What colour Grading system should I upgrade to?

[Adam Eden]

So Dermot, If I work in AVID DNX 36, does the DS link back properly to the original RED footage? Does DS support RED ROCKET yet? And what RESOLUTION are you converting the RED files or whatever to work in 2K on DS?

And finally how to you finish? To you export a DPX sequence or then make a DCP directly from the DS?

All the AVID dealers in Sydney suck and don't know DS unfortunately.

 

[Michael Lindsay]

We were discussing compositing/grading combo, not finishing app.
As far as finishing combo, I just heard, that Autodesk is finishing a brand new grading portion of the Smoke and Flame, replacing CC and CW with reworked Lustre. Couple of months from now there will be no more Lustre. It's discontinued. But new Smoke and Flame would be THE finishing app. May be out by IBC. I'm salivating at the thought...

Very interesting... Makes sense. Is Smac getting the cc upgrade?

 

[jake blackstone]

Very interesting... Makes sense. Is Smac getting the cc upgrade?

I hope so, but that is all the info I got.

 

[shashbugu]

We were discussing compositing/grading combo, not finishing app.
As far as finishing combo, I just heard, that Autodesk is finishing a brand new grading portion of the Smoke and Flame, replacing CC and CW with reworked Lustre. Couple of months from now there will be no more Lustre. It's discontinued. But new Smoke and Flame would be THE finishing app. May be out by IBC. I'm salivating at the thought...

if it appears as a module in smac, this will indeed be quote n quote a serious game changer, because the current cc and cw are really powerful as is. Lustre in Smac wow...

 

[Dermot Shane]

So Dermot, If I work in AVID DNX 36, does the DS link back properly to the original RED footage? Does DS support RED ROCKET yet? And what RESOLUTION are you converting the RED files or whatever to work in 2K on DS?

And finally how to you finish? To you export a DPX sequence or then make a DCP directly from the DS?

All the AVID dealers in Sydney suck and don't know DS unfortunately.

hey Adam!

I'll try to take them one-by one;

does the DS link back properly to the original RED footage?
[d] yes, links cleanly from an Avid timeline (not as good as Symp, but far better than anything else) , EDL or a washed Duck from FCS, AAF from CS5, conforms take a few min for the indexer to find the files, faster if there's fewer files, and if the files are on a fast array to start with, i usualy copy all needed R3D's to one of my arrays before i start

Does DS support RED ROCKET yet?
[d] yes it does

And what RESOLUTION are you converting the RED files or whatever to work in 2K on DS?
[d] I link to the R3D's @ full premium, i start with meatadata and RSX if any exist, then once the folks who need to see it, see it the way they have been looking at it for a while i often switch to camera RGB / PD985 to start the heavy lifting

And finally how to you finish?
[d] i do top to tail in DS usualy, so we finish in the box

And what RESOLUTION are you converting the RED files or whatever to work in 2K on DS?
[d] DS builds 2K uncompressed cashe files and those are what i work from, the cashes always refrence the RAW, so when it's time to output (or check 4K anything) i bypass them.. but then the AJA output is disabled so i'm limited to scaleing in GL mode to callibrated mon -or- viewing the unscaled on the desktop - i have the 30" HP for my desktop, so not too bad, the software scopes in DS are near useless tho, so i usualy sit in hardware I/O mode unless i'm judging scaleing/noise reduction/sharpening damage, then i switch to the 4K raw if i have any doubts

To you export a DPX sequence or then make a DCP directly from the DS?
[d] i export 16bit TIFF sequences for DCP creation @ a specilist, i can't see a reasonable chance of a return on an investment in first class DCP mastering, i seem to have misplaced my rose colored glasses, and my accountant warns me against getting a new pair...

I know of a couple of DS's in use in OZ in DI facilities with some great artists behind the wheel, there maybe some resources to tap into that doesn't involve the Avid dealer

Good luck!

d

 

[jake blackstone]

if it appears as a module in smac, this will indeed be quote n quote a serious game changer, because the current cc and cw are really powerful as is. Lustre in Smac wow...

