Jump to content

The ever changing world of Post Production


Mike Nichols

Recommended Posts

Here is my dilemma, I have searched older posts, but I feel like their relevancy might be outdated already! I am shooting a feature, about 50,000ft (there will be some HDCAM shot on an F-900). I don't have a distrib deal, so I don't know how I am going to finish. I need a cost effective HD solution!! I have a glorious Mac Pro set up with Raid Arrays and the AJA IO HD. With all that, would it be in my best interest to have my dailies on SR AND data on a hard drive in a compressed HD format (Pro Res/DVCPRO HD etc) This way, I can offline in either pro res or dvcpro hd (the higher quality used to have a leg up during screenings) and then do a tape to tape from the SR when a deal is in place? How well does Pro Res hold up as an intermediate codec? If need be, could I finish with Pro Res and have an acceptable Digital Master? If so, then can I skip transferring to SR all together? (Even typing that sounded like a bad idea...)

 

I am just trying to minimize costs, and never have to go to SD.

 

Thanks!

Link to comment
Share on other sites

  • Replies 66
  • Created
  • Last Reply

Top Posters In This Topic

How well does Pro Res hold up as an intermediate codec? If need be, could I finish with Pro Res and have an acceptable Digital Master? If so, then can I skip transferring to SR all together? (Even typing that sounded like a bad idea...)

 

I am just trying to minimize costs, and never have to go to SD.

 

Thanks!

 

Well first of all Pro res is only 4:2:2. So David has a good point, why shoot on 35mm and use a codec that is less quality than what you have on film?

 

You might be minimizing costs on the back end but you've already spent it on the film so why not use that investment rather than lose it?

Link to comment
Share on other sites

Not that 35mm isn't a good choice for material to be shown in HD broadcast or Blu-Ray... I was referring more in terms of using 4:2:2 HD as an intermediate step for 35mm material to be shown theatrically.

 

 

That was my question. I wasn't planning on using the pro res as an intermediate. I guess my post got a bit convoluted! I was asking if pro res was acceptable as an intermediate codec. Since the answer seems to be a resounding "NO!!" then I think transferring circle takes (or possibly all) best light dailies to HDCAM SR and then importing the selects as pro res for "off line." The catch is, my "offline" will be what I screen for buyers, so I want the best relative quality.

 

Am I trying to reinvent the wheel? Barking up the wrong (and expensive) tree? Should I just shoot RED????

Link to comment
Share on other sites

  • Premium Member

I've been fairly convinced for a while that this sort of thing could be done fairly effectively if you were cautious, attentive, and had a few tests shot out to film to establish at least a ballpark colour calibration.

 

In an ideal world you'd have it all transferred uncompressed to hard disks; 50,000 feet is only about 7TB of 10-bit 1080p24 and the disk storage for that is not particularly difficult - probably cheaper than the HDCAM-SR tapestock, in fact. Unfortunately, direct to disk transfers are strangely difficult - I suspect shadowy, behind-the-scenes types are actively trying to prevent this being offered because it opens up a lot of possibilities to people in your position. You could conceivably bounce it from SR to hard disk, at some considerable cost.

 

Then I'd make all my offline proxies and cut those.

 

I'd then figure out some sort of solution for creating the online from that stack-o-disks full of your full res material. I've found several ways of doing this in Adobe Premiere, but there's presumably an equivalent process for Final Cut. Then you can grade your full res stuff in Color, or whatever, and stick it back out to another disk or three for the attention of your friendly neighborhood ArriLaser.

 

This assumes you have enough full HD speed storage to hold at least a reel of your material. You can add that to a Mac Pro for, oh, $2000 or so. If you didn't want to invest in enough HD speed storage to grade, you could probably grade the offline if it was DVCPROHD and find that the same grading decisions would translate reasonably cleanly to your full res material.

 

I have never done this (I have done bits of it from HD masters). Technically it is more or less sound, notwithstanding color calibration and whatever you're going to do about an audio mix.

 

Phil

Link to comment
Share on other sites

I've been fairly convinced for a while that this sort of thing could be done fairly effectively if you were cautious, attentive, and had a few tests shot out to film to establish at least a ballpark colour calibration.

