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 | Alternative Gear & Lenses | Join Upload & Sell

1       2       3              26      
27
       28              43       44       end
  

Archive 2011 · Still no love for the Ricoh GXR?
  
 
douglasf13
Offline
• • • • •
Upload & Sell: Off
p.27 #1 · p.27 #1 · Still no love for the Ricoh GXR?


Hi. Could you explain what do to achieve curve reversals? Thanks!


Feb 09, 2012 at 01:17 AM
theSuede
Offline
• • • •
Upload & Sell: Off
p.27 #2 · p.27 #2 · Still no love for the Ricoh GXR?


First: Thank you denoir for your work. I've seen lots of your comparison exercises, and they are generally very well executed, and illustrate differences very well :-) I would spend just a few minutes more on making the image contrasts and colour more equal between frames though, if you care for some constructive criticism. If not, ignore me I know how much time it takes to execute those kind of comparisons.

Douglas, I get sidetracked all to easily...
Will try to do a reversal /curve adjusted sharpening tutorial soon. It's not to hard, it's three steps in PS, but I feel that some background knowledge is imperative.



Feb 09, 2012 at 01:41 AM
denoir
Offline
• • • •
Upload & Sell: Off
p.27 #3 · p.27 #3 · Still no love for the Ricoh GXR?


theSuede wrote:
Thought you had the 5N? No matter.


I've had both, kept the C3 for its size and don't use it at all It's my worst case scenario travel backup M-mount camera.


We are now comparing a 400 (four hundred) € camera to a 6000 (six thousand) € camera.


Indeed, but we're also comparing 18 vs 7 megapixel. Plus, the lens can't be ignored - a 50 Lux ASPH IIRC - which adds quite a bit to the price of the C3. So the performance/price ratio isn't quite as dramatic as one would initially think.

If we look from it the other way, this is the M9 shot uprezzed to ~36 megapixel vs the C3's 16 megapixel:








theSuede wrote:
First: Thank you denoir for your work. I've seen lots of your comparison exercises, and they are generally very well executed, and illustrate differences very well :-) I would spend just a few minutes more on making the image contrasts and colour more equal between frames though, if you care for some constructive criticism. If not, ignore me I know how much time it takes to execute those kind of comparisons.


You're welcome as is the constructive criticism In this case the defense is that the original test was supposed to show the results with as little post processing as possible. Once you go down that road, it's like you say - the guy with the wacom decides the outcome



Feb 09, 2012 at 01:43 AM
michaelwatkins
Offline
• • •
Upload & Sell: Off
p.27 #4 · p.27 #4 · Still no love for the Ricoh GXR?


Minor detour from the interesting conversation:

Ricoh reshuffle sees Pentax take control of cameras
http://www.fredmiranda.com/forum/topic/1084195

As you were!



Feb 09, 2012 at 04:36 AM
kosmoskatten
Offline
• • • •
Upload & Sell: Off
p.27 #5 · p.27 #5 · Still no love for the Ricoh GXR?


Since Denoir is comparing sensor technology and not price, I think the comparison is interesting. Leica prices are NEVER favorably comparable.

In all of the above shots I prefer the M9, the scaled C3 shot looks crude and strained in comparison.

MW: will check link.



Feb 09, 2012 at 05:25 AM
edwardkaraa
Online
• • • • •
Upload & Sell: Off
p.27 #6 · p.27 #6 · Still no love for the Ricoh GXR?


Totally agreed with you, Mitch! That's the way I see it too, even though I still have some unanswered questions about the true gender of the angels

Mitch Alland wrote:
I have been following the above discussion on AA filters with as much interest as would a medieval plowman a canonical disputation on how many angels can stand on the head of a pin. Also, a joke comes to mind about a man who comes home and finds his wife naked in bed with another man: she says, "Darling, what are you going to believe what I tell you or your eyes?"

I believe my eyes. The bottom line for me is that files from the M8.2, M9 and GXR-M look substantially better and print better than files from
...Show more



Feb 09, 2012 at 07:07 AM
Mitch Alland
Offline
• •
Upload & Sell: Off
p.27 #7 · p.27 #7 · Still no love for the Ricoh GXR?


Of course I should have written with more precision: the comparison that I have seen with my eyes is the much better resolution (and color accuracy) of the GXR-M over the A12/28 and and A12/50 camera units (i.e. no AA filter vs AA filter) which, as a result, I sold a week after getting the GXR-M.

—Mitch/Pak Nam Pran
Pak Nam Pran



Feb 09, 2012 at 07:59 AM
Toothwalker
Offline
• • •
Upload & Sell: Off
p.27 #8 · p.27 #8 · Still no love for the Ricoh GXR?


denoir wrote:
100% Crop comparison when the pano is resized to the native size of the M9:
http://peltarion.eu/img/comp/pano/B_comp_crop.jpg


Did the 5D2 fail to register the onset of autumn, or does the M9 deliver irreducible color artifacts in that tree?




Feb 09, 2012 at 04:18 PM
denoir
Offline
• • • •
Upload & Sell: Off
p.27 #9 · p.27 #9 · Still no love for the Ricoh GXR?


Color artifacts, although not irreducible. If you put a gaussian blur filter on the a & b channels in Lab mode they go away.








ACR is really bad at dealing with this automatically while other developers such as C1 do it effortlessly.

Here's another example:








And ACR vs C1:









Feb 09, 2012 at 04:28 PM
kosmoskatten
Offline
• • • •
Upload & Sell: Off
p.27 #10 · p.27 #10 · Still no love for the Ricoh GXR?


I have stopped using ACR entirely as I find it does a poor job with any camera I have files from to work with, AA-filtered or naked.

Toothwalker; for most of the problems that AA filter-less cameras are bashed for, there are solutions that are quite easy, as Denoir showed. I find it easier to deal with the potential artifacts of AA-filter less cameras than some of the inherent mushiness of AA-filtered cameras, on top of the inherent problems of the individual cameras.

But, I also think the M9 does create more artifacts than the GXR M, I'll tell you that. In a short period I managed to find occasions where the M9 would add some artifacts, a little more than I thought. I am not as versed with my Wacom when it comes to dealing with it as Denoir but for the most part it posed no real problem. But I did think it was annoying at times of course.

On the GXR M however, the artifactery is on such a small scale that I have yet to find it a nuisance in any way. Quite the contrary I have found shooting situations with small detail that none of my AA-filtered cameras have coped with, but the GXR M has, like the pearl in a deep pearl metallic color, the minute detail/grit surface in the center of a plastic tweeter where the Nex 5N simply could not reproduce the small detail. Looking at the GXR M files it can easily be regarded as "noise" but when actually looking at the tweeter it is in fact a surface grit with a smooth plastic edge around it. On the GXR M files you can see the grit and the smooth surface, the latter with no noise, and on the 5n files that were shot with the same lens you can't see the detail, it has been effectively killed by the AA filter and cannot be put back together again.

With the GXR M, post pro sharpening is very easy and quite predictable as it ranges from no sharpening to very little, on all subjects and shooting conditions.

With some of the other cameras I have seen or had I think that restoring sharpness is not as easy or predictable. Recombobulating a file from a camera that does introduce mush, with the promise that you can "demush it in post pro" is just about as tiring and I find that even when someone as technically gifted as the Suede has a go at it as per the exercise above, the end result can be quite coarse and I still find the M9 files to be more pleasing. With a lot less work, and despite the inherent drawbacks.

Still, I am not on a crusade against AA-filtered cameras. For the most part I had a really hard time telling which file was with the Nex 5N and which was with the GXR M, they are both that good. I find that I would have been perfectly happy with the 5N performance if the camera had been better in actual use and if there had been no issues with the alt wide angles I have.
Also, I think the NEX 5n handles highlights better and I think the sensor is something I hope Ricoh will wrap their brains around for the coming modules. Yes please.

Recently I have had some time to tinker more with the GXR M files and I find that I am really pleased with the workflow, the imagery and the way the images look. When I find that people are advocating against AA filter-less cameras I have to question it as most of them seem not to have worked with one but back their arguments on technical explanations of how it is "supposed" to work.

Regardless of camera sensor layout we have to put in an effort in post processing, the question is how much, and in what way do we have to do it to get to where we are satisfied.

I was extremely pleased with my Sony A900 files (for lower ISO) and found it to be a helluva camera. Quite a step up in all aspects from the 5D I used to shoot with which I really never grew to like. I am equally pleased with the GXR M as I have been with the Sony A900, which I first thought I would find a bit lacking in comparison. But after some time with the camera I am really happy with it and would not trade any of the traits for a higher MP count, or an AA-filter, if it meant that I would not have the same out of the camera files that I have now.

The GXR M is not a perfect camera, but it does a marvelous job and and has a superb user interface. There is currently no other camera I would trade it for for my kind of work. There are many things I'd like to see improved but none that I feel are forcing me to seek other solutions.

Should I have to shoot more action or a lot of telephoto work I would not reach for the GXR as my primary camera.

Also, I find it kind of disappointing that the new Zoom module hasn't got a manual zoom barrel, but is restricted to the zoom button on the camera, which I have dedicated to other functions.
Why?!



Feb 09, 2012 at 05:05 PM
 

Search in Used Dept. 



theSuede
Offline
• • • •
Upload & Sell: Off
p.27 #11 · p.27 #11 · Still no love for the Ricoh GXR?


Kosmos; the main AESTHETICAL problems with moire and color faults can be "fixed". You can "fix almost anything" at least superficially with clever enough gadgets and some manual hands-on labor. What you can't ever fix is that you are now the owner of an image about as related to reality as Donald Duck.
Aliasing is on a different level, that is a problem that can't even be fixed "aesthetically". It's there, and it stays.

AA filters muck up angle-dependent issues like corner problems and very large aperture problems. Other than this being without an AA filter does with almost no exceptions remove your image from reality by far more than than having an AA filer installed. This has to be weighed against size/ergonomic issues (without an AA filter you can make 60mm and shorter lenses more compact) and corner issues. Depending on what your priorities are, with/without is as I said before - a personal choice.

If someone "prefers" the AA-less image, fine. I prefer red to white - no matter. Claims like "closer to reality" and "more detail" and "higher resolution" and so on makes me itch however. Because they're simply not true - quite the opposite.



Feb 09, 2012 at 11:06 PM
Mitch Alland
Offline
• •
Upload & Sell: Off
p.27 #12 · p.27 #12 · Still no love for the Ricoh GXR?


Henrik,

I pretty much agree with your view of the GXR-M, although I would like to stress two aspects First, the way it is transparent to the lens used, showing the same characteristics of the lens that one is familiar with from film use, excepting of course the aspects of the outer edges eliminated through the 1.5x crop factor — that being the only reason for which I want a full frame camera.

Second, the color accuracy that substantially exceeds that of the M8 and the M9: generally the M8 and M9 files require more post-processing than those of the GXR-M. Sean Reid had said that M9 files have some more richness in the mid-tones, but this is not something that I have noticed. While I have never shot with the a NEX-5N or a NEX-7, the examples of shots that I see on the web seem to me not to have colors that I like as much as those from the GXR-M, but I have no idea whether that stems from the NEX files themselves or from the photographers.

My conclusions are based on "developing" DNG files with RPP, which, as I written before, I find to the best raw developer.

—Mitch/Pak Nam Pran
Wild Beasts of Botwsana



Edited on Feb 09, 2012 at 11:23 PM · View previous versions



Feb 09, 2012 at 11:16 PM
denoir
Offline
• • • •
Upload & Sell: Off
p.27 #13 · p.27 #13 · Still no love for the Ricoh GXR?


theSuede wrote:
Claims like "closer to reality" and "more detail" and "higher resolution" and so on makes me itch however. Because they're simply not true - quite the opposite.


It isn't a black and white issue. First, what is reality? A photo certainly isn't an accurate representation - if nothing else, it misses one dimension. We're talking about a pretty arbitrary projection on a planar surface with a huge number of factors affecting the end results - the AA filter (or lack thereof) being one of the least significant ones.

More detail is another complex issue - yes, you do get more detail but it also includes false detail. Then again, what details in a digital photo are not false? A pattern sampled below the nyquist limit is false detail in a very theoretical sense, but in practice since none of it is an accurate representation of reality the issue becomes purely academic - and as you mention, a question of personal preference.

You do of course get higher resolution if you don't low-pass the signal - which the AA filter does. At the expense of artifacts, some that most people find annoying (color artifacts) and some that are a question of personal preference (the "grainy digital" look).

Ultimately it boils down to whether you more prefer low-pass filtered mushy rendering or crisp with artifacts. For practical use of the images I don't really think it makes much difference. It's mostly what you are used to seeing when looking at 100% and if you prefer mushy rendering or artifacts. I'm no print expert, so I'm not going to say anything about that except that the largest I print is ~100cm x 70cm (poster size) and I would probably not be able to tell my M9 from my 5DII shots based on image quality. I think the printing process affects the quality much more than AA vs no AA.


Edited on Feb 09, 2012 at 11:24 PM · View previous versions



Feb 09, 2012 at 11:20 PM
Mitch Alland
Offline
• •
Upload & Sell: Off
p.27 #14 · p.27 #14 · Still no love for the Ricoh GXR?


theSuede wrote:
.
...Aliasing is on a different level, that is a problem that can't even be fixed "aesthetically". It's there, and it stays...If someone "prefers" the AA-less image, fine. I prefer red to white - no matter. Claims like "closer to reality" and "more detail" and "higher resolution" and so on makes me itch however. Because they're simply not true - quite the opposite.


This is the "how many angels can stand on the head of a pin" aspect that i referred to, because it's not what my eyes tell me.

—Mitch/Pak Nam Pran
Pak Nam Pran


Edited on Feb 09, 2012 at 11:30 PM · View previous versions



Feb 09, 2012 at 11:22 PM
Mitch Alland
Offline
• •
Upload & Sell: Off
p.27 #15 · p.27 #15 · Still no love for the Ricoh GXR?


...Ultimately it boils down to whether you more prefer low-pass filtered mushy rendering or crisp with artifacts. For practical use of the images I don't really think it makes much difference. It's mostly what you are used to seeing when looking at 100% and if you prefer mushy rendering or artifacts. I'm no print expert, so I'm not going to say anything about that except that the largest I print is ~100cm x 70cm (poster size) and I would probably not be able to tell my M9 from my 5DII shots based on image quality. I think the printing process affects...Show more

My feeling is that the difference does comes through in prints of 18x24 and 24x36 inches (about 60x90cm), which are sizes I often print.

—Mitch/Pak Nam Pran
Scratching the Surface



Feb 09, 2012 at 11:30 PM
kosmoskatten
Offline
• • • •
Upload & Sell: Off
p.27 #16 · p.27 #16 · Still no love for the Ricoh GXR?


theSuede wrote:
Kosmos; the main AESTHETICAL problems with moire and color faults can be "fixed". You can "fix almost anything" at least superficially with clever enough gadgets and some manual hands-on labor. What you can't ever fix is that you are now the owner of an image about as related to reality as Donald Duck.
Aliasing is on a different level, that is a problem that can't even be fixed "aesthetically". It's there, and it stays.

AA filters muck up angle-dependent issues like corner problems and very large aperture problems. Other than this being without an AA filter does with almost no exceptions remove
...Show more


I don't get that comment at all, that an AA-filterless camera image is as close to reality as Donald Duck. Since we are talking artifacts on a small level in both instances (AA vs non AA) I think you are taking it to the hilt unnecessarily. That is a childish comment in a serious thread.

The "closer to reality" is actually not something taken out of the blue, with that I mean I come closer to reality from the get go, not after thirty minutes of tinkering. The end result may very well be exactly the same and I haven't made any claims that other cameras are NOT close to reality in the final image. I just find that I get there a little bit easier with the GXR M than what I have compared it to, which is Leica, Canon, Fuji, Nikon, Panasonic and Olympus. That's all.

My experience contradicts your statement about having to make a lot of hands on labor with AA filter-less, and I use no gadgets other than a decent RAW converter. I am wondering if you have even seen a GXR M RAW file? I don't think you have as your comments are not holding water.

I agree that claims that something is radically better is stuff that make you itch. It makes me itch too. It really does. That is a hard boiled claim that you make though. Especially since I find the color accuracy excellent, much more so than from say, Canon, which I never came to terms with. Canon and Fuji colors are far more "Mickey Mouse" in comparison.

An aliased image that looks great in print I don't see as a problem. Things that can't get fixed are, of course. The tiny bit of extra detail I have seen in my comparisons are, in my experience, probably not visible in print, but it does not degrade from the image either, when working with it.
Moiré is the single biggest problem with AA filter-less cameras but I have much less trouble coping with it than you try to make it out to be. The GXR M is very forgiving in that regard and even shots with very minute patterns have come out not just fine, but excellent.

Again, I find that making aesthetical images, fairly true to reality, are not that difficult with any of the cameras I have used. If you have trouble doing so you need to work on your post processing and printing skills.

My next camera might or might not have an AA-filter. It is only part of the equation, as I try to see the sum of all parts in relation to money thrown in the pit. I threw money in the Leica pit, got cold feet and managed to get that back. The amount of $ I have thrown into the GXR M is far less than any other dSLR (or mirrorless) I have had prior to it and given the limitations of the system I find it is the best digital investment I have made so far.

In the future I will not red flag my next potential camera based on it having an AA-filter or not.

It will be interesting to see what the Nikon D800 vs D800E turns out to be like in practical reality. I don't think the D800E will produce Donald Duck like images just because it lacks the AA-filter of its sibling, nor do I think that the D800 will be seriously outperformed by the D800E. But Nikon obviously felt that something warranted two versions. Perhaps just hyperbole and market demands, but let's not dismiss either camera just yet.




Edited on Feb 10, 2012 at 05:17 PM · View previous versions



Feb 10, 2012 at 04:58 PM
AhamB
Offline
• • • • •
Upload & Sell: On
p.27 #17 · p.27 #17 · Still no love for the Ricoh GXR?


@Henrik: I think what TheSuede/Joakim is saying is his view as an engineer (not so much as a photographer).


Feb 10, 2012 at 05:14 PM
kosmoskatten
Offline
• • • •
Upload & Sell: Off
p.27 #18 · p.27 #18 · Still no love for the Ricoh GXR?


Yes, I have thought so all along. I just disagree with some of his conclusions, as presented by him, as I find that in pure practical terms the technology as implemented in both camps is sufficient and working - with neither being perfect.

From a photographers stand point that is.

Digital has come a long way the last couple of years. I remember not too long ago when I went to a press photo exhibit that had been printed digitally by a respectable photo finishing lab and I was sickened by the lack of color fidelity, the lack of resolution and the apparent digital look. I thought that it was maybe "just me" and didn't mention it but I had some friends that were there that weren't photographers and even they were commenting that they thought the pictures looked "false". I thought they looked like pure crap, all of them, so much that I even lost interest in the actual images and what they were trying to represent. But that was an example from the bottom of the barrel.

Things have come a long way since then, thank goodness.

The digital transition was not smooth, it was stair stepped. That's not just a fact, its an artifact.



Feb 10, 2012 at 05:31 PM
douglasf13
Offline
• • • • •
Upload & Sell: Off
p.27 #19 · p.27 #19 · Still no love for the Ricoh GXR?


I don't want to speak for Joakim, but I believe that he is just saying that no AA-filter isn't any closer to reality than having an AA filter, which I'd agree with. Which looks better is a personal choice.

Honestly, I'm sitting here looking at raws from the GXR and 5N with the same lens and scene (thanks Sam) in Lightroom 4, and they're pretty close when I make some contrast and detail panel adjustments. The GXR may have a slight advantage, but it also has noticeable artifacts in some spots, but I could probably fix that in LR4 for the most part, too. However, that is all at the center of the frame. With this Cv 15/4.5, the GXR clearly has less vignetting and color shift, and a little more detail than the 5N file at the edges, and that is the point of removing the AA filter in this camera. Ricoh didn't suddenly decide that no AA filter was always better, but they made a trade off for edge performance. Heck, Fuji is going to great CFA lengths to do the same thing with possibly less artifacts, because they want to yank the AA for edge detail, too. It isn't as if all these camera companies are all adding a relatively expensive AA filter in most of their cameras for no reason. I'm sure they'd love it if we all wanted the AA to be yanked and accept the artifacts. It'd be cheaper for them!

I want to yank my AA, but that is because I'm using rangefinder lenses. The choice would be less clear for me, otherwise. Of course, the NEX-7's pixel density may go a long way in reducing artifacts, so that could be the best candidate for MaxMax...if they'd return yet another email that I sent them.




Feb 10, 2012 at 05:46 PM
curious80
Online
• • •
Upload & Sell: Off
p.27 #20 · p.27 #20 · Still no love for the Ricoh GXR?


theSuede wrote:
...... Other than this being without an AA filter does with almost no exceptions remove your image from reality by far more than than having an AA filer installed. ....


That would really depend on the frequency content of the scene. Those parts of the scene which have data at greater than nyquist frequency will obviously get aliasing and false information if there is no AA filter. However the argument in favor of AA-less cameras has to do with parts of scene which have content @ less than nyquest frequency. Lets say we have an image area with content juts below nyquist. The AA-less camera will be able to sample it properly without aliasing and capture all the detail. The camera with AA will blur away some of the high frequency content and get less detail. So in this case AA-less camera will get extra "real" detail.

Obviously if you could have an ideal low-pass filter then this won't be an issue but real low-pass filters are not ideal. I don't know the frequency response of the AA filters used in the cameras, however they look like simple averaging filters. Averaging filters are rather mediocre low-pass filters and the filtering will likely have to be rather strong to sufficiently eliminate aliasing. That means blurring away a lot of frequencies below nyquist which could have been captured properly in the AA-less system.

So at the end of the day it is a compromise. If you remove the AA, you will get more detail in parts of the scene and false data due to aliasing in other parts.



Feb 10, 2012 at 09:47 PM
1       2       3              26      
27
       28              43       44       end




FM Forums | Alternative Gear & Lenses | Join Upload & Sell

1       2       3              26      
27
       28              43       44       end
    
 

You are not logged in. Login or Register

Username   Password    Reset password