dermot.shane

if there's a  good uni level art school nearby, see if you could audit their version of color theory 101 course?

also check out what is nearby in computer science/color and medicine/color at any local uni,there may be courses on offer that can be of great value to a colorist if allowed to audit

it's been many years since i last used a LUT beyond reverse engineering it, the math errors of some of the stuff the cleints drag through the door are amazing, and not amazing in a good way either...

sure, i set my output to match the display,

normaly i'd grade in 709 so screeners are somewhat accurate, the path of least resistance + it's as poor a represantation of the master as i will have to show.... about client managment,, working on HDR or P3 and then sitting down for a trim pass to 709, it just seems sad, grey and gloomy, hard to have love for that, but that''s what most people will be seeing.. so starting with 709, then it only gets better....

when 709 is in a happy place, then set ODT to match proj at P3, and fix what needs to be fixed, tweak what wants to be tweaked

rinse and repete for the favored flavors of HDR

cleaner math,

look at a transform & a lut over a grey scale

contrast and compare

not all LUTs are crap, but...  far too many are

ACES + trim pass here

Avid|DS,

eol'd a few years ago, i have two licences running, so far there's nothing better for my workflow,

next best is Flame,

and maybe by version 17, Resolve

it really only needs Fusion to be a bit more stable and responcive, parent /child timelines & sane / reliable audio mapping, not too far given where it was only a few years ago

but if i had to mothball the DS's for any reason, i'd turn to Flame today

broadacast = production sends me a DRP + media, i send them roundtripped media + XML/AAF

Featiures = production sends me a DRP + Media, then more DRP's as VFX trickle in

i grade and runout an uncompressed colortimed master file seq @ largest raster

that goes into a finishing system for tweaks/fixes and all that not so fun stuff, do the roller and then create the deliverables from there

often i start the finishing process early by running out a dummy file seq and when scenes are finished gradeing they overwrite the placeholder seq, and appear in the timeline of the finishing system automagicly

if production cannot sort out getting me a DRP, i ask them to send the project and media to an assist, and they deliver a DRP to me fully sorted and ready to rok, that's at the producer's expence, but cheaper than tieing up a gradeing suite looking through the rushes for a missing shot

the finishing system i use offers parent /child timeline relationships, so a change to the main timeline ripples through all timelines, a change to the master HD downconvert timeline only ripples to the HD delvierables, on and on.. a huge safety valve for dealing with QC notes and mutiple deliveries with differing mixes and show runtime's

oh, and really simple and reliable audio mapping

@ Thomas

No i had not seen that, i knew that something was in the works, but any talk of that was a long time ago, and the FCX trainwrecks keep appearing

last one cost the preditor two days of edit assist time to get a timeline that was clean enough to start conforming, like i mentioned i do not even touch a FCX project anymore, instead asking for a DRP / XML / AAF from Resolve, Pp or MC, and let the preditor get that sorted before it darkens my doorstep

Chapter threeeeee wheeeeeeeeee

no real issues with Pp & MC projects, do not see FCp7 anymore, never seen a Lightworks projec

good suggestions, many thanX for looking at this

is this what i might send on to a FCX preditor?

"simplify and Organize the timeline:

flatten video to primary storyline

break apart compound clips."

- i mentioend i don't see how FCX software could get it so very wrong, so very often, i'm thinking it's the training or lack thereof, coupled with the entry point focused user base that's the root of the pain - so your input would be valuable if you can edit in FCX  and finish to a high standard / pass QC in anything else

- as to the free Resolve idea, i do often suggest that, and it's never been a workable solution, it's appears to be far to hard to figure out for a person who only knows FCX, or their machine cannot run it...

- flattened timelines are not workable for me, i usualy deliver DCP for theatre and prefer to work from cam orig or VFX plates to control the quality of the image and the project's color managment, no matter the software used in editorial

many / most broadcast projects are roundtripped back to Pp or MC for finishing so effects come back home cleanly, for theatrical i do the finishing, and eye match as best possiable, although i do not use Resolve for finishing, it's getting better every version, but the weaklinks are too weak of a link currently for my comfort, and i have options at hand that are dead reliable already

a few issue wit h Pp and MC is to be expected, nothing major, all the shots show up in the correct place

so nothing near the level of nuttyness that appears from FCX typicaly

i'm thinking it's mainly due to the FCX crowd...  the folks i deal with on MC projects are usualy union assists, and are super well orgnised top to tail

the folks i deal with on Pp projects are usualy pretty good, not up to IATSE standards, but clean timelines appear

the timelines i have seen from FCX are layers of overlapping media, no order for priority, something that is wanted to be seen is on v3, and something useless and not needed is covering it on v17, on and on....

thinking it more about the lack of training, only folks with next to zero money + next to zero experience seem to be cutting in FCX, and then their film gets some legs and they have to finish it to a reasonable standard, meaning outside FCX, usualy preditor's not editors

the other issue is i have a long list of Pp and MC trained assists that can sort pretty much anything, and no resources to turn to localy for FCX, it's like the blind leading the blind

a good union assist can clean up a film cut in FCX in a few days, but that makes it more expensive than buying a perm licence of MC and getting it right in the first place

i would love to have a cheat sheet i could hand off to a FCX preditor, save them the expense of getitng their trainwreck sorted, and i can focus on getting their film to shine like a diamond

many thanx for listening, sorry for the bad news right up front

i see the vast majority of FCX conforms being total trainwrecks, to the point that i ask for the project be sent to an experienced assit to sort / conform and deliver me a DRP, so i don't have to get near it with a 1000 foot pole

to that end, any cheat sheets / check lists for an editor without experience handing over a project for gradeing would be vastly appreacated

i don't touch audio, i ask for all audio tracks to be removed before making the XML, if they still appear in my XML i strip them out before going one inch further, then delete all audio only tracks from the media pool

i ask for a refrence with scratch audio, and use that when screening

normaly the audio has gone to a professional who will send back to the editor what ever the deliverables call for, and that is laid up against the XML from color

in this case if the editor is doing the mix inside his system (yikes!!)  then he should be exporting the mix as per deliverables, and importing that into the project with the new graded media

not Kevin, but i have been doing this a fair bit

they editor opens a clean project, opens the XML and it automagicly links to the graded media and builds the timeline with graded media and effects that are embedded in the XML

one more thought, render with source rez checked, and bypass sizeing so the image does not get scaled twice, any scaleing  embedded in the XML will still be in the XML output from Resolve

i seem to remember a million or so years ago importing AAF into fusion,  and re-creating title media - this would be circia 2006, and i could be completly wrong...

so my questions are;

-do i remember correctly?

- can it be done in Fusion9 or ReFusion?

i think de-interlace is only active in a progressive project, but i only rarely see interlaced footage, and can't remember last time i saw a interlaced project, so i'm not of much use am i?

i've got two frontline machines, no issues doing supervised UHD sessions with a 2 x 2690's /1x 1080ti, no real diff between that and the machine with 2x2697v2's and 2x 1080Ti's in a session, they both respond pretty much the same from the surface, need to cache third party OFX in either,  don't have to cache native OFX if they are GPU acelearated, never have seen a out of mem error @ UHD

the diffrence comes into play when rendering out after the client leaves 😉

in Resolve use "re-conform from media pool bin", and target only the cam orig bin?

and another Q  for the smart guy...

is there something like "commit mutli cam edits" in FCX?

and a way to flatten the timeline?

i don't get alot of FCX, a few shorts a year

mostly the commericals, movies & series are cut in MC, and the indie features and some commericals and shorts in Pp

i kinda dread FCX conforms as they seem to fall of the rails far too often... like one i had recently with a timeline  @24, and refrence Qt at 23.98 (or was it the other way around?) mutiple tracks overlapping each other in seemingly random order...  was a nightmare best forgotten... or learned from - hence me asking you 😉

one more note, pg84 of the Resolve 15.2.1 manual explains how to map Mac drives to a windoze drive letter, can make conforms go a bit faster, not a huge deal on a one-off, but for series it's pretty useful if the media arrives mapped to the client's server time after time...

the Tuxera thing was just because ti was bundled with a external drive, as long as the drive was pluged in the software would licence it's self permenatly on any number of machines... so i've kept that disk around for years now (think it's a 320gig)

we bought the software years ago anyway because of bond requirments to have a licence for all software used

@Kevin

great summary for those of us who get asked for FCX to anything workflow and have no idea about FCX...

wanted to do a shout out for Paragon's "HFS+" - i've have it on mutiple windoze machines for years now, read and write HFS (and now APFS) trouble free

on the mac side we have "Tuxera" to write to NTFS, again been useing for years now with out ever an issue to be found.

it's in the transitions bin,

don't expect wonders, (but that's true for Pp and Avid's morphcut as well)

heavy lifting needs to go to Flame or Nuke,

humm.. might be a good tutorial if Fusion's tools are up to the task

nothing happens when you press them... so far

may change someday, but for now - no joy

some thoughts;

- really depends on the chip used, not all red camera's need the same processing, a Helium is not a RED1....

- i do not find any modern RED to be noisier than any other good camera, at the least, not objectionably noisy

- i tend to avoid noise reduction unless really, really needed, and then i turn it up until i can see the scopes start to clear up, then cut the values in half

- the exception being using Clairity in Nucoda

- Paul Dore has a usefull OFX that displays noise per channel

some questions

1- how does this scene fit into the story arc?

2- what is the main deliverable; theatre / web delivery / tv daytime / tv primetime / tvc ?

3- are there any weak links in the rest of the scene that shots have to match?

4 - how much time can production afford to throw at the show / how many days do you have to get the entire project complete?

depending upon the answers... i'd approach each option diffrently

suppose it goes back to one's prefrences to work in display refered, or scene refered... 

i prefer scene refered, and find matching these camera's to be pretty straight forward with the correct IDT's, no luts needed (or wanted really) in my choice of workflow

in Resolve my default node tree has a node set to HSL, with Ch1/3 bypassed, so i have control of saturation on the rings for tweaks in regions

most of the heavy lifting gets done in L*a*b nodes tho