 

In an ideal world you'd have it all transferred uncompressed to hard disks; 50,000 feet is only about 7TB of 10-bit 1080p24 and the disk storage for that is not particularly difficult - probably cheaper than the HDCAM-SR tapestock, in fact. Unfortunately, direct to disk transfers are strangely difficult - I suspect shadowy, behind-the-scenes types are actively trying to prevent this being offered because it opens up a lot of possibilities to people in your position. You could conceivably bounce it from SR to hard disk, at some considerable cost.

 

Then I'd make all my offline proxies and cut those.

 

I'd then figure out some sort of solution for creating the online from that stack-o-disks full of your full res material. I've found several ways of doing this in Adobe Premiere, but there's presumably an equivalent process for Final Cut. Then you can grade your full res stuff in Color, or whatever, and stick it back out to another disk or three for the attention of your friendly neighborhood ArriLaser.

 

This assumes you have enough full HD speed storage to hold at least a reel of your material. You can add that to a Mac Pro for, oh, $2000 or so. If you didn't want to invest in enough HD speed storage to grade, you could probably grade the offline if it was DVCPROHD and find that the same grading decisions would translate reasonably cleanly to your full res material.

 

I have never done this (I have done bits of it from HD masters). Technically it is more or less sound, notwithstanding color calibration and whatever you're going to do about an audio mix.

 

Phil

 

 

Believe me, I have given a lot of thought of Direct to disk, but it scares the CRAP out of me... I have heard some stories of less than desired work. The IDEA is fantastic. I guess it would work. I could be a guinea pig and document my trials for you guys!!

Link to comment
Share on other sites

Phil is correct. Here is the way to do it:

  1. Telecine everything to HDCAM SR 4:4:4 at 880 mbit/sec (FYI only the SRW-5800 will do 880, the 5000 and 5500 will only do 440 so the post house needs to have a 5800). Have it transferred "flat," to retain as much image information as possible
  2. Have the tapes captured to uncompressed 4:4:4 QuickTime on a Kona3 equipped Mac Pro. This will need to then be copied to several FireWire disks. You can opt to have the post house frameserve DPX files during the layoff to FW disks if you'd rather have DPX instead of QuickTime files
  3. Render your own offlines in FCP to any codec you want, DVCPROHD, ProRes422, DV, Cinepak, whatever (joking about Cinepak, btw)
  4. Cut the movie using the offline on a MacBook or some cheaper system
  5. Export an XML from FCP of the timeline, then re-link your online files (which will have the same TC if you did this correctly)
  6. Color correct in Color if you have an appropriate grading setup

Tape to tape is a complete, total, absolute miserable waste of money. Everyone is moving to software based grading systems. Even EFILM uses Lustre for DI work. There's nothing a da Vinci can do that can't be done with a Mac and some rendering time.

 

Even going to HDCAM SR is not the best way to do it, but there are not many facilities that go straight to disk for a reasonable amount of money. That's why we still have to use this method. At 880mb/sec it's very high quality, so at least we're not losing much going to tape. Plus, you have a tape backup.

 

I agree with Phil that the tape step is unnecessary and will eventually be done away with.

 

I've already done everything here and it works. The goal is to eliminate the expensive post house from the equation as soon as possible. The old school method of using window burns and transferring selected takes from keycode is a thing of the past. Don't waste your money.

Link to comment
Share on other sites

Phil is correct. Here is the way to do it:

  1. Telecine everything to HDCAM SR 4:4:4 at 880 mbit/sec (FYI only the SRW-5800 will do 880, the 5000 and 5500 will only do 440 so the post house needs to have a 5800). Have it transferred "flat," to retain as much image information as possible
  2. Have the tapes captured to uncompressed 4:4:4 QuickTime on a Kona3 equipped Mac Pro. This will need to then be copied to several FireWire disks. You can opt to have the post house frameserve DPX files during the layoff to FW disks if you'd rather have DPX instead of QuickTime files
  3. Render your own offlines in FCP to any codec you want, DVCPROHD, ProRes422, DV, Cinepak, whatever (joking about Cinepak, btw)
  4. Cut the movie using the offline on a MacBook or some cheaper system
  5. Export an XML from FCP of the timeline, then re-link your online files (which will have the same TC if you did this correctly)
  6. Color correct in Color if you have an appropriate grading setup

Tape to tape is a complete, total, absolute miserable waste of money. Everyone is moving to software based grading systems. Even EFILM uses Lustre for DI work. There's nothing a da Vinci can do that can't be done with a Mac and some rendering time.

