Low light printing issue – advice wanted please

Low light printing issue – advice wanted please

Author
Discussion

FurtiveFreddy

8,577 posts

237 months

Friday 17th June 2016
quotequote all
Simpo Two said:
Because the most compression happens in the areas of least information - in this case, the sky.

I have no idea about Lightroom; Photoshop, or at least my version, has 12 levels of compression.
The levels of compression in PS is not the same as selective compression. PS used to have selective compression but did away with it a few versions back.

The point is, in the areas of least information where there is the most compression, you'll see less evidence of compression, so why would there be visible artifacts there?

That's why I don't think whatever is visible on the print is likely to be jpg compression artifacts otherwise you would also see them in very detailed areas of the print, unless there's some form of selective compression being used.

The other way we can maybe see what's happening is if the OP allows us to look at the jpg sent to the printer and then we can see how it looks on our monitors.

RobDickinson

31,343 posts

254 months

Friday 17th June 2016
quotequote all
How long was the exposure? Even at low ISO you get noise when you have dark/shadow areas, more so at longer exposure times , I'd use LENR for anything over 50 seconds or so on th 5d3.

What have you done in processing? any pushing of shadows will exaggerate these small changes, far more so when in 8bit working space.

If you want to work on some of the image look at masking edits or luminosity masks (or blend if).

I would reprocess the raw into a 16bit file in adobeRGB (or bigger), then duplicae the final edit and carefully noise reduce the sky and use 'blend if' to bring the NR into the final image just in the shadow areas.

Then save as a full sized tiff in 16bit flattened.

Edited by RobDickinson on Friday 17th June 01:04

brman

1,233 posts

109 months

Friday 17th June 2016
quotequote all
It is a 2 second exposure according to the exif. I would not of thought that should push the noise up much but then I don't know Canon cameras.

A couple of questions:
Is it the same jpg that you have on flickr or was there any other processing on the print version?

Is this the blocking you are talking about?



It looks to me like sensor noise increased by quantisation and compression but I don't claim to be an expert. Why it is visible on the print I don't know though as I had to massively increase the exposure and contrast to see it on my (calibrated) screen. As others have said, I would be asking the printer for advice.


FurtiveFreddy

8,577 posts

237 months

Friday 17th June 2016
quotequote all
brman said:
It is a 2 second exposure according to the exif. I would not of thought that should push the noise up much but then I don't know Canon cameras.

A couple of questions:
Is it the same jpg that you have on flickr or was there any other processing on the print version?

Is this the blocking you are talking about?



It looks to me like sensor noise increased by quantisation and compression but I don't claim to be an expert. Why it is visible on the print I don't know though as I had to massively increase the exposure and contrast to see it on my (calibrated) screen. As others have said, I would be asking the printer for advice.
That is definitely not jpg artifacting if that's what the OP is seeing. As Rob and I already said, look at the BPC settings first.

Also, have a look in colour settings in whatever software you've been using. There's a setting in PS to use dithering on 8 BPC images during conversion. That may make a difference as you're saying you can see banding on the screen in some areas. You may find some other settings and checkboxes in colour settings which are worth playing with to see if they make a difference in the future.

Edited by FurtiveFreddy on Friday 17th June 10:29

toasty

Original Poster:

7,466 posts

220 months

Friday 17th June 2016
quotequote all
Thanks for all the input guys, I'm learning a lot going through all the possibilities.

I've now taken a photo of the print and posted a crop of the jpeg I see on the screen compared to a close up of what was printed. The print was on the windowsill which will exaggerate the light (the print looks darker in normal conditions) but shows what I am trying to describe.

In answer to some of the questions, the Flickr photo is lower resolution than what was sent to print. I only have LR6 which came bundled with the camera. I'm considering PS but struggle to justify it at the moment. Colour space was sRGB, AdobeRGB (1998) and ProPhotoRGB are also options.

JPEG


Photo


I'll send an email with the same back to Loxley to see if they can suggest anything.

brman

1,233 posts

109 months

Friday 17th June 2016
quotequote all
like I said above, I am no expert on printing but that does look like something in the print process to me.

FurtiveFreddy

8,577 posts

237 months

Friday 17th June 2016
quotequote all
I can't see anything in your jpg crop which looks like the print, even after maxing out the brightness, contrast and saturation.

That's not jpg artifacting on the print. I agree that something seems to have happened between the jpg file you sent and the print output.

brman

1,233 posts

109 months

Friday 17th June 2016
quotequote all
I'll be curious to know what the printers say so please report back wink

toasty

Original Poster:

7,466 posts

220 months

Friday 17th June 2016
quotequote all
Loxley are on the case and will pick up the print by courier on Tuesday. I'll send another update once I hear back from them