Frank Wylie
-
Posts
52 -
Joined
-
Last visited
Content Type
Profiles
Case studies - Free
Case studies - Premium
Resources
Insider
Courses
Forums
Store
Posts posted by Frank Wylie
-
-
Just my 2 cents...
To my eyes, it has a color pallet that is equally-split between Ektachrome 100D/Fujicolor Provia 100 reversal , but with well managed highlights, avoiding clip and retaining detail.
Blacks tend toward crushing at the extreme, all colors appear to be destaturated, but reds are desaturated a bit more with a tiny shift to magenta and overall exposure is well below clip.
Looks like a lot of very delicate qualification work has been done on faces and specific background details to enforce the look; I don't see a generic LUT being able to encompass the entire ambient representation in a linear fashion from full dark to full bright...
Of course, I am looking at the clips on a generic computer monitor in crappy lighting conditions; YMMV.
-
Another way to think of it; in traditional film timing, RGB controls your color/density and your LUT is your film processor and stock characteristics and you only can have ONE LUT for the entire reel...
-
(Hope this is the right place to post this; I guess a Colormaster is a form of grading "app", albeit a rather large one!)
Apologies to Stig and all for being so scarce lately, but I have just spent the last year searching for, procuring and waiting for delivery of a replacement film analyzer.
Not east to obtain in this day and age but it finally arrived! For the last 3 hectic days I was assisting Media Migration Technology (formerly RTI) install our new system. It replaces one of our HFC 300D Hazeltine analyzers that was inoperative and allows me to return to electronic grading of negatives instead of grading them by eye over a light table.
Being that there's no real analog grading forum, I hope at least this could be a diverting trip down memory lane for anyone who remembers film to film production...
-
4
-
-
Not as of yet. I know we loan a lot of prints to the BFI and others in the UK and will post an update if I am notified of a screening.
-
I rarely get a chance to "blow my own horn", but here are a few show dates around the World for B&W 35mm prints I have timed. Please understand that these are prints timed from either the camera original negative or the best available archival film element. These are NOT digitally restored films, but are straightforward prints from film elements that can be 30 to 90 years old so they are not perfect, but are a pure film experience.
Of course, our entire lab staff is to be credited for the work, not just myself, so keep that in mind if you have an kind words for our work!
NOTICE: This list will be updated as new screening dates are announced.
July 25th, 2018
"The Maltese Falcon" (1941), 35mm B&W at the UW-Madison Cinematheque
http://cinema.wisc.e...er/35mm-forever(No Play Dates determined) (1940) D. Garson Kanin MY FAVORITE WIFE in Locarno Switzerland (Festival in Locarno starts 8/1),Cine Francaise (September), possibly Kino Rex in Berne (October),
August:
Heights Theater
Columbia Heights, Minnesota
http://www.heightstheater.com/
8/2/18 "Gold Diggers of 1933" (1933) D. Busby Berkeley and Mervyn LeRoy
Film Forum
NYC, NY
9/27/18 "M" (1951 remake) D. Joseph Losey
September:
Arsenal; Berlin, Germany
https://www.arsenal-...in.de/home.html
9/17/ 18 "Craig's Wife" (1936) D. Dorthy Arzner
9/22/18 "M" (1951 remake) D. Joseph Losey
October:
Melbourne Cinematheque
Melbourne, Australia
http://www.melbourne...no-trailblazer/
10/10/18 "The Hitch-Hiker" (1953) D. Ida Lupino
Film Forum
NYC, NY
10/1/18 "Hotel Imperial" (1927) D. Mauritz Stiller
Festival Internacional de Cine de Morelia, October 20-28, 2018
Morelia, Mexico
https://moreliafilmfest.com/en/
(TBD) "In Caliente" (1935) D. Edward Chodorov
Filmpodium
Nüschelerstrasse 11, 8001 Zürich, Kinokasse
(TBD) "M" (1951 remake) D. Joseph Losey
November:
Kansallienen Audiovisuaalinen Instituutti
Helsinki, Finland
(TBD) “The Road Back” (1937, Restored Long Version) D. James Whale
-
1
-
-
Ah, that's a hard one...
