FS7: nasty artefact on clear blue sky with FS7 kit lens

moldcad

moldcad

Veteran
OK, so here is my first "is it a bug, or a feature" question: today when filming a contrasty scene in bright sunshine, when watching the clip (de-logged to REC709 in Browse - nothing else manipulated with it), I noticed an ugly artefact across the clear, blue sky (perfectly exposed btw.; I was in CineEi mode and using S-log3/S-Gamut; I increased aperture to check the theory about less noise in shadows, so that with ISO=800 the EVF picture looked perfect with REC709(800) MLUT).

The artefact I could only described as an effect similar to what can happen DURING electronic exposure change; sort of semi-level, pulsating and faint "scan lines". It was only present with my SELP 28-135 lens in fully manual mode; the same scene captured with my 2 vintage Canon primes (100% mechanical) was free from it. The only thing that was ON in the kit lens was OSS...

I'm quite disappointed with this; could anyone shed some light on what might have been the reason? TIA

Piotr
 
Well - when I applied a curve in Vegas to bring it out (spoiling exposure, but that's not important now), what was manifesting itself as "pulsating lines" turned out to be a nasty macroblocking! Just take a look how bad it is... Disclaimer: this was done on my laptop, so a 10 bit file in the 8 bit pipeline; but why it's almost invisible with the other lenses?
 

Attachments

  • macroblocking.jpg
    macroblocking.jpg
    20.6 KB · Views: 0
not sure how a lens can be responsible for this, maybe the OSS?
i would also try a different grading software just to make sure...
 
Paul schefz said:
not sure how a lens can be responsible for this, maybe the OSS?
i would also try a different grading software just to make sure...
Click to expand...

Either that or the electronic iris is ramping a little bit - like from F15.99 to F16.01

We all know (im sure) that electronic iris can cause flicker in stills time-lapses.

S
 
Guys, my laptop is terribly slow with Vegas at 32 bit, but I managed to display the scene and why there is no macroblocking, I can still se those "pulsating lines" (as if some interference - maybe from OSS circuitry?). I will check again and perhaps post some screen when I get to my office..
 
OK, let me be clear: shooting those tests away from my 10-bit capable PC, with just a laptop, I was not even going to "edit" - but this artefact was also visible in the camera LCD, so I hurried to see it on a bigger screen. The picture I posted is not representative as it mainly shows macroblocking/banding of a 10-bit, wide gamut picture in an 8-bit system!
 
That said, in 32-bit mode I can clearly see the same pulsating artefact I did on the camera LCD, and ONLY with the Sony kit lens. The picture shot with my 2 mechanical lenses are perfectly quiet.
 
OK, so I applied a 709 LUT in Catalyst Browse, and saved a full frame with indication where to look in another screen fragment (blown up 4x). I don't know how to upload pictures here in their full size, but even in the miniatures below one can see the "line" I'm talking about; please remember that in live video, it's pulsating which makes it even more disturbing. Also, I just realized my zoom servo was active (unused though at the moment shown in the pics), so there is a possibility of either OIS or zoom servo interference... What do you think? Has anyone noticed such artefact in their footage with the kit FS7 lens (28-135)?
 

Attachments

  • Browse full screen.jpg
    Browse full screen.jpg
    20.1 KB · Views: 0
  • Browse 400 percent.jpg
    Browse 400 percent.jpg
    2.3 KB · Views: 0
Upload the original un-cut clip for us to evaluate for you - you can go to my website, click on the "Uploader" button and I'll host it temporarily and publish the link back on here.

www.matobo.co.uk

Ben.
 
Thanks Ben. Here is a full frame; please look for an almost white line below the distant cable across the sky... It's pulsing at what I think is the framerate frequency (25 fps).

Piotr.
 
moldcad said:
But iris (like everything else, apart from OIS) was off (manual)!
Click to expand...

Im not an expert electronic iris but I think it works something like this.

electro magnets pull the iris closed on current delivered by the camera through the lense connections, this current is controlled by the settings on the camera (a dialed stop or auto)

- this system is known to be mildly variable. (aka hitting F15.9 or F16 or F16.1)

just because you have the iris set to manual there is still a conversation going on between the lens and the camera and the command is coming from the camera often..

Frame 12345 iris? F16 (iris closses to 15.9)

Frame 12346 iris? f16 (iris closes to 16)

Frame 123457 iris? f16 (iris closes to 16.01)

Other settings the convo would be
iris? auto F11.
iris? auto F8

This conversation happens at least once per frame maybe way more quickly. So if the cheap lens does not quite hit its marks you would get a varience in exposure between frames.. this would then be exasabated by the compression algorythm making different commpression choices frame to frame

The phrase you need to google is [h=1]Electromagnetic Diaphragm[/h]
Im not a tech.. but I bet you a beer that is your issue.

S
 
Last edited:
This sounds probable - in my first post, I did say the artefact resembles what happens when you aim the lens to the sky (from shadow area), and - with Auto Iris - there is this pulsing... I hoped the 28-135 G lens, chosen by Sony as a kit lens for the FS7, would be better than this :(....
 
Last edited:
Piotr - can you upload the whole original MXF video clip? A png won't reveal the oscillation you're talking about.

moldcad said:
Thanks Ben. Here is a full frame; please look for an almost white line below the distant cable across the sky... It's pulsing at what I think is the framerate frequency (25 fps).

Piotr.
Click to expand...
 
moldcad said:
I hoped the 28-135 G lens, chosen by Sony as a kit lens for the FS7, would be better than this :(....
Click to expand...

Sony is always a double edged sword.. sometimes too clever for their own good.

Very efficient compression VS odd artifacts sometimes
Lens 5Xcheaper than proper eng lens (that has no AF or IS) VS pulsing iris and AF and IS

take your choice. but dont expect a free lunch

etc etc.

You IMO have to learn to take and leave from Sony - so you might use that lens for crashing about in an interview but not use it for fine landscapes etc etc.

I have the 18-250 lens (the worst one) .. I use it 1 in 100 shots, but for that one shot it is excellent.

Usually the stuff made from brass and gun metal that costs the same as a small car is what you need if you have top end standards.

S
 
Last edited:
bengiles said:
Piotr - can you upload the whole original MXF video clip? A png won't reveal the oscillation you're talking about.
Click to expand...

A few seconds worth of the original, untouched clip is uploading as I type, Ben. Of course it needs at least a 709 LUT; I'd also increase contrast somewhat so that the sky is darker blue - the "pulsing line" is visible better then. Cheers

Piotr
 
You must log in or register to reply here.
Back
Top