Reading mode: Light Dark

D4 x 5D3 ISO series in ACR, IR raws (large file)

Go to last post
Gomenasai, OP but I take your processed images with a grain of salt. I'll give you an "A" for effort in trying to prove the "superiority" of the Nikon flagship.

The 5D3 hasn't been released so the RAW conversion quality may not be optimal. I will wait for "impartial" controlled tests before making any hasty conclusions.

But looking at the imaging-resource jpeg file comparison table between the D4 and the 5D3, the images are comparable.

Your posting history is biased towards Nikon although you seem to be all over the Canon board.

My DSLR is a Canon but my film SLR was a Nikon. For me, both the 5D3 & D800 are good tools. One is not better than the other in absolute terms. Some users would find the camera A better for their needs/wants, while others would prefer camera B.
 
Your observations about OP is very accurate. Besides, the firmware used is still a beta-version as is raw converter. So, it is too early to make a definitive judgment.
--
rr
 
though it should be rare everything go one direction but we have 25% of chance here.
I'd wait for DxOMarks.
 
+1

A similar post of mine, I was asking to Iliah Borg: http://forums.dpreview.com/forums/read.asp?forum=1032&message=40841938

What do you think about visual comparison of jpeg images? For example, if we shoot same scene (which includes a 18% gray card) with two cameras and set same exposure as per standard outpout sensitivity (SOS), the cameras can show photos of different brightness and tone curve for the same manufacturer indicated ISO (for example ISO 6400 selected on both 5D3 and D4). Will it be a fair comparison if we change the brightness of the both photos and match the brightness of the 18% gray cards in PP (we make the 18% gray card as 118 in sRGB)? Or should we left the images as is and compare these images of different brightness and say 5D3 gives darker image and D4 gives brighter image etc?
DSPographer said:
The exposures must be the same for a fair comparison , but the Nikon used a faster shutter speed. You also need the image brightness to match which looks close: as you note the gray patches have about the same leve_l. But the gray levels being the same doesn't mean the exposure was the same: only that the default conversion resulted in the same brightness . If the lighting levels are the same and the shutter speeds accurate then this test will handicap the Nikon camera by using a lower exposure value for it.
Click to expand...
 
first thanks for your post.

Second, I would be more interested in setting the default chroma noise reduction (25) of acr, since it does not come at the cost of loss of detail, and thats what I always use. Much more interested in lumi noise and retention of detail.

3rd, I think a more fair and even comparison would be upresing the D4 to 23 MP. the detail advantage of 5d3 is partly lost by downresing the images.

in any case, they are very close. I think when it comes to noise vs detail, the d800e will actually be the master, at least up to iso 3200 or so. this will all be in the context of upresing every image to 36mp and applying same amount of noise reduction to all of them and seeing who ends up ahead. I bet up till 3200, d800e will win.
--
http://razzi.me/kevindar/photos
 
Joofa said:
bobn2 said:
The statement I make is that read noise falls in relation to pixel area under strict scaling, which is true. As a historical trend read noise values tend to bear this out. However, there is no reason to assume that human pixel designers produce different pixel designs at all times by strict scaling, especially when they are up against process limits, as canon seems to be. Canon's practice seems to be to use the same active pixel design in a generation with different size photoreceptors, which is not strict scaling.
Click to expand...
LOL, what a baloney! :-). Translation of the above is as below:

"On paper and pencil, I (Bobn2) can come up with any scaling I desire, but in reality or actual practise DSPographer is right."
Click to expand...
How many chips have you designed?
--
Bob
 
boels069 said:
By using ExifToolGUI from:
http://u88.n24.queensu.ca/exiftool/forum/index.php/topic,2750.0.html
you will find some interesting detail in "MakerNotes".

Such as, "Measured EV", "real" shutterspeed (Canon), aperture setting and WB-RGGB values.
Click to expand...
But one cannot decide if it was 5D3's calibration or lighting issue with IR session. I'd say likely latter, plus some difference in calibration for midtone metering, but not a sample issue. Given that, mybe add about 1/3 stop in advantage to D4 on top of the visual effect.

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
david01234 said:
Gomenasai, OP but I take your processed images with a grain of salt. I'll give you an "A" for effort in trying to prove the "superiority" of the Nikon flagship.
Click to expand...
? Should I have compensated the apparent 2/3 stop handicap thet 5D3 seems to be having? Or just let it at 22MP? ;)
david01234 said:
The 5D3 hasn't been released so the RAW conversion quality may not be optimal. I will wait for "impartial" controlled tests before making any hasty conclusions.
Click to expand...
True part 1. "Impartial". Why don't you just download the RAWs and convert them yourself? You don't really need anybodys help. My job was just crop and put on the canvas for easier viewing.
david01234 said:
But looking at the imaging-resource jpeg file comparison table between the D4 and the 5D3, the images are comparable.
Click to expand...
Well, jpegs don't tell about images potential If you are a jpeg shooter, then ok, but they are vey adjustable in-camera, so the test again won't tell potential.
david01234 said:
Your posting history is biased towards Nikon although you seem to be all over the Canon board.
Click to expand...
I did this for the Nikon forum. A couple of people here had asked for such test. Why is that related to the facts?
david01234 said:
My DSLR is a Canon but my film SLR was a Nikon. For me, both the 5D3 & D800 are good tools. One is not better than the other in absolute terms. Some users would find the camera A better for their needs/wants, while others would prefer camera B.
Click to expand...
To me, in real life, I'm really more interested in base ISO DR, I rarely shoot above 1600. D800 will be my next camera.

