Ad
Ad
Ad
Pages: [1] 2 »   Bottom of Page
Print
Author Topic: 5DII Vs 1DsIII iso test  (Read 42284 times)
Derryck
Full Member
***
Offline Offline

Posts: 170



WWW
« on: November 30, 2008, 08:46:49 PM »
ReplyReply

This morning I shot with both the 5DII and the 1DsIII at iso's ranging from 50 to 3200 and then the 5DII at 6400 and 12800iso. To enable me to light the scene over such a large latitude I used daylight through the studio windows and monitored the light levels using a handheld light meter. The image was made with a 135mm f2 lens @f11. Live View 10x was used to check focus and both cameras were set to mirror up and 2sec self timer. The upper right hand image was taken at -3stops from the normal exposure to simulate shadow areas. Both sets of images were processed with ACR with no added contrast, saturation or sharpening added. Once the images were opened in Photoshop I applied a light smart sharpening of 60% with a radius of 0.6.

Both cameras' Noise Reduction was turned off. All images were made from RAW files.

I noticed that my 5DII was producing images that were about 1-2% brighter than the 1DsIII based on the reading off the greycard in Photoshop.

This comparison doesn't show the best you can get out of these two cameras but the bare minimum. Don't write and tell me that I should have used more saturation, sharpening or noise reduction.

For some the people the results may come as a surprise. I'm not a professional reviewer and only did this for my own purposes. There may be things that I could have done better but I think for a basic comparison the images below give you a good idea of what these cameras are capable of. I'm just happy that I've managed to buy a backup camera that is on par with my workhorse, that I can take away on holidays and now start to make feature length films with :-).

Now I really do have to get back to work. I think I'm starting to give the impression that all I do is sit around the studio surfing photography forums.

Cheers,

Derryck.
« Last Edit: November 30, 2008, 10:22:45 PM by Derryck » Logged

tony field
Newbie
*
Offline Offline

Posts: 33


WWW
« Reply #1 on: November 30, 2008, 09:25:56 PM »
ReplyReply

Excellent "neutral unbiased" test.  There seems to be no useful difference between the two cameras in terms of basic noise performance.  I was expecting a slight advantage towards the 5D-II since it uses a less dense CFA.  

This sort of implies that the 1Ds-III, when properly processed for noise reduction, can work very well as a high ISO camera.  Canon has worked magic in firmware when shooting JPG files on the 5D-II when noise reduction is turned on and extreme ISO values are desired.  This should be of some advantage to some JPG shooters.  
Logged
Derryck
Full Member
***
Offline Offline

Posts: 170



WWW
« Reply #2 on: November 30, 2008, 09:40:39 PM »
ReplyReply

Yes I must admit I was also expecting there to be a greater difference in favour of the 5DII at higher iso's. It still represents great value if you can get your hands on one. Thanks to the huge advances by Nikon this year we Canon users have a backup camera that produces the same resolution, DR and iso performance as it's bigger brother.

Canon I've sent you an email with my bank details.
Logged

Peter Gregg
Jr. Member
**
Offline Offline

Posts: 62


« Reply #3 on: November 30, 2008, 09:40:49 PM »
ReplyReply

This is very interesting. In my personal testing of the Nikon D3, Canon 1Ds MK3, and the Sony A900 with the final results being PRINTS sized at 16x24 and using ISO 3200 I rank the cameras in the following order of quality:

1. Canon 1Ds MK3
2. Nikon D3
3. Sony A900

ISO 1600 bears the same results. This is different than seen on the screen in "pixel peeping" at 100 percent. For me, prints is where the final definition of a camera's quality level is at and not 100 percent screen images. Too few people can actually interpret the 100 percent image examples and apply that data mentally to their intended output. I say that respectfully, but it does need to be said because it seems to much comparison is made with no particular goal in mind as very few people really look at images at 100 percent on computer screens.

In my conversations with Chuck Westfall, he indicated at least a 2 stop advantage of the new 5D MK2 over the 1Ds MK3. I don't see that happening here, it may be possible pertaining to JPG images and not to RAW files, which is not of interest to me. I see the images as a wash between the 2 cameras. If that bears out as a truism then the choice of body you select to buy can be determined by price, features, focus systems, and weight rather than by image quality.

