p.8 #1 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
alundeb wrote:
Unfortunately we cannot judge anything about that with these images either. I see only a huge difference in the subject and the light, in addition to the jpg processing, which alone makes it impossible.
Anders, thanks for having a look and taking one for the forum.
p.8 #2 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
artd wrote:
I'm not sure about Canon files being "more raw" than other files. But even if they were, that doesn't mean necessarily mean they are more malleable.
I probably should have said they have a "different" malleability.
Here's a couple RAW's from IR ... what can we glean from the diff's here.
p.8 #4 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
I wasn't intending for it to be a guessing game, but rather how can we read this to tell us something about DR or what we can expect from working these @ their different starting point ... or is it not useful at all.
But, as long as you started the game ...
Since you mentioned Canon & Nikon specifically ... I can neither confirm, nor deny either of those. But, I will say that it is within the possible set of those two brands @ Left vs. Right =
1) Nikon vs. Nikon
2) Canon vs. Canon
3) Nikon vs. Canon
4) Canon vs. Nikon
And will add that they are both contemporary sensors, even if not the absolutely newest.
p.8 #5 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Well, those have to be images converted from raw, not the raw files themselves. They're pretty much useless for any base of comparison. Dynamic Range? No. The very limited dynamic range of the old style ColorChecker would not be a challenge for any modern digital camera. The differences in color can easily be from different input profiles used in the raw converter, and are meaningless here. There's no way for a viewer to know which one is more accurate, although human nature will lure people toward the more saturated version by default.
p.8 #6 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Peter Figen wrote:
Well, those have to be images converted from raw, not the raw files themselves.
I just downloaded the native files (.cr2 or .nef as the case may have been) ... ACR was all parameter's @ zero (rat's, I forgot to zero the sharpening defaults) going into PS for the crops.
Peter Figen wrote:
There's no way for a viewer to know which one is more accurate,
p.8 #7 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Ah, I was running a colorchecker over the two images and the one to the right has a slightly 'greener' (CIE L*a*b a channel negative value) tinge and higher luminosity readings in neutrals (bottom row) than the image to the left. So I'm still going with Canon L and Nikon R, but wouldn't be surprised if both are Nikon as the neutrals lean towards the green tinge and that's where a lot of luminosity data is carried in the RGB colorspace.
Just an observation - There is colour fringing in both images and if this was the from the centre circle of the lens I'd be disappointed, as I would be if this was from the corners on lens stopped well down. Thinking aloud the drop in luminosity of the left image could be down to vignetting.
p.8 #8 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
RustyBug I'm not sure what kind of value you'd be looking to derive. Different sensors produce different results but also a raw converter will interpret the data differently...this can happen even in the case of the same sensor being used in different cameras. Im not an expert on the topic, but i believe this is because indivdual camera ISO values are more or less arbitrarily assigned by the manufacturer and are not fixed values based solely on sensor characteristics. Based on where a particular camera has its ISO point set, the raw converter can apply different tonal curves to make sure the exposure still comes out right.
p.8 #9 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Well that again raises the question of what is/can to be gleaned from these.
I mean if the response is going to be ... you can't tell anything from them or you can't go by it because there are more variables than the sensor ... then why do folks go to such painstaking efforts to produce them?
Again, what CAN we glean from them?
And if the answer is truly NOTHING ... then that would seem that there is nothing to discuss @ sensors, if we're going to say that the RAW output is not a valid comparison.
The alternate question would then be WHAT IS a valid comparison from which to glean or derive something for comparison of different camera / sensor?
The one thing that does stand out to me about them is that there are about four or five diffferent swatches where the image on the right has a more "pure" RGB value with a "0" value while it's counterpart on the other image has more of a three component to the color than a two component one. Not sure what that means @ better or worse or different or
p.8 #10 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Well, I suppose it depends on the context of the conversation. Are we still talking about comparing "cooking" of raw files? Are we talking about which camera has the best colors when run through a raw converter at default settings?
There is probably something to be gleaned regarding color reproduction. But that would also need to take into account the color filter array in the camera, not just the sensor and not just the raw files. But again, my interest would be less about how do camera A and B look different at default settings in the raw converter; rather I'd be more interested in knowing if I can make the raw files from camera A look like camera B if I want to, or vice versa.
p.8 #12 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Well said.
RustyBug wrote:
Well that again raises the question of what is/can to be gleaned from these.
I mean if the response is going to be ... you can't tell anything from them or you can't go by it because there are more variables than the sensor ... then why do folks go to such painstaking efforts to produce them?
Again, what CAN we glean from them?
And if the answer is truly NOTHING ... then that would seem that there is nothing to discuss @ sensors, if we're going to say that the RAW output is not a valid comparison.
The alternate question would then be WHAT IS a valid comparison from which to glean or derive something for comparison of different camera / sensor?...Show more →
p.8 #13 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
RustyBug wrote:
Well, that was my objective for the comps, but apparently what I thought would extend that dialogue seems deemed as an irrelevant and futile effort.
