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 | Canon Forum | Join Upload & Sell

1      
2
       end
  

Archive 2017 · 6DII blamed for bad censor. 5DIV any better?

  
 
Daan B
Offline
• • • • •
Upload & Sell: Off
p.2 #1 · p.2 #1 · 6DII blamed for bad censor. 5DIV any better?


I did see banding in some 5D4 RAW files (processed in LR) after not so heavy editing. Very disappointing to me as I had hoped to buy/use the 5D4 for doing landscape/interior photography. While getting better, the Canon sensor are still no match to Nikon/Sony sensors. Even the D810, which is a few years old, performs way better for editing high contrasty shots in post. It would be unfortunate if the 6D2 also shows banding when editing high contrast scenes.


Jul 12, 2017 at 02:37 AM
Mikehit
Offline
• • •
Upload & Sell: Off
p.2 #2 · p.2 #2 · 6DII blamed for bad censor. 5DIV any better?


When you say 'some files' do you mean banding is common to all the 'not so heavy editing' on the few times you use it, or do you mean when you have done the 'not so heavy editing' it happens on some (not all) of those images?


Jul 12, 2017 at 03:04 AM
Guest

Guest
p.2 #3 · p.2 #3 · 6DII blamed for bad censor. 5DIV any better?


charlyw wrote:
What a horrible result (and no, I don't talk about the banding!)- editing in the wrong direction totally ruined a potential nice shot, that didn't need pushing, it needed pulling back exposure to make her a silhouette. And if you really want the inside to be seen, pushing still isn't the right tool (a flash or two would be the one and only correct choice), it only makes your images look like a cartoon rendering gone bad - because your lighting will be a mess to start with and even without banding the lack of tonality in the pushed areas
...Show more

Oh, thanks man. Surely it is not the camera defective. It is me being defective seeing the capture in the wrong (different) way.
And surely it's better to say that no one needs to push shadows than saying Canons are not capable in this compared to other brands (Nikon/Sony/Fuji). Nice point, mate!



Jul 12, 2017 at 05:18 AM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #4 · p.2 #4 · 6DII blamed for bad censor. 5DIV any better?


charlyw wrote:
What a horrible result (and no, I don't talk about the banding!)- editing in the wrong direction totally ruined a potential nice shot, that didn't need pushing, it needed pulling back exposure to make her a silhouette.


I gotta disagree with ya on this one. While one certainly could take it in the direction of the silhouette (retaining the creative vibe) ... the information revealed from the lift is a game changer in the message it can convey to the viewer. I agree that a silhouette has it's place, but depending on the message you are trying to convey to your viewer, the direction you take your image can vary.

For shooting interior / exterior (such as this), I typically would bring some additional lighting into the interior space to better balance / narrow the gap in the exposure delta between interior vs. exterior. Alternatively, for static interiors, one can bracket. However, one doesn't always have the option to add lighting or bracket appropriately. When that occurs, we are faced with the prospect of blowing out the exterior, or lifting the interior.

I am of the opinion that there are better ways (i.e. lighting / bracketing, etc.) to yield the most optimal results ... but, there are times that optimal approaches aren't pragmatically available to us. I'm not a big fan of using lift the shadows to the max as a matter of routine, in lieu of other approaches (for optimal results), but I do acknowledge that as the ability to cleanly lift the shadows improves, the viability for doing so also improves.

I think someone wrote "better is better" ... which may not mean it is the holy grail of approach to optimal results (i.e. other methods), but it is kinda hard to refute (philosophically at least) "better is better" to the degree that it is at least welcome (if the penalty isn't increased), even if still isn't most ideal.

As to the image, I may not have pushed / lifted it quite that much, leaving the interior a slight be darker (conveys the diff between interior vs. exterior), yet still revealing the information that was previously hidden. But, I still think that the goal of the message being conveyed is what guides the direction (and original exposure) for processing.



Jul 12, 2017 at 05:52 AM
ggreene
Offline
• • • • •
Upload & Sell: Off
p.2 #5 · p.2 #5 · 6DII blamed for bad censor. 5DIV any better?


charlyw wrote:
And if you really want the inside to be seen, pushing still isn't the right tool (a flash or two would be the one and only correct choice), it only makes your images look like a cartoon rendering gone bad - because your lighting will be a mess to start with and even without banding the lack of tonality in the pushed areas does make this an exercise in futility!


Agreed, the low ISO DR brigade take this to the extreme far too many times. You can see it on a lot of shots on this site and others. Sadly, it's become an accepted norm. That washed out over processed HDR look is just not something that appeals to me.



Jul 12, 2017 at 06:37 AM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #6 · p.2 #6 · 6DII blamed for bad censor. 5DIV any better?


charlyw wrote:
And if you really want the inside to be seen, pushing still isn't the right tool (a flash or two would be the one and only correct choice), it only makes your images look like a cartoon rendering gone bad - because your lighting will be a mess to start with and even without banding the lack of tonality in the pushed areas does make this an exercise in futility!


ggreene wrote:
Agreed, the low ISO DR brigade take this to the extreme far too many times. You can see it on a lot of shots on this site and others. Sadly, it's become an accepted norm. That washed out over processed HDR look is just not something that appeals to me.


Agreed on both counts that more light is the "better" way (so noted in previous post) and that the over-processed HDR look isn't my preference ... but, when it's all ya got to work with, well sometimes it's all ya got to work with (imo, just don't push it so hard).



Jul 12, 2017 at 06:53 AM
TeamSpeed
Offline
• • • • •
Upload & Sell: Off
p.2 #7 · p.2 #7 · 6DII blamed for bad censor. 5DIV any better?


I have pushed shadows up 3-4 stops with no apparent banding on the 5D4. I can push the 7D2 by about 3 stops, once I get to 4, there are traces of banding starting to show.

Banding and DR are two different, but somewhat related, issues. DR in pushing shadows has more to do with how electronic noise is sitting in those shadows, banding is a different situation, has more to do with characteristics of the noise.

If I have 2 images with exactly the same noise, exposed identically, and one has noise evenly dispersed across the image, it is more salvageable than the one that has the noise banded together. Said differently, if I have 2 cameras that tout the same DR range, but one has banding with its noise in the shadows and one doesn't, then I will use the latter over the former, despite the seemingly same DR rating, because the images are more usable.

I really like the 5D4's Sensor, and hope for the best on the 6D2, hopefully people don't Censor their findings!



Jul 12, 2017 at 07:21 AM
charlyw
Offline
• • •
Upload & Sell: Off
p.2 #8 · p.2 #8 · 6DII blamed for bad censor. 5DIV any better?


AlexDROP wrote:
Oh, thanks man. Surely it is not the camera defective. It is me being defective seeing the capture in the wrong (different) way.
And surely it's better to say that no one needs to push shadows than saying Canons are not capable in this compared to other brands (Nikon/Sony/Fuji). Nice point, mate!


It doesn't matter what brand, the results do look equally horrible because of the lack of tonal distinction for which no processing in the world can compensate! And it is indeed the photographers fault for forgetting the basics of exposure and how to work with the light - or bring his light if needed.



Jul 12, 2017 at 07:49 AM
Guest

Guest
p.2 #9 · p.2 #9 · 6DII blamed for bad censor. 5DIV any better?


charlyw wrote:
It doesn't matter what brand, the results do look equally horrible because of the lack of tonal distinction for which no processing in the world can compensate! And it is indeed the photographers fault for forgetting the basics of exposure and how to work with the light - or bring his light if needed.


It's very kind of you to remind the basics of exposure.
But you know there are situations that you might not meet in your shooting practice when you can't decide which part of exposure to put in priority: shadows or highlights - because details in all areas matter. In general this genre of photography is called landscape. In my particular case it is night cityscape. And I can't take enough light for the building exteriors and the sky
Here is a sample of 1 exposure converted raw file that doesn't contain details in highlights because of limited camera DR (6D indeed) and my decision to prioritize shadows. The second part of the image shows the resulting picture of 2 exposures manual blending (the highlights priority exposure taken with -4 EV compensation). Is it unnatural looking for a manipulated HDR shot lacking "tonal distinction"?
Having 1 exposure with details both in shadows and highlights I can make the same looking resulting picture (on the right) with less efforts and time. So this image is a question of time but seldom limited DR turns into a question of missed opportunity.





Jul 12, 2017 at 08:29 AM
charlyw
Offline
• • •
Upload & Sell: Off
p.2 #10 · p.2 #10 · 6DII blamed for bad censor. 5DIV any better?


AlexDROP wrote:
Having 1 exposure with details both in shadows and highlights I can make the same looking resulting picture (on the right) with less efforts and time.


And here you are neglecting the fact that the underexposed shot doesn't have the tonal graduations in the midtones which make or break your sample (or any photo to be precise) - thus the less effort and time has a significant cost in terms of the image quality! Not everything that sounds promising in the first place does work out in the end and these butt ugly one shot HDR are such a thing because of the horrendous lack of tonal graduation in all areas but the highlights.