Yes, it would be a game changer indeed, except... In order to write Prores, they would need to do a MAJOR rewrite of the underlying code for importing and exporting images for Flame and Smoke. The Gateway is a total mess and, unfortunately, it doesn't look like this will change any time soon. And with that, the idea of one app finishing concept is not to be still. What a shame...

 

[Peter Chamberlain]

I think its fair to say that we often search for a post production system to do everything perfectly and that suggests one company has all the expertise in each field of post to make such a product. Then offer it at a price everyone likes. The reality is, developing products for each of the post processes requires quite different expertise and its unlikely that you can acquire this expertise in any product development area in the short term. Red have done an amazing job in a short time working within their area of expertise but we would not expect them to be instant experts in all other areas of production or post. Just as Resolve uses knowledge gained over 25 years for color correction, you cant learn that all in a year. Its the same for editing and VFX manufacturers.

For that very reason, picking the right toolset that integrates with those from different companies, using industry standards, seems a logical way to get the best value for your post $. This is of course a moving target as each developer adds new features, but I do find it hard to see how one company in our field, can have ALL of the best post products for ALL of the fields of expertise.

We would not expect the DOP to be the audio guy, or VFX artist, or producer and products are not different.

Establishing a post process is mainly about providing a service to your clients with a toolset that fits their needs and fills a void in your local market. Find the void the clients need to have filled and the budget they can afford to pay you for the service, then finding the toolset becomes easier. Seems to me that with each of the main editing applications and Resolve being around $1000, you can pick the hardware to run these apps, with shared storage, and have a three or four room facility, or put them all in one room; for what some companies offer as their 'all in one' suite. This way you can have the flexibility of multiple systems, each can be leaders in their field, and not have the risk of the one box in the room falling behind your new competition down the road that has invested less but has more flexibility.
Peter

 

[jake blackstone]

I think its fair to say that we often search for a post production system to do everything perfectly and that suggests one company has all the expertise in each field of post to make such a product. Then offer it at a price everyone likes. The reality is, developing products for each of the post processes requires quite different expertise and its unlikely that you can acquire this expertise in any product development area in the short term. Red have done an amazing job in a short time working within their area of expertise but we would not expect them to be instant experts in all other areas of production or post. Just as Resolve uses knowledge gained over 25 years for color correction, you cant learn that all in a year. Its the same for editing and VFX manufacturers.

For that very reason, picking the right toolset that integrates with those from different companies, using industry standards, seems a logical way to get the best value for your post $. This is of course a moving target as each developer adds new features, but I do find it hard to see how one company in our field, can have ALL of the best post products for ALL of the fields of expertise.

We would not expect the DOP to be the audio guy, or VFX artist, or producer and products are not different.

Establishing a post process is mainly about providing a service to your clients with a toolset that fits their needs and fills a void in your local market. Find the void the clients need to have filled and the budget they can afford to pay you for the service, then finding the toolset becomes easier. Seems to me that with each of the main editing applications and Resolve being around $1000, you can pick the hardware to run these apps, with shared storage, and have a three or four room facility, or put them all in one room; for what some companies offer as their 'all in one' suite. This way you can have the flexibility of multiple systems, each can be leaders in their field, and not have the risk of the one box in the room falling behind your new competition down the road that has invested less but has more flexibility.
Peter

I wholeheartedly agree with you Peter... if it was a few years ago.
Times of a narrow specialist are gone. Not long time ago, if you were a colorist, all you were doing is grading, period. Now, I can hardly remember, when last time that was the case. Editing, compositing, adding effects, marring sound with pictures, titles etc are now expected from the colorist, or should I say- finishing artist, as a routine matter. Resolve 8 made great strides toward easing the roundtripping from and to FCP. Kudos to BM, where credit is due.
FilmLight always impressed me as a company, that looks for a solution in the most innovative ways. I'd like to see BM displaying a little forward thinking like that as well. Here is an example of non traditional solutions. Even with these Resolve 8 improvements, you still need to do a render every time you need to hand it off to another app. Not with FL. Do your grade in BL and without rendering, just open the same session in FCP and all grades are there. No rendering... Well, at least that is what it looked like at NAB
Another example. FilmMaster. At the end of the session with the FM, the session is rendered in the background already. At the end of Resolve session you still need to do rendering.
Scratch. Import sound, grade and render QT with sound. No roundtrip needed. And so forth...
BM innovation? A free version of Resolve:-(((
Resolve 7 is an excellent app and the new version is even better, but I'd like to see a bit more of "out of the box", innovative thinking toward speeding up the handoff to other apps. And please, a little more attention to the interface improvements...

 

[Peace Villow]

Times of a narrow specialist are gone. Not long time ago, if you were a colorist, all you were doing is grading, period. Now, I can hardly remember, when last time that was the case. Editing, compositing, adding effects, marring sound with pictures, titles etc are now expected from the colorist, or should I say- finishing artist, as a routine matter.

Jake, is this your opinion based on what you see around LA or a fact about post production globally?

Even with these Resolve 8 improvements, you still need to do a render every time you need to hand it off to another app. Not with FL. Do your grade in BL and without rendering, just open the same session in FCP and all grades are there. No rendering... Well, at least that is what it looked like at NAB

Jake... you still need to render in FCP before you can playback the clip

Another example. FilmMaster. At the end of the session with the FM, the session is rendered in the background already. At the end of Resolve session you still need to do rendering.

You never use 'Render Cache' Jake? It's exactly the same thing.

 

[M Most]

Jake, is this your opinion based on what you see around LA or a fact about post production globally?

I'm not Jake, but I do live and work in Los Angeles. I would say that the answer to that depends on what market segment you're working in. In network level television work, it is generally not the case. Specialists (i.e., a colorist, an editor, a VFX artist are all different people with different skill sets) are usually used for each of those duties in part because of the rather brutal turnarounds, but also in part because that's the way it's always been done and it's the facility model. In other market segments, such as music videos and smaller indie projects, what Jake is talking about is happening with some regularity.

Of course, I also believe that the facility model is very likely to either change significantly or be replaced in the future. But that's another discussion...

 

[Mark Wilkinson]

Yes, it would be a game changer indeed, except... In order to write Prores, they would need to do a MAJOR rewrite of the underlying code for importing and exporting images for Flame and Smoke. The Gateway is a total mess and, unfortunately, it doesn't look like this will change any time soon. And with that, the idea of one app finishing concept is not to be still. What a shame...

Thanks for your comments, Jake.

Why is it a major rewrite - just seems like a simple transcode on the way out to disk... Seems like a crazy thing to leave out.

Though I must say I wouldn't avoid Smoke just because I had to send a file through compressor now and again - Smoke is a very powerful platform. And if they add Lustre that's a lot of punch.

Please explain further...
Thanks.

 

[Gavin Greenwalt]

We were discussing compositing/grading combo, not finishing app.

Like I said though, without being able to see a grade in context of the edit how are you going to match-grade shots and apply grades to entire sequences? Nuke is great, but the workflow of grading more than one shot in it would be a serious PITA. I graded a short once in Combustion. I love combustion's CC and grading tools but the workflow was horrendous.

 

[paulherrin]

i'm grading a short in nuke right now, but i don't have any better options available. don't particularly love the workflow, but i do love the results!

as far as matching, i just pick my favorite shot to grade, read in the others from the scene, and check them all out with the grade applied, and tweak each shot as necessary. pretty simple stuff, i just don't like doing the file conversions and all that, but it's not a big deal.

 

[Adam Eden]

i'm grading a short in nuke right now, but i don't have any better options available. don't particularly love the workflow, but i do love the results!

as far as matching, i just pick my favorite shot to grade, read in the others from the scene, and check them all out with the grade applied, and tweak each shot as necessary. pretty simple stuff, i just don't like doing the file conversions and all that, but it's not a big deal.

See thats my problem!! Symphony is the best at colour matching and applying numerous grades to multiple clips/sources.

I just need that in a DI workflow.

I just downloaded AVID DS 10.5 to have a play with and already it is giving me pain trying to load a AFE that links to the media from the UNITY

 

[jake blackstone]

Jake... you still need to render in FCP before you can playback the clip

Yes, but you still save a render on the rountrip to FCP. Apply the text, speed changes, sound, any other effects and THEN render

You never use 'Render Cache' Jake? It's exactly the same thing.

No it is not. As it requires selecting each clip for the cache render. In FM it's a default action, working unobtrusively in a background on the whole project. Not the same thing by a long shot. Would it difficult to implement in Resolve? I don't think so. But that is precisely what I meant by being proactive.

 

[jake blackstone]

Thanks for your comments, Jake.

Why is it a major rewrite - just seems like a simple transcode on the way out to disk... Seems like a crazy thing to leave out.

Though I must say I wouldn't avoid Smoke just because I had to send a file through compressor now and again - Smoke is a very powerful platform. And if they add Lustre that's a lot of punch.

Please explain further...
Thanks.

Because present I/O (Gateway) was written specifically to accommodate R3D and at the time it was only able to read. It was never intended to write files. Gateway has evolved since, but fundamental underlying code is still incapable of accommodating writing Prores in it's present form. As I said, they would have to completely rewrite the I/O.

 

[Dermot Shane]

I just downloaded AVID DS 10.5 to have a play with and already it is giving me pain trying to load a AFE that links to the media from the UNITY

Humm... should just automagicly work, when you hit "config alt sources" does it scan your R3D files? Then not link the timeline to them? Or does it not find the files?

There's a load of things this can be, maybe rather than figure out DS stuff on a RED forum, maybe ask at the DS google list;
http://groups.google.com/group/ds-list?hl=en&pli=1
or Avid's DS forum;
http://community.avid.com/forums/34.aspx

It should be working, i've done it so many times it happens in my sleep by now.....

Ta

d

 

[Peace Villow]

No it is not. As it requires selecting each clip for the cache render. In FM it's a default action, working unobtrusively in a background on the whole project. Not the same thing by a long shot. Would it difficult to implement in Resolve? I don't think so. But that is precisely what I meant by being proactive.

Yes it is. You can choose your cache mode in Resolve.
'Cache U' is the one that only do the render cache for the clip that you already mark.
'Cache +' is the one that render cache your transition.
'Cache All' is exactly the same thing as the default background render in FilmMaster.
You just need to turn it on, since it is not a default config in Resolve.

 

[jake blackstone]

Yes it is. You can choose your cache mode in Resolve.
'Cache U' is the one that only do the render cache for the clip that you already mark.
'Cache +' is the one that render cache your transition.
'Cache All' is exactly the same thing as the default background render in FilmMaster.
You just need to turn it on, since it is not a default config in Resolve.

And Resolve would keep rendering in the background as you're grading the timeline and keep updating as you make changes? Ooops, not really Resolve caching designed for real time playback, when your GPU real time rendering is not up to task. FM background caching is a standard way of operation. Big diff...

 

[M Most]

Because present I/O (Gateway) was written specifically to accommodate R3D and at the time it was only able to read. It was never intended to write files. Gateway has evolved since, but fundamental underlying code is still incapable of accommodating writing Prores in it's present form. As I said, they would have to completely rewrite the I/O.

It should also be noted that the design of all current Autodesk systems products is based on dealing only with uncompressed image sequences internally. One purpose of the gateway is to allow for new file format integration without having to change the core code. This allows the core operations to occur without the burden of compressing or decompressing multiple layers in either a timeline or an Action composite. And yes, changing that would be a major rewrite and would likely affect performance.