Peter
« Last Edit: November 30, 2008, 09:43:35 PM by Peter Gregg » Logged
Ray
Sr. Member
****
Online Online

Posts: 8947


« Reply #4 on: November 30, 2008, 10:19:46 PM »
ReplyReply

Quote from: Derryck
Both cameras' Noise Reduction was turned off.

You were shooting in jpeg mode? I'm a bit confused here. These test results look so close to me that any differences are negligible in practical terms. If these shots were taken in RAW mode, then DWdallam should be very happy that the 5D2 has not trounced the 1Ds3   .
Logged
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #5 on: November 30, 2008, 10:45:15 PM »
ReplyReply

Quote from: Peter Gregg
In my conversations with Chuck Westfall, he indicated at least a 2 stop advantage of the new 5D MK2 over the 1Ds MK3
I guess this is the same misunderstanding as regarding the 50D, compared to the 40. Westfall obviously did not mean, that the noise with the 5D2 @ ISO 6400 will be the same as with the 1DsMkIII @ ISO 1600.

Nikon has introduced with the D3 ISO 6400 (real) as well as 12800 and 25600 (fakes), and Canon "had to" follow suit, so the 5D2 offers now three fake ISOs (6400, 12800, 25600). This is useful for JPEG shooters, and even many if not the majority of raw shooters believe, that they gain something by using those ISOs.
Logged

Gabor
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #6 on: November 30, 2008, 10:54:26 PM »
ReplyReply

Derryck,

do you mind uploading the -3 EV raw files? Yousendit can be used without registering, but only up to 25 MB file size. Up to 100 MB it requires free registration.
Logged

Gabor
Derryck
Full Member
***
Offline Offline

Posts: 170



WWW
« Reply #7 on: November 30, 2008, 11:06:15 PM »
ReplyReply

I have just shot a quick comparison between the two cameras with Long exp. NR turned on and High iso NR turned on and will post shortly.

As for uploading Raw files, uploading anything from China can be extremely slow due to The Great Firewall of China. I'll see what I can do.

Cheers,

Derryck.
Logged

Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #8 on: November 30, 2008, 11:25:04 PM »
ReplyReply

Quote from: Derryck
As for uploading Raw files, uploading anything from China can be extremely slow due to The Great Firewall of China. I'll see what I can do.

If you do it, pls upload only one first, to verify if it is sufficiently underexposed.

Thanks anyway,
Logged

Gabor
Derryck
Full Member
***
Offline Offline

Posts: 170



WWW
« Reply #9 on: December 01, 2008, 12:57:16 AM »
ReplyReply

I have just shot comparison images using the two cameras at 1600iso with High iso and Long exposure Noise Reduction settings turned on and off. The 5DII has two settings in High iso speed noise reduction and so I have tested both Standard and Strong. I wasn't sure if altering these settings when shooting in RAW mode changes anything but after looking at the results it appears that at the least the Long exposure noise reduction (where the camera records a black frame) does.

Same setup as before this time exposure time was much longer at 13 seconds. I'm not sure how much you should read into the dark -3stop frame in the top right hand corner as the exposure time was only 1.6seconds (though it still activated the Long Exp NR).
Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #10 on: December 01, 2008, 01:40:32 AM »
ReplyReply

Quote from: Ray
You were shooting in jpeg mode? I'm a bit confused here. These test results look so close to me that any differences are negligible in practical terms. If these shots were taken in RAW mode, then DWdallam should be very happy that the 5D2 has not trounced the 1Ds3   .

Shut up Ray.

All kidding aside, I've let it go. I bought the last 8, 000 dollar Canon camera, but I've been using it like crazy. If I had waited for the 5D2 I'd still be using my 5D. On the up side, all cameras will start becoming much cheaper than the 8K landmark, which is good for all of us. I mean 8K? Get a life, right? I remember I bought a Minolta 7000i when they first came out back in 1991 or so and it was 700.00 dollars and one of the best cameras you could buy. It was the first to start using computer chips I think too, just reminiscing here.

Anyway, it's a really good thing. Nikon just released their new 25MP camera too. Now I'm sure Canon will announce their new 1DS this next year at around 30MP. And I'll still be happily using my 1DS3. I'm really am happy with it. It's resolution makes it more of a hybrid to use and get good image out of it, but yeah, it's kick ass. It's just not kick ass for 8K--lol.
Logged

Hägar the horrible
Newbie
*
Offline Offline

Posts: 40


WWW
« Reply #11 on: December 01, 2008, 05:14:36 AM »
ReplyReply

Thanks Derryck,
Did you see any banding in these high ISO images? as we know the 1ds3 has some issues there at very high ISO
Logged
David Anderson
Sr. Member
****
Offline Offline

Posts: 522



WWW
« Reply #12 on: December 01, 2008, 05:36:06 AM »
ReplyReply

Thanks for posting the samples.

I'm still in two minds about getting a 5DII because I already have a couple DsIII's and I'm not unhappy with them at high ISO.

I guess having the same performance at the much lighter weight could be a real bonus for some work and it would certinly be a better camera to pack along on a fishing trip..  




Logged

Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #13 on: December 01, 2008, 10:27:38 AM »
ReplyReply

Quote from: Derryck
I wasn't sure if altering these settings when shooting in RAW mode changes anything but after looking at the results it appears that at the least the Long exposure noise reduction (where the camera records a black frame) does
The High ISO NR does not affect the raw data. When using DPP, the setting is applied but it can be changed. ACR does not honor the setting.

The Long Exposure NR creates a second image with the same shutter time but closed (a black frame) and the second will be substracted from the first one in raw already, thus it does affect even ACR.
Logged

Gabor
Yakim Peled
Full Member
***
Offline Offline

Posts: 174


« Reply #14 on: December 01, 2008, 04:24:13 PM »
ReplyReply

Well, I am not impressed. I expected more. Much more. If I'd felt I must have a FF EF today I'd gladly pay the added cash for the 1Ds Mk III.
Logged

Happy shooting,
Yakim.
thewanderer
Full Member
***
Offline Offline

Posts: 106


« Reply #15 on: December 01, 2008, 05:04:21 PM »
ReplyReply

any chance of throwing a 1DS mark II shot in there at about 800 -1000......my primary interest is wildlife, and i dont see much golden light at 9 at night, so the high iso is not so important,, and i present use the m11n and m11,  but may be some more DOF on long lenses to catch 2 sets of eyes on different planes,, like a cow and calf where it might be nice to catch a litle more focus,, the dsII seems to still have some real followers,,

if no picutres, any further thoughts on the dsII
Logged
Jack Flesher
Sr. Member
****
Offline Offline

Posts: 2595



WWW
« Reply #16 on: December 01, 2008, 05:27:11 PM »
ReplyReply

Thanks for posting these!  It would be really cool to see the 3200 and 6400 / 12800 from the 5DII with NR on.
Logged

Derryck
Full Member
***
Offline Offline

Posts: 170



WWW
« Reply #17 on: December 01, 2008, 06:16:02 PM »
ReplyReply

Sorry guys I can't work on this for the time being. I am shooting five days this week and start a 14 day location shoot on Sunday. This has been a fun exercise but I have to get back to the part that pays for my new toys.

Cheers,

Derryck.

Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #18 on: December 02, 2008, 05:26:14 AM »
ReplyReply

Quote from: Panopeeper
The High ISO NR does not affect the raw data. When using DPP, the setting is applied but it can be changed. ACR does not honor the setting.

The Long Exposure NR creates a second image with the same shutter time but closed (a black frame) and the second will be substracted from the first one in raw already, thus it does affect even ACR.


Are you saying that if you shoot RAW, the ISO Noise Reduction has no effect?
Logged

Ray
Sr. Member
****
Online Online

Posts: 8947


« Reply #19 on: December 02, 2008, 09:41:25 AM »
ReplyReply

Quote from: dwdallam
Are you saying that if you shoot RAW, the ISO Noise Reduction has no effect?

I think the noise reduction (when enabled) only kicks in with long exposures in excess of a couple of seconds or so, I forget precisely. On the 5D it's called 'long exp. noise reduction', Custom Function #2. As Panopeeper mentioned, in this mode and with long exposures, the camera takes a dark frame shot of equal exposure time (with the mirror down of course). The results are very noticeable and worthwhile, but it can get very tedious when taking 30 second shots because the camera is engaged for a whole minute.
Logged
Pages: [1] 2 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad