Ad
Ad
Ad
Pages: « 1 2 [3] 4 5 »   Bottom of Page
Print
Author Topic: Comparing Apples to Grapefruit IQ180/D800E Crops  (Read 22028 times)
Petrus
Sr. Member
****
Offline Offline

Posts: 509


« Reply #40 on: May 25, 2012, 01:38:55 AM »
ReplyReply

I definitely think that adds to the look in a big way (maybe the only way?). Since MF inherently uses longer focal lengths to achieve the same angle of view from the same position, it will always be one step ahead in that respect. Another way of wording it that helps my understanding is that MF always allows you to be closer with a longer lens. Aside from stitching, there's no way to equalize that, and something like 4x5 takes the concept further still.

Huh

What does the focal length matter, we just adjust the picture angle to the sensor size by choosing the lens. If the pictures are taken from the same spot and framed identically the perspective is identical no matter what focal length is used in each case. Basic descriptive geometry. Physical impossibility to get different projections, the only difference is the quality of the lens.
Logged
marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #41 on: May 25, 2012, 01:48:50 AM »
ReplyReply

Yeah but that's not what you'd actually pay if you were a smart shopper  Grin      Plus if you're working with it, you can deduct this from your taxes etc. so the price isn't exactly sunk.  The IQ 180 used for your comparisons - was it yours? Will you sell it now? 

Anyhow I don't mean to say the D800 isn't a fine camera, but I do believe there are areas like color and depth that the MFDB's still are superior.

Yes it is my IQ180 and the cost really is a bit more than $39k, the back I acquired by trading in my P45+ that I had bought privately at a good price (trade in value was 2X my initial cost} but the WRS plus 3 Rodenstock lenses (2 T/S) aren't cheap, plus an upgraded tripod/head etc etc. But that's the way technology is, it costs 10X for a 10% improvement at the high end it's not linear. Oh and no you can't pry my IQ180 or my D800E out of my hands!
The processing was just straight out of C1 with basic settings so color etc are very different nothing optimized, I was trying to post a comparison of resolution on both a pixel level (the D800E wins) and on a FOV level (the IQ180 wins).
Marc

PS I drive a 1995 Ford Explorer with 200k miles on it, nice car/cheap camera or nice camera/old car, everyone has their priorities.
Logged

Marc McCalmont
torger
Sr. Member
****
Online Online

Posts: 1496


« Reply #42 on: May 25, 2012, 02:20:52 AM »
ReplyReply

Even though it's a portrait and can't be directly compared to the images in this thread, here's a shot that really says "MF" to me: http://www.davideustace.com/photography/characterproject/files/page5-1008-full.html
The subject is clearly isolated and stands out from the background, but I can still count the boards on the broad side of the far building. Those subtle depth cues are something I only tend to see in MF shots.

This is an interesting aspect. 135 digital with f/1.2 lenses can actually do shorter DOF than typical MF lenses, but I have realized that it is not ultra-short DOF that the MF look is about, it seems rather to be about a softer transition from in focus to out of focus in quite deep but not all-sharp DOFs, which provides a subtle layered 3D look. Possibly also better bokeh quality, due to that shorter DOF can be achieved with smaller apertures and thus less optical correction.

I don't know if it is true though, but now when D800 is up there resolution-wise it is something to investigate.

It can also be noted that the difference in size between 44x33 (IQ140) and 54x41 (IQ180) is similar to the difference between 36x24 (D800) and 44x33, that is a similar difference in this aspect would be noted between entry level MF and high end MF.
Logged
torger
Sr. Member
****
Online Online

Posts: 1496


« Reply #43 on: May 25, 2012, 02:29:32 AM »
ReplyReply

What does the focal length matter, we just adjust the picture angle to the sensor size by choosing the lens. If the pictures are taken from the same spot and framed identically the perspective is identical no matter what focal length is used in each case. Basic descriptive geometry. Physical impossibility to get different projections, the only difference is the quality of the lens.

It is not about perspective, it is about subtle differences in how depth of field behaves. I am myself not sure if it is so subtle that it is imagination rather than a real difference though, but something that I would like to find out eventually.
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8883


« Reply #44 on: May 25, 2012, 06:02:40 AM »
ReplyReply

Hi Marc,
That's an interesting comparison. The results tend to confirm the accuracy and relevance of DXOMark tests. If one compares the graphs for the D800 with the IQ180, in screen mode, one is comparing pixel quality, as you are attempting to do with the two crops of images taken with the same lens at the same focal length.

