Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

[Mikehit]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Yep.

 

[Loswr]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Nope.

 

[Loswr]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Well, sort of... Yes, the DR of the downsampled 6DII image will be higher, but downsampling doesn't change the DR when the image is captured. IMO, it's the latter that matters – if you have a 13-stop scene, and 12 stops of DR, you've clipped one or both sides. If you then downsample that image until you have 13 stops of DR, you don't get back what you clipped at capture.

 

[x-vision]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Yeah buddy, it definitely has better IQ than the 80d. That one stop DR advantage is the only advantage the 80 will have,at any other metric it'll be handily bested. If you're in doubt pls compare the original 6d to the 80d.

Sorry, pal, you either didn't read my post - or you didn't comprehend it.

Let me summarize/simplify it for you:
All that I said was that I'm not paying the FF premium unless image quality is better in every metric.
But since the 6DII is not better at every metric than the 80D, I'm not paying the FF premium.
I never said that the 80D has better image quality.

Was that simple/short enough for you?

 

[Don Haines]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Well, sort of... Yes, the DR of the downsampled 6DII image will be higher, but downsampling doesn't change the DR when the image is captured. IMO, it's the latter that matters – if you have a 13-stop scene, and 12 stops of DR, you've clipped one or both sides. If you then downsample that image until you have 13 stops of DR, you don't get back what you clipped at capture.

makes sense to me......

When I hit a static scene where I know I am going to be clipping, I turn on exposure bracketing ( 2 or 3 stops), shift exposure to where I am clipping on both ends, let er rip for the three pictures, and merge them in lightroom.... (Left image is the centre exposure of the three images, right image is the assembled lightroom image)

 

[Cthulhu]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Yeah buddy, it definitely has better IQ than the 80d. That one stop DR advantage is the only advantage the 80 will have,at any other metric it'll be handily bested. If you're in doubt pls compare the original 6d to the 80d.

Sorry, pal, you either didn't read my post - or you didn't comprehend it.

Let me summarize/simplify it for you:
All that I said was that I'm not paying the FF premium unless image quality is better in every metric.
But since the 6DII is not better at every metric than the 80D, I'm not paying the FF premium.
I never said that the 80D has better image quality.

Was that simple/short enough for you?

Sure mate, except for the part where you didn't say that in your original post and definitely said you wouldn't be buying the 6d2 since it has worst image quality than your 80d...that aside it's simple enough.

However, the reason to get a FF camera in the first place is for the superior image quality.
That's why FF cameras command a premium.

Thus, I surely won't be paying the FF premium for inferior image quality than my 80D.

Unless DR at ISO100 is the end all be all of image quality you're at odds with your own statements. If it is you should probably sell your 80d and buy a d7200.

 

[ecka]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

"dynamic range isn't everything"
Not feeling pain isn't everything, so have some, buy a 6D2

 

[x-vision]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Sure mate, except for the part where you didn't say that in your original post and definitely said you wouldn't be buying the 6d2 since it has worst image quality than your 80d...that aside it's simple enough.

Look, fella, here is what I said as well:

• Without unconditionally superior image quality, the FF premium is not worth it for me.

The 6DII doesn't have unconditionally superior image quality vs the 80D, since the 6DII dynamic range at ISOs 100-200 is worse than that of the 80D.
Therefore, I will not be paying the FF premium to get the 6DII.

My post ended with 'YMMV', which is usually interpreted as "people may have a different opinion or experience to yours".
That is, if the FF premium for the 6DII is not worth it for me, it doesn't mean that it's not worth it for other people.
In fact, just the opposite - I fully realize that other people may differ.

 

[Cthulhu]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Sure mate, except for the part where you didn't say that in your original post and definitely said you wouldn't be buying the 6d2 since it has worst image quality than your 80d...that aside it's simple enough.

Look, fella, here is what I said as well:

• Without unconditionally superior image quality, the FF premium is not worth it for me.