 

Even going to HDCAM SR is not the best way to do it, but there are not many facilities that go straight to disk for a reasonable amount of money. That's why we still have to use this method. At 880mb/sec it's very high quality, so at least we're not losing much going to tape. Plus, you have a tape backup.

 

I agree with Phil that the tape step is unnecessary and will eventually be done away with.

 

I've already done everything here and it works. The goal is to eliminate the expensive post house from the equation as soon as possible. The old school method of using window burns and transferring selected takes from keycode is a thing of the past. Don't waste your money.

 

When you say Telecine (bear with me, Telecine to me still means getting best light DigiBeta sessions at commercial rates...) you're talking about transferring ALL my shot film SR@880 via the 5800 and to do it flat. I think I got that. I have the sweet killer Mac Pro Pro set up WITH the Kona 3 already, but I am still pricing a nice Array to work with that sort of footage. Wouldn't it be wiser and easier for me to ONLY capture selects Full Rez and proxy THAT footage? 7TB of FAST storage for 4:4:4 1080p24psf capturing is a LOT OF CASHOLA. I still think I need (at least for the deck!) the post house for an awful lot!

 

edit: I see, you're talking about having the POST house dump the SR captures onto Firewire...

Edited by Mike Nichols
Link to comment
Share on other sites

Wouldn't it be wiser and easier for me to ONLY capture selects Full Rez and proxy THAT footage? 7TB of FAST storage for 4:4:4 1080p24psf capturing is a LOT OF CASHOLA. I still think I need (at least for the deck!) the post house for an awful lot!

You don't need fast storage for the entire production. Only enough to work on roughly 20 minutes of online at a time (20 mins per reel). The rest of the storage can be FireWire drives.

 

edit: I see, you're talking about having the POST house dump the SR captures onto Firewire...

Yes. Your own Kona is really only good for monitoring via SDI out, unless you somehow get access to a deck. Depending on how much you need to bring into the system, it might be more cost effective to rent a deck (even with insurance) for a weekend than to have to pay a post house to both 1. capture the footage, and 2. lay it off to FW disks.

Link to comment
Share on other sites

  • Site Sponsor
Phil is correct. Here is the way to do it:

 

Tape to tape is a complete, total, absolute miserable waste of money. Everyone is moving to software based grading systems. Even EFILM uses Lustre for DI work. There's nothing a da Vinci can do that can't be done with a Mac and some rendering time.

 

 

I've already done everything here and it works. The goal is to eliminate the expensive post house from the equation as soon as possible. The old school method of using window burns and transferring selected takes from keycode is a thing of the past. Don't waste your money.

 

 

One could argue that the mac is allot of rendering time and that a davinci (esp. resolve) can do some things color can not... But for this kind of work a properly equipped Mac-Pro can do it, just keep in mind that a cinema display is not a reference grade monitor....

 

 

I think compared to transferring everything to SR a SD transfer to D-Beta (or DvCam or BetaSP) with keycode is a tremendous savings, then you could spend the money for a good selects scan to DPX files instead of the limitation of a Spirit, etc. to SR. A Arri-Scan or Northlight scan to DPX will blow a Spirit or DSX transfer to SR away I feel that the negative scan is probably the most important part of the process as it is the step where you can lose the most fidelity...

 

 

-Rob-

Link to comment
Share on other sites

  • Premium Member

For some reason a Northlight scan will probably cost more than the Spirit, but it will of course go directly to data by default.

 

You will want to have some sort of backup of it because the scan will be a bit pricey to do again if you drop a disk down the stairs.

 

P

Link to comment
Share on other sites

  • Site Sponsor
For some reason a Northlight scan will probably cost more than the Spirit, but it will of course go directly to data by default.

 

You will want to have some sort of backup of it because the scan will be a bit pricey to do again if you drop a disk down the stairs.

 

P

 

 

These things are too hard to simply calculate but I would think a SD Keycode xfer and a selects scan might be close in price to a complete scan to SR on a Spirit. Deals can be made on all this stuff depending on the time, place and project, YMMV.

