« It's Bigger, But Is It Better? Part II | Main | This Inevitable Day »

Sunday, 02 March 2008

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00df351e888f883400e550a1cc828834

Listed below are links to weblogs that reference The Fujifilm FinePix S5 Pro:

Comments

Experimentation is a great key to constant learning in this world. Consider yourself a student of life and knowledge and happiness will follow.

A very interesting read and another vindication of the alternative approach of TOP reviews, thank you.
After a few lines it all began to sound familiar - the raw conversion and colour problems took me back to the days when I struggled with a Fuji S2. Wish I'd stuck with the S1, it only gave tiffs (no raw!) and the underexposure was consistent - looking back now they're some of the nicest looking digital files on my hard drive.

Cheers, Robin

Dear Oren,

A bit of case of locking the barn door after etc., etc., but...

Three ways to deal with errant color, which can be automated.

1) In Photoshop, use a hue/saturation layer and make full use of each of the available color channels.

2) In Photoshop, use Image\Adjustments\Selective Color in much the same fashion as (1).

3) Use the built-in profile builder in Picture Window Pro to construct a custom color profile for your camera from a photo of a Macbeth or IT8 chart.

pax / Ctein

You have not considered the 5D?

And so the search for the perfect camera continues . . .

I always feel uneasy when I see reviews of automatic-mode behaviour; this article is a particularly prominent example. When I see talk of exposure or dynamic-range issues, I instantly wonder whether manual+spot wouldn't have a large effect on the reviewer's outlook; I really worry about talk of changing camera hardware wholesale without mentioning the possibilities of fixing a custom whitebalance such as the underside of a cloud on an overcast day.

Seems this could be a never ending process for you Oren. I hope you find what your looking for.

I found it hard to take serious to be honest. You used the word snapshooting or snapshot so many times I was beginning to think it did not really matter.

I believe your observations to be honest and real but I can't figure out if you really need what your looking for (due to the fact it's all just for snapshots) or if you just like buying cameras.

Hi Oren,

Have you considered the Sony Alpha 700?

I have found it's dynamic range to be excellent. I have used both Nikon and Fuji in the past and wouldn't go back. I have found the Dynamic Range Optimiser on the A700 to be a great benefit to my work. Mike J. was thinking of getting an A700 (don't know if he has yet) why not ask him?

I don't think that you will get comparable results to the dynamic qualities from any current digital camera, but, I think the A700 is worth considering.

Regards,

Chris

You sound like the perfect candidate for Lightroom, which, though it still has the ACR RAW converter, also has close at hand all the color controls you'd need to make your adjustments; and it would clean up your workflow significantly.

I empathize with your point of view on snapshots. When I have time, I go out and "photograph." But I also use cameras in my work, taking record and reference shots. I just want to put the thing on A or P and hammer away and not think about it too much, because I'm thinking about something else. I need good color, focus, and high ISO performance, but don't care about fast sequence shooting. I have a D3, and three days ago, like you, picked up a D300 (for its size) and have great hopes for it. I do carry an Expodisc in my bag for tricky color balance situations, but don't use it much.

JC

Oren, It doesn't sound like you spent much time in the darkroom making color prints. Had you, you would realize that digital is sooooo much better no matter what camera you are using. The magic is in your skills to get the best results, not the camera. Film was good, but never gave you correct colors. No two people will ever see the same colors in the same way.
The science of photography is very large and take many years to learn just a small part of it. After 40 years in the dark room, I have to relearn everything in digital and forget D-76 or C-41 etc. A camera is just a tool, like a hammer and nail--if the nail bends when you hit it do you through the hammer away or the nail, or learn how to use both to get the job done. There will never be a right camera or lens for every job, except and adapt and do the best you can with what you have. Creating great images is not the camera's job, it's yours.

"After twenty-odd years of photographing with Tri-X, I found the dynamic range limitations of the K110D endlessly frustrating."

Are you sure it's the dynamic range that's giving you problems? Sounds like an exposure latitude issue to me. Most testing, and my own personal experience with several DSLRs suggests digital easily matches the dynamic range commonly possible from print film. What I do see is people who are used to the very forgiving exposure latitude of print film having to get used to the more E6-like exposure latitude of digital.

"Are you sure it's the dynamic range that's giving you problems? Sounds like an exposure latitude issue to me. "

What's the difference? What do you call the ability of a camera to record the full range of brightness in a scene?

I'd think that would be dynamic range and latitude would reflect the ease of having slightly different exposures all yield images that don't cut off any part of the scene's brightness.

I think the last few posters are being too harsh on Oren who has a legitimate concern.

Matt (and others) - That is what I thought when reading this article. I have been told that the dynamic range of good sensors is, in most cases, a match for even the best black and white film. Does anyone happen to have a reference for comparison? I couldn't find anything.

Thanks...

David Bostedo

Dear Matt,

Exposure latitude is a consequence of dynamic range. People say a recording medium has "wide latitude" when it has a dynamic range that substantially exceeds the luminance rsnge they'd like to record/print. They say it has short latitude when when it doesn't.

Obviously, exposure latitude will depend upon what you photograph and what tonal placement you expect in that photograph. Dynamic range doesn't; it's a physical characteristic of the system. So it's the appropriate metric for Oren to be using when he's comparing recording media and cameras.

pax / Ctein

Dear Charlie,

Some of us demand a LOT from our "snapshots." Snapshot doesn't have to equate to shoddy image quality.

Me, I find it much less worth my time to make a photograph if the print doesn't give me genuine pleasure (I mean psychophysical, not because of the subject matter).

I used an Olympus Stylus Infinity/Epic for most of the 90's as my snapshot camera. With slow film, it was genuinely good. With fast film, the results were, umm, 'acceptable.'

Eventually I realized that I wasn't getting PLEASURE from those available-light photos. Whereas my Fujica GA 645 did give me pleasurable photos at even ISO 800.

It became my standard snapshot camera, and I started really enjoying making snapshots again. Sold all the little-bitty 35mm's last year-- I hadn't used them in a long time.

Now I admit I'm way off the norm in my standards. Dye transfer is my normal good print. Custom color-balanced, contrast-masked, and dodged-and-burnt 8x10's are 'work' or 'reference' quickies for me. What I expect from a snapshot exceeds what most photographers expect from professional prints.

But it's a snapshot nonetheless.

pax / Ctein

My comments sound more harsh than I meant them to. I was just giving you a hard time Oren.

;-)

Oren, just did a quick test on how many stops of contrast my Canon 40D and my 1Ds could handle using an Eastman Kodak Calabration Step Tablet--The 1Ds looked a little better than 8 stops of contrast range and the 40D a little less--when your in the mud its kind of hard to tell--regular grade #2 B&W paper could handle about 4 to 5 stops before you had to burn and dodge the heck out of the print. I think technically to pull more out of a digital camera, it's going to take more technical work.
I think a silicon cell can only produce 1.3 volts max in sunlight. So half that 8 times and you can see your some where down to .005 volts that has to be amplified to give you an image. The big question is how low can they go before you run out of photons. Also film only gave you more latitude on the top end. Unless you exposed for the shadows and developed for the hi lights, film was about the same as digital when it came to making the print. There are a thousand more tricks with digital so I'll take digital over film any day.

Hello,

As far as dynamic range is concerned, you can get greater dynamic range by taking a series of images at different exposures (2 stops apart, or in some cases 1-and-a-third stops apart) and blending them into a high dynamic range image using Photoshop or, my favorite, Photomatix. I was used to a view camera when I started in digital, and once I started taking several images with my digital camera to blend, it actually began to feel more like using a view camera again. With most DSLRs you can you use the auto-bracketing feature to take the shots pretty easily, though you may at times want to take more exposures than auto-bracket is set up for, and Photomatix makes the blending pretty painless, a gives you more options than PS.

Something must have been set up wrong. I've been using the S5 for a year, since it arrived, and I've never had any color issues at all. And in fact, I believe it produces far nicer skin tones than any camera on the market. And I find that almost any camera tends too warm in common incandescent lighting (typically about 2700 K). For starters, are you sure you weren't using one of the "film simulation" modes? Also, it could be the actual saturation ("color") setting was set higher than normal. Anyway, my experience with the camera is polar opposite to yours when it comes to image quality. And indeed, in my own tests, the D300 only looks just as good as the S5 in spite of it's newer vintage.

I can't offer any help but I do have a question/comment to toss 'out there'. Why do Fuji, with all their excellent lens-making and camera making know-how use Nikons as a platform? I've used there middle-format film rangefinders and their point-and-shoots and have big respect for them. If Hasselblad uses Fuji for their platform (AND lenses) why do Fuji use Nikon? Is it just a way to market to a larger professional base?
Thanks,
M

Oren can speak for himself here, but perhaps I should have asked him to make the following clearer: his normal medium is large format (whole plate and up). So for him, "real" or "serious" photography (or some equivalent term that's not loaded, if there is one) is large format black-and-white. It's in that context that a digital camera is a "snapshot" camera. It's not a judgment of the camera, as much as a description of what he does with it.

