matching Z280 and FX6

[Tom Van den Berghe]

I always go for a preset. It's too difficult for me to white balance on stage.

So I asked now which lights there are using the most. In the 2 stage events I filmed this week
it was mostly 3200K and 4000K. It seems to work good when looking at the footage.

But the biggest problem I have is the shutter speed. I push direct menu and scroll to the shutter speed for manual control. It was on 1/100 (pal country here) and the diafragma was mostly between F2.8 and F6.

There was sometimes much less light (bethroom scene) so I scrolled the side wheel quicly to shutter 1/50
to avoid much of gain kicking in. My unmanned FX6 was on 1/50 in low base and there it mostly F2.8 - 4.5 when I sometimes checked it.

I read here somewhere on the forum that the Z280 can go quite soft when going above F4 aperture...

So how do you control this? I don't think I will have to use the variable nd for stage show?

[Tom Van den Berghe]

I hope someone can help me out how to shoot indoor when switching manual shutter to control iris.
Outside no problem with the variable nd filter.

[Tom Van den Berghe]

Had yesterday again a stage show. pxw-z280 settings: 25p, F2.8, 1/50 shutter, auto iso (but was mostly on 0 db) and turned on the variable nd filter the have the fixed F2.8. Otherwise the F stop would go beyond F4.

In darker scenes I have switched off quickly the nd filter.

It seems the Z280 is lacking some sharpness against the FX6. I had no control about my FX6 because it was on the side and I was standing between the audince with the Z280.

FX6 was also on 1/50 25p and auto F stop.

So is the lack of sharpness (what I may be telling myself) of of the choosen profile?

See attached images.

And a link to the test video from yesterday.

repelsteeltje test - YouTube

forget to mention. Also the autofocus (face priority) was not always flawless... I checked the backfocus the day before and there was no issue.

[Christopher Young]

This is a bit long-winded, but I think it will go some way to explaining what you are experiencing.

This is a subject that comes up now and again, and it all really comes down to the physics of the sensor, the lens and the size of "Airy disc" created at any given aperture with regard to the actual pixel dimensions of the sensor involved.

I shoot with a number of cameras, FF, S35, APSC, 2/3" and 1/2". We are talking 1/2" cameras here, like the Z280 and comparing it to another 1/2" camera. The Z280's predecessor, the PMW-300. The Z280 is a 4K 4096 x 2160 camera that has pixel dimensions of 1.56µm. the PMW-300 is an HD 1920 x 1080 camera with pixel dimensions of 4.4µm.

Bear with me as this has relevance further down as to how, with reference to the previously mentioned "Airy disc". Now, how is our signal from the sensor measured out. The percentages are. Light Measure of brightness. The measured value of brightness; reflected light measured on motion picture screens as footlamberts or candelas per square meter.

The video luminance signal is formed by combining a proportion of 30% red, 59% green and 11% blue from the colour signal.

This combined output becomes the luminance (brightness/ monochrome) signal. It is written as Y. In TV usually derived from the RGB signals originating from a camera.

In 99% of cases when doing measurements in the video domain, most measurements are based off the green wavelength as it is the major component of the video signal. It is the Y, luminance signal in the YUV video signal.

If we measure the Airy disc size generated at F/2.8 on a theoretically perfect lens, we will see that the focus spot size is 3.55µm. This greatly exceed the size of the pixels on the Z280 sensor, which has a pixel dimension of 1.56µm. The result is the focus spot size is bigger than the target pixel. Therefore spills over onto the adjacent pixels. As this is happening all across the sensor, all focus spots tend to a degree blur into one another. No distinct boundary per pixel, therefore limiting both the resolution and very importantly the contrast of higher resolution smaller pixel sensors. Lower contrast makes things look less sharp. Look at the attached Airy disc size chart that is attached, read its summary in relationship to how the Airy disc size can surpass pixel size. Also attached is the sensor calculation chart for the Z280, demonstrating the size of each pixel at 1.56µm. Even shooting at F/2.0 on the Z280 we still have an Airy disc spot size of 2.54µm which is still 60% aver the Z280s pixel size of 1.56µm. Of course, shooting stage shows you generally need at least three meters DOF on the stage to allow for movement and an acceptable range of perceived sharpness in front and behind your critical focus point. Which you are struggling to do when shooting from the back on the long end of the lens to a theatre stage that is many meters away from the camera positions.

Now we come to the HD PMW-300 with its 1/2" sensors. Same size 1/2" sensors with the same dimensions as the Z280 but having an HD resolution of 1920 x 1080 only. Herein lies the difference. The pixel dimension on the PMW's 1/2" sensor is 4.4µm. So what does that tell us. That a lens at F/2.8 delivering a focus spot size 3.55µm is going to land nicely within the boundaries of the 4.4µm pixels. No bleeding over the edge onto the adjacent pixels. Therefore, not overpowering the resolution of the sensor or diminishing its contrast. Look at the attached graphic of the Airy disc impacts on pixel size for a visual understanding.

To sum up. I had a reasonable understanding of the principles outlined above because of my technical as well as shooting background in television. But the 100% confirmation for me was having it confirmed on a number of occasions when using said two cameras, the Z280 and a PMW-300 side by side on certain shoots, especially lower light shoots such concerts and stage shows. The 300 consistently delivered sharper, higher contrast images. In post, I've had to increase contrast and apply detail to the Z280 images to match the two cameras.

Hope this helps explain what is going on.

Chris Young

[Pete Cofrancesco]

I think you're concentrating on the wrong things. Technically the image looks fine from both cameras. I personally would never us ND inside and most cameras are sharp enough up to F16. For simplicity sake I would only use the iris to control exposure. I'd also leave the camera in manual focus. This stage in particular is quite shallow and shouldn't need AF.

The main issue I see is your ability to smoothly pan and zoom. You're zoomed in too close. Most actors don't want to be shown that closely. You're missing the context by zooming in that much. If you zoom in to head and shoulders while actor is talking, you're missing the other actors reaction. If you are selling this to parents every time you zoom in that much you are excluding the other actors who their parents want to see. This is a stage performance not a movie. The video should give you the experience as if you attended the performance.

I would invest in a better tripod head because the camera movement isn't smooth and is distracting. The farther you zoom the hard it is to get smooth pans and tilts. To me the camera work of stage performances should be transparent complimenting whats going on not distracting. I would concentrate more on what is occurring on stage and less about your camera settings.

[Christopher Young]

Pete. The problem with stopping down past F/5.6 on the Z280 brings on quite severe diffraction with a pronounced softening of the image. Seen it quite often even in bright daylight.

Why his FX6 looks way sharper is the fact that it has enormous pixels of about 8.36 µm, so absolutely no diffraction or Airy disc diffusion onto other pixels. Big light sensitive pixels that are pretty immune to diffraction until you get down below about F/11.0

Chris Young

[Tom Van den Berghe]

thx Christopher you for long answer! I had to read it several times to understand it better.

But how do you shoot stage shows like this? If I leave it on auto shutter I will have horizontal lines (banding) in my image. Never of my previous camcorders (panasonic dvx200, sony NX3, sony FS100, canon XH-A1) had such banding (horizontal lines) when leaving it on auto shutter when shooting stage events.

For the stage show I had posted here my shutter was on 1/100 fixed but even then my F-stop was going above F4.

And always scrolling through the wheel to have other shutter speeds is not easy because it's not 1/50 multiply. You have 1/60 1/100 1/125 ... so several shutter speeds not "pal" based.

So I used the variable nd and a fixed shutter of 1/50 so my F-stop would stay at F2.8 I had chosen.
When it become to dark in some scenes I switched it off.

@Pete: the Z280 has normally great face autofocus so why would i go for manual focus?
I understand the not smooth pan and zoom but this was a quick test I made.

[Pete Cofrancesco]

Originally Posted by Christopher Young

Pete. The problem with stopping down past F/5.6 on the Z280 brings on quite severe diffraction with a pronounced softening of the image. Seen it quite often even in bright daylight.

Why his FX6 looks way sharper is the fact that it has enormous pixels of about 8.36 µm, so absolutely no diffraction or Airy disc diffusion onto other pixels. Big light sensitive pixels that are pretty immune to diffraction until you get down below about F/11.0

Chris Young

looked up the z280 and I stand corrected even at f16 this camera suffers from distraction. So for a bright performances using the lowest ND isn't a bad idea.

I just ran a test on my 2 camcorders
Panasonic X2 1" sensor doesn't show any appreciable sharpness difference in any aperture but it is limited to 3.5-11
JVC HM600 1/3" sensor has a much larger range of aperture 1.8 - 22. f22 awful, f16 soft but usable , f8 much better, 5.6-3.5 the sharpest, 1.8 a little softer like f8

[Christopher Young]

Pete. The numbers you've just quoted for the various sensor sizes is pretty much in keeping with my experiences with similar sized sensors.

Chris Young.

[Tom Van den Berghe]

@Pete: thx for testing this out with your camcorders!
@Christopher: so how do you shoot stage shows like this to avoid diffraction?

[Pete Cofrancesco]

Here's a diffraction test I filmed this weekend. In confirms what Chris said small sensor cameras are susceptible to it while 1" cameras and larger are not.

JVC HM600 1/3"
JVC 600 Iris test

Panasonic CX350 1"
Panasonic CX350 Iris test

[Pete Cofrancesco]

Originally Posted by Tom Van den Berghe

@Pete: thx for testing this out with your camcorders!
@Christopher: so how do you shoot stage shows like this to avoid diffraction?

You have 3 options to reduce the exposure without using the iris above f5.6
1. ND
2. Higher Frame rate 60p/120fs (This will reduce the light by one stop)
3. Negative Gain. This camera can go down to -3db.

In my experience I've never filmed a stage event brighter f8 @ 0 gain. Using 2 & 3 would bring you below f5.6. Most performance aren't very bright but if you have any doubt ask the lighting guy to put on the brightest lights before the show to determine what you need to do to lower the exposure without going over f5.6. If it bothers that much I'd sell the Z280 and get a 1" camcorder they don't suffer from any diffraction at any aperture.

[Tom Van den Berghe]

thx for the tips Pete. Never used the negative gain. will have a look at it.

Bothers me? it is just something I experience now with this camcorder. I wanted this camcorder already for years and finally bought like new second hand on a very good price last year.

I think I will keep it but a 1inch is a good alternative like you said.

[Pete Cofrancesco]

I've used many camcorders and they all have different annoying issues. You usually only learn about them on a real job.

[Doug Jensen]

Originally Posted by Pete Cofrancesco

If it bothers that much I'd sell the Z280 and get a 1" camcorder they don't suffer from any diffraction at any aperture.

Not true. All cameras can suffer from diffraction problems at small apertures. I can even make it happen with my super35 and full-frame cameras. Furthermore, I guarantee that there is no 1" camera you can buy that will outperform the Z280.