 

 

Furthermore I feel that a 4:4:4 Spirit classic transfer is a complete waste as the Spirit has half res color ccd line arrays so all additional color sampling (0:2:2) is interpolated. A DSX or Millenium will have true 4:4:4 but watch out for tube life... The newer Spirit 2K and 4K are full bandwidth color but many people feel they do not produce as good a picture as a Arriscan or Northlight.

 

 

As with all digital information there is no really good long term backup, maybe LTO tape and a Raid 5 or 6 disk array to transfer to... move quick for thy data will be dust long before the film starts to fade...

 

 

-Rob-

Link to comment
Share on other sites

These things are too hard to simply calculate but I would think a SD Keycode xfer and a selects scan might be close in price to a complete scan to SR on a Spirit. Deals can be made on all this stuff depending on the time, place and project, YMMV.

Yes, the biggest factor is what type of deal you can score. I've gotten reasonable deals on telecine so I've adopted the HDCAM SR route.

 

Also, keep in mind that doing it my way eliminates the second visit to the post house. It also eliminates EDLs, Flex files, window burns, etc. Also, you can have an HD version of the picture to cut (DVCPROHD, for example), which makes judging fine detail (like focus) easier. And you can generate as many different versions of the picture to as many formats you want, since you are in control of the online.

 

The goal is now to get the data straight off the telecine to a drive, fully uncompressed. Someday... ;)

Link to comment
Share on other sites

Here is my dilemma, I have searched older posts, but I feel like their relevancy might be outdated already! I am shooting a feature, about 50,000ft (there will be some HDCAM shot on an F-900). I don't have a distrib deal, so I don't know how I am going to finish. I need a cost effective HD solution!! I have a glorious Mac Pro set up with Raid Arrays and the AJA IO HD. With all that, would it be in my best interest to have my dailies on SR AND data on a hard drive in a compressed HD format (Pro Res/DVCPRO HD etc) This way, I can offline in either pro res or dvcpro hd (the higher quality used to have a leg up during screenings) and then do a tape to tape from the SR when a deal is in place? How well does Pro Res hold up as an intermediate codec? If need be, could I finish with Pro Res and have an acceptable Digital Master? If so, then can I skip transferring to SR all together? (Even typing that sounded like a bad idea...)

 

I am just trying to minimize costs, and never have to go to SD.

 

Thanks!

 

 

:unsure: :rolleyes:

 

I don't get it all. You have done your offline right. Your whole feature does not run a full 50000 ft I presume. The final edit time for Online frames storage. How long is it?. If it is some 90 to 100 min it is going to take about some 2to3 TB if you get your work scanned into dpx files. Say one frame of 1920x1080 dpx file will hold upto 10 or 12 MB. You do the math for the total running time. You can either process them on some Online edit machines like Smoke or Lustre or keep them stored until you could get a deal. I do believe cost of 1 TB hard drives have come down.

 

Or have them converted into D cinema or E cinema.

 

D-Cinema is a minimum standard defined by Hollywood where 2K or 4K resolution projectors are used with defined contrast ratios, brightness on screen and color gamut. Compression is in JPEG2000 at 250 Mbits. The Digital Cinema Initiative (DCI) defines these specs. D-Cinema Customers include cinemas that play Hollywood content across the world and multiplexes and popular cinemas in top cities. They use also VC-1 or wm9 codecs too.

 

E-cinema is a stripped down format that has resolutions for 1.3K. It is used in smaller theatres.

 

Both these cinemas use DLP projectors for display. Eg Barco 2K or 4K projectors. The cost of getting your film into this form is 6 or more times lesser than getting into film print.

 

I may not be entirely correct with the above data. Others do help to correct this.

Link to comment
Share on other sites

  • Premium Member

As far as I'm aware you can't cut a D-cinema MXF, nor is it intended to be used for that purpose, although packaging material for digital distribution is something I know very little about.

 

Mike -check your PMs.

