Someone had to do it

[Shashin]

That is a long-edge print from an un-cropped Nikon file of a touch over 36 inches, which at this aspect ratio is 24" on the width - the largest print that almost anyone can make at home or even in their professional studios without having to outsource.

??? I have a 44" printer in my studio. It is a pretty common size--Canon, HP, and Epson have been making them for a long time. Paper rolls up to 44" are also common. Anything over that needs to be outsourced as there are no consumer printers available.

 

[tashley]

??? I have a 44" printer in my studio. It is a pretty common size--Canon, HP, and Epson have been making them for a long time. Paper rolls up to 44" are also common. Anything over that needs to be outsourced as there are no consumer printers available.

Oh i know you can get them, I considered one myself, but the vast majority of people I know with large format printers have the 24" version and outsource anything larger... I've got a Canon 6300 and had an HPZ3100 before that.

Let's have a poll on it!

 

[glenerrolrd]

Tim

First ..you have to be brave to post tests like this ! Several things keep coming up .

1. The plane of focus needs to be exactly the same and it does not appear to be in the test . Looking at the blue and white bowl and the pattern on it ...you can see that the focus of the Q180 is slightly back or the D800 . This makes comparisons of items in front of the clear vase inconsistent . To avoid this you need a set up where you absolutely can see the plane of focus ..which means a focus point and some detail in front and back of the target . The area just in front and just in back of the target should look “equally sharp” .

2. Each image should have the same exposure and white balance . This looks pretty good but if I was using camera calibration ..I want those color charts as close as I can get . This is a point for debate ...but achievement of a finished file includes both in camera and post processing . I take it one step further myself and try to match the overall scene contrast and sharpen to optimum.

You can debate ..straight out of the camera ..but a good S2 file has a linear contrast curve ..and no life until I apply a preset . I believe this is to preserve the absolute maximum DR out of the camera . Sharpening corrects for the any blur caused by glass between the lens and the sensor ...minimal on MF ..not sure on the D800E .

3. The subject should include organic items ...glass,plastic etc have few tones . Skin of course is a good example but some fruits work . If the target is a single tone ..I can t see anything. This is the only way to see tone range a key factor in MF IQ.

Separate point on Guy s observation about mid tones on the D800. There has been speculation since the D3 that Nikon compressed the mid tones to allow for head room in the highlights and depth in the shadows . I have observed this in a major way with the D3 in sports on players shirts and on shooting cars . A good preset probably tones down the saturation(maybe luminosity). and adds clarity . The D3X was way better than the D3 as it had more DR to work with . I have no idea if this is fact but it is the way the files appear to me.

 

[tashley]

Tim

First ..you have to be brave to post tests like this ! Several things keep coming up .

1. The plane of focus needs to be exactly the same and it does not appear to be in the test . Looking at the blue and white bowl and the pattern on it ...you can see that the focus of the Q180 is slightly back or the D800 . This makes comparisons of items in front of the clear vase inconsistent . To avoid this you need a set up where you absolutely can see the plane of focus ..which means a focus point and some detail in front and back of the target . The area just in front and just in back of the target should look “equally sharp” .

2. Each image should have the same exposure and white balance . This looks pretty good but if I was using camera calibration ..I want those color charts as close as I can get . This is a point for debate ...but achievement of a finished file includes both in camera and post processing . I take it one step further myself and try to match the overall scene contrast and sharpen to optimum.

You can debate ..straight out of the camera ..but a good S2 file has a linear contrast curve ..and no life until I apply a preset . I believe this is to preserve the absolute maximum DR out of the camera . Sharpening corrects for the any blur caused by glass between the lens and the sensor ...minimal on MF ..not sure on the D800E .

3. The subject should include organic items ...glass,plastic etc have few tones . Skin of course is a good example but some fruits work . If the target is a single tone ..I can t see anything. This is the only way to see tone range a key factor in MF IQ.

Separate point on Guy s observation about mid tones on the D800. There has been speculation since the D3 that Nikon compressed the mid tones to allow for head room in the highlights and depth in the shadows . I have observed this in a major way with the D3 in sports on players shirts and on shooting cars . A good preset probably tones down the saturation(maybe luminosity). and adds clarity . The D3X was way better than the D3 as it had more DR to work with . I have no idea if this is fact but it is the way the files appear to me.

Hi Roger,

For all the above reasons, I only ever do these tests for my own curiosity and then share the results, including RAW files, for other people to play with if they think that'll be useful. I was only testing for shadow detail and noise here and though equal planes of sharpness, which with sensors of different sizes you will only get (assuming the same FOV) with a flat target (there are almost no flat targets in real life) would have been useful, they don't change the particular part of the results I was looking at.

As it happens I was indeed trying to get focus on the neck of the most central thin glass vase and, true to my real world approach, I gave it one go on the D800E in live view and four goes on the Phase and in any event I wasn't doing a test for detail at all. Merely for shadow depth.

As for PP, that is very much season to taste and wasn't the point here, nor was the mid-range tonal expression, which can easily be tweaked in LR.

Per WB, all were WB'd off the chart and beyond that, it's purely up to the RAW converters and their various profiles. I have shown to my own satisfaction (and I think that Guy agrees) that the home brewed Xrite profiles are less useful than the canned ones: but again this was not about colour fidelity...

As you say, you have to be brave. Or stupid. Or both!

 

[tashley]

Roger,

Wasn't Tim's disclaimer enough for you? It's difficult to disagree with your points, but Tim clearly stated that his comparison was not "scientific enough to prove anything". Even so, I thought the comparison was interesting.

Joe

Thanks Joe. I agree. You don't need a Large Hadron Collider to show that it's dark inside a poorly lit big tube...

 

[T.Karma]

Looking at both files in LR4 the difference is like looking at a .tif or .psd (Phase) versus a compressed .jpeg (Nikon).

I do like the more organic feel of the Phase better, .... which comes at a price in money and handling.

The test is fair under the given circumstances altough it would be great to compare skin tones and organic stuff as well.

..... thanks a bunch for sharing this.

 

[tashley]

Looking at both files in LR4 the difference is like looking at a .tif or .psd (Phase) versus a compressed .jpeg (Nikon).

I do like the more organic feel of the Phase better, .... which comes at a price in money and handling.

The test is fair under the given circumstances altough it would be great to compare skin tones and organic stuff as well.

..... thanks a bunch for sharing this.

It's a pleasure - each to make of it what (s)he will - but I can't say that my experience of looking at the files in LR matches yours! I prefer the D800E file, but then I have spent a lot more time developing D800E files in LR than I have Phase files, which I always do in C1...

 

[T.Karma]

Tim,

Just not be misunderstood: The Nikon file allows more exposure of the shadows, but once that is done, the overall appearance is somewhat is flatter.

Expressed in painting technique it is like comparing oil technique to acrylics, ..... well at least a bit.

But I confess I am not a professional with LR4 (yet).

 

[Vivek]

You don't need a Large Hadron Collider to show that it's dark inside a poorly lit big tube...

:ROTFL:

..but as a never ending thread proves, sometimes, things are not as clear as black and white.

 

[Jan Brittenson]

The Nikon shot has too much red in the WB.
The Nikon shot is a little underexposed.
The lens on the Nikon has more veiling flare.

Fix the WB by pulling down the red channel 3-4 points (e.g. PS curves, red channel, set a point with input 4, output 0) and it looks like a slightly murky (veiled) and underexposed version of the IQ shot.

The detail levels where focus coincides between the two look fairly similar.

 

[tashley]

The Nikon shot has too much red in the WB.
The Nikon shot is a little underexposed.
The lens on the Nikon has more veiling flare.

Fix the WB by pulling down the red channel 3-4 points (e.g. PS curves, red channel, set a point with input 4, output 0) and it looks like a slightly murky (veiled) and underexposed version of the IQ shot.

