Preorder: Canon EOS RP camera and kits

Quarkcharmed

EOS 5DMkIV
Feb 14, 2018
291
134
Australia
www.michaelborisenko.com
The frequency with which ~1 extra stop of sensor DR makes a meaningful difference in output is very low.

The frequency with which the molehill of that 1-stop difference becomes a mountain on the Internet is very high.
The difference between 5DIV and PR is about 2 stops, if I get it right. The frequency with which it'll make a meaningful difference in output hugely depends on your individual typical shooting scenarios. I'm scratching the limits of my 5DIV quite often. That's me and i don't claim everyone needs a huge DR. But claiming that low DR is fine and even makes you a better photographer also sounds strange to me.
 

Quarkcharmed

EOS 5DMkIV
Feb 14, 2018
291
134
Australia
www.michaelborisenko.com
I provided both edited with lifted shadows and unedited. With edited, I can push the shadows further without getting noise - using the original raw CR2, of course.

You provided just one image and it's a jpeg. 100% cropped fragment doesn't make it a new image. Lifting shadows in it causes an awful result, but you have, I assume, the original raw file and you can lift the shadows on your side and show us the result.

If you try to push the image you provided, you will see that the details are "lost".


I don't understand what you mean by "recoverable" in this case. If you don't like where the shadows currently are, you can play with the attached 100% crop image.
 

Jack Douglas

CR for the Humour
Apr 10, 2013
5,883
499
Alberta, Canada
I am not a pro and I am not presently engaged in this degree of shadow lifting etc. I do however have a sense of what's appealing and I can say that for my taste I get this impression. I know what looks natural when I walk around in nature and I see deep shadows with no detail and likewise highlights that my eye can't discern information within. When a photo has this detail in it, my brain goes into, "there is something wrong here, this is not natural", mode. In other words I don't find it appealing.

Now maybe the issue is seeing HDR that has not been well done. One scene that an unnamed person was praising had so much detail it was even confusing why the sun was illuminating areas that it would never reach. I could say the similar thing about portraits with all the blemishes removed. I guess it depends on what final outcome a person wants or thinks others want.

I suspect that my interests being mainly in wildlife places me in a different category than those that want the maximum DR. For me, I need/want high ISO IQ. Just my thoughts.

Jack
 

Kit.

EOS 7D MK II
Apr 25, 2011
758
311
Is it with the lifted shadows on a raw file? Not much of a lift and quite noisy, to be honest.
What did you expect me to do, turn a dark wet stone into a white one? It is as noisy as a 100% crop is... surprisingly, the "lifted shadows" area in your picture doesn't show more details than this one (it wasn't a 100% crop, wasn't it?) and still suffers from color noise.
 

neuroanatomist

I post too Much on Here!!
Jul 21, 2010
23,865
1,032
The difference between 5DIV and PR is about 2 stops, if I get it right. The frequency with which it'll make a meaningful difference in output hugely depends on your individual typical shooting scenarios. I'm scratchin the limits of my 5DIV quite often. That's me and i don't claim everyone needs a huge DR. But claiming that low DR is fine and even makes you a better photographer also sounds strange to me.
It’s ~1.6 stops per DxO. I think you’re missing the point, although the way you phrase it is a commonly employed strawman. It’s not that ‘low DR is fine’ - more DR is always better. But it’s rare that scenes have a DR that exceeds 11 stops but does not exceed 12.6 stops. Not impossible, but rare. More commonly, scenes have DR that either fits comfortably below either sensors’ capability or exceeds both, and in the latter case something additional is required if you want a high-quality image representing the scene’s DR (merging multiple exposures by various methods, GNDs, etc.).
 

jjesp

I'm New Here
Dec 30, 2016
19
9
Really hope that Canon makes it 24mm for RF.

I can live with f/2.8, but not without 24 (or shorter).
I don't think it is possible to make a pancake lens in 24mm full frame. As I heard it can only be made around 35-40mm view.
 

Kit.

EOS 7D MK II
Apr 25, 2011
758
311
I don't think it is possible to make a pancake lens in 24mm full frame. As I heard it can only be made around 35-40mm view.
There is a "body cap" Perar 24 lens for Leica M (27.8 mm flange distance), but it is f/4. RF is 20mm flange distance, so maybe... although it won't be easy, because of the sensor microlenses.
 

Quarkcharmed

EOS 5DMkIV
Feb 14, 2018
291
134
Australia
www.michaelborisenko.com
What did you expect me to do, turn a dark wet stone into a white one? It is as noisy as a 100% crop is... surprisingly, the "lifted shadows" area in your picture doesn't show more details than this one (it wasn't a 100% crop, wasn't it?) and still suffers from color noise.
That's absolutely not true, if I lift shadows more on my original raw file, there's virtually no noise (check below). It's simply a different league from your 100% crop https://www.canonrumors.com/forum/index.php?threads/preorder-canon-eos-rp-camera-and-kits.36718/page-6#post-765521
Your 100% crop with the recovered shadows is simply not usable.
 

Attachments

Kit.

EOS 7D MK II
Apr 25, 2011
758
311
That's absolutely not true, if I lift shadows more on my original raw file, there's virtually no noise (check below).
But I see noise there, and I absolutely don't like it.

The same with mushy underexposed overpushed Sony sensor images that people use to "prove" Sony's "superior" dynamic range. They may have less color noise, but they still look bad.

Your 100% crop with the recovered shadows is simply not usable.
Both are usable if they are the shadows. Both are unusable if they are the main subject. Both after "recovering shadows" show clear signs of underexposure.
 

Quarkcharmed

EOS 5DMkIV
Feb 14, 2018
291
134
Australia
www.michaelborisenko.com
It’s ~1.6 stops per DxO.
According to this measurement, it's almost 2 stops, although apparently they use different methods:
http://www.photonstophotos.net/Charts/PDR.htm#Canon EOS 5D Mark II,Canon EOS 5D Mark IV,Canon EOS 6D Mark II

Note 5DII and 6DII perform almost the same and we've seen the noisy shadows from 5DII in the 100% crops above.

I think you’re missing the point, although the way you phrase it is a commonly employed strawman. It’s not that ‘low DR is fine’ - more DR is always better. But it’s rare that scenes have a DR that exceeds 11 stops but does not exceed 12.6 stops. Not impossible, but rare.
Rare for whom and for which shooting scenarios? With controlled strobe/flash light, it's almost never. For concerts/sports, it becomes more significant. For nature/landscapes, I'm pushing the limits of my 5DIV very often. I do use grad ND sometimes and still feel I'd like to have a bit more of DR. With 6DII sensor, I'd have used blending much more often, and many shots would have not been possible.

I've already given this shot https://500px.com/photo/290893861/bombo-quarry-splash-2-by-michael-borisenko as an example where HDR is impossible and successful blending is highly improbable, and the shot uses all available DR.
 

Quarkcharmed

EOS 5DMkIV
Feb 14, 2018
291
134
Australia
www.michaelborisenko.com
But I see noise there, and I absolutely don't like it.
Sorry this conversation gets ridiculous. There's no noise, full stop. Everyone can see it form the 100% crop I provided. There's grainy rock texture, same as on the highlighted parts, but it's not the noise. There's darker smooth wet rock texture at the bottom and you can clearly see there's no noise on it. Also noise reduction slider was at 0.
When I push exposure further by ~2 stops, only then the shadow starts showing some subtle noise.
 

neuroanatomist

I post too Much on Here!!
Jul 21, 2010
23,865
1,032
According to this measurement, it's almost 2 stops, although apparently they use different methods:
http://www.photonstophotos.net/Charts/PDR.htm#Canon EOS 5D Mark II,Canon EOS 5D Mark IV,Canon EOS 6D Mark II
From Bill Claff’s data that you link, it’s 1.7-stops at ISO 100, and slightly less than 1.6-stops at the maximum DR of which the 5DIV and 6DII are capable (which isn’t ISO 100 for the 6DII). ‘Eyeballing’ photographic composition works well, but that’s not the way to read and interpret quantitative data.

Sorry this conversation gets ridiculous.
Yes, even if your statement of ‘almost 2 stops’ was correct, please realize that you’re quibbling over less than half a stop. That absolutely qualifies as getting ridiculous.
 