 

P

Link to comment
Share on other sites

I think compared to transferring everything to SR a SD transfer to D-Beta (or DvCam or BetaSP) with keycode is a tremendous savings, then you could spend the money for a good selects scan to DPX files instead of the limitation of a Spirit, etc. to SR. A Arri-Scan or Northlight scan to DPX will blow a Spirit or DSX transfer to SR away I feel that the negative scan is probably the most important part of the process as it is the step where you can lose the most fidelity...

 

 

-Rob-

 

I wonder if Arriscan to SR might be an acceptable 'economy' alternative --- ?

 

I totally agree the front end machine is critical.

 

-Sam

Link to comment
Share on other sites

I don't understand the reasoning of HDCAM SR as a 'transport format'. Any good scanner can output uncompressed 10bit Log DPX files that are easily copied to any IT container such as hard drives, LTO3, DVD (small quantities).

 

Why would you want your scanner to output to an expensive HDCAM SR VTR in order to recover the same digital files on the other end using an expensive (again) HDCAM SR? I can understand using SR tapes for final deliverables but not for transporting digital files between computers.

 

I use LTO3 and get a speed of about 10 frames per second reading/writing HD format DPX files. A good LTO3 drive costs maybe 1/20th of an SR machine and will cost considerably less in maintenance and tapes.

Link to comment
Share on other sites

  • Premium Member

> Why would you want your scanner to output to an expensive HDCAM SR VTR in order to recover the same

> digital files on the other end using an expensive (again) HDCAM SR?

 

You wouldn't, and that's exactly the point Mr. Worth and I have been trying to make. The problem is that:

 

1) Conventional transfer houses using realtime machines like Spirit will not do it, even if you offer to provide the equipment and engineering expertise, and

 

2) Scanners such as Northlight, despite being cheaper to buy and slower to use than Spirit, for some reason cost more money to rent - often enough money that the pointless, quality-sapping and expensive HDCAM step is actually cheaper in the long run.

 

It occurs to me that with 50Gb writable blu-ray discs now easily available, that might be another way to go.

 

P

Link to comment
Share on other sites

  • 2 weeks later...
Guest tylerhawes

A couple tweaks to Thomas workflow, which I like and use a lot myself:

 

- I'm not convinced 880Mb transfer is worth it. The main problem is, like Thomas said, you're limited to only the latest studio deck, which means less rental options and post houses you can work with. This might raise costs, and I haven't seen any proof that the 880 image has any important quality advantage over the standard 440Mbit image (even though 2X the datarate is a lot, it seems that the HDCAM-SR compression scheme may not actually benefit that much from it). If someone can point me to some good tests that show otherwise, I'll be humbled. But my own tests did not show any tangible advantage for anything but special uses.

 

- After you've TKed to HDCAM-SR, instead of capturing ALL the footage through KONA 3 to uncompressed QuickTimes, capture it all to a good compressed codec like DVCPRO-HD. Use that to cut and do previews, etc. Once picture is locked, only then would I do a batch recapture of the selects as uncompressed. Now, instead of needing 7TBs, you only need about 2TBs, and the system will offer you all the realtime goodies while doing the edit.

 

- In an ideal world, cost no object, Thomas' way is technically better on both counts above. I mention those two points because they MAY be the more practical choice once you crunch the numbers, without sacraficing anything you'll notice.

 

- Another note is you may want to avoid a Best Light transfer. For DI, what we do is transfer 10-bit log just like a film scan (although we often cheat the white point higher for greater SNR, but only if we've made sure that the other post houses in the production are able to work with a custom curve without trouble - because the higher quality is not any good if the other guys can't work with it right...). So essentially you are transferring with the same characteristics as a film scan, you just happen to be using a telecine to get it onto 10-bit SR tape...

 

- As has been pointed out, the big gaping holes in this workflow so far are A) how are you going to color manage your viewing environment B) what physical devices are you going to monitor with and C) who is going to actually grade the footage. Maybe you are a talented colorist and have a 2K projector, or maybe you're content to work on a CRT (I'm not :), which leaves the color management as the big thing. Even if you're only intending for video distribution, you really still need a 3D LUT managed sytem that will restrict your colorspace, because without it software DI systems will merrily go about creating unrealistic super-saturated reds, neon greens, and grotesque magentas that scream VIDEO despite the fact that you shot film... this is something even a lot of "DI" companies don't understand and makes the difference between a DI that perserves all the qualities of film and one that looks like a digital finish... so my point is the steepest learning curve in this whole thing (aside from grading the footage if you're not an experienced colorist) is learning about the color science and figuiring out how to create an end-to-end color managed workflow that will keep you working with a filmic color palatte...