The detail levels where focus coincides between the two look fairly similar.

This is a test for nothing but shadow noise...

 

[tashley]

Tim,

Just not be misunderstood: The Nikon file allows more exposure of the shadows, but once that is done, the overall appearance is somewhat is flatter.

Expressed in painting technique it is like comparing oil technique to acrylics, ..... well at least a bit.

But I confess I am not a professional with LR4 (yet).

Hiya,

If you haven't already, try using Camera Neutral as the starting point for the NEF file: my version was Camera Standard, and if you downloaded the NEF and opened it in LR, it will have defaulted to Adobe Standard, which is not great and which does cramp the upper mid tones...unless you told it to do otherwise...

 

[Guy Mancuso]

Tim you should try bringing them both in C1 now. My bet you'll like the Nikon even better.

 

[tashley]

Tim you should try bringing them both in C1 now. My bet you'll like the Nikon even better.

Thanks Guy - the last set I posted (dark bowl crops) are C1 but despite having processed a zillion Phase files in it, I never feel as confident as I do in LR. But the colours are better straight off the bat and I must say that when I compare the Xrite patches to the screen representations of these files on C1, they look more accurate. C1 is also more accurate about histograms and clipping.

BTW Am I the only person who finds that the clipping warning on the D800 and E kick in waaay to soon? It's a problem because in most scenes with bright sky and indirectly lit foreground, the camera overexposes. I set -1/3rd to -2/3rds stop quite often but using the clipping warning in camera isn't useful. I do get sharper shots though, cos it means faster shutter speeds ;-)

And why can't the camera meter properly? Everyone else seems to think it does apart from me but my 800 and my E are the same. The matrix metering might have 60,000 scenes in memory but none of them seem to involve a bright sky and a moderately lit foreground!

 

[Guy Mancuso]

Mine seems to underexpose a little. The D800 that is, I may just put in a little exp. comp. in, Also i agree it does indicate blown highlights when they are not. My guess only a guess but like anything else Nikon/Canon/Sony the OEMS do this on purpose. Given the numbers of units worldwide they want protect people from blown highlights. This is nothing new been going on for years. It's the same PR nightmare that they would deal with on Moire if they did not call out a disclaimer. Again imagine the numbers of complaints or forums of folks ranting how they have blown highlights. I have come to expect this kind of customer protection over the years and much more so than with Phase,Hassy, Leaf, Leica and so on as they know more experienced users are using there gear. At least this is the theory I have had for years on this stuff. I think I'm on to something since we see it a lot. That's okay though us experienced users can read into that more. I have had very few blow outs with the D800 metering setup.

 

[Guy Mancuso]

Funny mine under exposes and yours over exposes. I think the reason is I am dealing with really bright light and in most cases your shooting either in more diffused light or more cloudy skies. It's tring to bring up the values on you is my guess.

 

[Guy Mancuso]

You can change the metering pattern also in the menus. Might be worth looking at.

 

[tashley]

I think that's right Guy. In the South of France last week, well-lit daylight scenes were ok. It's medium cloudy days that screw it up. Also I think that LR tends to skew the histo a liitle to the left and you can be clipped in a channel without seeing it.

 

[glenerrolrd]

Roger,

Wasn't Tim's disclaimer enough for you? It's difficult to disagree with your points, but Tim clearly stated that his comparison was not "scientific enough to prove anything". Even so, I thought the comparison was interesting.

Joe

Oh I see sorry ...I misunderstood that the intent was only to check noise in the shadows .

 

[Toomanyartists]

I know this is an old post now, but I wanted to add something I'm surprised nobody's mentioned yet.

Most IQ180 owners say that they slightly overexpose to "push the histogram to the right" which appears to be the opposite of what you've done here. I bet if you tried it the other way then you'll find your shadow noise much much better when you boost it like in those sample pics. I suspect it has something to do with the natural ISO being 50 on the IQ, maybe?