only sharper and without motion artifacts, it's also widescreen, which is usual in the UK now.

 

> You need to define what you think "looks rotten" and why.

 

Most of Deep Space Nine. Soft and with visible motion artifacts from the conversion, particularly pathological on starfields. It's due to the NTSC transfer; I know what it looks like, and that's it.

 

> If you're talking about color correction, there are creative choices being made that

> are based on American visual aesthetics.

 

Personally I prefer the kind of high contrast, high saturation material which you refer to as an American asthetic. This makes me popular with young independent filmmakers who have noticed that British TV and film looks far too much like the grey reality they live in, but almost unemployable among the current establishment of ex-BBC technical types who like to see detail in everything from shadow to hilight.

 

> I would like to know what you're referring to, because almost every PAL

> conversion I've seen in the last 10 years, at least from 24fps material, has been

> almost completely transparent to the HD or NTSC original.

 

If you're talking about HD downconversions then yes, clearly it looks great, but on the subject of standard def NTSC to PAL transcodes I've seen far too much of it to overlook the fact that it looks soft and juddery, especially with the transcode piled on top of 3:2 pulldown. The reason it sticks out is that it's strange to watch this highly expensive, clearly high production value TV series which has really quite severe technical QC issues, especially when you know they're shooting film and could just as easily retransfer it to PAL and have it look great if they considered it a worthwhile investment. It's like watching a 4:3 mastered DVD that's been transferred from a print with burned-in subtitles - well, not quite that bad, but you're aware you're watching it through a layer of... well.. crud.

 

This is what almost all material from the US looked like until recently. Now, recent Star Trek, ER, West Wing, CSI [location], all look absolutely fantastic because they're posted HD and downconverted, not forgetting the fact that they're actually lit rather than blasted with 2Ks for exposure and shot, which is how most stuff here is produced.

 

It's interesting that the soft-and-smeary look of US stuff has almost become a look of its own in PAL countries - as people learn to associate it with high production value television in the same way 3:2 pulldown is seen in the US.

 

Phil

[Michael Most]

The reason it sticks out is that it's strange to watch this highly expensive, clearly high production value TV series which has really quite severe technical QC issues, especially when you know they're shooting film and could just as easily retransfer it to PAL and have it look great if they considered it a worthwhile investment.

 

It's not that simple. The vast majority of US television programs have been posted solely on videotape for quite some time (since 1985 for some, since at least 1990 for most of them). In most cases (i.e., shows not produced by Warner Bros. or Universal) there are no assembled negatives and the video master is the only element available. In the case of the Star Treks, there would be no point to attempting to resurrect the negative because all of the visual effects have been done solely in video since the very first episode of Next Generation - and even now, on Enterprise, some of the CG shots are still done in standard definition and uprezzed for the show master.

 

One shouldn't throw around terms such as "easily retransfer" when they don't understand the circumstances of the production they're referring to.

[Phil Rhodes]

Hi,

 

I'm aware of the difficulties, it just seemed particularly insane to me to shoot 35 then end up with these grotty-looking dubs.

 

Phil

[Landon D. Parks]

As to the subject of D.I:

 

I have yet to see a film in theaters that looks horrible. I'v seen tons of 2k scans, 4k scans downrezed to 2k and 4k scans that sat 4k.

 

4k looks a lot better, but 2k don't look bad either. I am always to interested in the story to bother to look for compression artifacts, skin tons, ect.

 

I think some people take that stuff to hard. Why not go to the theater to see a great story? Not complain about it being a little soft?

 

DI has its advantages, Optical has its advantages. But both have a lot more dis-advantages than advantages.

 

Personally, I would post DI. This is because in my opinion its easier to work with. Other opinions may very from mine, and I'm not saying my opininion is right, its just my opinion.

[Brad Grimmett]

DI has its advantages, Optical has its advantages. But both have a lot more dis-advantages than advantages.

 

<{POST_SNAPBACK}>

Could you expand a little on the dis-advantages that you mention above? I'm curious what the dis-advantages are.

[Michael Most]

Personally, I would post DI. This is because in my opinion its easier to work with. Other opinions may very from mine, and I'm not saying my opininion is right, its just my opinion.

 

There is nothing simpler than the standard film post production path. DI is far more complex, in part because there is no standard for it (every company has their own set of rules and procedures, and practically every project invents its own post path to some degree) and because it involves an awful lot of data manipulation, data transfers, eye matching and confirmation of editorial accuracy, and color/quality accuracy issues all along the way. Flexible, yes. Simple, no.

 

My guess is that your opinion is based on no practical experience in either methodology, but simply things you have read. Opinions on things like motion picture post paths need to be based on experience and knowledge, not on press hype.

[John Sprung]

> You need to define what you think "looks rotten" and why.

 

Most of Deep Space Nine. Soft and with visible motion artifacts from the conversion, particularly pathological on starfields. It's due to the NTSC transfer; I know what it looks like, and that's it.

<{POST_SNAPBACK}>

DS-9 goes back to 1992, 1"C and non-3-2-removing PAL conversions were still in use at that time. Certainly much better PAL could be made from the original NTSC masters today. But it'll only happen if a customer refuses to buy the show without a better PAL conversion. I'd guess that would cost maybe $600 - $700 per episode, no problem at all if it mean making a sale. Mike Most probably knows far better than I what the conversion price is today.

 

 

 

-- J.S.

[Mark Allen]

There is nothing simpler than the standard film post production path.

<{POST_SNAPBACK}>

 

I've done the post path with film post and with HD intermediate (shoot film / HD post / back to film), but never a full film with DI. However, in my experience using the HD intermediate was simpler than standard film post production.

 

It's certainly easier to get the look you want (unless you just want it to look like what you shot). Negative cutting and preparing for that is a big hassle. Also it just seemed like less "anomalies" would occur.

[Wendell_Greene]

My guess is that your opinion is based on no practical experience in either methodology, but simply things you have read. Opinions on things like motion picture post paths need to be based on experience and knowledge, not on press hype.

<{POST_SNAPBACK}>

 

 

Bravo! Mr. Most.

[Rik Andino]

There is nothing simpler than the standard film post production path.

<{POST_SNAPBACK}>

 

 

Well DI can be simpler if you're doing alot of Digital FX in post

And these days most movies are

[Michael Most]

I've done the post path with film post and with HD intermediate (shoot film / HD post / back to film), but never a full film with DI.  However, in my experience using the HD intermediate was simpler than standard film post production.

 

It's certainly easier to get the look you want (unless you just want it to look like what you shot).  Negative cutting and preparing for that is a big hassle.  Also it just seemed like less "anomalies" would occur.

 

An HD intermediate is not necessarily what I'm talking about. When you do an HD intermediate, you're basically using a television post path with a tape to film transfer once you're done. This is simple, due to the use of time coded videotape and familiar methodologies for assembly and color correction. A DI, however, is a different beast that requires scanning frame by frame (which requires having all of the negative that was shot identified, cleaned, and available at the time of scanning - not to mention accurately logged during the daily telecine transfer), embedding metadata in those scans to be able to correlate it back to an electronic edit, use of specific naming conventions at different points in the process, assembling frame by frame on specific equipment dedicated to the task, a dustbusting pass, color correction in log space in a properly calibrated environment, and recording on a film recorder specifically calibrated to the lab you're using. If you're not an assistant editor, a post supervisor, or an employee of a digital intermediate facility it may seem simple. It isn't.

 

I have been involved in negative cutting for years, from both film editing (using work picture) and electronic editing, and it is not and never has been a "big hassle" if those involved know what they're doing. The nice thing about it is that the terminology and methodologies used are completely standardized, so there is no question as to what something means, as there always is in a DI. You have key numbers - and that's basically it. No correlating one thing to another, no translating between incompatible formats, no needing to embed numbers electronically that may or may not read properly down the line. I would call that pretty darned simple. The only real downside is the inability to recover from a miscut, a problem that is totally solved by electronic assembly.

 

As far as getting the look you want, I would say that any competant cameraman is capable of shooting in a way that this is achieved in production. In many ways, it is often easier to "hit the mark" on an answer print than it is in a DI suite, where the colors are not exactly the same (especially flesh tones) and where the eye of an electronic colorist has often been honed by timing a lot of television material. If one wants a look that is not available by photochemical means, this is a good reason to do a DI. But getting "what you shot" on the first pass is more often achieved in the lab than it is in a DI suite, at least in my experience and that of most of the cameramen I know.

[Michael Most]

Well DI can be simpler if you're doing alot of Digital FX in post

And these days most movies are

 

More flexible, perhaps. Faster, in some cases. Cleaner, in most cases. Not necessarily simpler.

 

Visual effects and DI are not the same thing and are not done in the same environment. The advantages of doing a DI when visual effects are involved are primarily the ability to better match the visual effects material to the non-effected original material, and the elimination of the need for a final film out of the visual effects shots (they are usually delivered to the DI vendor as Cineon or DPX files). This allows for a slightly faster turnaround when required, as in the case of a tight delivery date or a meddling, indecisive, or excessively anal director. It also allows for less handling of the original negative, since in a film posted situation it would have to be opened up for insertion of the effects shots as they're delivered. In a DI, the negative is basically handled only twice - once for daily telecine, and once for scanning. This is indeed an advantage of that approach. However, as I mentioned earlier, if you're involved in dealing with any of the problems or details, it is not "simple."

[Michael Most]

DS-9 goes back to 1992, 1"C and non-3-2-removing PAL conversions were still in use at that time.  Certainly much better PAL could be made from the original NTSC masters today.  But it'll only happen if a customer refuses to buy the show without a better PAL conversion.  I'd guess that would cost maybe $600 - $700 per episode, no problem at all if it mean making a sale.  Mike Most probably knows far better than I what the conversion price is today.

 

As I recall, most if not all of the DS9 episodes were converted by Encore using Deft. There might have been some 3:2 anomalies due to odd pulldowns generated by video based speed changes, as well as some effects shots done at rates other than 24fps, but for the most part these conversions were done with proper 3:2 removal and a very high quality interpolation. I also recall them looking pretty darned good. I don't recall the original mastering format, but I would trust John's recollection of that. At some point the Deft process was switched over to D1 for both playback of the NTSC version and recording of the PAL version, but I don't think it was as early as 1992. I could be wrong, though.

 

As far as the cost, as I recall in those days the cost was around $1000 per hour program (maybe $1200), it's probably less now, simply because everything has been pretty commoditized.

[Guest Jim Murdoch]

My guess is that your opinion is based on no practical experience in either methodology, but simply things you have read. Opinions on things like motion picture post paths need to be based on experience and knowledge, not on press hype.

<{POST_SNAPBACK}>

Yes but if you enforced that rule, you'd eliminate probably 97% of the people who post here :P

[Keith Mottram]

I think there is alot of confusion over what a DI is. A DI can be simple or complicated as you want (or as necessary). Mark I think you make it sound far more complicated than it actually is. True if you are editing with pulldown then convert an edl then reconform from scans you end up with this long and complicated (with a fair amount of room for human error) route, but this does not have to be the case. You can easily edit at native 24 and reconform at 2 or 4k with little hassle apart from the expense of equipment and storage. At HD this is achievable on a very modest system and from conversations I've had with Apple 2K will be similarly achievable in the next year or so. The new version of Color Finesse (still in Beta stage) allows Lustre like capabilities on a mac or pc and there a number of other companies who are working on real time 2k correction and conform. In my opinion DI is simpler cheaper and more controllable (if you are finishing at HD) than neg cut conform and as I said soon 2K will be the same. It is not nearly as scary as you make out as long as you fully understand the workflow and have an editing team that is fully up to date with technology (this is the hardest part).

 

Keith

[Mark Allen]

<<Mark I think you make it sound far more complicated than it actually is.>>

 

If this was me, Mark, I actually don't think DI's are complicated at all. I actually like the idea of avoiding cutting negative though. While I realize pros get things right, you're not always guaranteed a pro at every stage.

 

 

Related to the subject of this thread - I just posted the headlines of my thoughts on seeing a screening of the Genesis footage today here:

 

http://www.cinematography.com/forum2004/in...showtopic=4791#

[Michael Most]

You can easily edit at native 24 and reconform at 2 or 4k with little hassle apart from the expense of equipment and storage. At HD this is achievable on a very modest system and from conversations I've had with Apple 2K will be similarly achievable in the next year or so.

