weird noise w/A7Riii. Looks like banding but diagonal?

[bwud]

I shot some pretty underexposed shots today (playing around with stacking technique, this isn't ordinarily how I'd shoot), and when pushing them in lightroom (and the same thing appears in Capture One).

 

Could it be heat from the card slot? I've not seen this is any other situation, but again don't typically underexpose so significantly. Anyone else seeing this?

 

Attached are: default conversion & +100 shadows +2.14 exposure (Lightroom)

 

Other details: compressed raw, silent shutter (so I've lost precision), burst to a single card, every shot in the burst has the same noise.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[bwud]

large crop of affected area:

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[bwud]

and pushed beyond reason (+100 shadows +4.5 exp), with borders depicted

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Chrissie]

Could it be heat from the card slot? I

 

 

If it was heat-related, then all areas of the image should be affected somehow. But since only the darkest areas are affected, I'd guess these could be artifacts introduced by compression. Would be interesting to see the histogram of each individual channel while you're pushing.

[Singingsnapper]

What I am seeing in out of focus areas is like a lot of square pixels.  At first I thought it was poor rendering in Lightroom, but it is still there when I view from Phase One Express for Sony.  I tend to shoot in compressed raw (never had an issue like this with the MKII).  It's not present in the parts of the image that are in focus...it's a bit like it's been printed on canvas...

 

You can see the effect on the image on my flickr page at 2048px

 

Hi Bella 1 copy by singingsnapper, on Flickr

[bwud]

If it was heat-related, then all areas of the image should be affected somehow. But since only the darkest areas are affected, I'd guess these could be artifacts introduced by compression. Would be interesting to see the histogram of each individual channel while you're pushing.

 

I thought maybe heat since the area affected is nearest to the car slot, which was handling a lot of bandwidth and with circuity potentially dissipating a lot of power. If it is localized heat, it's a little weird that they're straight diagonal lines and not curves, though.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[bwud]

What I am seeing in out of focus areas is like a lot of square pixels. At first I thought it was poor rendering in Lightroom, but it is still there when I view from Phase One Express for Sony. I tend to shoot in compressed raw (never had an issue like this with the MKII). It's not present in the parts of the image that are in focus...it's a bit like it's been printed on canvas...

 

You can see the effect on the image on my flickr page at 2048px

 

Hi Bella 1 copy by singingsnapper, on Flickr

 

interesting!

 

I never saw anything like this with my rii either (by again don’t typically underexposed so drastically).

[Guest Jaf-Photo]

I've never seen blocky diagonal banding like that. Thinner diagonal streaks can occur due to optics, electronics or software.

 

The dog photo looks a bit different, more like a faux film grain effect in the bokeh, which is also unexpected.

 

You should definitely send sample pictures to Sony and ask for an explanation. See if you have the banding in a more normal exposure, otherwise they might say it's not a problem because you're not supposed to underexpose and push.

[bwud]

I've never seen blocky diagonal banding like that. Thinner diagonal streaks can occur due to optics, electronics or software.

 

You should definitely send sample pictures to Sony and ask for an explanation. See if you have the banding in a more normal exposure, otherwise they might say it's not a problem because you're not supposed to underexpose and push.

 

I’ve not seen it in any other photo (aside from the few in this series). I don’t think it’s software (manifests in multiple converters) or optics (not seen in other photos using the lens which is EF24-70GM).

 

If I can figure out to to submit one to them I will do so!

 

They might say that, and I’ll say “I’m trying to replace the smooth reflections functionality which isn’t supported by my new camera.”

[Guest Jaf-Photo]

If you bought the camera from a photography dealer, they'll know a Sony rep who can look at the pictures.

[bwud]

B&H...

[bwud]

I thought maybe heat since the area affected is nearest to the car[d] slot

I take it back. Given the inversion from the optics, the affected area is actually the top left of the sensor (viewed from the photographer's perspective), putting it far from the flash components.

 

Some of the bands appear to match the orientation of the processor clocked at a 45­° angle (per https://kolarivision.com/sony-a7r-iii-dissasembly-teardown/), but why would it be worse towards the edges?

[Username]

`    

   

Edumacated guess:  

   

You might need a faster card. The "grainy" noise level suggests 

a rather high ISO, which increases the processing load, and the 

high MP count of the "R" camera is itself already quite a load. 

  

Just my semi-smart guess. Worth at least what you paid for it :-) 

[bwud]

`    

   

Edumacated guess:  

   

You might need a faster card. The "grainy" noise level suggests 

a rather high ISO, which increases the processing load, and the 

high MP count of the "R" camera is itself already quite a load. 

  

Just my semi-smart guess. Worth at least what you paid for it :-) 

 

 

This was shot with a 300MB/s UHS-II U3 Class 10 card at ISO 100

[LiveShots]

The banding areas look like aperture blades, I wonder if you have some light leaks on the lens that are being exaggerated by the stacking.

[bwud]

The banding areas look like aperture blades, I wonder if you have some light leaks on the lens that are being exaggerated by the stacking.

The above is from a single image (not yet stacked).

 

 

Interestingly I took another burst just afterwards with even less focal plane exposure (all else equal but f/8 instead of f/4.5), and when pushed to roughly the same brightnesses they do not show banding, so maybe this was a unique problem.

[Guest Jaf-Photo]

The banding areas look like aperture blades, I wonder if you have some light leaks on the lens that are being exaggerated by the stacking.

That's the best guess so far. Maybe it was intenal reflections from the sensor hitting the back of the aperture blades? That could explain why the bands are brighter but lack colour information.

[bwud]

^^ And why the next series didn’t band (aperture geometry changed)