Luminous Landscape Forum
Equipment & Techniques => Medium Format / Film / Digital Backs – and Large Sensor Photography => Topic started by: Dave Gallagher on May 28, 2009, 08:54:01 pm
-
Hello All,
The medium format digital back marketplace can be quite confusing, even for the people who are supposed to be the experts. The only way that you will truly understand the new technology is to shoot any new system side by side the older systems. With the arrival of our new P40+ the first test that we wanted to conduct was a shooting scenario on a technical camera with movements. We wanted to answer the question, ” will the P40+ produce an LCC that was unacceptable.” The answer is clear in our examples, however so much more can be learned from this comparison than just LCC information. We will leave it up the veiwer to determine which relationships are more significant to them. Enjoy!
Downloads
Download 100% JPGs at our website (http://www.captureintegration.com/tests/phase-one/)
Download Raw Files (EIP format) at our website (http://www.captureintegration.com/tests/phase-one/)
Equipment / Testing Procedure
- Cambo RS with Schnneider Digitar 35mm f/5.6 XL Lens
- 10mm of fall on rear standard
- 1/4 second , f/11.5 , ISO 100 on all DB’s
- Blue sky, direct sunlight, deep shadows, Capture Integration famous Alley
- Nothing optically or on the camera body changed between DB exposure
(http://www.captureintegration.com/download/R2-Scene%20Crop-smaller.jpg)
-
That alley is sexy like Paul Blart - Mall Cop (http://www.sonypictures.com/homevideo/paulblartmallcop/)
The next time, have an office bet and whoever pulls the short straw has to put on a dress and get out there for some fashion shots (guys only eligible). That'll spruce up those test shots.
John, you have a permanent invite to model whichever tube top you prefer....
-
i think it would be a helpful, from a "recruiting"/marketing standpoint, to add examples of 4x5 and/or 8x10 film next to those images for comparison.
there is much discrepancy on these topics, and there shouldn't be.
-
.
-
it sure looks like the P65 has a noticeable better DR, the other too look the same.
as an architectural and general photographer with a p25 where a view camera and WDS system are the norm, how much of a difference is that 1.3x crop? when i bought the p25 i debated the P30 for being cheaper and more MP, but decided to spend the extra money for the P25, i believe i didnt know it couldnt be used on VC/WDS but that problem seems to be resolved now.
-
as an architectural and general photographer with a p25 where a view camera and WDS system are the norm, how much of a difference is that 1.3x crop?
He just posted side by side examples of how much of a difference the chip size is, and linked thumbnails in his post. You don't even have to click a link to find out.
-
it sure looks like the P65 has a noticeable better DR, the other too look the same.
as an architectural and general photographer with a p25 where a view camera and WDS system are the norm, how much of a difference is that 1.3x crop? when i bought the p25 i debated the P30 for being cheaper and more MP, but decided to spend the extra money for the P25, i believe i didnt know it couldnt be used on VC/WDS but that problem seems to be resolved now.
P30 has microlens on CCD. So it is not suitable for large format with tilt and shift lens. There is color cast problem and files is not very sharp when tilt or shift.
I thought it is quite difficult to compare quality for these 3 images. I don't know why but it seems tilt or shift angle is a bit different.
-
I thought it is quite difficult to compare quality for these 3 images. I don't know why but it seems tilt or shift angle is a bit different.
We were very very cautious not to change anything when switching digital backs. The captures are as close to identical as possible. Overlayed in Photoshop I measure the variation between captures as less than +/- 0.2 degrees of rotation.
Doug Peterson (e-mail Me) (http://mailto:doug@captureintegration.com)
__________________
Head of Technical Services, Capture Integration
Phase One, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870 | Cell: 740.707.2183
Newsletter: Read Latest or Sign Up (http://www.captureintegration.com/our-company/newsletters/)
-
how much of a difference is that 1.3x crop?
(http://www.ishotit.com/Dr_Evil.jpg)
$1,000,000
B
-
Interesting question since I just went from a P25 framing to a P30 framing. It's bugging me a little and the only reason is I can see outside the crop lines but the crop lines are thin. If I did one or two things it may help . Make the crop lines a little bolder with a sharpie and/or I had another idea use matte tape to mask it but still let the light in since I THINK the meter works off the full frame . That needs to be confirmed though. Steve , Doug would love a answer on that. Other than that no real issues except my 28mm is a little less wide but also no issues with corner sharpness either, so a blessing in disguise.
-
Thanks for posting those files, Dave. As between the P65 and P45 files, I can see a real improvement in the dynamic range with the P65 compared to the P45. The P65 has better shadow detail and also seems to hold a bit more highlight detail. There are differences in the colors between the files, but my preference for the colors in the P45 file is obviously subjective. Of course, the colors in both files can be easily adjusted in the raw conversion or PS, but I do prefer to start with a color rendition that is more pleasing to me off the bat. What is difficult for me to assess is that the P45 file APPEARS sharper in some areas of the image than the P65 file, but in what might be a brittle/digital way. The P65 file seems "smoother" and "less digital" than the P45 file, which is generally a good thing, but I can't tell if that is at the expense of real sharpness. (Leaf files from Dalsa sensors are often described as "more filmlike", and I wonder if that is what I am seeing.)
