Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Fuji Forum | Join Upload & Sell

  

Archive 2017 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans

  
 
rsk7
Offline
• • •
Upload & Sell: Off
p.1 #1 · p.1 #1 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans


I know.. a lot of topics but it is all inter-related. Stick with me.

I took a picture of pre-dawn that was pretty dark in the foreground. I used a gradient to push the exposure +2.5 and got a bunch of hot pixels that I ended up having to clone out. Way more than I ever had on any of my canons (10D, 5D, 5D3). (But of course I could never push the Canons +2 without banding/blocking but that is a diff story and issue).

Having read about X-T2 and how great it was re: ISO Invariance I decided to try a lens cap test. Image at iso200, f/5.6, 2 second exposure. Imported the raf file into LR and created a copy then pushed the copy up +5 on the exposure with sharpening at 30. The un-pushed image I can see only 1 hot pixel at 100%. The +5 image I can see 80-100 red, blue, green white pixels. Is this normal for an X-T2 sensor? It seems very well controlled at exposure but when pushed holy moly.

So next thing I did was process the raf with X-Tran using all the default settings and then I pushed it +5. Uh oh.. I think if you X-Tran your RAF then you need to forget about ISO Invariance and pushing much or I am doing something seriously wrong to even try this. The whole image lightens up where the blacks become purplish and there is a massive "Purple Sand Bloom" (not sure what to call it).

So I'm interested in if anyone else has tried this.
1. Take a lens cap image with settings as above. Open in LR and set sharpening to 30 and push +5 on exposure. Compare the raf to the pushed copy and see if you get lots of hot pixels.
2. Then do an X-Tran conversion of the raf and push it +5 and see what you get.Do you lose all black in a field of purple sand? I can get rid of the sand by applying massive luminance noise reduction but don't think that is an option with a real image and the image is still purple and lighter than the RAF +5

I'm interested in hearing if anyone else has the same issue with so many hot pixels when pushing and if you think X-Tran basically kills the ability to push exposure in an ISO Invariance way as I have concluded (at least using default settings).

Thanks,
Rick




Left side LR, Right side LR+5 exposure






Left side LR + 5 Exposure, Right side XTran + 5 Exposure






default X-Tran settings






Left side X-Tran +5, Right side LR +5 showing lighter, purple in xtran side




Apr 22, 2017 at 07:51 PM
cputeq
Offline
• • • • •
Upload & Sell: Off
p.1 #2 · p.1 #2 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans


Your findings match the general performance characteristics of Iridient Xformer in general - it just doesn't do well with high ISO shots (which this essentially is after the push) because it handles raw files completely differently than LR.

If it's over ISO 1600, I actually prefer to use LR, as the noise becomes strange with Xformer.



Apr 22, 2017 at 08:20 PM
rdeloe
Offline
• • • •
Upload & Sell: Off
p.1 #3 · p.1 #3 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans


Rick, so I tried your test exactly as you explained and got the exact same result. But I'm not sure what this means, if anything!

If you want to test for ISO invariance, why wouldn't you use an actual scene and exposures that makes sense? I'm not sure what a massively underexposed test like this actually means.



Apr 22, 2017 at 08:22 PM
rsk7
Offline
• • •
Upload & Sell: Off
p.1 #4 · p.1 #4 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans


rdeloe wrote:
Rick, so I tried your test exactly as you explained and got the exact same result. But I'm not sure what this means, if anything!

If you want to test for ISO invariance, why wouldn't you use an actual scene and exposures that makes sense? I'm not sure what a massively underexposed test like this actually means.


Thanks for going to the effort to duplicate my results.

To me the fact that you got the same result means that there isn't likely something uniquely wrong with my sensor relative to at least one other sample. I'm not going to send my X-T2 in for servicing thinking I got a problem with my sensor.

The purpose of doing the test this way was to figure out why in my real scene and exposure I started getting so many hot pixels when I pushed the foreground +2.5.

It also tells me that maybe I'm better off using bracketing of exposure when I can rather than assuming I can push. In my real scene there were people in the foreground moving around but the one couple held still. Some of the people erased themselves by moving enough but others ghosted and I had to clone them out. Bracketing may or may not have helped but by not bracketing and thinking I could push in post I ended up cloning out a bunch of hot pixels after pushing.

It also tells me not to use Iridient if I am going to need to push exposure in post.

All this helps me understand the camera better and when I can bracket vs push for better results in post.

Thanks again for the help.

Rick







Apr 22, 2017 at 10:59 PM
rdeloe
Offline
• • • •
Upload & Sell: Off
p.1 #5 · p.1 #5 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans


Rick,

Happy to help. I tend not to shoot really high ISO where I need to do what you're doing, so (happily) I haven't noticed the stuck pixels!

Did you try "Pixel mapping" in the Fuji's menu? Does this help?

Also, others have noticed that Iridient X Transformer is not great at higher ISOs. I did extensive testing and reported in another thread, and Lightroom was the way to go for anything over 6400.

If you have a minute, consider sending a note to Brian (the developer) with your test results. He's very responsive and it is only Beta 3 so your findings may give him something to work with.

Rob

rsk7 wrote:
Thanks for going to the effort to duplicate my results.

To me the fact that you got the same result means that there isn't likely something uniquely wrong with my sensor relative to at least one other sample. I'm not going to send my X-T2 in for servicing thinking I got a problem with my sensor.

The purpose of doing the test this way was to figure out why in my real scene and exposure I started getting so many hot pixels when I pushed the foreground +2.5.

It also tells me that maybe I'm better off using bracketing of exposure when
...Show more




Apr 23, 2017 at 07:50 AM
Michael H
Offline
• • • •
Upload & Sell: On
p.1 #6 · p.1 #6 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans


Call me simple minded, but the point of doing all of this is wasted on me, and I suspect others. If you need to push an exposure 5 stops that image is probably not worth saving in the first place. I can't imagine any system being able to perform at an acceptable level with this (useless?) test.

Dig around and look for 'flaws' in anything and I promise you will find them.

We have systems today capable of amazing things. Capabilities in even low level systems that amaze me in what they can do. While I appreciate the technical digging into things I've tried really hard to keep things in a more real-world perspective.



Apr 25, 2017 at 10:38 AM
rsk7
Offline
• • •
Upload & Sell: Off
p.1 #7 · p.1 #7 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans


Read these articles and then maybe you will understand the relevance of this post:

http://improvephotography.com/34818/iso-invariance/
https://fujilove.com/isoless-photography-with-the-fujifilm-x-series/

Rick



Apr 25, 2017 at 11:02 AM
Michael H
Offline
• • • •
Upload & Sell: On
p.1 #8 · p.1 #8 · X-T2 hot pixels, LensCap test, Iso Invariance and X-Trans


rsk7 wrote:
Read these articles and then maybe you will understand the relevance of this post:

http://improvephotography.com/34818/iso-invariance/
https://fujilove.com/isoless-photography-with-the-fujifilm-x-series/

Rick


Thanks for the links. At first glance, I can start to understand some of this. When I get more time I will read more of the articles. Conceptually, I understand though.

That said, I suppose I still have a lot of old-school thoughts on creating images and post-processing. I can see where this might come into play, but time will tell where I go with that understanding. Honestly I'm probably using a bit of it without really realizing it when processing RAFs.



Apr 25, 2017 at 11:26 AM





FM Forums | Fuji Forum | Join Upload & Sell

    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.