Mike J.

Oren, I think you gave up too early. Use Lightroom and calibrate your camera as described here, the difference is remarkable:
http://www.luminous-landscape.com/reviews/accessories/dig-calib-profil.shtml

A friend of mine and I run a K10D, E-1 and 1DSMKII through it and each of these benefit from the calibration, especially where subtle hues are concerned, like the last rays of sunset light on a white wall for example.

Ctein, 'Some of us demand a LOT from our "snapshots."' is going to be my sentence of early 2008. Have to applaud you!

Dear David & Carl,

FWIW, my Fujifilm S6000fd checks out about the same at ISO 100-- 8 or so stops of dynamic range.

Thing is, while that's decent for color slide films, it's not good compared to B&W neg films, which can record several more stops (color neg does even better).

Also, the luminance range that can be printed on a typical Grade 2 paper from a standardly-developed negative is 8-9 stops, not 4-5.

Which about matches the dynamic range of the sensors we've been talking about... except...

-- the sensor clips the extremes in a way that film doesn't (and that is more usually visually unacceptable).

-- you can print on other grades of paper.

-- you can change the development of the film.

-- even modest amounts of dodging and burning in will add 2-3 stops to the luminance range you can accomodate.

Now, the best cameras out there do a lot better than a mere 8 stops; some of them hit 11-12 stops, and those would satisfy most B&W film users. But 8 stops? They're gonna keep bumping their noise on the luminance wall, and it hurts!

pax / Ctein

I've been using a Fuji S5 Pro as a backup camera to my Nikon D2X for professional landscape photography for almost a year, and find it to be an outstanding camera with great color reproduction and superb dynamic range, albeit with less resolution than the 12MP D2X. The Fuji S5 Pro offers a variety of color settings in-camera, and as several others have commented, you can always adjust the color later in Photoshop or Lightroom.

As for the "false color artifacts in edges adjacent to bright backgrounds", are you sure this isn't chromatic aberration, which
is fairly easy to correct in the ACR or Lightroom raw converters? And if you are only making "small prints" from your photos
would the artifacts even be discernible in the prints? I regularly make 16x24-inch Epson Ultrachrome inkjet prints from my Fuji S5 raw files on an Epson 9800 printer, and have never experienced any problems with color artifacts in these prints.

Both the snow photograph and the country lane image were taken under dull, overcast light, and hardly provide a worthwhile test of the dynamic range capability of the Fuji S5 Pro. In my experience, dull light = dull photograph, and sophisticated cameras and talent with Photoshop are no substitutes for an interesting subject, artfully composed, and photographed in great light!

The third photograph, which looks like a section of a construction site, has no bright highlights either. The JPEG looks sharp and the colors look fine on my Eizo CE240W LCD monitor. Since you don't reference any of the three photographs in your comments, I'm wondering why you included them with your review?

I've owned 2 Fuji S2 Pro cameras, an S3 Pro, and currently the S5 Pro, and dozens of digital photographs taken with each of these fine cameras have found their way into the last five of my fifteen published books of landscape and garden photography. The Fuji S5 Pro compares favorably with my Nikon D2X with respect to both color reproduction and dynamic range, and I'm hard-pressed to discern the difference, from a resolution viewpoint, in 13x19-inch Epson inkjet prints made from 12 MP raw files taken with both cameras. Since as a landscape photographer I have little interest in high ISO performance or shooting speed, I have not purchased a Nikon D3 or D300, though I'm eagerly anticipating a higher resolution version of the D3 (the D3X?) from Nikon sometime in the future. If Fuji ever decides to offer a higher-resolution version of the S5 Pro, I will certainly be interested in testing it for landscape photography.

Ian Adams

Whole Plate! On a glass base for stability I hope! Ha Ha!

JohnL

"On a glass base for stability I hope!"

No, special order cut film.

Mike J.

Thanks for all the comments, and especially to Ctein for jumping in with some very helpful clarifications.

Although I don’t care for the S5’s color, the take-home lesson is that it’s distinctive, not that it’s defective in any objective sense. If you have an S5 and like the color, that’s just fine – enjoy!