Reactions: bluediablo

M_S

EOS 80D
Jul 31, 2013
146
5
He didn't claim to.
He answered: "...nor probably for most potential buyers of the EOS RP." He used that sentence to support his argument about the importance of DR. In doing so he suggests that this statement is true, or at least for the most part, otherwise that part would have been useless to write. He suggests that he has the knowledge or a good insight that the other person doesn't have. It's short before writing "shut up". Hence my rant about this, as this can produce negative vibes. Would he have written "since I have the feeling that" or "looking at the sales numbers of the R or RP in relation to..." the argument would have been more self-reflective and substantial. Anyhow, I am done with this conversation, I have better things to do.
 
Last edited:

neuroanatomist

I post too Much on Here!!
Jul 21, 2010
23,865
1,032
He answered: "...nor probably for most potential buyers of the EOS RP." He used that sentence to support his argument about the importance of DR. In doing so he suggests that this statement is true, or at least for the most part, otherwise that part would have been useless to write. He suggests that he has the knowledge or a good insight that the other person doesn't have. It's short before writing "shut up". Hence my rant about this, as this can produce negative vibes. Would he have written "since I have the feeling that" or "looking at the sales numbers of the R or RP in relation to..." the argument would have been more self-reflective and substantial. Anyhow, I am done with this conversation, I have better things to do.
If you cannot comprehend facts and make a logical deduction based upon those facts, then yes...I am ‘suggesting that I have knowledge or a good insight that you don’t have’. Canon’s competitors have delivered better low ISO DR for a decade. Canon has not lost market share to those competitors. Canon is also the top seller of FF ILCs. Therefore, low ISO DR is not a critical factor in the purchasing decisions for the majority of ILC buyers. Facts and a logical deduction. Simple.

Enjoy doing those ‘better things’, hopefully they don’t require logic or deductive reasoning.
 

Michael Clark

Now we see through a glass, darkly...
Apr 5, 2016
450
158
ok... so I'm at a crowded bar taking pictures in low light of my band friends..... I'm too lazy to set up reflectors and its ok to annoy everyone with flash? interesting.
Key word in my previous comment: Sometimes.

But as one who does a lot of band photos in low light bars, sometimes you have more control than you think. Asking whoever handles the lights for the bar to turn up the green LEDs (when they use those nasty magenta/green/blue LED spots on stage) if only the magenta and blue LEDs are on, for instance, can make a huge difference in your results.

It's the difference between this:

201509205900 - Copy2.JPG

(5D Mark III + EF 50mm f/1.4. ISO 5000, f/2.2, 1/50)

(Which looked like this with auto WB before a lot of work with CT/WB adjustment as well as with the HSL tool)
201509205900 - Copy.JPG

And this (from the same stage):

201802107002LR.JPG

(5D Mark III + EF 135mm f/2. ISO 3200, f/2.2, 1/100)


So can moving around a bit to use the light you do have to an advantage. Sometimes you do have to resort to post-processing wizardry. But even then, my (according to some) "pitiful" and "outdated" Canon 5D Mark II and 5D Mark III sensors with "horrible" low ISO DR do okay.

This place doesn't even have a real "stage", much less any stage lighting, unless you count those strip lights.

201707213006LR.JPG

(5D Mark II + EF 50mm f/1.4. ISO 6400, f/2.2, 1/80. Lots of color correction.)

But there are some spots where you can get decent light angles from the TVs high on the wall above the mirrors behind the bar.

201707213049LR.JPG

(5D Mark III + EF 135mm f/2. ISO 6400, f/2.2, 1/100. Pushed 1/3 stop in post.)

Another band in the same place after the strip lights had been removed:

201807210001LR.JPG

(5D Mark II + EF 50mm f/1.4. ISO 5000, f/2.2, 1/125 and pushed 2/3 stop in post. Contrast and highlights reduced a bit and shadows pushed a bit more, and a lot of color correction.)

Sometimes instead of getting the shot you want (shadow on the face the entire night):

201811220023LR.JPG

(5D Mark III + EF 135mm f/2. ISO 5000, f/2.2, 1/100. Contrast reduced, highlights pulled a bit and shadows pushed a bit more.)

You have to take the shots the light will give you:

201811220010LR.JPG

(5D Mark III, EF 135mm f/2. ISO 5000, f/2.2, 1/100, Contrast reduced, highlights pulled a bit and shadows pushed a bit more.)

201811220004LR.JPG

(5D Mark III, EF 135mm f/2. ISO 5000, f/2.2, 1/80. Pushed 1/2 stop and a green filter applied.)
 
Last edited: