High ISO still life 5D3 vs. 1DX

Home · Register · Search · Software · Hosting · Software · Join Upload & Sell

FM Forum Rules
Canon SLRs, primes, and zooms lenses reviews

FM Forums \| Canon-mount SLRs \| Join Upload & Sell
1 2 3 4 end

rabbitmountain
Online
• • •
Upload & Sell: Off

p.3 #1 · High ISO still life 5D3 vs. 1DX

@Stan
I like taking a look in the future cameras, because the 1Dx is a future camera for me (say I pick up a used one when the 1DxmkII comes out

) so I follow these topics with interest. Even though the 1Dx is significantly better in the ways you describe, I'm still eagery awaiting my 5DmkIII.
Congrats on some fine gear and we will hopefully see lots of work from you with them. The Thin-DoF-Girl-With-Spider is still my all-time favourite and it inspired me to get a 35L. So thank you for that!

Jul 09, 2012 at 06:40 AM

rabbitmountain
Online
• • •
Upload & Sell: Off

p.3 #2 · High ISO still life 5D3 vs. 1DX

stanj wrote:
rabbitmountain wrote:is it correct that the 1Dx is less silent than the 5D3, single frame mode?
By far. The 5D3 is much much quieter, both single frame and sequence (something the 1DX can't do at all).
OK thanks. Seems like that can do without that poor and lousy 1Dx after all...it's a piece of junk...can't even take photographs silently

Stay good.
Ralph.

Jul 09, 2012 at 06:42 AM

zogster
Offline
• •
Upload & Sell: Off

p.3 #3 · High ISO still life 5D3 vs. 1DX

Stan, thanks so much for taking the time to post this information and thanks to those who participated in the cordial and informative discussion thus far. This is exactly the reason why I come to FM!

Jul 09, 2012 at 03:39 PM

Beni
Offline
• • • • •
Upload & Sell: On

p.3 #4 · High ISO still life 5D3 vs. 1DX

garyvot wrote:
snapsy wrote:
FYI, Adobe ACR appears to be adjusting the exposure of 5D3 images up by about 1/3 stop when I compare the 1DX vs 5DM3 raws, when both are using the Adobe Standard profile. I've seen this same behavior on the D3s vs the D800. So to produce comparable images the raws need to be processed using ACR's Camera Neutral profile (which mimics Canon's Neutral DPP profile). Without this normalization the 1DX will have an unfair advantage over the 5D3 when comparing the images side-by-side.

It's quite interesting how different ACR is rendering the files from the two cameras when using certain Adobe cloned Canon profiles. 'Camera Standard' for instance causes the 5D3 file to be more saturated and applies a contrastier tone curve.

I took a look at all the images in DPP and the color, contrast, and exposure rendering was near identical, as you would expect.

However, I *stongly* dislike Canon's new default sharpening approach, which creates visible halos and edge artifacts and generally makes the images look really artificial.

So if using DPP to evaluate images, you may want to switch everything back to the legacy sharpening algorithm. (Unfortunately, there is no way to make this an application default.)

That must have been what I was noticing re tonality, wish I had the curve expertise to neutralise the difference.

Jul 09, 2012 at 03:46 PM

ben egbert
Offline
• • • •
Upload & Sell: On

p.3 #5 · High ISO still life 5D3 vs. 1DX

To me, the 1DX at 1600 is the top iso limit based on these samples. The rest are too noisy. The 5D3 at 1600 would be ok with more NR. All of them would be fine for historically valuable shots but not for artistic shots.

As you may conclude, I don't like noise so your tolerance for it may be different.

Jul 09, 2012 at 03:55 PM

snapsy
Offline
• • • •
Upload & Sell: On

p.3 #6 · High ISO still life 5D3 vs. 1DX

Beni wrote:
That must have been what I was noticing re tonality, wish I had the curve expertise to neutralise the difference.

You can create your own Adobe profiles with custom curves via the DNG profile editor and apply that profile to either specific images in ACR/LR or make that the default profile so that it's applied to all future images. The easiest way to get identical curves is to specify a linear curve in the profile editor.

Here's the process you can use to create a linear curve version of the Adobe Standard profile for a given camera:

Download and install the Adobe DNG profile editor and DNG convertor
Convert any raw file from the camera in question to DNG via the Adobe DNG convertor.
Launch the DNG profile editor application and load the converted DNG file
In the "Color Tables" tab, select "Adobe Standard (your camera model)" as the base profile
Go to the "Tone Curve" tab and select "Linear" for "Base Tone Curve"
Export the profile. This will put the .dcp profile file in the CameraProfiles directory. On my Windows 7 64-bit system the directory is C:\Users\snapsy\AppData\Roaming\Adobe\CameraRaw\CameraProfiles
You can now apply this profile to a given raw image for that camera within ACR/LR. Be sure to restart the apps if they are already running so that they will detect the new profile.

