-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Subtle tonality/hue/color differences - what Camera does it best?
- Thread starter MB_W124_200D
- Start date
5
57even
Guest
People like different cameras for different reasons. They only get into trouble when they start trying to justify it.
Exactly. It's mostly subjective. Get the camera you like and be happy.
so how do you explain at least 3 folks were compelled to leave a comment that a then-new camera had the same look as then a 12-year-old body. would they not have to notice something ?.has that ever happened before with a different camera ...why do some say there is something special about the 5D ?
yes but an unprecedented amount of folks do say there is something special about the photos of the 5D..
what trouble?
5
57even
Guest
When people start saying 'it's the colour filter array' or 'CCD is better than CMOS' or some other technical reason for colour differences, it's mostly based on poor research, or 'something they read on the internet'.
For example, the sensor in the 6Dii bears no relation at all to the one in the 5D, although in the DPR studio scene the JPEGs do indeed look very similar.
However, the raws look far less saturated, and much more similar to the 5D iv, which makes sense as both used the new dual pixel sensor technology, and likely have a very similar CFA spectral response, as is common on contemporary model lines.
At the same time, JPEGs on the 5D iv are somewhat less saturated, more in keeping perhaps with the intended audience. No doubt, one could increase saturation and create a better match.
In other words, the similarities between 5D and 6Dmk2 are almost certainly attributable to the Digic image processing in-camera, and have very little to do with colour filters, AA filters, or any other hardware parameters.
So my point is that if you like something, there is no need to invent a lot of pseudo scientific reasons for your preference. Canon clearly felt that it would be a popular look and they were clearly right, given the response you reported.
are you saying the 5Dmki has not got a strong CFA and a week AA filter?? which i claimed in an earlier post,,if so what are the characteristics of the CFA and AA filter in the 5D?
i at no point did i say why the 6Dmkii and 5D have a similar look to them ..just that i noted they do and others commented the same looking at the same D6mkii photos
i whuld be REALLY interested to know what is the pseudoscientific reasons i have invented
spider-mario
Senior Member
I would be quite surprised if a weak AA filter helped with this. Aliasing sounds exactly like an artifact typical of bad digital sampling. Which of the two versions of the image at the top of this article looks more natural to you: https://www.red.com/red-101/resolution-aliasing-motion-capture
Last edited:
5
57even
Guest
What I said is that it has nothing to do with the similarities, and I explained why.
So why did you mention AA filters and colour filters? What do they have to do with it?
Quoted, not invented. I didn't disagree with the similarities, only the reasons you quoted.
ggbutcher
Senior Member
- Messages
- 1,842
- Solutions
- 4
- Reaction score
- 1,487
Thank you, 57even, for taking on the beleaguered discourse on why any difference in cameras' renditions needs careful consideration of ALL the contributions.
I've done a lot of poking around with respect to cameras' spectral performance and color management in general in the past 5 years, I'm not a color scientist by any measure but I've come to a mechanic's understanding of how color is managed in digital images, And, in that work I've come to realize that the most significant contribution to color quality is white balance - if you don't make the camera encoding respect the light which bathed the sensor, you don't have proper colors as a starting point.
After that, it's the inevitable transform from the camera RGB encoding to what I call "rendition colorspace", the colorspace that respects the capabilities of the rendition medium, either electronic display or print. This is the phase of workflow that got me digging into color management, trying to make blotchy blue LED spotlight rendition look more like what I saw in the scene.
And finally, the least significant contributor has been the color filter array design of the sensor. I've seen significant consistency in particular manufacturers, and observable but not significant differences manufacturer-to-manufacturer in recent models I've either measured or have obtained SSF data from another source. The deltaE numbers I posted in my previous post show this: the differences in those magnitudes are not significant in terms of what we observe.
Soooo, when we look at these JPEGs in this thread and see differences, the contributions by all of the above need to be individually considered to understand the difference.
5
57even
Guest
Yep. Garbage in, garbage out...
Good way to waste a few weekends... ;-) You might add that if you don't calibrate your display, you are not even seeing the right colours to start with... or as close as possible given that display colour spectra are not monochromatic.
In the end, at low ISO, your display is likely to be far more limiting that your camera.
Indeed. Cameras are all a lot more similar than film ever was. It's generally specific colours that give them away, like bright reds and dark blues, where minor differences are most apparent. But it's seldom a major shift.
What most people seem to notice is saturation and contrast more than actual colours, and some cameras ramp that way up in the JPEGs.
They can be optimised, but yeah. The JPEGs reflect the manufacturer's signature look, but what we see is more processing than hardware.
It's not cherry picking, I specifically chose to show this data because the OP mentioned that he liked the tonal gradation of the 5D. This spectrograph explains why.Nice bit of cherry picking. What makes you think this is still true (the 5Dmk 2 is an old camera)?Great question. It's easiest to visualize via a comparative spectrograph that shows how much overlap there is between the color channels. Here we see the Canon 5D Classic plotted next to the Canon 5D Mk2.
"Weaker" means that there is more overlap of wavelengths of the CFA installed in the 5D Mark2. The color channels are less specific than the 5D Classic allowing more false green color in the red and blue channels. As to the effects this had on color discrimination I would suggest this link:
http://www.fredmiranda.com/forum/topic/1013683/2&year=2011
How do we know that the 5D Mk2’s weaker CFA isn't an aberration? Phase One, who manufacture cameras of their own and profile hundreds more for C1 Pro, confirmed in 2018 that the entire camera industry had moved to weaker CFA's shortly after the move from CCD to CMOS. Read my synopsis of their commentary here:
https://www.dpreview.com/forums/post/60658069
Well, I'm glad we can at least agree that the 5D beat the 5D Mk2 for color as evidenced by both their SMI scores as well as by their spectrographs!
Regarding why the the M8 scored such a low SMI, obviously it's because DxOMark tested it for color without an IR-Cut filter. Consequently all of the colors were significantly thrown off.
Again, the SMI of the M8 can't be used as a comparator because of the IR-Cut filter issue.
But let's talk about the 5D Mk4 and those other modern cameras. I notice you prefer to only list their "daylight" results, but DxOMark actually have two SMI scores for each camera. The second one for incandescent lighting is actually more revealing.
Here’s the reason. Full spectrum daylight is an easy CFA test for cameras because it is balanced and complete. However, partial spectrum lighting (like that from incandescent, fluorescent, and LED bulbs) is much harder test for the reverse. Therefore, it is the SMI for CIE-A incandescent lighting that really tells us about the strength of the CFA.
So if we look at the complete picture the 5D Mk4 is actually pretty weak at 85,78… again the second number representing its incandescent SMI performance. Same with the Nikon 7Zii at 83,78 and the Sony A7iii at 82,78. The 5D Classic handily beats both on both SMI scores with an SMI of 84,80 thanks to its legacy CFA.
It is interesting to note that some CCD cameras were able to score equally well on both illuminants. The lowly Nikon D40 has an SMI of 85,84 for example. So, in addition to the other things I’ve demonstrated here, the rumor that CCD was better than CMOS for color has definitely not been laid to rest, haha!
Firstly, a weak AA filter has nothing to do with color discrimination. It has everything to do with acuity and sharpness. No one claimed otherwise.
Secondly, you can disagree with my definition of microcontrast all you want, but I've provided a link to my rationale and visual examples. Until you offer proofs against that your opinion for what is correct or incorrect counts for little.
You’re telling us the M8 has a lower SNR than than the A73? LOL, tell us something that we don't already know. You are comparing an old CCD sensor against a modern CMOS sensor... of course the readout will be cleaner for a modern camera.
But you are missing the point. At base ISO the color signal vastly exceeds the noise floor for every camera sensor ever made. As such, it is CFA strength that differentiates which color signals are registered. With an abundance of light and a noise floor that is inconsequential at base ISO, camera's like the M8 and 5Dc absolutely rule for color.
It's above ISO 640 on the M8 that I's say the noise outcompetes the signal for color photography. See, I can admit CCD isn't perfect for every use case. ;-)
fPrime
--
Half of my heart is a shotgun wedding to a bride with a paper ring,
And half of my heart is the part of a man who's never truly loved anything.
Last edited:
At the bottom of this M8 review page there is this user comment:
villagranvicent: 10 years later and my M8 still delivering the goods, never give me any issues just pure photographic satisfaction and great image quality. Along with my Canon 5D 12.8 are my 2 favorite cameras.
Isn't it ironic that both the OP and I have also praised these two cameras for color? :-D
fPrime
Sorry. The lower SNR is not a factor at base ISO and the lower SMI is explained by DxOMark not testing the M8 with IR-cut filter. See my other post.
fPrime
It's just not us three on this thread that feel this way about the 5D, mate. There are thousands who still shoot the 5Dc in 2021 for its sublime colors and unique look. Even 57even admitted that the 5D was lauded for its colors. Our friend robert1955 simply isn't as gracious.
fPrime
Except that in this thread some people have presented SMI data measured by DxOMark, comparative spectrographs that match that SMI data, and camera industry references who confirm that CFA's in modern cameras were indeed weakened.
Oops, I guess not all of these people are that easy to dismiss. :-D
Too funny. What's written above is actually more representative of what pseudoscience is, LOL. It's basically all personal conjecture without data nor industry reference to confirm anything said.
Talk about the kettle calling the pot black! :-|
fPrime
--
Half of my heart is a shotgun wedding to a bride with a paper ring,
And half of my heart is the part of a man who's never truly loved anything.
Last edited:
photonut2008
Veteran Member
- Messages
- 6,577
- Solutions
- 3
- Reaction score
- 1,856
It's easy enough to find a hundred people who believe the earth is flat or any other ludicrous proposition you can think of.
And it’s also easy enough to change your handle from tonybeach to photonut2008 as well, apparently. :-D
fPrime
--
Half of my heart is a shotgun wedding to a bride with a paper ring,
And half of my heart is the part of a man who's never truly loved anything.
Last edited:
you are not making sense..its almost like you are replying to someone else ..read the last 3 posts, yours than mine than yours
your last reply has no bearing on the previous 2 posts
i think you are trying to give the impression you know more than you really do
i say the 5D has a strong CFA and AA filter and it may be a reason why it produces nice photos but at no point did i say what the 6Dmkii CFA or AA Filter maybe like
it is also said that newer CFA are weaker to increase the sensitivity of the sensor and that the RWA files are baked to compensate
now i am not going to re sarch the net looking for the articles from some very clever folk who think/know this but if you think you are as clever as you are trying to be please explain why this would be incorrect
you say What do they have to do with it?...well obviously it will change things.. colour science is very complicated as Evan a massive company like Sony with billions of R&D have yet to master it...getting better
again this reply is like it is for a different post and yes you clearly say the word invented .....so i will recap..
YOU say to ME and i quote >>>there is no need to invent a lot of pseudo scientific reasons for your preference.<<<<<
and i say and i quote >>>>i would be REALLY interested to know what is the pseudoscientific reasons i have invented<<<<<
and you reply with some random stuff..
i would like to know what is the pseudoscientific reasons i have invented, please tell me
so far you really have said nothing of substance
--
.
.
.
.
Attention Dislexsic i mean dyslexic person... This post will have many although spell checked, spelling and grammatical errs ..its The best its going get so no need to tell me it is bad I know it is .....................................................................................................
the EOS M is not dead and wont be for a long time ....as long as you don't want a flagship camera with a VF...if that's the case it died sometime ago
My 5D IS a MK1 classic
.........................................................................................................
There is no argument for FF vs APS-c (or m43) with shallow DOF..as it's a law of physics and a very subjective personal thing if you want to make use of the shallow DOF only FF can offer
.....................................................................................................
If you wait for a camera that will tick all your boxes ....by then you will have more boxes to tick..... so the wait continues .....David Appleton
Last edited:
photonut2008
Veteran Member
- Messages
- 6,577
- Solutions
- 3
- Reaction score
- 1,856
Hey, I would happily use the former, but having the latter sit dormant for several years because I wanted to go by my actual name somehow got the attention of the moderators and they took the former away and forced me to go back to using this one.
Did you have anything substantive to say here, or are you just trying to blow more smoke while selling your snake oil?
Similar threads
Keyboard shortcuts
- f
- Forum