In principle a camera profile might help. The Chromaholics calibration script that Andreas kindly mentioned is dependent on ACR and so not of much use this time. I certainly would like to see all raw converters provide a documented “socket” for user-generated camera profiles. If Fujifilm should see fit to upgrade HyperUtility in that way, I might take another look at the S5. But I don’t expect to see that anytime soon. In the meantime, in the absence of a straightforward way of implementing an alternative profile in an acceptable raw converter, the post-processing effort demanded by the S5 to achieve less than satisfying results defeats my purpose for owning another DSLR. I’m also not interested in multi-exposure HDR techniques, for similar reasons.

The color artifacts that bother me in ACR conversions of S5 RAFs aren’t likely to be CA. ACR lens correction tools don’t fix them and they’re almost entirely absent in HyperUtility conversions. My guess is that the color is determined by which channels are blown in the surround, but I didn’t try to test that in a controlled way.

I ended up with the D300 as a replacement for the S5 in part because I already had the 17-35 AF-S. Otherwise it would have been a very close call between the D300 and the Canon 40D. The A700 lost out because of both handling and lens preferences. I expect, though, that when it comes to dynamic range the differences among them will amount to splitting hairs.

I'd like to add to Ctein's last comments about the (dynamic? luminance?) range of film. Many negative films (B&W or color) can capture way more tonal information than can be easily tamed in the darkroom. This changes with good film scanners. With a film scanner I am now shooting scenes with a subject brightness range that I never would have considered for darkroom printing and that are also well beyond the capability of a DSLR. And it is liberating not to be always fretting about film development time, knowing that my scanner can usually dig out those extra stops of exposure. For this and other reasons (e.g. lack of micro-contrast), my DSLR is going on eBay.

Oren i like your thoughts and tests. Maybe you are a Pentaxian ;-). Your S5 is gone..., but there are many out there using it regularly. I am not working with a Fuji DSLR but i am using SILKYPIX DEVELOPER STUDIO. I don't know if ACR, Lightroom and all the other RAW Converters can handle the Honeycomb Design of Fujis Super CCD, but i know SILKYPIX 3 can handle them in a special way (...as Ichikawa Soft Laborities wrote me some time in a newsletter..) For color corrections and editing it is also a very fine tool, also for the extra-pipette "skin tone" in WB.
For deeper understanding of the Super CCD-Handling of Silkypix i don't know anything. I just want to give this information as a tip. They have a 14 Day trial & a free Version to test it.

Oren, you said "Many negative films (B&W or color) can capture way more tonal information than can be easily tamed in the darkroom. This changes with good film scanners. With a film scanner I am now shooting scenes with a subject brightness range that I never would have considered for darkroom printing and that are also well beyond the capability of a DSLR.".

My question is: do you change the way you expose and develop, when you do it for scanning? What are your best practices in this respect?

carloch, that comment is latent image's, not mine. My own impression, based on limited experience so far, is that the 35mm scanner I use - a Nikon 5000 - likes a thinner negative than I'd usually prefer for silver printing. But I haven't yet tried to figure out which combinations of film, exposure and development will work best for scanning.

Carloch,
You've quoted me, not Oren. So, if you don't mind, I'll respond: "do you change the way you expose and develop, when you do it for scanning?"

In short, I'm still working this stuff out. I only starting investigating it in a serious way quite recently, though I'm surprised I didn't twig to it a long time before.

I've done some fairly rigorous tests with B&W film (Delta 100), and I am amazed at how well my film scanner (Canonscan FS4000 US) handles an extreme negative density range, as you long as you configure the scanner software (I use VueScan) correctly. An alternative to working through the scanner software is to work on the raw file directly in Photoshop. Whatever you do, make sure your black point is 0.00 and your white point is 0.01 or less.

I'm still working out my practice, but I'm not worrying much about combining "normal" development with either normal or contrasty lighting as long as I record shadow detail adequately. My real worry is in the opposite direction: i.e. negatives that lack density range due to a combination of soft lighting, under-development, and/or under-exposure. My impression is that digital expansion of thin negatives amplifies grain and noise in a way that is truly ugly. I just need to confirm this systematically.

I should mention that I intend to stay clear of films that have a pronounced shoulder (e.g., FP4+) if I want to relax about exposure and development in contrasty situations.

Oren, latent image, sorry for my confusion. I think the subject of exposing and developing for scanning vs wet printing does not receive enough attention. Could we ask Mike to start something about it?.

There are discussions about this on some of the specialty forums. Here's an example from the large format board:

http://www.largeformatphotography.info/forum/showthread.php?t=19967

I've done some fairly rigorous tests with B&W film (Delta 100), and I am amazed at how well my film scanner (Canonscan FS4000 US) handles an extreme negative density range, as you long as you configure the scanner software (I use VueScan) correctly. An alternative to working through the scanner software is to work on the raw file directly in Photoshop. Whatever you do, make sure your black point is 0.00 and your white point is 0.01 or less.

I was very intrigued by this review, as it took a different approach from many of those I've read of the S5. Given the complaints about the limited d-range of the Pentax, I anticipated (correctly) that the reviewer would be pleased (though not overall, and ultimately not) with the S5.

Apart from what is meant by the word "snapshot" (one person's snapshot is another's "decisive moment"), I believe that this can be an excellent method for evaluating equipment, as it tends to give us a kind of uncritical average of what the equipment can do when we're NOT optimizing its use (which is far more 'real world' than what we might get otherwise). Of course, it is possible to use the equipment incorrectly or to minimize the particular advantages it may offer. I agree with some of the previous posters that Oren's experience might have been more positive with somewhat different technique.

I will tip my hand by saying that I'm a long time Fuji user, but I do recognize some of the complaints - particularly about color: yes, the Fujis tend to produce a warm, magenta-ish cast, though the S5 is more easily corrected than, say, the S2. All of the pro Fujis, however, from S2 onward, are capable of producing excellent color - but, with certain kinds of shots or under certain lighting, may require more adjustment or fine-tuning that other cameras (assuming RAW capture in all cases).

However, in my experience (long time film shooter and, briefly, owner of Nikon D200), I have not found any camera that produces close to the d-range of the Fuji or, as an overall professional tool, is as well suited to a range of photography from portraiture to event, wedding, and a host of other types (though not sports - at least professionally).

As an aside: Way too much has been made of the Fuji's speed, buffer, etc. Suffice to say it is fast enough or faster than what most of us will use it for, unless that need is specialized. The only speed complaint is the shoot-to-preview time, for those moments when a lighting or other check is needed (best, though, to keep one's eyes on the subject and away from the screen!).

While other cameras are starting to catch up in terms of d-range, the Fuji still holds a substantial lead - an opinion based on close examination of files from current Nikon and Canon cameras (I have not closely scrutined the Sony, but DXO's specs suggest it is good but not Fuji-like).

What I think may be missed by many who compare the Fuji's range with other cameras is just how profoundly important a soft highlight rolloff (to use a film term) can be in producing credible highlight detail and color. D-range is not simply about whether a sensor can capture the Nth detail at so many stops above Zone V, but rather what actually happens at those limits. That is where the Fuji's advantage is not just technical or theoretical, but all-important.

I say this as a wedding photographer, who must often shoot outdoors, sometimes in direct sun as well as someone who loves the look of how color negative film handles highlights.

With the other digital cameras I have used, as the sensor approaches its upper limit of highlight capture, color turns ugly; even if the image is converted to black and white, there is a washing out that says, in a word, "digital." The long search for the holy grail of wide tonal range (a quest that has obsessed film shooters for a century or more) comes to an abrupt halt in trade for the convenience that digital offers (and it is a huge one--no doubt).

While the Fuji does not "solve" the tonal range problem completely, it takes us far closer to a solution than any of the other current DSLRs whose output I have seen.

Again, I'm not a Fujista, but I do recognize superior technology, especially when I've had a chance to use others.

As a quick footnote: I worked extensively with images from both the Nikon D200 and D300 (yes, the D300 is much better with noise but only slightly better with d-range). I like Nikon color, generally, and the cameras are better in a number of ways (all the way back to the excellent D100 - a real ground breaker, I think, and better in some ways than the D200). In comparison to the Fuji, working exclusively in RAW with recommended workflows, I find the Nikon's very difficult to use. Why? Simply because of the difficulty getting the tonal range, and specifically the highlight rolloff that doesn't announce that you're shooting with a digital camera.

I would be delighted to hear others' perspective on this question. Perhaps a seperate essay or thread?

I just picked up the S5 Pro and have enjoyed it so far. I've encountered one problem though, and can't seem to figure out what caused it-- when I open images in phhotoshop and click on "image>image size", it shows the resolution as only being 72dpi. I have the image quality set to FINE and the recording pixels set to Large. I've searched through the different menus, but haven't been able to come up with a solution. Any ideas?

The comments to this entry are closed.