I went ahead and did this for both the 1DX and 5DM3. You can download the profiles here:
https://www.transferbigfiles.com/f676915d-4a5d-4ef6-8662-b8708591a9b7?rid=ncYWBdBcH2k0RcDqrYiKsA2

The above link is only good for 5 days. Drop those folders in your CameraProfiles directory.

Jul 09, 2012 at 04:55 PM

Beni
Offline
• • • • •
Upload & Sell: On

p.3 #7 · High ISO still life 5D3 vs. 1DX

Snapsy, really appreciate it!

Going to have a play and I'll report back. I don't own either the 1Dx or 5D3 but would love to have a 1Ds3 look to my 5D files, been chasing that elusive goal for years now.

Jul 09, 2012 at 05:15 PM

rabbitmountain
Online
• • •
Upload & Sell: Off

p.3 #8 · High ISO still life 5D3 vs. 1DX

ben egbert wrote:
To me, the 1DX at 1600 is the top iso limit based on these samples. The rest are too noisy. The 5D3 at 1600 would be ok with more NR. All of them would be fine for historically valuable shots but not for artistic shots.

As you may conclude, I don't like noise so your tolerance for it may be different.

You must be joking. Do you apply proper Noise Reduction, like Lightroom 4.1? Like I said, when viewing the RAWs in Lightroom on a 30 inch monitor full screen, I start noticing the noise for the first time when the 12800 ISO file comes along. If you are seeing what I'm seeing and you think ISO 1600 is the limit, then yes, you are sensitive to noise. And if so, there's nothing wrong with that, let me say that straight out.

Jul 09, 2012 at 05:22 PM

Beni
Offline
• • • • •
Upload & Sell: On

p.3 #9 · High ISO still life 5D3 vs. 1DX

Well had a play around, I see what you've done. Problem with the linear is that you don't have a decent working baseline for the images. They don't look the same even with the linear curve anyway.

I wish there was a way to equalise the look between two cameras. There is a way to apply the profile from one camera to another using the DNG converter, I just did it, didn't help.

Have to be honest, I'm thinking it's more the file than the raw converter at this point. I'll have a play in C1 and see if I am still seeing the same difference. EDIT I would but C1 doesn't support the 1Dx yet

Jul 09, 2012 at 05:24 PM

snapsy
Offline
• • • •
Upload & Sell: On

p.3 #10 · High ISO still life 5D3 vs. 1DX

Beni wrote:
Well had a play around, I see what you've done. Problem with the linear is that you don't have a decent working baseline for the images. They don't look the same even with the linear curve anyway.

I wish there was a way to equalise the look between two cameras. There is a way to apply the profile from one camera to another using the DNG converter, I just did it, didn't help.

Have to be honest, I'm thinking it's more the file than the raw converter at this point. I'll have a play in C1 and see if I am still seeing the same difference.

Matching images of different cameras to the precision necessary to compare High ISO performance is rather difficult. One complication is that the nominal ISO calibration between camera models is different, sometimes up to 1/3 stop. That's why I usually just wait for the DxoMark numbers, since their graphs account for the nominal ISO differences and their custom raw methodology avoids the usual profile difference issues.

Jul 09, 2012 at 05:30 PM

Beni
Offline
• • • • •
Upload & Sell: On

p.3 #11 · High ISO still life 5D3 vs. 1DX

Not comparing ISO snapsy, just tonality/look/colour. Thanks.

Jul 09, 2012 at 05:35 PM

R. Eisenberg
Offline
• • •
Upload & Sell: Off

p.3 #12 · High ISO still life 5D3 vs. 1DX

Thanks once again Stan. You observe above that the metering on the 1Dx is better. Would you care to elaborate?

Jul 09, 2012 at 05:47 PM

Scott Stoness
Offline
• • • • •
Upload & Sell: On

p.3 #13 · High ISO still life 5D3 vs. 1DX

mohamed alfari wrote:
Thanks Stan and Benny for your contribution. The difference is clear but not by much, however at 25k it might be worth it seeing a full image?

Ralph, until Stan answer your question I will give a real life example that I wished for 1DX. I was on boat to feed eagles some fish and let me tell you that despite 5D3 have good buffer it really fills fast with a lot of action everywhere around the boat. which wasn't the case with 1DIV. Some where my mistake for not managing the buffer limit, but there were a lot of birds and action that I missed the peak action. after that I was planning to get 1DX but felt that these situation isn't my everyday photography like that sports or birds photographer. Still I dream of getting one but with that price its not an easy decision to make.

Mine was limiting on burst too, on 5diii, until I bought the lexar 1000 32gb and now it has a dramatic improvement.

25 vs <19 and fast recovery and when you hit the limit it slows to about 2 per sec, whereas before it just choked.

Try an new card.

You might not be comparing apples to apples.

Scott

Jul 09, 2012 at 06:16 PM

stanj
Online
• • • • •
Upload & Sell: On

p.3 #14 · High ISO still life 5D3 vs. 1DX

R. Eisenberg wrote:
Thanks once again Stan. You observe above that the metering on the 1Dx is better. Would you care to elaborate?

Well, I was shooting extremely high and some low contrast scenes all day, and not one single image was what I would consider poorly exposed. No other camera that I have used before had such a track record. It underexposed a picture of a white plate with a small piece of steak on it, but only by a little bit. More importantly it didn't overexpose a single "fun in the pool" picture.

Jul 09, 2012 at 07:09 PM

Beni
Offline
• • • • •
Upload & Sell: On

p.3 #15 · High ISO still life 5D3 vs. 1DX

and the 5D3 isn't the same?

Jul 09, 2012 at 07:13 PM

stanj
Online
• • • • •
Upload & Sell: On

p.3 #16 · High ISO still life 5D3 vs. 1DX

Beni wrote:
and the 5D3 isn't the same?

No. The light meter between the two cameras is very very different, and it shows.

Jul 09, 2012 at 07:32 PM

R. Eisenberg
Offline
• • •
Upload & Sell: Off

p.3 #17 · High ISO still life 5D3 vs. 1DX

stanj wrote:
R. Eisenberg wrote:
Thanks once again Stan. You observe above that the metering on the 1Dx is better. Would you care to elaborate?

Well, I was shooting extremely high and some low contrast scenes all day, and not one single image was what I would consider poorly exposed. No other camera that I have used before had such a track record. It underexposed a picture of a white plate with a small piece of steak on it, but only by a little bit. More importantly it didn't overexpose a single "fun in the pool" picture.

When I briefly used the camera in December at a CPS sponsored event, I was impressed at both the light metering and the white balance capabilities of the camera under trying artificial illumination. But I only had limited access to the camera and no means to examine its files more closely. They did weigh the 1Dx for me, since at that time its specifications with battery were unavailable, and I recall that its weight was identical to the 1Ds2. At the time, there was no 5D3 at hand for comparison. Hence your observations are 'priceless', but they may also prove to be pricey!

Jul 09, 2012 at 08:11 PM

garyvot
Offline
• • • •
Upload & Sell: On

p.3 #18 · High ISO still life 5D3 vs. 1DX

snapsy wrote:
So to produce comparable images the raws need to be processed using ACR's Camera Neutral profile (which mimics Canon's Neutral DPP profile). Without this normalization the 1DX will have an unfair advantage over the 5D3 when comparing the images side-by-side.

Beni wrote:
That must have been what I was noticing re tonality, wish I had the curve expertise to neutralise the difference.

Beni, I think you can just do what Snapsy suggests above. The 'Camera Neutral' profile does seem to normalize the tone curve differences (though you may find the color rendering less pleassing according to your taste). Its fine though when comparing noise levels, as here.

Jul 09, 2012 at 08:12 PM

garyvot
Offline
• • • •
Upload & Sell: On

p.3 #19 · High ISO still life 5D3 vs. 1DX

(Apologies to our Russian friends!)

Jul 09, 2012 at 08:17 PM

rscheffler
Offline
• • • •
Upload & Sell: On

p.3 #20 · High ISO still life 5D3 vs. 1DX

Stan, thanks for doing the comparison.

A bit OT from ISOs but I hope you don't mind. I'm curious about the 1DX's buffer, especially compared to the 1DIV. Shooting RAW, how many frames at a moderate ISO and full fps can you get with your fastest card before hitting the buffer wall? Once it hits the wall and you continue shooting, at what fps does it continue? Lastly, how long does it take for the buffer to clear if you hit the buffer wall and stop shooting at that point?

Jul 09, 2012 at 08:23 PM

1 2 3 4 end
FM Forums \| Canon-mount SLRs \| Join Upload & Sell 1 2 3 4 end