-
Thanks for posting those files, Dave. As between the P65 and P45 files, I can see a real improvement in the dynamic range with the P65 compared to the P45. The P65 has better shadow detail and also seems to hold a bit more highlight detail. There are differences in the colors between the files, but my preference for the colors in the P45 file is obviously subjective. Of course, the colors in both files can be easily adjusted in the raw conversion or PS, but I do prefer to start with a color rendition that is more pleasing to me off the bat. What is difficult for me to assess is that the P45 file APPEARS sharper in some areas of the image than the P65 file, but in what might be a brittle/digital way. The P65 file seems "smoother" and "less digital" than the P45 file, which is generally a good thing, but I can't tell if that is at the expense of real sharpness. (Leaf files from Dalsa sensors are often described as "more filmlike", and I wonder if that is what I am seeing.)
I keep reading about this "leaf being more like film" and "phase looking more digital".
I cannot express judgement of Leaf files, since I have used them seldom, but to me the Phase files, in general I mean, look like the "real thing". On a 30 inches monitor it is like being in front of the photographed subject in real life. that may not appeal to everyone, but that is how good they are.
then if we want to get the look of film, that should not take more than a few clicks per files.
Or am I missing something?
-
I keep reading about this "leaf being more like film" and "phase looking more digital".
I cannot express judgement of Leaf files, since I have used them seldom, but to me the Phase files, in general I mean, look like the "real thing". On a 30 inches monitor it is like being in front of the photographed subject in real life. that may not appeal to everyone, but that is how good they are.
then if we want to get the look of film, that should not take more than a few clicks per files.
Or am I missing something?
Without being particularly knowledgeable in this, what you are saying doesn't seem to have any contradictions. "like real life" is one thing, and "film-like" is something else. I suppose the Phase files are quite neutral, whereas the Leaf files have colour which is more "designed". Film is not neutral, by any means, the colours were designed to be pleasing in some particular way.
I suppose that any file is a few clicks away from any look, give or take some clicks. That pros use Canons instead of Leafs or Phases is solid proof for this. I don't think anyone would claim that the Canon files are as good, but after a few clicks, few can tell the difference.
I suppose it is all in how you want to work. If you like the file out of the box, there is more time for photography. The less the native look matches your preferences, the more you will sit in front of the computer.
-
Isn't there just part of you that wants to see a Comparison Test like this, but with:
A. Hasselblad version of 50MP, whatever it's called.
B. P40+ or P65+, either one.
C. Nikon, whatever the big one is called, the eight grand one.
D. 5D Mark II.
Nobody's ever really done that, except maybe once, with that obscure Tech place in LA, years ago.
What I'd like to do is ignore the Chip Size, and just pick a size, on the short side. Any size. Maybe 11 inches wide. And rez all the files to that same size.
Shoot all four, with NO sharpening, and with the default color profile, and get a model on a set, with strobe, so there are no variables with ambient changing. (And no Cowboy Drunks).
I'd love to shoot them all tethered, with someone with each camera/back that knew how to run the software, and REALLY knew how to get the best out of each camera/back in processing settings. No monkey business in the processing. No vignetting or whatever.
Have some guy there in a black/white striped shirt to be a Referee, and be able to throw a flag if Dougie tries to add some sharpening.
Shoot them all at about f11, with strobe, and bracket in third stops.
And then, in the end, make a nice Epson 7900 or 3800 print, about 11" wide image area on the short side, and put them tacked to the wall, cropped the same proportion, and then ask a bunch of normal people to come in and judge them. Not photographers.
Wouldn't that be funny?
I'd love to have a non-invested Observer to sit in a big high back Directors Chair, and watch all the shenanigans, and then watch the hassle factor of tethering and processing, and then judge the prints, and then walk over into one corner of the boxing ring at the end of the night, and hold up an arm of The Winner.
-
Have some guy there in a black/white striped shirt to be a Referee, and be able to throw a flag if Dougie tries to add some sharpening.
Wouldn't stop me. I'd just use clarity instead :-P.
Anyone is welcome to come to our studios and try any gear we have on hand. We have all the Phase backs and all the Canon cameras. Bring your own Nikon/Leaf/Hassy.
It's really really really a nightmere though trying to decide what is "fair". Just for instance your choice of f/11. On really good dSLR glass that hits diffraction. Should you instead match the apparent DOF; shoot each lens at each f-stop? Should we use a Phase One back on a Phase One body or an H2? Maybe we should put all the digital backs on a tech camera. When lenses don't match up between systems. How do you account for the different aspect ratios (match the diagonal FOV, horizontal FOV or vertical FOV).
