Weird color noise(pattern?) on X100T

LR demosaicing can expose the moire problem more than other converters (it's camera-dependent) and although LR anti-moire brush can help yet it's also a good idea not to cause the problem if possible in the first place. My CPA shows more moire in LR than in RawTherapee, Capture One or Capture NX-D because of the demosaicing differences.

Yake said:

tesilab said:

Post processing is not the most effective strategy to deal with moire patterns. At best you can use the tools to attempt to mitigate the issue.

If you don't have an adequate low pass filter on your sensor, your best bet is to attempt to avoid exciting moire in the first place by not focusing on very fine patterns. For example, if you are shooting pictures of a bride, avoid focusing on the veil.

Click to expand...

The OP is shooting buildings, not brides. So that strategy won't work.

Click to expand...

Exactly. There is a reason that nearly all sensors used to be equipped with AA filters, since it was simply the best available option for the problem.

oookay... so there's no effective way of removing the moire in pp and I'll just have to avoid it occurring in the first place?

I only use built-in Photos app on mac. Maybe I should try using LR.

thank you. and i'm sure it was shot in MS because I remember I couldn't use the L100 iso when I shot it.

I don't know much about the sensor or the processor, but me too! I've never seen anything like this and I always found my X100T excelling a6000 in every way.

umm.. do i lose image quality in F11? I don't usually stop down below F8

I have the raw file but I don't know how I can upload it here.

I use default Photos app in my MacBook, and I think it's the app's internal conversion algorithm that emphasizes the moire. I just installed lightroom mobile and checked it out again, and surprisingly it removed the moire(not completely, but a bit). So maybe it was the program. Thank you!

Diffraction causes a kind of blur which in effect is similar to what AA filters do, so yes, you'll lose some sharpness in other areas. You could diffraction-bracket: shoot one shot for maximum sharpness, the other for anti-moire, and then combine the two shots in PS. That's probably too much of a hassle for normal photography -- the LR moire brush is certainly the easier option.

If you provide the RAW file I'll test it with ACR Adobe and Raw Therapee plus Silkypix

That will tell if it is the demosaic process which is involved

Look at :





and:





you can see that ACR RAW engine has removed completely the moiré in both camera

I think here i s the explanation

--
Good judgment comes from experience
Experience comes from bad judgment

baobob said:

Look at :



and:



you can see that ACR RAW engine has removed completely the moiré in both camera

I think here i s the explanation

Click to expand...

The "RAW" designation, as far as I understand, stands for JPEG images that were derived from raw data through processing by ACR using default values. The "JPEG" designation stands for sooc-JPEGs. So, it's the other way around, the internal processing has removed the moiré, while the ACR processing has not (likely because moiré removal isn't a default operation in ACR).

It is not clear at all, I agree look at the size of 2 JPEGs the one close to RAW and the other,

which is which ....

DrCastle said:

oookay... so there's no effective way of removing the moire in pp and I'll just have to avoid it occurring in the first place?

Click to expand...

Moiré is a fact of digital photography whenever there is a subject pattern that is close to the density of pixels in the sensor. There are almost always two approaches to dealing with problems in digital photography: correct in post, or avoid in the first place. Correcting problems in post almost inevitable leads to detail loss and other IQ reductions but the result is also likely more appealing than without correction. In critical cases, though, it's best to avoid any issues upfront. With moiré one can try a couple of things, most notably getting farther away from the subject. Using a camera with higher resolution also helps Blurring methods aren't the best solution in IMO, because blurring amounts to a reduction in IQ. That includes use of an AA filter, deliberately invoking diffraction, etc. But if you can't change the subject distance these might still be acceptable. Correcting in post is preferable IMO because it doesn't degrade regions outside the problem regions like a global method (e.g. diffraction) would. It helps to train yourself recognize potential moiré situations, take a couple of test shots to see if it really is a problem, then react accordingly.

Regarding Apple's Photos, the underlying demosaicing machine is very good, but Photos lacks the advanced tools that other software provides. It's a pity Apple has gotten out of the serious-photography business. I am still using Aperture, and it's great. Alas, that may not be an option for you. Over time, hopefully, some of these tools may make it back into Photos.

That's not moire there: it's high ISO, colour noise. OP talked about having the moire problem at ISO 200 with X100T, so that would be a better illustration. Also, choose a different part of the DPR studio scene to illustrate moire. This part actually shows that moire shouldn't be an issue with this camera, and the OP has since admitted that the moire in the screenshot was caused by the resizing done by the viewing app rather than inherent in the file.

below my RAW dev at 12800 ISO





clearly even with ACR engine there is nearly no moiré which again suggests that in the OP the raw engine could be involved

In my already long career with Xtrans cameras I think I got limited moiré in 1 to 3 shots among many 100000 shots

--
Good judgment comes from experience
Experience comes from bad judgment

and with RT5, slightly better







--
Good judgment comes from experience
Experience comes from bad judgment

DrCastle said:

I have the raw file but I don't know how I can upload it here.

Click to expand...

What most folks do is sign up for a free account at dropbox.com and upload the file there, then provide a link to the file in a posting here at dpreview. I haven't personally used dropbox myself. I'm an Amazon S3 user, so I've always used S3 to share large files.

Sorry but moiré appears more frequently where there is the pattern I have choosen

If you look carefully at the first posts ther is a blend of color noise plus moiré that you can distinguish by its diagonal pattern

it's always in this zone that you diagnose moiré the best

--
Good judgment comes from experience
Experience comes from bad judgment