 

--

Tyler Hawes, DI Colorist

Liquid DI, Santa Monica

www.liquidcompanies.com

tyler -at- liquidcompaneis.com

Link to comment
Share on other sites

  • Premium Member

I'm not sure I'd characterise those as gaping holes, so much as known tradeoffs. You do not have to work in a Truelight-calibrated, Christie-projected suite to create commercially saleable DVDs, although there would probably be some chicanery involved in getting to a viewable filmout.

 

There are several approaches, most of them involving getting some test frames filmed out and lining up your grading display to approximate the projected results - which, frankly, even the best systems have to do to achieve identical results. To get it really right, yes, I'd expect to have to spend a few hours in a real suite tweaking, but I suspect you could get to a point where that would be an optional nicety.

 

But the point is that what you're doing is accepting a tradeoff. It's my supposition that you could probably achieve prints that were reasonably good-looking, even if it isn't precisely identical to what you saw in the grade. There are people doing filmouts more or less by mail order at the moment and you can put them in a projector and press start and pictures do appear on the screen.

 

And at the end of the day, the elephant standing in the corner is that you can go and buy the Mac Pro and Color and a reasonably decent DLP and do this for 20k, whereas what you'd probably consider a "proper" DI is going to cost six figures just for the grading time, let alone whatever the scanning and recording costs. If you're Steven Spielberg, fine. But otherwise, that's a lot of money for a small production which could, I feel, be much better applied elsewhere.

 

Talent, of course, is down to the individual, but again, if you're just going for professionally reasonable results, there are a million and one After Effects guys out there who have to know how to grade plates as just a small part of their skillset. It would depend what you were doing - a hard edged sci fi with lots of blue light and serious muzzle flash would probably be easier than a soft, naturalistic romantic comedy.

Link to comment
Share on other sites

  • Premium Member
But why shoot 35mm then if you're just going to end up with Pro-Res HD quality?

 

 

Why bother to shoot any features on 35mm if all they do is end up as MPEG-2 dvds or VHS tapes? Becasue the inherent look is not lost and the general rule is if you can shoot on the best format you can, do so. If he has the money, I say why not? HD and 16mm will simply not look as rich as 35mm, whether transferred to pro, pro rez or prosumer. :)

Link to comment
Share on other sites

Guest tylerhawes
But the point is that what you're doing is accepting a tradeoff. It's my supposition that you could probably achieve prints that were reasonably good-looking, even if it isn't precisely identical to what you saw in the grade. There are people doing filmouts more or less by mail order at the moment and you can put them in a projector and press start and pictures do appear on the screen.

 

I'm making the assumption that we're talking about a production where the standard is higher than having an actual picture appear on the screen. :) But seriously, I know you don't mean that literally and I hear your point. Suppose you just grade for Rec.709 and you do some tests and print to film. I've done that before when I haven't had a choice, and I've been able to get good results in the film print that everyone was happy with. It did require more testing than usual, but we got there.

 

The thing is, it's not just about the film print. This still leaves the problem of the digital masters. When I use a 3DLUT to emulate the film colorspace, it's like having a fence built around the color palatte of the image that won't allow me to push colors into non-film digital values. Take away the 3DLUT "fence", and now you will have non-film colors that make it look like video or at least digitally-processed. Things like super-saturated reds, neon greens, bright magentas. Some would argue that they want to be able to get these colors, but most people still want to emulate the look of film and find these undesireable unless they're doing the latest CGI cartoon.

 

Now as a colorist I can be aware of this issue and compensate, but it is a PITA and wastes time. I'll tell you why. First of all, I have to be much more careful of how far I push an image, to the point that I may frequently have to back off from where I really wanted to take something because it introduces those undesireable colors, or that I am having to "split the difference" sometimes. Or, I might take it where I want it, but have to add on some secondary controls that are selecting the most saturated colors and toning them down. Either way, it's a creative compromise and also a waste of my time and energy in the grading suite.