 

I have to say that this post really identifies the mindset of most of the people who have posted on this topic. It basically says that "do it yourself" is the mentality. Having worked in the industry for a long time, I really don't understand that attitude. Collaboration with experienced professionals at every step of post production is the best way to achieve the best result. A creative editor is not a finishing/conforming person. An editor is not a colorist. And a colorist is not a color scientist (at least not most of the time). And a cinematographer is none of these things. All of these disciplines come into play in a true DI environment, but what I realize now is that most of the people here are not talking about a real DI environment. They seem to be talking about equipment being more important than those who operate it, which is not what I believe to be true. Low cost is not worthwhile if the end product is not what it could be given the input of a sharp finishing editor, a good colorist, and an experienced technical person to manage the scanning and recording.

[Phil Rhodes]

Hi,

 

The other way to look at this is that it's better than calling the DV your online. Which it is.

 

Phil

[Tenolian Bell]

Low cost is not worthwhile if the end product is not what it could be given the input of a sharp finishing editor, a good colorist, and an experienced technical person to manage the scanning and recording.

<{POST_SNAPBACK}>

 

I agree. Since I've begun doing work in Los Angeles, colorists here have redefined for me the telecine room. I've had the pleasure of working with some highly talented people who have totally blown my mind with what is possible with a well exposed negative.

 

I'm actually in no rush to do a DI. Not that it has come to me as an option. Yes its easy to grade a look in digital. But the real challenge is getting that exact same look up on the screen in a predictable and repeatable fashion. That part is not simple at all, and many people are right now trying to figure that part out.

 

One flaw in the chain we know already is 2K scanning, its not good enough.

[Keith Mottram]

I have to say that this post really identifies the mindset of most of the people who have posted on this topic. It basically says that "do it yourself" is the mentality. Having worked in the industry for a long time, I really don't understand that attitude. Collaboration with experienced professionals at every step of post production is the best way to achieve the best result. A creative editor is not a finishing/conforming person. An editor is not a colorist. And a colorist is not a color scientist (at least not most of the time). And a cinematographer is none of these things. All of these disciplines come into play in a true DI environment, but what I realize now is that most of the people here are not talking about a real DI environment. They seem to be talking about equipment being more important than those who operate it, which is not what I believe to be true. Low cost is not worthwhile if the end product is not what it could be given the input of a sharp finishing editor, a good colorist, and an experienced technical person to manage the scanning and recording.

<{POST_SNAPBACK}>

 

Exuse me but I am talking from the experiece of someone who oversaw the final edit and post production (this would make me the 'sharp finishing editor' you reffered to, I assume) on a studio feature film last year. This final DI was completed in house as was the final conform of the feature from 2k DPX. We may have finished on SR but the workflow that I devised for Mirrormask (just screened at Sundance) would have worked equally well for a 2K finish. I was not offering my thoughts idly- as I assume you dont, if you have a problem with my thoughts address them individually. I did not once state that roles should be removed from the process, I merely stated that there are various ways to the DI route and some of them are not as expensive as the less inlightened believe. There was a time when you needed rediculously expensive custom built systems to run 3D pacages like Maya, now a G5 or a speedy Wintel system can surfice. I think the point you are trying to make is correct but muddled, the operator is key not the expense of the system. If you need me to clarify this please ask, but dont try and bundle me into some kind of hair brained no budget idealist tribe. I have been working in post for several years in both editing and effects and I think my experience entitles me to comment on these matters.

 

Keith

Edited February 4, 2005 by keith mottram

[Mike Brennan]

Having worked in the industry for a long time, I really don't understand that attitude. Collaboration with experienced professionals at every step of post production is the best way to achieve the best result. A creative editor is not a finishing/conforming person. An editor is not a colorist. And a colorist is not a color scientist (at least not most of the time). And a cinematographer is none of these things.

 

But times are changing. We all know getting 4 experts to do their job is not always possible.

 

For those on lower budgets there are workflows that can be achieved with out expert help.

"Dust to Glory" is a good example posted on Adobe, and mixing HD, DV 16mm and 35mm.

Would these projects benifit from expert involvement, sure, but can they afford it, not always.

 

 

In my view a colourist role as we know it today will be a very nich occupation in a decade replaced by the DPs vision being delivered by a completely digital chain that needs fewer experts with black art.

 

 

 

Mike Brennan