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

Moderated by: Fred Miranda
Username   Password

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Canon Forum | Join Upload & Sell

  

Archive 2012 · 40D 5Dclassic 1DsmkII ISO test
  
 
rabbitmountain
Offline
• • •
Upload & Sell: Off
p.1 #1 · p.1 #1 · 40D 5Dclassic 1DsmkII ISO test


I finally found some time to do a ISO test with the 1DsmkII I bought last week. Set it up against the 40D (that I am going to sell) and de 5Dclassic, that I will keep as second/backup body.

Many said the 5Dclassic has slightly bettre IQ. I partially agree. It's a tad better at handling noise, but I prefer the 1DsII files for their colours and contrast.

Please note that this test was done in ideal shooting conditions, with good exposure and few dark areas. I do know that real assignment images in dark venues will contain more noise. But it is good to compare the three bodies.

Now about the test setup:
- I shot the same scene at ISO 400-800-1600-3200-3200(-1)-3200(-2)-3200(-3). The last three simulate ISO 6400, 12800 and 25600.
- I loaded all images in LR4.1
- I exported the images using the following settings:
* JPEG (sRGB) Quality 90
* Resize to fit 2424 pixels
This way I get a file with 24" width on my 30" monitor, exactly as wide as a spread in one of my wedding albums. So I can judge the noise performance in a real size.

From these files I took screen shots. You can find them here:

s1245.photobucket.com/albums/gg597/rabbitmountain/ISO test 40D 5Dc 1DsII/

I will add the screen shots below as well.

The scene:






----------------------------------------------------------------------------------------------------------------

Each screen shot has 40D on top, 5Dclassic in the middle and 1DsmkII bottom.
Please note again that these are not 100% crops!


Conclusions:
5D classic has the best high ISO noise performance, but practically the 1DsII is just as good. The 40D is clearly worse, as I expected. But with the cleanup software, the 40D is not far behind the other two.

When I start cropping, the 40D is falling behind more, because of the small resolution. The screen shots are too wide for this web site, so please follow the link above and view the crops there.

ISO 400:






ISO 800:






ISO 1600:






ISO3200:






ISO 3200, with Topaz DeNoise5, setting "Light":






ISO 6400 simulated:






ISO 6400 simulated, with Topaz DeNoise5, setting "Light":






ISO 6400 simulated, with Topaz DeNoise5, setting "Moderate":






ISO12800 simulated:






ISO 12800 simulated, with Topaz DeNoise5, setting "Moderate":






ISO 25600 simulated:







Now I took the ISO3200 RAW file and cropped it to 70% width and 70% height, leaving 49% of pixels. I then followed the same procedure as above. This is to mimic what is left of the resolution when images are cropped before exporting them, and to judge if spreads can be made.

[Edit: I see that these are too big for this web site. Please judge following the links provided above]




Aug 27, 2012 at 01:52 PM
jcolwell
Online
• • • • • • •
Upload & Sell: On
p.1 #2 · p.1 #2 · 40D 5Dclassic 1DsmkII ISO test


Hey Ralph,

Thanks for your efforts in taking and posting these results.

I generally find it most useful to look at 100% crops, but that's complicated by the two different sensor sizes, of course.

Cheers, Jim



Aug 27, 2012 at 08:22 PM
rabbitmountain
Offline
• • •
Upload & Sell: Off
p.1 #3 · p.1 #3 · 40D 5Dclassic 1DsmkII ISO test


Hi Jim,
I could, of course, post 100% crops. If you'd like me to, I will. However, that is not what this test was about. I wanted to know if I could see a difference when viewing at the size I would view a wedding album spread. So just imagine yourself looking at a 24" wide image and viewing just a small cut out. Just look at the images from about 15 or 20 inches (the same distance you'd watch an album) and see if there is noise that bothers you. That is what I wanted to know.

I've looked at 100% and it's all hideous over ISO800. But it doesn't mean much if I can't see that noise in actual print.

My conclusion is that I can easily use ISO3200 on the 1DsmkII and 5Dclassic. In extreme emergencies I could even do ISO6400 simulated. With appropriate NR software with carefully selected settings, I would use that in an album (though probably not as a spread, but at 8 or 10", why not?).

