"The Nikon D3x offers the finest image quality in a DSLR the world has yet seen"

[Ray]

Ray, nearly any raw converter...rpp, raw therepee, rawmagick, bible, capture one, etc is better than ACR, outside of highlight recovery. I'm using C1 currently.  ACR is still fine for most, but if you care about the slight nuances between these cameras, using acr is contradictory.

Wow! You've actually compared all those? Did it take long?

If you admit that ACR is better at highlight recovery, then it's difficult to claim that the others are better at shadow recovery since you would have to give a shorter exposure to avoid blowing highlights. The shorter exposure cause the shadows to be noisier so the other converters need to be better at shadow recovery to achieve the same result as ACR.

In what respect are these other converters better than ACR? Color accuracy? Sharpness and detail? Can you show some example images?

[dwdallam]

Okay, I think you're incompetent too -- no offense intended.

The problem with the word is that like "pitiful" its vernacular usage has deviated from its actual definition. As such, people tend to think "pitiful" is a derogatory or put down remark, when it's actually a form of compassion. You're using the word here, in my case, to show how it can be used a a rude remark, but only if you take it in the wrong way in the first place.  Of course you have also used the word out of any context, unlike Lepig, because you have not stated in which way--and proved it--that I am "incompetent."

 In other words, you need to show in which way someone is incompetent, which goes beyond your subjective point of view, which is what Lepig was trying to explain using scientific criteria. If you do not state in which way, the first intelligent response to any assertion of incompetence would be, "In which way?" Lipeg used the word correctly in that he applied it contextually to faulty research and methodology, and his assumption is that the person was not up to the task. There is another option in that the person could have been up to the task, but simply made methodological errors. Either way, it's not meant a a put down the way Lepig used the word, and he used it quite correctly.

I think the word you are confusing with "incompetent" is "charlatan."

Incompetent: 1. not legally qualified 2. inadequate to or unsuitable for a particular purpose 3. lacking the qualities needed for effective action 4. unable to function properly

Charlatan: 1. quack 2. one making usually showy pretenses to knowledge or ability 3. fraud 4. faker

Similar, but very different.

In either case, make sure to back up any assertions of incompetence (as Lepig did) or a charlatanism, because if you don't, you may diminish your own credibility.

[OldScotch]

To my eyes the A900 knocks the spots off the 'horrible' film images. What the author takes as extra detail in the Velvia image is in fact noise. Add speckle noise to the A900 image, and it will 'improve' it to the level of the Velvia one.

No doubt, in these tests the a900 outclasses the films. My point was that they're still competing even at the 24 mp level, and reasonably will outresolve the 6, 10 and 12  sensors.  

Ah, found the old thread on dpreview - http://forums.dpreview.com/forums/readflat...4253&page=1

Naturally it eventually descended into petty bickering, and at best, film really only competes with a900 resolution under perfect conditions with select, very low iso film and a perfectly calibrated drum scanner. And even then, the film exhibits more grain.

I'm not arguing that 35mm film is better than the d3x/a900, the only thing film would have against these beasts is dynamic range - it's interesting though that there is still a comparison. And I've no doubt that it will easily outresolve those lower mp sensors.

[Slough]

No doubt, in these tests the a900 outclasses the films. My point was that they're still competing even at the 24 mp level, and reasonably will outresolve the 6, 10 and 12  sensors.  

Ah, found the old thread on dpreview - http://forums.dpreview.com/forums/readflat...4253&page=1

Naturally it eventually descended into petty bickering, and at best, film really only competes with a900 resolution under perfect conditions with select, very low iso film and a perfectly calibrated drum scanner. And even then, the film exhibits more grain.

I'm not arguing that 35mm film is better than the d3x/a900, the only thing film would have against these beasts is dynamic range - it's interesting though that there is still a comparison. And I've no doubt that it will easily outresolve those lower mp sensors.

You make very valid points, but is resolution the best measure of IQ? I would argue that the A900 is easily outresolving the film in your tests, though I do agree that the film is doing well in that regard. But in terms of IQ, I find the film horrible. (Subjective I know.) I did my own comparison here:

Velvia 100F versus D200

Take it with the measure of salt intended. How does your film compare to Velvia 100F?

I find it incredible how far we have come in less than 10 years.

"Naturally it eventually descended into petty bickering"

A dpreview thread descended into bickering? Crikey, now there's a first ...

[OldScotch]

You make very valid points, but is resolution the best measure of IQ? I would argue that the A900 is easily outresolving the film in your tests, though I do agree that the film is doing well in that regard. But in terms of IQ, I find the film horrible. (Subjective I know.) I did my own comparison here:

Velvia 100F versus D200

Take it with the measure of salt intended. How does your film compare to Velvia 100F?

I find it incredible how far we have come in less than 10 years.

"Naturally it eventually descended into petty bickering"

A dpreview thread descended into bickering? Crikey, now there's a first ...

Abosolutely not, resolution is significant, but hardly defines IQ. Overall, the very clean grain and iso performance of digital hold such a significant advantage over film, that it's easy to compromise with dynamic range and compete with resolution. That's not getting into colour, saturation, contrast, etc that can get into a whole other discussion  

[douglasf13]

Wow! You've actually compared all those? Did it take long?

If you admit that ACR is better at highlight recovery, then it's difficult to claim that the others are better at shadow recovery since you would have to give a shorter exposure to avoid blowing highlights. The shorter exposure cause the shadows to be noisier so the other converters need to be better at shadow recovery to achieve the same result as ACR.

In what respect are these other converters better than ACR? Color accuracy? Sharpness and detail? Can you show some example images?

  I'm specifically referring to detail, sharpness, and noise "blotchiness..." really the whole demosaicing process.

  Regardless, here is an interesting thread on RAW processing, and it's the direction I plan on taking.

  flat raw conversion

[Ben Rubinstein]

I think that by the time you are quibbling over which RAW converter to use just to show ANY difference between the cameras, the choice is no longer pure resolution, it's systems, usage, etc. You might be willing to live with a lesser body/system for superior IQ (the 5D in it's heyday) however when the superior IQ is so hard to get at that it's pretty much only visible with specific converters which may have an inefficient workflow in comparison anyway, the ability to use better lenses, weather proofing, superior AF, dual card, etc should be far more important a consideration than an extra 3% better resolution possibly visible using specific software only. The choice between a 1Ds mkIII or a D3x is far more the choice of system than a possible insignificant difference in resolution. Ditto the 5D mkII vs the Sony.

[ErikKaffehr]

Hi,

A couple of comments.

1) Slide film has very narrow DR about 5 steps IMHO, that's much less than digital which is the nine+ steps of territory.
2) Drum scanning is not exactly cheap. Top class MF slide scanners are quite expensive.
3) With digital and raw format it is possible to emulate almost any color (within the limitation of the color filter array and its spectral characteristics).

Best regards
Erik

Abosolutely not, resolution is significant, but hardly defines IQ. Overall, the very clean grain and iso performance of digital hold such a significant advantage over film, that it's easy to compromise with dynamic range and compete with resolution. That's not getting into colour, saturation, contrast, etc that can get into a whole other discussion  

[douglasf13]

I think that by the time you are quibbling over which RAW converter to use just to show ANY difference between the cameras, the choice is no longer pure resolution, it's systems, usage, etc. You might be willing to live with a lesser body/system for superior IQ (the 5D in it's heyday) however when the superior IQ is so hard to get at that it's pretty much only visible with specific converters which may have an inefficient workflow in comparison anyway, the ability to use better lenses, weather proofing, superior AF, dual card, etc should be far more important a consideration than an extra 3% better resolution possibly visible using specific software only. The choice between a 1Ds mkIII or a D3x is far more the choice of system than a possible insignificant difference in resolution. Ditto the 5D mkII vs the Sony.

  Very good point.  I'm looking at it in similar terms.  By the time you get to the point about quibbling between the difference of theses cameras, choice of RAW converter and lens, and overall RAW technique, has as much to do with IQ as the camera itself.

  Interestingly enough, I've been reading some stuff about lens DR, which I'd never really considered.  Apparently, many lenses tend to top out in DR around 11 stops, but I'm new to this stuff and have little understanding.

[Tony Beach]

In either case, make sure to back up any assertions of incompetence (as Lepig did) or a charlatanism, because if you don't, you may diminish your own credibility.

Well, I don't think Lepig did back up his assertions of incompetence and I couldn't care less what you think of my credibility; I have been ignoring you for a long time in these forums (I click on replies from people on my "Ignore User" list to see what they are saying about me and if my name doesn't come up I don't read them).  I'm not going to go searching for the reason why I started ignoring you, but it was obvious to me when I did it that you had nothing to offer me, just as I now consider your pedantic defense of Lepig a lot of BS and a waste of my time to read.  I backed up my critique of Lepig by linking to some bizarre claim of 300% improvement he claimed for the D3x over its competitors, so by your own reasoning I was not rude either.  Write whatever you want in reply, but realize that I will not be reading it as I'm done caring what you write about me in this thread.

[eronald]

Enough. You people have the collective IQ of a clam. Let me be clearer: Individually you are all nice guys and bright cogent practitioners of the art of photography, and i have the highest respect of your skills, far superior to mine. Together you are a bunch of querellous tykes, and you spout insulting nonsense. Go home, get a drink. Kiss your wife and tell her you love her and she makes you happy every day. Get a life. Stop polluting this forum.

This was a public service broadcast. Thank you for tuning in.

Edmund

Well, I don't think Lepig did back up his assertions of incompetence and I couldn't care less what you think of my credibility; I have been ignoring you for a long time in these forums (I click on replies from people on my "Ignore User" list to see what they are saying about me and if my name doesn't come up I don't read them).  I'm not going to go searching for the reason why I started ignoring you, but it was obvious to me when I did it that you had nothing to offer me, just as I now consider your pedantic defense of Lepig a lot of BS and a waste of my time to read.  I backed up my critique of Lepig by linking to some bizarre claim of 300% improvement he claimed for the D3x over its competitors, so by your own reasoning I was not rude either.  Write whatever you want in reply, but realize that I will not be reading it as I'm done caring what you write about me in this thread.

[Ray]

I'm specifically referring to detail, sharpness, and noise "blotchiness..." really the whole demosaicing process.

  Regardless, here is an interesting thread on RAW processing, and it's the direction I plan on taking.

  flat raw conversion

Don't ask me to wade through the posts at dpreview. There are sufficient here   .

The idea that a particular converter can do a better de-mosaicing job than another, and extract more detail than ACR can, has its appeal. We all want to get the most out of our RAW images.

What I've found in the past when comparing converters is that some converters do indeed produce 'better' results than ACR with a few clicks and slides of the control options. ACR does tend to produce a slightly blotchy result, even at its best, when compared with certain other converters, including Canon's DPP.

However, I get the impression that the reason for the blotchiness may be due to the fact that ACR performs some noise reduction which cannot be switched off. As a consequence, the price to pay for the additional detail that another converter may provide, is greater noise in the shadows and the lower mid-tones. This what I have found with the trial version of Bibble which I downloaded yesterday.

A few years ago I shot a couple of hundred bracketed shots of the city of Brisbane for the purpose of making a gigantic panorama. I used the Canon 100-400 IS at 400mm and F11 with the 20D. The atmospheric turbulence and dust, combined with the slight softness of F11 produced rather flat images that need a lot of sharpening and local contrast enhancement. I put the project to one side. Too much work.

I wondered what Bibble would make of these images. Bibble has an interesting feature called "Perfectly Clear" which optimises contrast and color with one click. It really seems to work.

Basically, I wanted to know if Bibble could extract more detail than ACR from the slightly blurry and blotchy foliage, so I selected one of the 200 shots from my panorama project and converted it in ACR and Bibble with zero noise reduction but maximum sharpening in both converters. The starting positions were the 'Auto' setting in ACR and the 'Perfectly Clear' setting in Bibble. I made further adjustments in Photoshop to get the color and contrast matching as closely as possible.

It seems clear from the results that Bibble is able to extract more detail than ACR and produce more naturally looking highlights, when sharpening is at its maximum. But it's also clear that ACR produces less noise in the shadows and the lower midtones. Any attempt to apply noise reduction in the Bibble software, destroys the detail advantage over ACR. Any attempt to produce smoother highlights in ACR by reducing sharpening, increases the smearing and blotchiness of the foliage.

These are simply different results. Neither is necessarily better than the other, taking everything into consideration. I suppose I should take this experiment one step further and make prints.

Judge the results for yourself.

[attachment=10899:8315_full_image.jpg]  [attachment=10900:8315_com..._foliage.jpg]  [attachment=10901:8315_com...uildings.jpg]  [attachment=10902:8315_com...t_cranes.jpg]

[ErikKaffehr]

Hi,

I downloaded Bibble yesterday, too. Bad news is that it doesn't read my DNGs, neither does DxO, so it seems that a DNG-based workflow is a one way solution.

I didn't have the time to play around with Bibble yet. I have used Bibble earlier and at that time "Perfectly Clear" introduced some artifacts, mostly double contours around bright edges. I have not seen this on the few pictures I checked since yesternight.

Thanks for observations

Erik

Don't ask me to wade through the posts at dpreview. There are sufficient here   .

The idea that a particular converter can do a better de-mosaicing job than another, and extract more detail than ACR can, has its appeal. We all want to get the most out of our RAW images.

What I've found in the past when comparing converters is that some converters do indeed produce 'better' results than ACR with a few clicks and slides of the control options. ACR does tend to produce a slightly blotchy result, even at its best, when compared with certain other converters, including Canon's DPP.

However, I get the impression that the reason for the blotchiness may be due to the fact that ACR performs some noise reduction which cannot be switched off. As a consequence, the price to pay for the additional detail that another converter may provide, is greater noise in the shadows and the lower mid-tones. This what I have found with the trial version of Bibble which I downloaded yesterday.

A few years ago I shot a couple of hundred bracketed shots of the city of Brisbane for the purpose of making a gigantic panorama. I used the Canon 100-400 IS at 400mm and F11 with the 20D. The atmospheric turbulence and dust, combined with the slight softness of F11 produced rather flat images that need a lot of sharpening and local contrast enhancement. I put the project to one side. Too much work.

I wondered what Bibble would make of these images. Bibble has an interesting feature called "Perfectly Clear" which optimises contrast and color with one click. It really seems to work.

Basically, I wanted to know if Bibble could extract more detail than ACR from the slightly blurry and blotchy foliage, so I selected one of the 200 shots from my panorama project and converted it in ACR and Bibble with zero noise reduction but maximum sharpening in both converters. The starting positions were the 'Auto' setting in ACR and the 'Perfectly Clear' setting in Bibble. I made further adjustments in Photoshop to get the color and contrast matching as closely as possible.

It seems clear from the results that Bibble is able to extract more detail than ACR and produce more naturally looking highlights, when sharpening is at its maximum. But it's also clear that ACR produces less noise in the shadows and the lower midtones. Any attempt to apply noise reduction in the Bibble software, destroys the detail advantage over ACR. Any attempt to produce smoother highlights in ACR by reducing sharpening, increases the smearing and blotchiness of the foliage.

These are simply different results. Neither is necessarily better than the other, taking everything into consideration. I suppose I should take this experiment one step further and make prints.

Judge the results for yourself.

[attachment=10899:8315_full_image.jpg]  [attachment=10900:8315_com..._foliage.jpg]  [attachment=10901:8315_com...uildings.jpg]  [attachment=10902:8315_com...t_cranes.jpg]

[dwdallam]

Well, I don't think Lepig did back up his assertions of incompetence and I couldn't care less what you think of my credibility; I have been ignoring you for a long time in these forums (I click on replies from people on my "Ignore User" list to see what they are saying about me and if my name doesn't come up I don't read them).  I'm not going to go searching for the reason why I started ignoring you, but it was obvious to me when I did it that you had nothing to offer me, just as I now consider your pedantic defense of Lepig a lot of BS and a waste of my time to read.  I backed up my critique of Lepig by linking to some bizarre claim of 300% improvement he claimed for the D3x over its competitors, so by your own reasoning I was not rude either.  Write whatever you want in reply, but realize that I will not be reading it as I'm done caring what you write about me in this thread.

Sounds like for some reason you have a lot of anger and hate in your life at this time for some reason. I'm sorry to about that. Good luck to you.

[dwdallam]

[attachment=10899:8315_full_image.jpg]  [attachment=10900:8315_com..._foliage.jpg]  [attachment=10901:8315_com...uildings.jpg]  [attachment=10902:8315_com...t_cranes.jpg]

Seattle--!!!

[Ray]

Seattle--!!!

Brisbane, Australia. Capital of the State of Queensland. Bigger than Texas.  

[dwdallam]

Brisbane, Australia. Capital of the State of Queensland. Bigger than Texas.  

LOL--those cranes looks just like the ones in Puget Sound, Seattle. I can see the difference now!

[OldScotch]

Hi,

A couple of comments.

1) Slide film has very narrow DR about 5 steps IMHO, that's much less than digital which is the nine+ steps of territory.
2) Drum scanning is not exactly cheap. Top class MF slide scanners are quite expensive.
3) With digital and raw format it is possible to emulate almost any color (within the limitation of the color filter array and its spectral characteristics).

Best regards
Erik

1 - True, but negative film still usually has more dynamic range than digital (although digital is catching up very fast). Using Kodak Ektar gives you almost the same resolution of Velvia 50 and the benefit of a negative's much higher dynamic range. I still love those Velvia colours though, heh.
2 - Yes, to do a proper conversion to digital is expensive. The resolution is still there though and available to use.
3 - Also true. With film a shot that's in black and white is in black and white and that's it. And a shot with tungsten film is always going to be a shot with tungsten film, you can't just change it afterward.

I'm not arguing that film is better than digital. In the 35mm world, the advantages of digital now are so strong that it really doesn't make sense except for certain niche shooting styles where people might prefer the film.
It is a misconception though, that the resolution of film was surpassed "a long time ago", that's all I'm trying to say

[lovell]

The Nikon D3X is precisely $28.65 better in image quality then a Canon 5D Mark II.

Therefore anything over $2,728.65 is way too much to pay for the Nikon D3x.  Sure it offers faster, and more features, but if one determines value by image quality alone, then the D3X is > 1000% over priced.

[dwdallam]

The Nikon D3X is precisely $28.65 better in image quality then a Canon 5D Mark II.

Therefore anything over $2,728.65 is way too much to pay for the Nikon D3x.  Sure it offers faster, and more features, but if one determines value by image quality alone, then the D3X is > 1000% over priced.

Yeah that's what I think too. You can buy 3 5D MKIIs for the price of one DX3, and have 500.00 left over to buy equipment. If I were a pro and were worried about image quality plus reliability, I think 3 5D MKIIs would beat one DX3.