Well, the dialogue does seem to have been extended from my perspective at least. But the question of whether or not the raw files are more or less "cooked" I don't know how that gets answered from the images. So my proposition is if we have a question that can't be answered, can we ask a question that can be answered. In this case, what kind if latitude do we have with the raw files in the converter? If we have sufficient latitude to make either image look like the other to any degree we wish, then the question of if the files come out of the camera cooked becomes academic. If we run into limitations, then maybe it's cooking, or maybe it's the sensor itself, or the CFA, or whatever, but at least we know what the limitation is from a system standpoint.
p.8 #14 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Jon Tainton wrote:
Sigh. You couldn't provide links to any of your positive FM posts about Canon then? Say 5 in your last 1,000
Surprise I did not catch this earlier. Sorry Jon, I'm not here to pat you or your gang or canon in the back with my posts. Like any other canon shooter, most of us here is sharing our experiences with canon consumer, are you not your response is nothing to do with gear but sound like a troll thread. Stick with the facts, would you. TIA
p.8 #16 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
mttran wrote:
Surprise I did not catch this earlier. Sorry Jon, I'm not here to pat you or your gang or canon in the back with my posts. Like any other canon shooter, most of us here is sharing our experiences with canon consumer, are you not your response is nothing to do with gear but sound like a troll thread. Stick with the facts, would you. TIA
Deeper sigh, I was hoping to establish some facts on your positive posts about Canon, from a Sony MILC user perspective. Since this is a gear forum, everyone is freely entitled to raise positive/negative experiences, there's no doubting I've posted a few gripes at Canon in my posting history, but I think, I've made the point only once and then moved on.
I'll make that point again, in case you miss it, I haven't kept banging the drum through thread after thread, at every opportunity, my gripes about Canon, as in my eyes, it would serve to disrupt the thread and constitute trolling behaviour. Your posting history has compelling evidence to suggest you struggle with this concept.
Oh, btw, If my responses to your negative posts seem like trolling behaviour, please feel free to use the hide button. TIA
p.8 #17 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
mttran wrote:
Surprise I did not catch this earlier. Sorry Jon, I'm not here to pat you or your gang or canon in the back with my posts. Like any other canon shooter, most of us here is sharing our experiences with canon consumer, are you not your response is nothing to do with gear but sound like a troll thread. Stick with the facts, would you. TIA
Michael, I appreciate your contributions and references where due. However there are two points that I want to give you clear response to as a user of both Canon and Sony cameras.
The first is that transitions of tones from highlights to shadows is not something that varies between sensors. All sensors in the context here have a response very close to linear. What we see in images is a result of the subject, light and tone curve. Period. In the knowledge of this, when you come with sample images from Canon that look "worse" than sample images from Nikon, it comes across as "Hey Canon folks, don't you think this Canon image looks bad?". Knowing that what we see does not in any way represent the characteristics of the sensor, we wonder what the motivation is.
The second is the repeated posting of graphs showing dynamic range. The graphs only have a loose connection to the subject of the thread. They don't serev any other purpose than repeating that Canon sensors are not market leading in low ISO dynamic range. I think we all know this, and especially the Canon engineers know this.
p.8 #19 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
+1 @ the thread topic ...
Comparative to CANON improvements beyond the former CANON offering.
Not a Canon vs. the world or Canon not the highest absolute DR. Relative to within Canon.
That's not to say that the Canon forum doesn't have a place for comparative to Sony, etc. ... but this thread was specific @ relational to Canon vs. Canon. (Apologies if I went OT a bit myself).
p.8 #20 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Sometimes the discussions of "tests" of image go in very strange circles. To wit (with creative license):
"I think my camera produces better [thing about image quality] than your camera."
"My camera does just fine. Look at the results I get from it."
"But still, there is something off about hour camera. It should be better. Like mine."
"My photographs look great from my camera. Whatever it is you are seeing can't be important."
"OK. I'll do a test. To ensure the validity of the test, I will eliminate all variables but the one I'm testing, the performance of [thing x] on the two cameras. See [thing x] is [.24 units of something] better on my camera than on yours."
"But I never make photographs that way! And no one goes looking for [thing x] by staring at two samples magnified to 100% on the screen side by side for an hour."
"Yes, but there might be some situation some time in which it might be visible if we look just the right way and know what to look for."
"OK, but that's only in controlled tests. In the real world you couldn't tell photos apart reliable based on that."
"Are you saying that making cameras better doesn't matter? Why don't you shoot with a 6MP Rebel, or a Brownie."
"I'm not saying that at all, just that measurements are different than real photography and that is where they should be judged."
But the question of whether or not the raw files are more or less "cooked" I don't know how that gets answered from the images. So my proposition is if we have a question that can't be answered, can we ask a question that can be answered. In this case, what kind if latitude do we have with the raw files in the converter? If we have sufficient latitude to make either image look like the other to any degree we wish, then the question of if the files come out of the camera cooked becomes academic. If we run into limitations, then maybe it's cooking, or maybe it's the sensor itself, or the CFA, or whatever, but at least we know what the limitation is from a system standpoint.