Despite the IQ180 pixel being marginally larger than the D800 pixel (5.17 microns as opposed to 4.7 microns), the quality of the D800 pixel is as 'close as matters' in respect of SNR at 18%, Tonal Range and Color Sensitivity.

For example, the Color Sensitivity of the D800 pixel is described as being 22 bits. That of the IQ180 is 21.5 bits. In other words, the Color Sensitivity of the D800 is actually better by 0.5 bits. However, DXO state that a difference of less than 1 bit would likely not be noticeable. The values for Tonal Range and SNR at 18% are even closer.

But there's one glaring difference, and that's DR. The DR of the D800 pixel is claimed by DXO to be 1.34 EV better than the IQ180 pixel. That is a difference that should be clearly noticeable in a very high contrast scene.

If I have one criticism of your comparison, it would be that you haven't chosen a scene with some significant areas of deep shadow containing detail, so we could see if the D800 pixel really does have noticeably better DR.  Grin

Cheers!

PS. The EXIF data is a bit puzzling. The IQ180 appears to have been shot at ISO 35 and 1/8th sec exposure. No mention of F stop. The D800E shot is at ISO 100, F2.8 and 1/500th sec exposure. What's going on here? Why the huge difference in exposure?
« Last Edit: May 25, 2012, 06:11:15 AM by Ray » Logged
marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #45 on: May 25, 2012, 06:32:11 AM »
ReplyReply

Hi Marc,
That's an interesting comparison. The results tend to confirm the accuracy and relevance of DXOMark tests. If one compares the graphs for the D800 with the IQ180, in screen mode, one is comparing pixel quality, as you are attempting to do with the two crops of images taken with the same lens at the same focal length.

Despite the IQ180 pixel being marginally larger than the D800 pixel (5.17 microns as opposed to 4.7 microns), the quality of the D800 pixel is as 'close as matters' in respect of SNR at 18%, Tonal Range and Color Sensitivity.

For example, the Color Sensitivity of the D800 pixel is described as being 22 bits. That of the IQ180 is 21.5 bits. In other words, the Color Sensitivity of the D800 is actually better by 0.5 bits. However, DXO state that a difference of less than 1 bit would likely not be noticeable. The values for Tonal Range and SNR at 18% are even closer.

But there's one glaring difference, and that's DR. The DR of the D800 pixel is claimed by DXO to be 1.34 EV better than the IQ180 pixel. That is a difference that should be clearly noticeable in a very high contrast scene.

If I have one criticism of your comparison, it would be that you haven't chosen a scene with some significant areas of deep shadow containing detail, so we could see if the D800 pixel really does have noticeably better DR.  Grin

Cheers!

PS. The EXIF data is a bit puzzling. The IQ180 appears to have been shot at ISO 35 and 1/8th sec exposure. No mention of F stop. The D800E shot is at ISO 100, F2.8 and 1/500th sec exposure. What's going on here? Why the huge difference in exposure?
None of the exif data is valid The 70mm Rodenstock HR is a "technical Lens" no electronic connection to the camera, I set the IQ180 to its base ISO 35. For the D800E I set it at base ISO 100 and used a Leica R lens that is not "chipped" again no valid exif data transferred to the camera. The only accurate data would be ISO and shutter speed on the Nikon. I manually set f5.6 on both lenses and ETTR'd
Marc
Logged

Marc McCalmont
Ray
Sr. Member
****
Offline Offline

Posts: 8883


« Reply #46 on: May 25, 2012, 06:53:31 AM »
ReplyReply

None of the exif data is valid The 70mm Rodenstock HR is a "technical Lens" no electronic connection to the camera, I set the IQ180 to its base ISO 35. For the D800E I set it at base ISO 100 and used a Leica R lens that is not "chipped" again no valid exif data transferred to the camera. The only accurate data would be ISO and shutter speed on the Nikon. I manually set f5.6 on both lenses and ETTR'd
Marc

Okay! I understand. I experienced a similar situation using a Nikon/Canon adapter on my 5D. Now, to complete the comparison, all you need to do is reshoot a very high contrast scene with lots of deep shadow detail.  Grin
Logged
CalvinHobbes
Newbie
*
Offline Offline

Posts: 3


« Reply #47 on: May 25, 2012, 12:09:28 PM »
ReplyReply

As stated by the OP, this is comparing apples to Grapefruit! I am much more interested in seeing what the best can be obtained from each captured file. Using some type of minimal default to compare the two captures is at best a waste of time.