 

And at the end of the day, the elephant standing in the corner is that you can go and buy the Mac Pro and Color and a reasonably decent DLP and do this for 20k, whereas what you'd probably consider a "proper" DI is going to cost six figures just for the grading time, let alone whatever the scanning and recording costs. If you're Steven Spielberg, fine. But otherwise, that's a lot of money for a small production which could, I feel, be much better applied elsewhere.

 

It's not all that different from our strategy. We use Color and 8-core Macs and 15-ft screens with DLP projection. The other considerable costs for us are the SAN and the room and the color management. And as far as client costs, we're able to do the scanning, grading and print for under the six-figure price tag you mentioned. So I guess it's a question for each production, do they really want to go to the expense and learning curve and risk of trying to build their own setup to do this, or are they better off partnering with a veteran DI botique like ours that is sharply focused on features?

 

Talent, of course, is down to the individual, but again, if you're just going for professionally reasonable results, there are a million and one After Effects guys out there who have to know how to grade plates as just a small part of their skillset.

 

With all due respect to motion graphics artists everywhere, I think this is really underestimating the specialization required for grading features. Obviously, as a colorist I'm biased. But I've also been a visual effects artists and am part-owner of our company where we employee many VFX artists and we do thousands of VFX shots every year. We definitely can and have taken VFX artists who have demonstrated a talent and motivation for color grading and helped them transition into being colorists. But in my experience, it takes some mentoring over the course of several films for them to get to a level of confidence to be able to work directly with a client in a supervised grading session. And definitely only a subset of compositors to me seem to really have the mindset and/or drive for color grading. IOW, just because someone is a compositor who knows how to manipulate color controls to match a plate, that doesn't mean they will be good at grading. The compositor's needs for grading are more forensic in nature, where they are matching a plate, or making an element look like it belongs to a scene, etc., etc.

 

If your point is that you could always find a compositor who understands what pushing button X will do, and can basically be your trained monkey who will operate the system for you while you tell them what to do, OK. Or if you have lots of time for trial-and-error for them to hunt down the look with you, and somehow that is not translating into a lot of additional cost, OK. I just think that isn't the experience most filmmakers want. I can see the case of neccessity of cost driving this decision, it just seems like a lot of compromises for a little bit of savings if we're talking about a single film.

 

From my own experience, I have an analagous situation. Last year I produced a very low budget indie film shot on Super 16 (total production cost was under six figures). Because of my involvment here with Liquid, we were able to do all the DI and about 400 effects shots for a song. However, when it came to getting the sound mix done, I worried about it matching the quality of our picture. I am not a sound pro and my company doesn't touch sound, but I am an audiophile and huge home theater nut and definitely know enough to be dangerous. The thought crossed my mind that I could go out and by a Pro Tools or Nuendo system, I already have a good JBL 5.1 sound system, I could hang some acoustic panels, buy a few sound libraries, and have a reasonably competent setup to do sound design and mix for my independent film. Sure, I would never get it "perfect" like I could in a big room with a golden eared mixer. But I think I could get it to not sound "bad" and maybe even be a little "good". Plus, it would be fun for a while, since I've always wanted to do some sound work. Of course, it would take me forever and I knew it would be a horrible distraction from my day job.

 

Ultimately, we had an opportunity to work with Gary Bourgeois at Sony on one of their best stages for a couple weeks. It was not my first time at a sound mix session, but it was my first time with what I'd consider an A-list mixer. Internally I wondered if I would leave reaffirmed that "yeah, I could do this", or if I'd be humbled. I was definitely humbled. I know how to wire sound rooms and a fair bit about acoustics. I probably even know more than Gary does about some technical arcanery in sound. But that doesn't translate into knowing how to mix. I was amazed at how fast and effortless his work was, at how quickly we reached the right decision. I could've spent six months tweaking my decisions in my homebrew sound studio, and I would still not have reached the level of finish he gave us in eight days. So why would I want to try?

 

I now relinquish the soapbox! :)

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Forum Sponsors

Visual Products

Film Gears

BOKEH RENTALS

CineLab

CINELEASE

Gamma Ray Digital Inc

Broadcast Solutions Inc

Metropolis Post

New Pro Video - New and Used Equipment

Cinematography Books and Gear



×
×
  • Create New...