abrupt highlights clipping

[Christopher Barry]

Piotr, I don't judge you on camera ability. I too have my EX1 learning to come.

I loaded those frames in PS and reduced the saturation value to -50, then the trees and sky don't stand out so much. Perhaps the funk is a combination of these particular settings, coupled with exposure? Hisat on, perhaps try another?

[Steven Thomas]

Piotr, I believe what their saying is that the EX1 has more latitude; therefore, how it rolls into over exposure and how it's perceived is going to be different, than say your V1E.

I'm not knocking your opinion, it's just that you seem to be the only one concerned with this.

I've taken great day light shots with plenty of sky holding decent latitude allowing good exposure for the bottom half of the image.

[Piotr Wozniacki]

Originally Posted by Steven Thomas

Piotr, I believe what their saying is that the EX1 has more latitude; therefore, how it rolls into over exposure and how it's perceived is going to be different, than say your V1E.

I'm not knocking your opinion, it's just that you seem to be the only one concerned with this.

I've taken great day light shots with plenty of sky holding decent latitude allowing good exposure for the bottom half of the image.

Sure, the problem is that you have no real benefit of this latitude: here is the grab perhaps some half-a-stop less exposed that image54 in my previous post; this is how the sky really looked (well, almost: there still is a single, unnatural patch of light blue there). Why no gradation between the blue tones, but the abrupt change in colour? This simply shouldn't happen - nobody will tell me one of these two grabs is poorly exposed (if anything, perhaps slightly off-focus, as it was a walking shot). And when you watch those changes in the actual video, it looks even worse because the patches appear and disappear constantly all over the sky.

[Christopher Barry]

Saturation, over.....

[Piotr Wozniacki]

Originally Posted by Christopher Barry

Saturation, over.....

Yes, it's all with Hisat matrix. But the Standard one looks muddy - and again: on the V1E, when I choose Cine1 gamma and Cinecolour on, I have everything at least as saturated as with the Hisat on the EX1, but still no patches when slightly overexposed.

I'm going to give the Cine matrix with one of Cine gammas the last try tomorrow.

[Steven Thomas]

Piotr,
Are you believing that the closely paced branches in the sky view do not reduce some of the sensor illumination at those given points?

I believe Alexander and a couple others have mentioned this. I'm not sure why you think it should not. Also, since I'm away from the office, I did not bring the image in and analyze it, but the sky sure looks over exposed in some locations. I know you doing the over exposure on purpose, but I'm not sure where this is going....

[Steven Thomas]

Piotr, if you think there's a problem with your camera you may want to return it for exchange. If you feel it's across all EX1 cameras, I suggest asking for a refund if they will still honor it.

[Piotr Wozniacki]

Originally Posted by Steven Thomas

I know you doing the over exposure on purpose, but I'm not sure where this is going....

Exactly Thomas - I am over and under exposing on purpose, to learn the new camera limitations and discover its strengths. Where is it going? Well, I thought I said many times: I'm trying to find a way to avoid abrupt colour changes (nasty patches) where the camera is just on the verge of, or slightly crossing, the clipping boundary. So far I haven't succeeded; I agree that tree foliages are enough to limit the amount of light just like some sort of ND filtering, but the resulting patches are washed out of any colour gradation, as if the colour resolution was insufficient (I'd call it near-clipping banding). The V1 doesn't exhibit this, so please don't tell me it's normal.

[Piotr Wozniacki]

Originally Posted by Steven Thomas

Piotr, if you think there's a problem with your camera you may want to return it for exchange. If you feel it's across all EX1 cameras, I suggest asking for a refund if they will still honor it.

Too late - I've just sold my V1E:)

Seriously though, I still hope I'll find a solution. BTW I am trying to be as neutral in this topic as possible, even though have been attacked a couple of times myself as a "unfuctional" operator, so let's just forget it for a while.

Nobody even commented on the vignetting caused by the ND filter - could it be something quite separate from the infamous vignetting problem?

[Michael H. Stevens]

Piort:

To get you out of this rut, let the past go and start afresh like this, then see if you have same problem:

1. DO NOT USE STD curves. They are not made for the likes of us. There are just there for comparison or to add some bells et. Also with the CINE curves knee is automatic so that goes out of the equation.

2. Try CINE4 it gives most detail in the blacks.

3. Choose HiSAT but leave all other settings flat.

4. DONT EXPOSE BY THE HISTOGRAM UNLESS YOU HAVE CALBRATED IT. On my camera it grossly over exposes even when the LCD look reasonable. YOU MUST USE ZEBRAS.