Jul 12, 2017 at 09:06 AM
Guest

Guest
p.2 #11 · p.2 #11 · 6DII blamed for bad censor. 5DIV any better?


charlyw wrote:
...and these butt ugly one shot HDR are such a thing because of the horrendous lack of tonal graduation in all areas but the highlights.


If you are talking about one shot HDR taken with a Canon camera, you are absolutely right. And I support your point completely.
But Nikon/Sony shots are more capable for retaining gradual transitions of tones in all areas even after heavy (compared to Canon cameras) post-processing. You may restore connection with reality simply checking out DPreview reviews and test shots of Canon and Nikon FF cameras (e.g. D810 review) https://www.dpreview.com/reviews/nikon-d810/14



Jul 12, 2017 at 09:21 AM
charlyw
Offline
• • •
Upload & Sell: Off
p.2 #12 · p.2 #12 · 6DII blamed for bad censor. 5DIV any better?


AlexDROP wrote:
But Nikon/Sony shots are more capable for retaining gradual transitions of tones in all areas even after heavy (compared to Canon cameras) post-processing.


Nope, they don't - because the tonal graduations are lost due to the severe underexposure. I have yet to see a single one shot HDR that doesn't bear the hallmarks of a shot gone bad. Even the highly praised (by the camera user himselfl) shot of the tulip field DPreview showed as an example what can be done shows clearly that it shouldn't be done under any circumstances.


Let me give you an example. When you shoot at 5 stops under to push by 5 stops in postprocessing not only do you get more noise (5 stops means ISO-3200 instead of ISO 100) but you also get different noise and tonal graduation characteristics because you will be digitally scaling not only the signal resulting in distinct signal steps but also the noise. So if your sensor has a 14 bit ADC (and uses that, not like the Sony sensors which only employ 12 bit most of the time - if you only look at them at an angle they drop down to 12 bit) and let's for simplicity say that full well capacity of the sensor cell means that 1 electron is translated to 1 ADU at ISO 100, then your subject normally lives between 512 and 4096 electrons per cell - drop the exposure by 5 stops and you are at the 16 to 128 electron level. Instead of 22 to 64 electrons of shot noise you get between 4 and 11 electrons of shot noise - physical fact, unavoidable. Now you digitally scale back - resulting in distinct steps of 32 between tonal values and the noise will also be equally coarsely defined. On top of that you have greatly increased noise levels as well - to the tune of 128 to 352 in distinct steps of 32 around the signal value which itself is equally coarse in resolution. Yes you can do things to mask that but that doesn't really make things better, only less blatantly visible...

It is a fallacy to think that just because the technical noise doesn't interfere you can get away with such an underexposure and get a decent image out of it!

Edited on Jul 12, 2017 at 09:59 AM · View previous versions



Jul 12, 2017 at 09:41 AM
Guest

Guest
p.2 #13 · p.2 #13 · 6DII blamed for bad censor. 5DIV any better?


charlyw wrote:
Nope, they don't - because the tonal graduations are lost due to the severe underexposure. I have yet to see a single one shot HDR that doesn't bear the hallmarks of a shot gone bad. Even the highly praised (by the camera user himselfl) shot of the tulip field DPreview showed as an example what can be done shows clearly that it shouldn't be done under any circumstances.


Check this link https://youtu.be/y8WgDamhTME
It's very fun to learn what modern cameras are capable for.



Jul 12, 2017 at 09:48 AM
KKFung
Offline
• • •
Upload & Sell: Off
p.2 #14 · p.2 #14 · 6DII blamed for bad censor. 5DIV any better?


The "under expose and then push 5 stops game" or say "over cook a photo to test camera ability" is really a funny game for forum discussion and never end, however in real life shooting, if the scene DR is really high, then 1 stop different between cameras will not help the situation in most case. Also how can you judge the one more stop DR of the camera can make you safe when you are going to capture the changing lighting of a beautiful scene on site? When we know ETTR not helping then we will go to bracketing no matter you are using sony, canon, nikon or even iphone right? And after we use these kind of techniques then the DR different between camera will make different?

I have many examples about DR matter in actual practice, I can say if we don't over cook the photo, 11-12 stops DR capability from modern camera is really good good good, the following photo showing a push on the shadow +50 and black +19, the foreground already brighter then when I look at it by my eyes in actual.

B14I5391-Pano-2 by KK Fung, 於 Flickr

B14I5391-Pano by KK Fung, 於 Flickr

未命名 by KK Fung, 於 Flickr

Enjoy shooting!