I have to say that the little noise you might see in ISO3200 and ISO6400 simulated will not be visible in print, provided it's printed by a pro lab.

So please if you want the 100% crops well tell me and I'll put them up.



Aug 27, 2012 at 08:30 PM
mttran
Offline
• • • • •
Upload & Sell: Off
p.1 #4 · p.1 #4 · 40D 5Dclassic 1DsmkII ISO test


rabbitmountain wrote:
My conclusion is that I can easily use ISO3200 on the 1DsmkII and 5Dclassic


+1, that is exactly what i said about the differences between canon old vs new FF models. Iso (100-3200) noise ratio should be identical with a touch of any 3rd party NR filter. For those who does not need higher than 3200, video...etc there are plenty great & cheap pro bodies out there to support your needs.



Aug 27, 2012 at 09:34 PM
jcolwell
Online
• • • • • • •
Upload & Sell: On
p.1 #5 · p.1 #5 · 40D 5Dclassic 1DsmkII ISO test


rabbitmountain wrote:
...So please if you want the 100% crops well tell me and I'll put them up.


Not necessary, but thanks. I often have my nose in the 100% crop mode to compare lenses and ultimate resolution, but you're right - it's the application that counts!

Look at me - I just picked up my new 1DX last week, and today I dropped off my old 1DsIII for a CLA - it'll be on the market soon. I guess I'm slidegrading from 21MPx to 18 MPx. What's up with that?!

Cheers, Jim



Aug 27, 2012 at 09:40 PM
jcolwell
Online
• • • • • • •
Upload & Sell: On
p.1 #6 · p.1 #6 · 40D 5Dclassic 1DsmkII ISO test


Ralph,

Follow-up question...

Did you notice that the 40D images consistently show a Left/Right separation of Light Grey/Dark Grey illumination on the phone screen, while the 5D and 1DsIII consistently show a dark phone screen? What's up with that?

Jim



Aug 27, 2012 at 09:49 PM
rabbitmountain
Offline
• • •
Upload & Sell: Off
p.1 #7 · p.1 #7 · 40D 5Dclassic 1DsmkII ISO test


^ ^
I decided to use the same lens (50/1.8II) for all cameras and because of the crop factor of the 40D I had to move the 40D back. As it is tilted a bit (it would fall all the time otherwise) you see a different reflection than with the 5D/1DsII.



Aug 28, 2012 at 03:17 AM
rabbitmountain
Offline
• • •
Upload & Sell: Off
p.1 #8 · p.1 #8 · 40D 5Dclassic 1DsmkII ISO test


mttran wrote:
+1, that is exactly what i said about the differences between canon old vs new FF models. Iso (100-3200) noise ratio should be identical with a touch of any 3rd party NR filter. For those who does not need higher than 3200, video...etc there are plenty great & cheap pro bodies out there to support your needs.


True. But...applying the same "touch of any 3rd party NR filter" to the newer bodies gives pulls them ahead of the old bodies. Based on some of your earlier posts, in particular the one where you underexpose by 2 stops and then correct by +4 stops in post (if I recall correctly), I got the impression that there was something extraordinary about the latitude in the 1DsII files, that other bodies don't have. Now in my findings, whatever I do with the 1dsII to clean it up, the 5Dclassic always stays even cleaner when applying the same treatment.

I'm not saying I'm disappointed by the noise performance of the 1DsII with the help of 3rd party software, on the contrary, it made me decide to keep this body because I know I can safely use ISO3200 and still print 24" spreads and clients just won't notice I used this high ISO on a 2004 technology camera. But if anything, it merely made me realise the value of 3rd party noise reduction software. I thought LR4 was just as good but it isn't.



Aug 28, 2012 at 05:43 AM
 

Search in Used Dept. 



mttran
Offline
• • • • •
Upload & Sell: Off
p.1 #9 · p.1 #9 · 40D 5Dclassic 1DsmkII ISO test


Hi Ralph,

I believe i did not say anything about simulated iso range. We only discussed the standard iso range 100-3200 in other post. You might mis-understood me by the duplication of Giusmelix's 40vs1ds2vs5d3 that i posted.



Aug 28, 2012 at 06:14 AM
elader
Offline
• • • •
Upload & Sell: Off
p.1 #10 · p.1 #10 · 40D 5Dclassic 1DsmkII ISO test


1dsmkII has more noise at 1600 and 3200 than the 5D when viewed at 100% but when the 1dsmkII image is scaled down to the 5D size, I found them very very similar. Both make lovely images.

I never liked the 40D, and yes, I owned one.



Aug 28, 2012 at 05:27 PM
rabbitmountain
Offline
• • •
Upload & Sell: Off
p.1 #11 · p.1 #11 · 40D 5Dclassic 1DsmkII ISO test


elader wrote:
1dsmkII has more noise at 1600 and 3200 than the 5D when viewed at 100% but when the 1dsmkII image is scaled down to the 5D size, I found them very very similar. Both make lovely images.


The images above were all scaled down to 2424 pixels. The 5D stays ahead in noise performance and when going up the ISO scaled bests the 1DsII more and more. @ISO25600 simulated the difference is very clear. Now these are taken with a white background and well exposed. Today I took some test shots in a dark church where a wedding will take place. They didn't turn the lights on so I tested in the worst case scenario that lighting was weak. I wound up with 1/60s ISO3200 f/1.4 on my 35L. Both the 5Dc and 1DsII performed quite well. However, dark corners become very noisy with both cameras. I'll post some samples here as soon as I have time.



Aug 28, 2012 at 05:51 PM
Leslie168
Offline
• • •
Upload & Sell: Off
p.1 #12 · p.1 #12 · 40D 5Dclassic 1DsmkII ISO test


I just thinking if need buying 1DS2, thanks for above and that result should make me more quickly to finding 1DS2 for wedding...


Sep 06, 2012 at 01:31 PM
PhilDrinkwater
Offline
• • •
Upload & Sell: Off
p.1 #13 · p.1 #13 · 40D 5Dclassic 1DsmkII ISO test


rabbitmountain wrote:
but I prefer the 1DsII files for their colours and contrast.


Realistically these are things that could easily be "fixed" in post.

I was always happy enough to use the 5DC at ISO3200 Amazing that older tech can do this..



Sep 06, 2012 at 02:40 PM
rabbitmountain
Offline
• • •
Upload & Sell: Off
p.1 #14 · p.1 #14 · 40D 5Dclassic 1DsmkII ISO test


^ ^
After some more 5D-1DsII comparing I conclude that IQ is practically tne same, ie i dpn't see differences in large prints. Basically the only benwfit I see in the 1DsII is shutter lag and dual cards' not IQ. That said, I would have cwrtainly kept it if the AF had been OK.



Sep 07, 2012 at 02:45 PM
bauerman
Offline
• • •
Upload & Sell: Off
p.1 #15 · p.1 #15 · 40D 5Dclassic 1DsmkII ISO test


You know what this shows me? The 40D is a GREAT high ISO performer and in real world shooting situations and creating normal sized prints, you would be hard pressed to tell the difference between the three cameras - other than the cash left over in your pocket if you were shooting a 40D.

Now back to your regularly scheduled pixel peeping.



Sep 09, 2012 at 04:23 PM
rabbitmountain
Offline
• • •
Upload & Sell: Off
p.1 #16 · p.1 #16 · 40D 5Dclassic 1DsmkII ISO test


Hi Bauerman,

The 40D is a good high ISO perfomer, but it really is no match for the 5D and 1DsII. Mind you - and I repeat it again - these are not 100% crops, but real size prints. When I look at the 40D images at ISO3200 I don't really like them. I think the noise is not flattering, it doesn't add character. 5D and 1DsII much more so (and less of it). The 40D feels more plasticky and this distracts from the image itself. This is why I sold it. I need to shoot at highest possible ISO regularly.

I don't know what you mean by "your regularly scheduled pixel peeping" but that is something that is meaningless for me. I want to simulate a real print (or print a real print) and see if I see differences between cameras. Only that is meaningful for me and that is why i did the test the way I did it. No 100% crops, but real size prints.



Sep 09, 2012 at 09:37 PM





FM Forums | Canon Forum | Join Upload & Sell

    
 

You are not logged in. Login or Register

Username   Password    Reset password