Might key it and pull a matte, reverse it and erode the magenta out, but that will typically leave a smooth, defined edge that looks artificial. If you could find a way to inject some noise along the matte edge, it might composite successfully over a replacement sky, but you'd probably have to grain/noise that up too in order for it to blend.
Hard to speculate without seeing the image...
-
Why isn't sky replacement an alternative when there is no detail to enhance?
In the distant past, cinematographers used to carry around large diapositive glass plates with images of clouds on them to cover bald skies. They would put them on a C-stand, position them in the frame to fill the sky and shoot.
-
2
-
-
I'm sitting here listening to the live NAB feed from Blackmagic while eye-ball timing a finegrain from the silent era...
The irony is palpable...
-
1
-
-
The .dpx format is strictly a wrapper; the "essence" (image and metadata) can be just about any color space, resolution, encoding methodology, etc...
Have a peek at the file headers to see if there are clues there, or, ask the file owner just what they are!
-
"Film print emulation" is what I have always heard, but I assume they all mean the same thing...
-
1
-
1
-
-
In any event, would assume the LUTs came from the film stock manufacturer, but I could be wrong...
-
I'd urge you to stick to the industry adopted standards.
This is nothing new under the sun; the same problems existed in motion picture film distribution. Theatrical venues varied widely from theater to theater in color temp and screen brightness (among other things) and even film labs didn't totally conform to agreed SMPTE standards, but at least everyone had a common nodal point around which to cluster in regards to what constituted an acceptable quality image.
If we start independently chasing the latest whim of TV manufacturers, there will be quite a mess.
-
1
-
-
I'll dodge your question directly and send you to Mr. Art Adams!
https://www.provideocoalition.com/the_not_so_technical_guide_to_s_log_and_log_gamma_curves/
-
1
-
-
There is news of a new iMac Pro Workstation, but after looking at the specs and the price, I'd have to say it's too little, too late...
https://www.engadget.com/2017/12/12/apple-imac-pro-launches-december-14th/
-
1
-
-
I don't know of an unsharp mask plugin (like in Photoshop) for Resolve; too bad.
Short of that, you might try splitting out the luma from the chroma and manipulating those independently, but truly out of focus material is really hard to recover.
Josh Petok shared a tip from Mike Seymore's Red302 class at FXPHD:
When sharpening images, sometimes you'll notice a color fringing around the edges. This image has sharpening set to 20.
Here's the same image zoomed in 200%. Notice the fringing around the edges.
In Resolve, by default, the sharpening algorithm is performed on both the luminance and chrominance of the image. Ideally, images could have cleaner sharpening if it was applied only to the luminance of the image. Here's how you do it:
Start by creating a Layer Node
In the first node, lower the saturation to 0. This will remove all the color information from this node.
In the second node, lower the luma gain to .01 (the lowest that it will go)
All that is left in this node is color information.
In the Layer Node, change the Composite Mode to "Add"
If you enable/disable all our nodes at this point, your image should look exactly the same as it did before. All we did was separate the color from the luminance.
Now that your color and luminance are discreet elements, add a serial node after your luminance only node.
Apply a healthy amount of sharpening
Notice that the color fringing is gone from the edges
-
1
-
-
Does the main grade still look the same on the IMac? Have only the stills changed?
Any chance the IMac was upgraded to the latest version of IOS and not the laptop?
High Sierra will also see Apple move to the H.265 video standard to support better 4K playback. H.265, also known as HEVC (High Efficiency Video Coding) compresses video 40% more than H.264, and means high-def videos will take up less space on your Mac. High-quality video streaming will be possible too.
I wonder if the stills function is somehow tied to the H.264 codec...
Sounds like a machine-dependent difference. Has something has changed on the Imac and not the laptop? Done any upgrades or recent software installs?
-
Thomas,
I spoke to BM about Revival about 3 years ago and they said they will not be retooling it for general release; just using some of the code for integration into the DR suite.
As to the dustbusting feature, as Remco notes, it only works on files in .DPX format, and I have found it to have a very real limitation on the size of the repair that can be made without introducing artifacts. You have a limited number of options to choose from which frame to steal pixels for the repair (prior to or after the target frame) and the frames you correct are permanently altered once you save your project. Undo is only available between saves and undo tends to undo ALL repairs prior to the last save, so you have to be careful when and where you do save the project.
I tend to make up to 3 repair before I save a project. I once made the mistake of making about 20 repairs and then hit undo (which is NOT the typical undo for DR, but a separate undo in the drop down menu) and was a very unhappy camper!
Also, be aware that if you are working with optimized media or proxies, the dustbusing repair is NOT shown on the active frame buffer, as you will be looking at a proxy that does not reflect changes in the original file! So, you will find yourself trying to repeatedly remove the same spec of dirt until you turn off the proxies, and then magically, it will disappear and show you the artifacts of multiple dust removal attempts!
You could, of course, re-generate proxies on a shot by shot basis as you go through the project to reflect the repairs in the active color grading buffer, but you'll need to keep careful notes as to which shots have been re-generated.
Hope this helps (and makes sense, only had one cup of coffee so far...)
Frank
-
2
-
-
Nicholas,
Lots of qualifiers here; are you speaking of rendering independent of copying (i.e., copying one set of files while a project renders in the background) or are you speaking of the rendering speed difference between reading/writing to the same disc Vs reading from one disc while rendering to another?
Are both discs internal?
What are the interfaces on each disc?
Too many questions to make a good guess here...
-
Nicolas,
I have been able to recover data from totally dead drives by simply replacing the driver board on the HDD.
Now, since I bought the drive that failed as one in a lot of 20, I had a lot of exact spares from which to attempt swapping the interface board.
If you don't have an exact match, try this place: http://www.hddzone.com/
"Most HDD PCBs' failures are caused by Motor Controller Chip burnt, then the TVS diodes burnt and Main Controller IC burnt."
This is MUCH cheaper than a commercial service, and, they probably try this first ANYWAY!
Frank
-
In traditional color grading, you don't mess with a fixed gamma that has been determined for the project; you adapt the images to that gamma.
Added to this is the logarithmic response of the print stock (which has it's own unique palette) the light valve values, trims of the printer and even the influence of local water supplies used in compounding the developer.
This causes some interesting problems for the timer, who has to establish a uniform methodology of bending the subject matter to the will of the director and DP.
The difference between digital color grading and film timing is both aesthetic and locked in the physical properties of the respective mediums.
-
2
-
1
-
-
In my book, the subject matter drives the look, so in my opinion, it's hard to "achieve a look" when you've got nothing to see!
Short of that, you could acquire a few image sequences and then study a few painters to try to emulate their styles and pallets.
Might I suggest John Singer Sargent, Pierre Auguste Renoir, Diego Rivera and Andrew Wyeth as a random starting point.
It always seems to go back to painting, somehow...
-
1
-
-
Err, ah... that should have read, "across the spectrum"...
-
As Bruno states, Kodak stocks tend toward the "warm" spectrum in color and Fuji is much more "neutral to cool".
It can be very subtle, but a simplistic method would be to just reduce the reds across the RGB spectrum for Fuji and boost them a bit for Kodak.
Frank
-
Abby,
So sorry for the delay in responding. I have gone both forward into Digital Color Correction and backward into literally timing like a silent lab technician at once! My Hazeltine Analyzer broke down and, due to a lack of available technicians to repair said beast, I have had to resort to timing by-eye over a light box, so my days have been rather full.
That and certifying a new Resolve System we just purchased...
Talk about your dichotomies...
I am finishing a print on "Finishing School" (1934) with Frances Dee, Billie Burke and Ginger Rogers, as well as certifying a show print on "Hotel Imperial" (1927) with Pola Negri, James Hall and George Siegmann.
Unfortunately, both titles have severe issues with generational loss, decomposition and damage; You do the best you can...
Positive elements (to make dupe negs) are not as hard to "eyeball-time" as negative (to make show prints), but I am somehow managing.
Needless to say, this development has slowed my throughput down considerably.
Hope your technical woes are less severe!
Frank
"The north water" look
in General Discussions
That is really difficult to judge from this monitor; I can see some tendency toward warmth within the lighter values but still can pick-out pure whites. If there is an overall warm cast, it can be "pierced" at a certain luma level with pure white.
Maybe the highlights have a roll-off of red toward clip on a curves adjustment?
Again, not judging from the best monitor possible, so just guessing...