The perspective is also a bit off and does affect the comparison photos. After all the color corrections, lens corrections and post processing can we really begin to compare what can be achieved with these files. These are only data that have been collected. What happens after the data collection is much more important than this simple comparison.
Logged
Ellis Vener
Sr. Member
****
Offline Offline

Posts: 1801



WWW
« Reply #48 on: May 25, 2012, 12:25:20 PM »
ReplyReply

It is not about perspective, it is about subtle differences in how depth of field behaves. I am myself not sure if it is so subtle that it is imagination rather than a real difference though, but something that I would like to find out eventually.

Making the assumption that a "normal" angle of view  focal length lens for the IQ 180 is a 100mm lens and for the D800 is a 50mm, and we are photographing a subject subject with depth at somewhat less than "infinite" distance for either lens, and using the same ISO, shutter speed, and f-stop, the depth of field with the 100mm will be shallower at all f-stops than with the 50mm. You could also see this if you use both a 50mm and the 100mm lens on the IQ180 camera and crop the photo shot with the 50 down to the same angle of view the full frame 100mm lens records.
Logged

Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.
marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #49 on: May 25, 2012, 03:05:39 PM »
ReplyReply

As stated by the OP, this is comparing apples to Grapefruit! I am much more interested in seeing what the best can be obtained from each captured file. Using some type of minimal default to compare the two captures is at best a waste of time.

The perspective is also a bit off and does affect the comparison photos. After all the color corrections, lens corrections and post processing can we really begin to compare what can be achieved with these files. These are only data that have been collected. What happens after the data collection is much more important than this simple comparison.

When I have a chance I'll give it a try probably this summer though when I'm back in Hawaii with some good subject material. Problem is if you start optimizing too much where does personal taste take over? But at the least I think C1 favors Phase files better to use Camera raw for comparisons and I should do some color correction. The bottom line here is that the D800E is an excellent DSLR that at the least intrudes into MFDB territory and is the first DSLR who's image quality doesn't leave me wanting. 
Marc
Logged

Marc McCalmont
Lightsmith
Full Member
***
Offline Offline

Posts: 111


« Reply #50 on: May 25, 2012, 08:28:48 PM »
ReplyReply

I was curious as to the difference in image quality between the 12MP APS-C sensor equipped D300 and the 12MP full frame sensor equipped D3. I shot a series of images using the Nikon 70-200mm f2.8 VR II lens of an adult sea lion at a distance of roughly 50 feet. Comparing the two images there was more "detail" captured by the D3 and it showed as more tonal variation in the animal's fur. The collapsing in part of the tones by the D300 resulted in lost detail.

Comparing the D800 to a MF or digital back I would want to do a similar test using lenses that provided roughly equal levels of image magnification.

I found Jim Brandenburg's comments regarding the D800 to be very interesting. He stated that he felt that the D800 goes beyond the territory of the MF camera and approaches that of a 4x5 camera and he was referring in part to an extremely large print he had recently made using a file from the D800.
Logged
eleanorbrown
Sr. Member
****
Offline Offline

Posts: 624


WWW
« Reply #51 on: May 25, 2012, 09:16:20 PM »
ReplyReply

Marc, thanks again for taking the time to do these shots.  On these two shots, while the 800e shot looks sharper (but with color fringing also), the Phase One file looks much much smoother color wise and tonally to me.  In this one I could pick out the Phase file right away. Eleanor

