[PiDicus Rex]

Premiere now has the Lumetri colour tool Stewart/PiDicus Rex,

I prefer Edius over Prem - just can't get used to that slow response to user inputs from the horrible GUI, and lack of realtime performance on playback (why don't the waveform monitors update during playback??)
Getting to grips with Resolve too, which shows just how bad the colour tools in Prem are,.. mind you, that needed a second GPU in SLi to make it run decently.

[Steve.Ledger]

My guess is that "highlight correction" and "shadow correction" which requires further in-camera processing will likely further degrade the editability of the footage.
This is why I see it suggested so often to disable all those in-camera processes so that you can end up less 'baked' files to post edit.
Having just recently dumped all my Pentax APS-C (and Q) gear (because it became clear to me that Pentax are not listening to users) to move to MFT, I can tell you that looking back on my Pentax vids now sends shudders down my spine.

[PiDicus Rex]

High-light / Shadow correction, if applied between the sensor and the CoDec, should be no different to the way a CCU operator can adjust a broadcast studio camera on the fly, same as Iris control, knee, black level, etc.
Yes, they will change the way the footage looks, but it shouldn't cause any more degradation then the Codec causes on any other footage.
Turning up (or down) the sharpening will affect the footage more as it's compressed.

Hope you kept the lenses Steve - an MFT adapter is pretty cheap, and the Pentax glass looks better then a lot of other options. MFT and E-mount are a lot more flexible then every other mount, due to the shorter FFD.
Now, if only Metabones would get off their bums and make a K-mount speedbooster,....

Did you look at the JVC LS-300 ? (I'm assuming you went GH4 or BMD.)

[Steve.Ledger]

High-light / Shadow correction slows down in-camera processing for stills, I'd assume it affects video for the same reasons, it's digital routines being applied to every single frame.

Hope you kept the lenses Steve - an MFT adapter is pretty cheap, and the Pentax glass looks better then a lot of other options.

Nope. Sold everything. Lock/Stock'n'Barrel. Would be pretty stupid of me to decide to downsize then shove an adapter + 55-300mm Pentax lens on a MFT body.

Did you look at the JVC LS-300 ? (I'm assuming you went GH4 or BMD.)

No I did not. I don't want a camcorder and I don't want 4K.

[PiDicus Rex]

Downsizing,.... I like the sound of that sometimes, I keep looking at an A7s, or the new a6300, the later being a good small camera, as well as fitting in with my kit to be a 'cinema' camera,... As I do need 4K, just for the people asking 'Can you shoot 4K', even though they're planning HD or 720p content delivery....

[Steve.Ledger]

a6300 without IBIS is just wrong.

[richandfleur]

a6300 without IBIS is just wrong.

Amen to that. So much potential...

[Mahadragon]

Thanks for your replies!
my problem is that I am shooting Herons in their nests in the trees and the scene is very contrasty - there are a lot of deep shadows and the heron's heads have bright white on their sides that get blown when they turn to a profile view. If I expose to save the highlights the resulting footage is extremely dark overall and useless in that way. I am thinking now that there is nothing to do about it short of shooting raw video into a external recorder and the being able to effectively process it. I can't afford that!
Do some cameras have settings that allow shooting video with a flatter response curve that allows for more effective exposure processing? Sony A7RII?
The K-3 can't do that, right?
I think I will have to be more selective with my recordings: shoot earlier or later in the day, shoot on overcast days, etc.
any other ideas?

I couldn't disagree more. Expose for the highlights and simply bring the shadows up to recover the rest. Final Cut Pro allows you to do just that. I would do the exact same thing with my KS2, it's easy to bring the shadows back.

---------- Post added 02-13-16 at 10:28 AM ----------

My guess is that "highlight correction" and "shadow correction" which requires further in-camera processing will likely further degrade the editability of the footage.
This is why I see it suggested so often to disable all those in-camera processes so that you can end up less 'baked' files to post edit.
Having just recently dumped all my Pentax APS-C (and Q) gear (because it became clear to me that Pentax are not listening to users) to move to MFT, I can tell you that looking back on my Pentax vids now sends shudders down my spine.

That's not my understanding at all. I thought those in-camera filters were being added to the image prior to the final raw file being rendered. I know Apple can do this with their iPhone's photos. I could be wrong, but I don't think any raw file coming out of my KS2 is any less editable due to filters being applied. I have all kinds of filters turned on including chromatic aberrations, perspective distortions, blown highlights, and so on.

[Steve.Ledger]

That's not my understanding at all. I thought those in-camera filters were being added to the image prior to the final raw file being rendered. I know Apple can do this with their iPhone's photos. I could be wrong, but I don't think any raw file coming out of my KS2 is any less editable due to filters being applied. I have all kinds of filters turned on including chromatic aberrations, perspective distortions, blown highlights, and so on.

Filters are applied to JPEG, not RAW.

[kadajawi]

Filters are applied to JPEG, not RAW.

Erm... the filters only work when the output is JPEG, and not RAW. That totally makes sense, because RAW is supposed to be the original data from the sensor (though it isn't completely, but close at least). Applying any effects onto that RAW file doesn't make any sense.

The question is if the camera processes the RAW file, compresses it to JPEG, then reads that JPEG, applies the filter and saves it again. In that case you wouldn't be able to do much with the file anymore. However what if the filter is applied earlier in the chain (which seems more likely). Perhaps even before the camera converts the data into an 8 bit image. I think it is quite possible that they do the shadow recovery for example to the 14 bit RAW data, and after that convert it to 8 bit and save it.

[Steve.Ledger]

Erm... the filters only work when the output is JPEG, and not RAW.

Isn't that exactly what I just said??

The question is if the camera processes the RAW file, compresses it to JPEG, then reads that JPEG, applies the filter and saves it again. In that case you wouldn't be able to do much with the file anymore. However what if the filter is applied earlier in the chain (which seems more likely). Perhaps even before the camera converts the data into an 8 bit image. I think it is quite possible that they do the shadow recovery for example to the 14 bit RAW data, and after that convert it to 8 bit and save it.

Quite likely, and would slow down writing, which it does.
However. we are discussing video actually, my point was that in video applying effects would degrade the video IQ owing to the extra processing. IOW: The least amount of in-camera processing to the video stream - the best IQ from the camera.

[kadajawi]

Video = JPEG as far as what we get is concerned. If the shadows are lifted before the 8 bit conversion (and even if not, since the h264 encode will lose data in the shadows and highlights as well) we are indeed getting more detail in the shadows, which can then be used to grade afterwards. Just make sure not to push it too far because being 8 bit and quite heavily compressed it will soon lead to banding.

[Steve.Ledger]

"Video = JPEG as far as what we get is concerned."
Yes, of course I know this.
As to which point in the pipeline the filters are applied, I don't see this making any difference to the overhead required to do so.
At the end of the day I would suggest 'before' and 'after' comparison clips be recorded to see how these filters in fact affect the results.

[kadajawi]

"Video = JPEG as far as what we get is concerned."
Yes, of course I know this.
As to which point in the pipeline the filters are applied, I don't see this making any difference to the overhead required to do so.
At the end of the day I would suggest 'before' and 'after' comparison clips be recorded to see how these filters in fact affect the results.

If the filters are applied before converting to 8 bit, it will result in better quality. Much like if you resize from the full sensor to 1080p, then crop to 720p and then scale up to 1080p, or if you crop to the same frame, then rescale to 1080p.

[Steve.Ledger]

I said 'overhead' - not IQ.
Can you just try it and show the results? Otherwise it's just academic..