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
bronxbombers said:
then again you get more details after downscale so you could hit the 5D3 with more NR and maintain the same detail and that would bring things closer than 2/3 stop away, maybe get it to 1/2 or even approaching just 1/3 difference???
Click to expand...
I agree. Don't forget about that when D800 is tested ;).
bronxbombers said:
hard to say

looks 2/3 worse ignorig that
Click to expand...
Same here. Some more technically inclined people are complaining about the exposure differences, but it's possible to know ehere it came from, if from difference in metering/ISO of 5D3 or from lighting used. I bet on the second option.

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
rhlpetrus said:
boels069 said:
By using ExifToolGUI from:
http://u88.n24.queensu.ca/exiftool/forum/index.php/topic,2750.0.html
you will find some interesting detail in "MakerNotes".

Such as, "Measured EV", "real" shutterspeed (Canon), aperture setting and WB-RGGB values.
Click to expand...
But one cannot decide if it was 5D3's calibration or lighting issue with IR session. I'd say likely latter, plus some difference in calibration for midtone metering, but not a sample issue. Given that, mybe add about 1/3 stop in advantage to D4 on top of the visual effect.
Click to expand...
The 5D3 "measured" a constant 9.63 EV, indicating the lighting is also constant.

Notice the difference in shutter speed (D4 shows a neat 10, 20, 40,.. series and 5D3 6, 13, 25, 51, 102, ..) and D4 Fnumber f/11 @51200 and f/16 @102400, 5D3 f/11 @102400.

The latter could influence the visual sharpness (DoF) at focus distance around 1.5m (5 ft).
IMHO a flat subject should be used.
 
bobn2 said:
rhlpetrus said:
Dave Largent said:
rhlpetrus said:
ACR default, chroma NR=0
Click to expand...
Just what I've been waiting for!

Does luma NR=0, as well?
Click to expand...
Yes, that's ACR's default.
Click to expand...
But NR=0 does not necessarily mean zero NR, nor does it necessarily mean the same thing for different cameras.
--
Bob
Click to expand...
True, but I don't have further access to controlling all the variables. For the enduser, knowing how ACR renders images is already relevant info. Not so for the ones digging further into the sensor tech aspects ;).
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
boels069 said:
The 5D3 "measured" a constant 9.63 EV, indicating the lighting is also constant.
Click to expand...
For that session, but that's different from the D4's session.
boels069 said:
Notice the difference in shutter speed (D4 shows a neat 10, 20, 40,.. series and 5D3 6, 13, 25, 51, 102, ..) and D4 Fnumber f/11 @51200 and f/16 @102400, 5D3 f/11 @102400.

The latter could influence the visual sharpness (DoF) at focus distance around 1.5m (5 ft).
Click to expand...
These tests show problems for detail analysis, I didn't intend the crops to be evidence of detail rendering, just of visual noise with ACR.

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
rhlpetrus said:
Only if Canon has crippled the 5D3's sensor/toppings/ADC somehow. Otherwise, at same printing size, 5D3 should be better, since it'd be same tech with more pixels.
Click to expand...
I'v never believed this line of reasoning personally, at least the idea that more pixels downrezed will automatically equal the ISO performance of fewer larger pixels.

The NEX 7 recently seemed like a pretty clear mark agenst this whole idea, in the tests I saw dowrezing to 16 MP to equal the NEX 5 it was pretty clear that the newer sensor offered inferior high ISO performance.

When you have a multi billion dollar industry saying one thing and some net geeks saying another I'm generally inclined to go with the former.
 
Renato - useful info....

sure looks like the Canon will be a great general purpose photographic machine: plenty of resolution, great low light ability, fully prof. AF (likely), decent frames per second, and priced well below "the bricks" ....

just speaking generally, it looks like we are arriving at a basic limit for additional gains in effective light gathering using the current technology..... ?
 
You must log in or register to reply here.

Similar threads

R
Z5II review
Replies
0
Views
319
rhlpetrus
R
R
Tamron 16-30 f/2.8
Replies
4
Views
1K
CNorre
CNorre
R
Which Auto WB mode for portraits on Zf?
Replies
7
Views
6K
rhlpetrus
R
R
HLs headroom on the ZF
Replies
0
Views
4K
rhlpetrus
R
H
Panasonic did something special with the Z6 III sensor
4 5 6
Replies
118
Views
35K
Leonard Shepherd
L

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top