The 6DII doesn't have unconditionally superior image quality vs the 80D, since the 6DII dynamic range at ISOs 100-200 is worse than that of the 80D.
Therefore, I will not be paying the FF premium to get the 6DII.

My post ended with 'YMMV', which is usually interpreted as "people may have a different opinion or experience to yours".
That is, if the FF premium for the 6DII is not worth it for me, it doesn't mean that it's not worth it for other people.
In fact, just the opposite - I fully realize that other people may differ.

Listen, bloke, your argument only makes sense if the +/- 1 stop dynamic range the 80d might have over the 6d2 is enough to bring it's image quality above that of the 6d2.

I'm not buying the 6d2 because I find it disappointing on the whole, but if I had to choose btw the two and image quality was my concern, the 6d2 will handily beat the pants of the 80d.

Or to make it short and simple for you: there's more to IQ than DR.

 

[Mikehit]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Well, sort of... Yes, the DR of the downsampled 6DII image will be higher, but downsampling doesn't change the DR when the image is captured. IMO, it's the latter that matters – if you have a 13-stop scene, and 12 stops of DR, you've clipped one or both sides. If you then downsample that image until you have 13 stops of DR, you don't get back what you clipped at capture.

I agree with that....sort of . As I understand it the DR discussion is about the level of shadow detail and at the moment I am not sure if this is defined as whether a sensor can pick up the signal at all, or if it is about signal above noise.
If you preserve the highlights, then shadow recovery depends on the artefacts you see when you do so. Some look to a higher DR sensor to recover the shadow (no downsampling needed) but you can also do it by downsampling which mean you can 'uncover' an until-then hidden signal by effectively lowering the noise floor (which I believe this is what the astrophotographers do).

[this is gleaned from bits-and-pieces read over the years so may be incorrect].

 

[Don Haines]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Sure mate, except for the part where you didn't say that in your original post and definitely said you wouldn't be buying the 6d2 since it has worst image quality than your 80d...that aside it's simple enough.

Look, fella, here is what I said as well:

• Without unconditionally superior image quality, the FF premium is not worth it for me.

The 6DII doesn't have unconditionally superior image quality vs the 80D, since the 6DII dynamic range at ISOs 100-200 is worse than that of the 80D.
Therefore, I will not be paying the FF premium to get the 6DII.

My post ended with 'YMMV', which is usually interpreted as "people may have a different opinion or experience to yours".
That is, if the FF premium for the 6DII is not worth it for me, it doesn't mean that it's not worth it for other people.
In fact, just the opposite - I fully realize that other people may differ.

I rather doubt that you are going to find a camera with unconditionally better IQ..... there is a lot that goes into image quality, and you will find that glass and AF system are far more important than sensor....

 

[snoke]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Huh, I see they're so "political correct", and what they say about 6D, they can apply the same to all the new release DSLR, mirrorless cam, nothing new, everything is as expected, for example "Brand new sensor and latest engine" -> wtf is that, of course it is obviously as say in their specs, why include in PROS, because they cannot find anything new to say about this 6D2.

Yes this right. If you not write good about Canon, Canon stop sending cameras to review. If no new camera for review, nobody buy magazine or goto website and reviewer lose money. Because Amazon, DPR too big/important, Canon can't ignore.

On the contrary. Many review websites had pure clickbait titles for years (bashing Canon) for generating profit of such website. Ironic, really, that websites bash Canon for making a profit, in order for those website to create a profit

Clickbait is everywhere. Ads here. Titles there. But for some serious photography website (almost no clickbait ads), Canon stop sending review camera when start be critical of Canon.

 

[BillB]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Well, sort of... Yes, the DR of the downsampled 6DII image will be higher, but downsampling doesn't change the DR when the image is captured. IMO, it's the latter that matters – if you have a 13-stop scene, and 12 stops of DR, you've clipped one or both sides. If you then downsample that image until you have 13 stops of DR, you don't get back what you clipped at capture.

I agree with that....sort of . As I understand it the DR discussion is about the level of shadow detail and at the moment I am not sure if this is defined as whether a sensor can pick up the signal at all, or if it is about signal above noise.
If you preserve the highlights, then shadow recovery depends on the artefacts you see when you do so. Some look to a higher DR sensor to recover the shadow (no downsampling needed) but you can also do it by downsampling which mean you can 'uncover' an until-then hidden signal by effectively lowering the noise floor (which I believe this is what the astrophotographers do).

[this is gleaned from bits-and-pieces read over the years so may be incorrect].

This issue seems to be at least partly a product of DPR's terminology and magic numbers used to describe how well the files produced by various cameras will respond to shadow lifting, which is a post processing technique, as is file downsizing. There are 5EV shadow lifts and there is screen DR and print DR, all of which generate images and magic numbers of dubious significance, which are very useful in generating all sorts of Internet craziness while attracting attention to DPR.

To me, it ultimately comes down to print quality, which does depend on post processing as well as what comes out of the camera.

 

[Sporgon]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Well, sort of... Yes, the DR of the downsampled 6DII image will be higher, but downsampling doesn't change the DR when the image is captured. IMO, it's the latter that matters – if you have a 13-stop scene, and 12 stops of DR, you've clipped one or both sides. If you then downsample that image until you have 13 stops of DR, you don't get back what you clipped at capture.

I agree with that....sort of . As I understand it the DR discussion is about the level of shadow detail and at the moment I am not sure if this is defined as whether a sensor can pick up the signal at all, or if it is about signal above noise.
If you preserve the highlights, then shadow recovery depends on the artefacts you see when you do so. Some look to a higher DR sensor to recover the shadow (no downsampling needed) but you can also do it by downsampling which mean you can 'uncover' an until-then hidden signal by effectively lowering the noise floor (which I believe this is what the astrophotographers do).

[this is gleaned from bits-and-pieces read over the years so may be incorrect].

This is the point, and the one that is misunderstood regarding "DR". It would seem that many people think that a higher "DR" camera such as the 810D can capture more EV range including highlights, which of course it can't. It's capturing pretty much the same EV range, it's just a question of when the shadow detail becomes a noisy mess, ultimately losing all tonality and definition. We don't discuss highlight in the same way because "highlight headroom" ( the ability to describe differing very light, almost white tones just before blowing out) is more subtle and much less noticeable than the other end of the scale.

Downsampling reduces noise, moving the lower acceptable point lower down, so we have greater "DR". But the "DR" at capture hasn't changed. You are right in that it is the way that dynamic range is defined; it's rather ambiguous.

So many things happen to an image when we downsample, this increase in "DR" is just one of them.

As a practical photographer I prefer to think in terms of EV range, and then the EV latitude for processing. The EV range for all these cameras, format for format is pretty much the same, it's the EV latitude in processing that differs, and it looks like the 6DII has the same latitude as the 5DII, III, 6D etc. For the vast majority of people that won't make any difference to their images at all. Just don't under expose !

 

[snoke]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

+1 of course. On another thread, I've likened DR Storm 2017 to (say) Honda putting out a new model of a car that is now brimming with new tech but it lost a little punch in first gear compared to its predecessor.

No, new Honda have less MPG than old Honda. MPG is something almost nobody care about but for small number it important. Nobody who buy Ferrari care about MPG

 

[tomscott]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

I have to say im going to eat my words here. After looking at early RAW files I thought it looked promising.

BUT...

dpreview have launched a gallery of images which you can download the RAW and JPGs.

You can find the images here:
https://www.dpreview.com/samples/5865039367/canon-eos-6d-mark-ii-sample-photos

This image caught my eye on the home page, as im a wedding/event photographer.

Download link here
http://download.dpreview.com/canon_eos6dmkii/IMG_0528.CR2

Although this image is not typically how I would shoot a pre-wedding (I assume this image is, could just be a snapshot). If I was shooting into sunset like this I would be using some sort of illumination be that flash or a large reflector.

Sometimes you dont have the time or the means if you are on holiday etc which is what this camera is aimed at - the enthusiast.

Obviously hugely underexposed, protecting the highlights. I know from experience that highlight retention is pretty good in canon files and less so with the shadows so probably would have shot this the opposite way round protecting the shadows, as these are the subjects.

I would have shot with an extra stop or stop and a half giving much more room to play with. It would seem it was a photographers decision to shoot this underexposed with -1.33 stops dialed in, which makes me think this was on purpose or they dont shoot Canon cameras regularly. But is a really easy mistake to make and one anyone could.

Anyway... Its a great test for anyone looking to shoot these type of images especially at base ISO.

Lift the shadows and up the exposure as preference and there is very little latitude here. It took me a good 10-15 mins to get the the point they have in the processed image above. Creating a good balance between exposure and acceptable noise.

An easy way to remove this issue is to crush the deep shadow which is what they seem to have done in the preview but with a bit of tinkering you can get there.

Colour noise with purple muddy shadows are also prominent very quickly and so is the pattern noise.

I found myself using an old profile I would use with my 5DMKIII to remove the noise and colour issues. Pretty aggressive 30 on the luminescence noise slider, 50 on the colour slider and 90 on the smoothness slider. Looks acceptable, using this sort of aggression on an ISO 100 files makes it quite soft.

I ended up with an acceptable image but it took a lot of work to get there. It can be done just like the 5DMKIII but you end up spending more time processing.

So I would say the early negative comments are valid. The 5DMKIV would give you more room.

What I will say...

Expose properly or use lighting etc and the files are excellent just like the 5DMKIII. Its that 1-5% of images that you may need the help and is the difference between a keeper and a binner.

It is disappointing that its very similar to a 5 year old camera, even if it is still excellent.

You just have to ask yourself would I ever shoot like this. I only really had issue with my 5DMKIII while traveling because I didnt have access to my lighting gear because it would have been too heavy.

At the end of the day even with this image pulled it still doesn't look great because you cant create light where it didnt exist in the first place. You cant bend the laws of physics. You have to introduce light to balance exposure.

I would say - skills like lighting are much more important than worrying about 1-2 stops of DR. I also understand why people are upset. The 5DMKIV would have produced a usable image here, although it still wouldn't be an ideal situation IMO.

 

[Khalai]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Clickbait is everywhere. Ads here. Titles there. But for some serious photography website (almost no clickbait ads), Canon stop sending review camera when start be critical of Canon.

Bollocks. That would create s**tstorm on the web, word would get spread quickly and Canon would receive pretty hefty backfire for that.

Look at Intel for example. Even the reviewers, who are openly critical against Intel still receive early samples and engineering samples? Why? Good advertising for both parties, simple as that.

I have to say im going to eat my words here. After looking at early RAW files I thought it looked promising.

BUT...

dpreview have launched a gallery of images which you can download the RAW and JPGs.

You can find the images here:
https://www.dpreview.com/samples/5865039367/canon-eos-6d-mark-ii-sample-photos

This image caught my eye on the home page, as im a wedding/event photographer.

Download link here
http://download.dpreview.com/canon_eos6dmkii/IMG_0528.CR2

Although this image is not typically how I would shoot a pre-wedding (I assume this image is, could just be a snapshot). If I was shooting into sunset like this I would be using some sort of illumination be that flash or a large reflector.

Sometimes you dont have the time or the means if you are on holiday etc which is what this camera is aimed at - the enthusiast.

Obviously hugely underexposed, protecting the highlights. I know from experience that highlight retention is pretty good in canon files and less so with the shadows so probably would have shot this the opposite way round protecting the shadows, as these are the subjects.

I would have shot with an extra stop or stop and a half giving much more room to play with. It would seem it was a photographers decision to shoot this underexposed with -1.33 stops dialed in, which makes me think this was on purpose or they dont shoot Canon cameras regularly. But is a really easy mistake to make and one anyone could.

Anyway... Its a great test for anyone looking to shoot these type of images especially at base ISO.

Lift the shadows and up the exposure as preference and there is very little latitude here. It took me a good 10-15 mins to get the the point they have in the processed image above. Creating a good balance between exposure and acceptable noise.

An easy way to remove this issue is to crush the deep shadow which is what they seem to have done in the preview but with a bit of tinkering you can get there.

Colour noise with purple muddy shadows are also prominent very quickly and so is the pattern noise.

I found myself using an old profile I would use with my 5DMKIII to remove the noise and colour issues. Pretty aggressive 30 on the luminescence noise slider, 50 on the colour slider and 90 on the smoothness slider. Looks acceptable, using this sort of aggression on an ISO 100 files makes it quite soft.

I ended up with an acceptable image but it took a lot of work to get there. It can be done just like the 5DMKIII but you end up spending more time processing.

So I would say the early negative comments are valid. The 5DMKIV would give you more room.

What I will say...

Expose properly or use lighting etc and the files are excellent just like the 5DMKIII. Its that 1-5% of images that you may need the help and is the difference between a keeper and a binner.

It is disappointing that its very similar to a 5 year old camera, even if it is still excellent.

You just have to ask yourself would I ever shoot like this. I only really had issue with my 5DMKIII while traveling because I didnt have access to my lighting gear because it would have been too heavy.

At the end of the day even with this image pulled it still doesn't look great because you cant create light where it didnt exist in the first place. You cant bend the laws of physics. You have to introduce light to balance exposure.

I would say - skills like lighting are much more important than worrying about 1-2 stops of DR. I also understand why people are upset. The 5DMKIV would have produced a usable image here, although it still wouldn't be an ideal situation IMO.

Honestly, that'd downright badly exposed picture to begin with. DR latitude should not replace poor exposure metering and lack of proper lightning. Such backlit shot would be unusable with any camera out there...

 

[Sporgon]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

I've had the 5D2 before 6D and I know how it behaves in similar lighting conditions. The low ISO noise wasn't that "in your face" straightaway out of the shadows. I needed to push the brightness up for that. Maybe it's the early RAW support problem, I hope. But I've noticed that DPR removed some of the worst DR samples from their gallery, I can't find them anymore. So, we'll see ... I still have those downloaded.

I would just be cautious of drawing too many negative conclusions from the various downloadable RAW samples at low ISO floating about: I have yet to find one which I would consider to be exposed optimally for a Canon of this type in raw. In other words there is more highlight room available than the photographer has used. This makes a huge difference to these "off chip ADC" Canons as I am sure you are aware.

I'm pretty sure the 6DII is not worse than the 5DII.

I have to say im going to eat my words here. After looking at early RAW files I thought it looked promising.

BUT...

Same comments as above Tom; I'd like to see some optimally exposed 6DII shots in high contrast situations. Looking at the few blown highlights around the sun in that wedding shot I think that may be an impossible exposure.

Honestly, that'd downright badly exposed picture to begin with. DR latitude should not replace poor exposure metering and lack of proper lightning. Such backlit shot would be unusable with any camera out there...

Agreed, or at least the exposure compromise has been too much towards the impossible highlights.

 

[tomscott]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

It is interesting to see what you can do and completely agree would be a binner and it is funny they would post an image like that to show how bad the camera is

On the same merit I was hoping like the 7DMKII 5DSR 5DMKIV and 80d the old purple shadow cast and colour noise would be a thing of the past but its not.

We are 20 odd pages in and everyone's arguing but nobody has posted an image.

Plenty out there have a go

Quite a few good high contrast images in the DP image slideshow IMG_0741.CR2 is worth a look.

Like I said the files look very familiar almost feel exactly like the 5DMKIII.

 

[privatebydesign]

Re: Canon EOS 6D Mark II Dynamic Range Talk & Sample Images

Depends what you want your muck ups to look like, but make no mistake this file is not exposed optimally, a -1.33 EC was dialed in yet there is at least one stop of head room unused!

Here is my two minute play in LR, mind you I prefer a 'natural' look over the radioactive HDR style Dan is trying to portray as optimal.