Jul 13, 2017 at 11:06 AM
bozziovai
Offline
• •
Upload & Sell: Off
p.2 #15 · p.2 #15 · 6DII blamed for bad censor. 5DIV any better?


if Canon Execs were to see this thread, they would be flattered. just proves to show how excited people are with the 6DM2, that even when it's not yet officially in the customers hands, people are doing some sort of "reviews" or pre-production model results.


Jul 13, 2017 at 02:15 PM
garyvot
Offline
• • • •
Upload & Sell: On
p.2 #16 · p.2 #16 · 6DII blamed for bad censor. 5DIV any better?


RustyBug wrote:
I gotta disagree with ya on this one. While one certainly could take it in the direction of the silhouette (retaining the creative vibe) ... the information revealed from the lift is a game changer in the message it can convey to the viewer. I agree that a silhouette has it's place, but depending on the message you are trying to convey to your viewer, the direction you take your image can vary.

For shooting interior / exterior (such as this), I typically would bring some additional lighting into the interior space to better balance / narrow the gap in the
...Show more

My problem with that shot (and the ridiculous shadow lift) is that if the photographer had wanted to balance the exposure properly he or she should have used some kind of fill, as skilled photographers have been doing for decades.

In the *absence* of a secondary light source, pushing shadows a bit can help, but only if done subtly, so that the image matches the visual perception one would have of such a heavily backlit scene.

Trying to *create* a secondary fill light by pushing shadows (as the photographer has done here) creates horribly artificial results. I think dynamic range has become a kind of crutch for too many, sadly.

Edit: I suppose this sounds kind of harsh, not my usual style. For some reason though, this topic (and those who judge the worth of all cameras by the dynamic range of the sensor) sets me off. Sorry.

Edited on Jul 13, 2017 at 03:30 PM · View previous versions



Jul 13, 2017 at 02:20 PM
garyvot
Offline
• • • •
Upload & Sell: On
p.2 #17 · p.2 #17 · 6DII blamed for bad censor. 5DIV any better?


Isaacheus wrote:
What other characteristics have you found from the 5dmk4 sensor?


Mainly that by default the sensor has a shorter "shoulder" in the highlights, and so is easily prone to highlight blocking. Also, Adobe's Camera Standard profile (and all of its camera-specific profiles for the 5D Mark IV) has crushed blacks. In combination, these two things lead to files that at first were very challenging for me to adjust.

I'll share my solution, in case it is beneficial for anyone else:

For improved highlight tonality, I now shoot the 5D Mark IV with Highlight Tone Priority enabled at all ISOs up to 32000. I also enable highlight blinkies and pay more attention than I have with previous Canon cameras. Makes a big difference in the malleability of the files, and ACR/LR handle Canon's HTP RAWs perfectly.

To fix the crushed blacks and certain color issues, I've reworked my ACR defaults for the 5D Mark IV to:

In Camera Calibration:
- Profile: 'Camera Standard'
- Red Primary Hue: -5
- Green Primary Hue: -3
- Green Primary Saturation: +10

In Tone Curve:
- Parametric Tone Curve: Shadows +20

Other:
- Lens corrections, sharpening, NR, and FX to taste.

These settings produce punchy files with pleasing (to me) colors that are a reasonable match for historical DPP output, and a good starting point for editing. (Again, for me--I'm sure someone will hate this.)

Adobe's default 'Adobe Standard' profile for the 5D Mark IV is next to useless for this camera, sadly.



Jul 13, 2017 at 02:38 PM
TeamSpeed
Offline
• • • • •
Upload & Sell: Off
p.2 #18 · p.2 #18 · 6DII blamed for bad censor. 5DIV any better?


I exclusively use DPP for raw processing, it seems to handle some aspects better with the raw to JPG/TIF, then I will do more work in photoshop. If you are a LR/Adobe savant, then you might figure out the magic sauce, but DPP takes less effort, especially if you nail a majority of settings properly in-camera.


Jul 13, 2017 at 03:38 PM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.2 #19 · p.2 #19 · 6DII blamed for bad censor. 5DIV any better?


+1 ... too much lift goes against what the brain expects to see.

Imo, more DR means cleaner lift when done judiciously ... not necessarily meaning we should extreme lift in lieu of other viable approaches.



Jul 13, 2017 at 05:54 PM
1      
2
       end




FM Forums | Canon Forum | Join Upload & Sell

1      
2
       end
    
 

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.