#1 correct
#2 Digaron-W 70mm & Leica 28-90 f2.8-4.5R
(according to Erwin Puts 1. the "good" Leica R zooms are sharper than the Leica R primes which are the equal of Leica M primes, and 2. the slower zooms that I have are the equal of the 2.8 zooms when stopped down. The reasoning is with more design flexibility in moving the elements in a zoom you can exceed the performance of a prime. Who am I to question the Tao of Puts!) but both at f5.6 and f8.0 the D800E/Leica is sharper than the IQ180/WRS/Rodenstock HR!
Marc

the next step is to verify infinity focus since the WRS does not allow you to shim the back but the lenses were calibrated to the WRS at the Cambo factory.
Logged

ErikKaffehr
Sr. Member
****
Online Online

Posts: 7419


WWW
« Reply #52 on: May 27, 2012, 02:55:01 AM »
ReplyReply

Hi Eleanor,

Just a comment...
To me the Phase One shot seems to be sharpest of the three. The 45 mm crop does show fringing, and a healthy amount of Moiré and also shows some fringing, which I don't see in the 70 mm crop.

I'm somewhat confused that the 70 mm crop is so much better than the 45 mm crop, at least IMHO.

I'd suggest that Marc can improve a lot on color rendering from the D800E shots, and I don't presume that his IQ180 will be put to pasture for a long time.

Best regards
Erik

Marc, thanks again for taking the time to do these shots.  On these two shots, while the 800e shot looks sharper (but with color fringing also), the Phase One file looks much much smoother color wise and tonally to me.  In this one I could pick out the Phase file right away. Eleanor

Logged

Fine_Art
Sr. Member
****
Offline Offline

Posts: 1087


« Reply #53 on: May 27, 2012, 03:44:38 AM »
ReplyReply

The biggest advantage I see on the IQ180 is the accuracy of the auto WB in C1 relative to the D800E. The price difference is huge.
Logged
KevinA
Sr. Member
****
Offline Offline

Posts: 898


WWW
« Reply #54 on: May 27, 2012, 05:56:01 AM »
ReplyReply

I don't know what the reason is, but I'd agree with Mike Yamin.  No question the IQ 180 is the better image here. The d800 image looks flat compared to the one by the IQ 180.  And it appears there are some big color differences too.

Yes I think so too, the more you look the bigger the difference. I wonder if this difference is the same with all DMF backs or do you need the 180 big boy to get it. If it's a ccd thing does the Leica M have it? The difference looks so big I just wonder if the lighting change a lot as the clouds have moved on out of sight from one picture to the other. We have no idea what is happening to the lighting out of scene, there could easily be less light bouncing around to light the shadows.

Kevin.
Logged

Kevin.
EricWHiss
Sr. Member
****
Offline Offline

Posts: 2412



WWW
« Reply #55 on: May 27, 2012, 11:32:31 AM »
ReplyReply

Yeah, Kevin, I wondered too if the lighting had changed enough to impact the look.  One could probably figure that out by comparing exposure data - for example if a cloud was behind the sun in one of the shots and not the other.
Logged

Authorized Rolleiflex Dealer:
Find product information, download user manuals, or purchase online - Rolleiflex USA
marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #56 on: May 27, 2012, 07:24:18 PM »
ReplyReply

shot within 5 minutes of each other but C1 doesn't have the same color settings available for the D800 as they have for the IQ180 (I think they favor their own products)
Marc
Logged

Marc McCalmont
Fine_Art
Sr. Member
****
Offline Offline

Posts: 1087


« Reply #57 on: May 27, 2012, 08:00:24 PM »
ReplyReply

So what happens on the D800E shot when you move the white balance slider to the same number as the IQ180? Or just pick 5000k as a starting point.
Logged
marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #58 on: May 27, 2012, 08:39:05 PM »
ReplyReply

I had shot both with WB set to daylight and processed them in C1 with daylight settings. I don't have C1 in front of me but there was no "film standard" selection for the D800 just linear. tomorrow is a day off I'll shoot a few more examples. Perhaps Camera RAW would be a less "prejudiced" converter not favoring the IQ180?
Again not trying to compare apples to apples just amazed at the resolution possible with a DSLR and the Leica lenses.
Marc
Logged

Marc McCalmont
marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #59 on: May 28, 2012, 04:56:49 PM »
ReplyReply

DR test
labeled them A and B, C1 shadows & Highlight sliders 100%, focus fixer .7
Marc
Logged

Marc McCalmont
Pages: « 1 2 [3] 4 5 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad