The Great Tetris Challenge
in Olympus E-System , Thursday, May 18, 2006
This is just a quick post, prompted by the release of Iridient RawDeveloper 1.5, and various writings over at the lair of the auspicious dragon...
This is NOT a test. I try things out, but I don't test them. I rarely read the instructions properly, I am not methodic, and I get bored quickly, so for < INSERT DEITY OF CHOICE HERE >'s sake, take it with an enormous pinch of salt.
Anyway, since I was one of the first to spot the famous Tetris effect which showed up E-1 RAW files developed in some software, back in early 2004, I continue to take some interest in the topic. Since Iridient RAW Developer (IRD from now on) is quite well rated by many, including me, I decided to see how well the new version, 1.5, deals with the issue. The task I gave it, and others, was to render a lone poppy captured in a field in the Marches region of Italy some time ago, and which featured in a previous post:
Not too challenging, although it is a touch over exposed, and is only vaguely acquainted with the concept of "in focus". Here, without further fanfare, is how IRD, CaptureOne Pro v3.7.4, Adobe Camera Raw v3.3 and Olympus Studio v1.5 coped. Please ignore the colours. I did back off the exposure a bit, but I certainly didn't bother with anything else. Sharpening is application default in all cases.
First up, RAW Developer 1.5
Ouch.
Let's see how the Danish entry performs: CaptureOne Pro 3.7.4, come on down.
Hmm. Danemark, nulle points. (for readers who don't get that, substitute "Ouch"). Back of the classroom for C1.
Ok, how does Thomas Knoll's meisterwerk manage ? Over to you, Adobe Camera RAW 3.3
A bit better, I think, but what gives with these blotches ?
Just for fun, let's see how iPhoto, and hence CoreImage, and hence, sort of, maybe, Apple Aperture would manage;
Hmm. A bit pink, Steve, but once you get over that, look, no Tetris! (Later edit: and look at those strange edge effects. iPhoto sharpening I suppose)
Finally, over to the Great Ninja itself, Olympus Studio 1.5
As expected, no Tetris here (the colour is pretty much spot on too...)
So... what ? Well obviously if I was exclusively printing photos of poppies at 200% (did I mention they're all at 200% ? No ? Ok, they're all at 200%), I'd obviously not be straying too far from ACR (or Studio if I had infinite time on my hands). I'm pretty much a CaptureOne fanboy, and whilst clearly one needs to keep an eye on things when their is a lot of red about, it still floats my boat. What I don't like about C1 is the enforced organisation of files into sessions, and IRD actually is extremely (even rashly) flexible in that area. It is actually far easier to use IRD with iView MediaPro as the catalog / browser than it is to use C1 in that way, despite industry alliances.
I'll probably stick with CaptureOne for now, mainly because I'm stubborn, but also because I find there's a little too much knob twiddling potential in IRD, and some of it is a touch opaque. For example, it is very nice having two extremely clever sound new sharpening methods, but just a few paragraphs of explanation to go with them, especially on the sliders, would be a comfort. And in Studio, there isn't enough knob twiddling (no black point adjustment, for example), and the inscrutability of some of the controls certainly reinforces cultural stereotypes. As for the usability, and did I mention the lack of EXIF in 16 bit TIFFs ? Oh yeah, I did... Well, both stink.
So, as far as the Tetris effect goes, it looks like Olympus Studio remains the boss.
Not too challenging, although it is a touch over exposed, and is only vaguely acquainted with the concept of "in focus". Here, without further fanfare, is how IRD, CaptureOne Pro v3.7.4, Adobe Camera Raw v3.3 and Olympus Studio v1.5 coped. Please ignore the colours. I did back off the exposure a bit, but I certainly didn't bother with anything else. Sharpening is application default in all cases.
First up, RAW Developer 1.5
Ouch.
Let's see how the Danish entry performs: CaptureOne Pro 3.7.4, come on down.
Hmm. Danemark, nulle points. (for readers who don't get that, substitute "Ouch"). Back of the classroom for C1.
Ok, how does Thomas Knoll's meisterwerk manage ? Over to you, Adobe Camera RAW 3.3
A bit better, I think, but what gives with these blotches ?
Just for fun, let's see how iPhoto, and hence CoreImage, and hence, sort of, maybe, Apple Aperture would manage;
Hmm. A bit pink, Steve, but once you get over that, look, no Tetris! (Later edit: and look at those strange edge effects. iPhoto sharpening I suppose)
Finally, over to the Great Ninja itself, Olympus Studio 1.5
As expected, no Tetris here (the colour is pretty much spot on too...)
So... what ? Well obviously if I was exclusively printing photos of poppies at 200% (did I mention they're all at 200% ? No ? Ok, they're all at 200%), I'd obviously not be straying too far from ACR (or Studio if I had infinite time on my hands). I'm pretty much a CaptureOne fanboy, and whilst clearly one needs to keep an eye on things when their is a lot of red about, it still floats my boat. What I don't like about C1 is the enforced organisation of files into sessions, and IRD actually is extremely (even rashly) flexible in that area. It is actually far easier to use IRD with iView MediaPro as the catalog / browser than it is to use C1 in that way, despite industry alliances.
I'll probably stick with CaptureOne for now, mainly because I'm stubborn, but also because I find there's a little too much knob twiddling potential in IRD, and some of it is a touch opaque. For example, it is very nice having two extremely clever sound new sharpening methods, but just a few paragraphs of explanation to go with them, especially on the sliders, would be a comfort. And in Studio, there isn't enough knob twiddling (no black point adjustment, for example), and the inscrutability of some of the controls certainly reinforces cultural stereotypes. As for the usability, and did I mention the lack of EXIF in 16 bit TIFFs ? Oh yeah, I did... Well, both stink.
So, as far as the Tetris effect goes, it looks like Olympus Studio remains the boss.
1 comment
Bernard Frangoulis May 18, 2006 - 10:20This is what Brian Griffith, the author of IRF, says about the R-L Deconvolution algorithm: "R-L Deconvolution is Richardson-Lucy, more an image reconstruction or re-focusing technique than a traditional edge contrast enhancing sharpening technique like the other 3 options. Richardson-Lucy gained some "fame" in use correcting early images from the Hubble space telescope. Most digital cameras have an anti-aliasing filter which basically blurs the image a bit, in theory R-L should be good at "undoing" this. It doesn't tend to emphasize edges a much as the others though and it can be very, very slow if the iterations are set high." In my (so far limited) experience, a setting of Radius = 1 and Iterations = 10 does OK with E-1 photos.
As for C1, I find that with version 3.7.4 the Tetris effect is more pronounced than with version 3.7.3, simply because Phase One has reduced the "base level" of Noise Suppression. With Noise Suppression at "Low", and Sharpening (standard) at 150/3 or so, the effect is almost always present, at least in the reds and skin tones. When increasing the Noise Suppression one notch up from Low, the Tetris effect disappears, at the expense of some sharpness. Checking the new "Pattern noise Suppression" checkbox does remove the Tetris effect as well, but sometimes makes it appear where it was not visible before, particularly at edges.
All in all, I like IRD very much, and am switching from C1 to IRF as my raw developer of choice.
-- Bernard