4. DO NOT USE ZEBRA 2, it is set at 100% which is way too high for this camera when in good light. Set your Zebra1 to 95 and expose so you see no zebras, yes NO ZEBRAS, in the sky.

Then tell us what you see in your NLE.

[Bill Ravens]

Bad advice, Stevens. I disagree with practically everything you've said.
I believe the STD curves may not look too nice on the VF, but, they capture the most detail; and, it's up to the editor to bring that out in post.
On the contrary, I've found that exposing to 100% zebra still shows detail in the white. I've never seen another camera work like this....100% zebra blows the hilights completely out on other cameras, but not the ex1. In fact, I've begun to shoot with my TLCS levels set to +0.5.
The histogram is calibrated for the capabilities of the camera CMOS, not necessarily the 0-100IRE range of an NLE. CINE3 shows the most detail in the blacks, but, at the expense of the middle grays and hi's. CINE4 is the best compromise for a high dynamic range scene. Cine 3 works for scenes without as much hi-lite, But CINE1 works very well for low dyanmic range.
The factory matrix, even in hisat, is off in the greens, cyan and blues. TC2 color matrix resets for more realistic colors in all modes.

[Michael H. Stevens]

Ravens:
As for you liking STD to give your more detail in post you forget that the sort of processing the CINE gammas do is much better than doing the same adjustments in post. The closer you can get to the look you want in the camera is always best.

As for 100% Zebras, all I can say is on my camera, using the CINE gammas, 100% Zebra blows a lot out. Now I know the camera claims to record to 108% and I do see that in my NLE, but my camera gets no detail in this range and is fully white blown, at anything over 95.

[Leonard Levy]

Michael,
i have to agree with Bill. In the real world of video production it is impossible to choose to shoot with nothing over 100% especially exteriors.
Thjat's just a fact of life in my world.

It may be that the cine gammas on this camera are the only viable option, but they shouldn't be and if theym are its a significant flaw that should be corrected. Cine gammas do a lot of compression which may or may not work for your shooting style and the subject but the camera should not be artifacting under normal gammas.
I'm going to look into this this afternoon and if I learn anything I'll post.
Lenny levy

[Bob Grant]

Originally Posted by Christopher Barry

Bob and Bill, the F900 Standard REC709 setting is up at around 150 IRE, likely the same in the XDCAM graph.
http://www.siliconcine.net/temp/XDCAM_Gamna_Curves.jpg

Alfonso Parra, DoP, has performed F900 tests on the Digital Praxis custom curves. If you look at this image/graph, at 400%, the F900 Standard curve runs up to 150 IRE. The other curves are the Digital Praxis customs, somewhat similar to the EX1's cine gamma, except closer to Cineon log and some ideal for intense post grading and film out. This provides some perspective of what the STD curve values may be, and IMO, can be applied to the EX1 STD curve, as approximations of the X and Y values. The EX1 is after all part of the CineAlta family.
http://www2.alfonsoparra.com/php/ima...nt/167/049.jpg

I'm going to have to ask the obvious question. If 110% is the point of digital clipping how does the camera record 150%?
I think I can understand how this in fact works but if so aren't we way outside Rec709?
If that's the case then NLE's such as Vegas are going to get the conversion to RGB wrong. Even more worrying is how that holds up in an 8 bit pipeline and recording system.
Might be interesting to hookup a 10bit scope to the HD SDI port.
Would I be way off the mark in suggesting that some of these gamma curves could be optimised purely for the 10bit HD SDI output and not used when recording with the 8bit XDCAM codec at all. I know trying to get something from Cineon into 8 bit can be difficult with banding problems pretty easy to come across.

[Bob Grant]

Originally Posted by Piotr Wozniacki

Too late - I've just sold my V1E:)

Seriously though, I still hope I'll find a solution. BTW I am trying to be as neutral in this topic as possible, even though have been attacked a couple of times myself as a "unfuctional" operator, so let's just forget it for a while.

Nobody even commented on the vignetting caused by the ND filter - could it be something quite separate from the infamous vignetting problem?

When I first got my hands on the camera I commented that the ND filter control was not good. That you've managed to only partially engage a filter comes as no surprise to me at all. The ND filter control and the power switch need a lot of care and attention in use. Yes I agree, not well done at all.