[Bruce Greene]

[Kevin Monahan] “Your computer monitor can’t display YUV”

My Sony broadcast monitor can’t display YUV either: It converts the YUV to RGB for display. Photoshop does this all the time as it converts into and out of LAB color space. I don’t think this is the problem. The challenge is that we’re thinking “the old way” is the only way.

I think this issue is becoming even more important as more content is watched on computer screens everyday. We should at least be confident that what we see in FCP will look the same on a properly calibrated computer screen elsewhere.

[Bruce Greene]

[walter biscardi] “If photoshop can do it, so should FCP.”

The problem with this train of thought is that FCP’s Viewer and Canvas images are purposely degraded because Apple assumes correctly that we all use external reference monitors to view our work. So if you truly want to know what your image looks like, you need to view it via an external video feed.”

I for one, would not need a full res image to color correct. I think this approach is a legacy of the weak computing power when FCP was new. I think today, one should have the option to see accurate images in the canvas at the price of fewer realtime effects.

As the price of production hardware keeps going down, more and more FCP users will be less financially endowed and will use FCP for finishing their projects as it will be the only alternative. The requirement of a very high priced broadcast monitor and I/O card seems outdated to me.

And so does anyone know the answer to my question: What’s the best way to set up my mac/FCP today so that the canvas image is as accurate as I can get it?

[Bruce Greene]

Thanks Glen for the long and well thought reply. Unfortunately, you’ve pretty much summed up what I already know and practice (thought I actually use a sempte c sony crt that’s in pretty good calibration.

I guess, with computer monitors so easy to calibrate with affordable probes and software, I think we should be able to view the image in FCP on a calibrated and profiled computer display as a standardized monitor.

If photoshop can do it, so should FCP.

A good computer monitor plus a computer should easily be able to equal a 20,000 dollar calibrated LCD production monitor.

So, does anyone know what gamma FCP assumes the computer monitor is set to?

…

[Bruce Greene]

I’ve never seen one of these up-resizing programs do anything that looked very significant. If it were me, I’d just use Photoshop to resize the jpg with perhaps, a little sharpening.

There really is no miracle here, when upsized, images don’t look as good as we’d like past a certain point.

[Bruce Greene]

in the dvx menu, try setting the gamma to “cinelike D”. This setting offers a little bit lower contrast and brightens the midtones that get dark as you close the iris to hold the highlights.

Other than that, I would use reflectors to fill in the shadows if you can.

Also, make sure that your camera set-up is not crushing the blacks with a pedistal setting that’s too low. I seem to recall that when using the cine gamma settings that the ped be set to something like -6 to make black expose at 0 IRE. If you’re ped is set to zero, you’re probably safe from crushing any blacks, but they might not be the blackest the camera can record. For your high contrast scene this may be a good thing.

Of course, test this out before you shoot.

[Bruce Greene]

Thanks CaptM!

[Bruce Greene]

Paul,

Yes, what you’re missing is the timing of the frames. When you shoot at 60p, you shoot one frame every 1/60th of a second. To get to 30FPS you’ll be ok by just dropping every other frame and you’ll get 30 evenly spaced frames for every second. To get 50 FPS from the 60P footage you’ll have to keep most of the frames, dropping one every once in a while if you will. Thus the motion will be uneven. It will be just like when you shoot 30FPS and try to convert it to 25FPS or 24FPS. The dropped frames are obvious by their timing.

The Varicam on the other hand will shoot 50 evenly timed frames per second when set to 50FPS (or 48FPS or any other setting) Duplicate frames are slugged into the tape that are flagged for disposal later (to fill the 60P requirement of dvcproHD), but the real frames will be shot every 1/50th second not every 1/60th sec like on the 900.

I hope this makes sense as written.

-bruce

[Bruce Greene]

CaptM,

Thanks, but I’ve tried this already and the problem is that I can’t apply the resize to the whole sequence because it keeps bringing me back to the original sequence and it’s underlying clips. IOW, there is no motion tab for the nested sequence, only for the individual clips in the original sequence.

Is there a filter that I can apply to all the clips to perform the resizing? Unfortunately, one can’t copy and paste motion attributes the way one can with filters.

[Bruce Greene]

Jay wrote: “Anybody run into these two things before? Bugs of the H? Or VariCam gone bad?”

Mine’s just like this one, not sure about the lens files but it’s always seemed tricky. Since I own one lens, it’s not the biggest issue to me:)

Varicam/Steadicam Owner
Los Angeles, CA
http://www.brucealangreene.com

[Bruce Greene]

Dear Mr. Carpetbagger,

I’ve noticed the same thing about the noise, but there is a way around it sort of.

First off, the noise is most obvious in Vid Rec mode. Vid Rec mode is the gamma menu mode which is most like the Cine Alta. At first I was really bummed because I own the Varicam and thought I’d made a mistake here.

But after some more thought, the cameras are a little more alike noise wise.

At 0 db gain on the f-900 the ISO is about 320. On the Varicam, 0 db gain is about ISO 800. Quite a bit of difference in light sensitivity. So, if shooting with the Varicam in Vid Rec mode, use -3db gain only to minimize noise. -3db gain still results in an ISO of 640 which would equal about +6db gain on the f-900.

That said, in Vid Rec mode on the Varicam even -3db gain is a bit noisy in the shadows. To improve on this further, one must shoot in the Film Rec mode.

The Film Rec mode is designed to shoot at higher dynamic ranges than normal. The idea is to capture a flat, low contrast image, and adjust the look in post or a film out. Film Rec mode allows a choice of four dynamic range settings: 200%, 300%, 400%, 500%. The idea is that 200% means that you record twice the dynamic range of standard video (with knee off) and 500% is 5 times the dynamic range etc. The limitation of this mode is that there is no Knee adjustment allowed (note that manual knee must be turned on in the menu to activate the dynamic range settings in Film Rec mode though).

For Film Rec mode, Panasonic seems to suggest that one expose as if the sensitivity of the camera is ISO 640. This will leave the most of the image at low IRE range on the waveform and allow a very large ceiling for highlights. The result is that images look dark on the monitor and need a significant gamma adjustment in post to make them look “normal”. This gamma correction will also magnify the noise back to the level seen when shooting in the Vid Rec mode (along with additional increase in the visibility of compression artifacts).

Another way to use the Film Rec mode is to try for a Wysiwyg approach. IOW, expose and adjust the gamma in the camera so that it looks “normal” on the monitor. And this method really reduces the noise to f-900 levels. For starters I would try setting the dynamic range to 300% and the gamma to 55 or even 60. This will produce a very noise free image. The drawback is that you’ll be shooting now at ISO 250 at -3db gain. At 0db gain, you’ll be shooting at ISO 320 which will still be noise free and the same sensitivity as the f-900. The 2nd drawback is that you will not have any knee adjustment, but you can always lower the gamma setting a bit and stop the lens down to capture highlights. On the plus side, you won’t see any of the funny color shifts one often sees at the knee point because there is no knee point.

About the “compression mode – dark” setting: To see it’s effect, you must record the scene onto tape and play it back. This setting does not effect the live image at all. I’ve played with this recently and have not really been able to play back the footage on set to really learn how much it helps the compression artifacts in the dark areas. I would definitely turn it on when shooting low key scenes. On a day exterior it would be hard to see, but might negatively effect compression in the highlights. Anyone with more experience here is welcome to comment on the compression mode setting.