Heck if you look at this test we chose ISO 100 for each of the four backs because the P30+ doesn't have an ISO50 (which we only used to show the LCC reinforcing a 30+ is not suited to tech camera usage). Is that fair or are we handicapping the three backs that start at ISO50?
etc etc etc etc
Doug Peterson (e-mail Me) (http://mailto:doug@captureintegration.com)
__________________
Head of Technical Services, Capture Integration
Phase One, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870 | Cell: 740.707.2183
Newsletter: Read Latest or Sign Up (http://www.captureintegration.com/our-company/newsletters/)
-
It's really really really a nightmere though trying to decide what is "fair". Just for instance your choice of f/11. On really good dSLR glass that hits diffraction. Should you instead match the apparent DOF; shoot each lens at each f-stop?
Diffraction, Schiffration -- big deal. Tell me this -- If you were in the middle of a real job, and you had eight more setups to do before you left that day, and the assistant took a meter reading, and he yelled out, "Uh, Dougie, it's f11", would you stop and relight the shot, and not do the shot at f11? Get serious. Don't make me drive down there.
I understand what you're saying, if we were both Molecular Scientists, and not photographers, but in the end, there are eight more shots to do before we have a drink. Can we forget about some stupid MTF Chart and get real?
I know there would be some variables, but if there were at least one guy there that KNEW each system, and there was a Referee, we could make it work, and learn something.
Another sub-test, if there was time: Take out a stopwatch, and do a shot with each camera, and time it on how it takes before you get a processed TIFF, after shooting the picture.
But in the end, all I want to look at on those monitors, and on those prints, are Skin Tone, basic sharpness, Shadow Detail, and overall color. We could find a model somewhere that could hit a pose, and repeat it exactly for all four cameras.
-
Diffraction, Schiffration -- big deal. Tell me this -- If you were in the middle of a real job, and you had eight more setups to do before you left that day, and the assistant took a meter reading, and he yelled out, "Uh, Dougie, it's f11"
[...]
But in the end, all I want to look at on those monitors, and on those prints, are Skin Tone, basic sharpness, Shadow Detail, and overall color. We could find a model somewhere that could hit a pose, and repeat it exactly for all four cameras.
I 100% agree with you for your usage/style/subject-matter. In that situation I wouldn't give a single thought to diffraction or most other technicals.
If I were a car shooter I, a landscape shooter, an interior/architecture shooter, a high end macro table-stop shooter (e.g. watches) then yes I would really care about the sharpest aperture and I wouldn't give a thought to "skin tone". If I were shooting black and white nudes I wouldn't give a thought to "overall color"
That gets the core of the problem of all such testing. Not everyone cares about (or should care about) the same image attributes.
That's why I say anyone is welcome to come to our studio and do their own test.
Doug Peterson (e-mail Me) (http://mailto:doug@captureintegration.com)
__________________
Head of Technical Services, Capture Integration
Phase One, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870 | Cell: 740.707.2183
Newsletter: Read Latest or Sign Up (http://www.captureintegration.com/our-company/newsletters/)
-
Isn't there just part of you that wants to see a Comparison Test like this, but with:
A. Hasselblad version of 50MP, whatever it's called.
B. P40+ or P65+, either one.
C. Nikon, whatever the big one is called, the eight grand one.
D. 5D Mark II.
and put even a Sony in the mix. with a prime zeiss it may deliver the best of the dslr lot.
-
He just posted side by side examples of how much of a difference the chip size is, and linked thumbnails in his post. You don't even have to click a link to find out.
ya i realize that, other than looking at crop alone, i was hoping to see what peoples real world experience has been. i obsess about the details too much. its not all that much difference unless you are backed into a corner shooting the inside of a house. that being said, it seems my clients want WIDE, i live off my 24mm for architecture insides.
-
"Nothing optically or on the camera body changed between DB exposure"
Yes but the lighting in the alley changed. If you look on the left wall you will see a horizontal line running down the wall. The shadow running along that line changes indicating that the light hitting the alley is changing. Also if you look at the cloud at the very end of the alley you can see that it changes too. So there might be clouds running across the sky also changing the lighting.
I somewhat agree with gwhitf, you must use things in the practical circumstances of what you shoot. But if money were not a consideration, I would buy a P65+ and get on with it.
-
That gets the core of the problem of all such testing. Not everyone cares about (or should care about) the same image attributes.
Dave's smart.
He knows that throwing up three files and using the word comparison is the fuel that drives these forum threads.
This comparison is slightly informative, also slightly flawed, but Like Hollywood has known forever there is no such thing as bad publicity and like I said, Dave's smart.
Actually to get everybody's motor running he should have shot a 5d2 with a Tamron lens and that would sent this thread into the 84,000 views category.
The test you want to see, does a Nikon or Canon work as well as a peefourty, peesitxyfive, or peefourtyfive, will never happen and even if it did, the internet noise factor alone would render any conclusion mute See that DXO thread for example.
B
-
Diffraction, Schiffration -- big deal.
EDIT
But in the end, all I want to look at on those monitors, and on those prints, are Skin Tone, basic sharpness, Shadow Detail, and overall color. We could find a model somewhere that could hit a pose, and repeat it exactly for all four cameras.
George you are such a dumbass.
You are the absolutely the very best photographer I know* (*as in "internet" know).
Medium format is different from 35mm.
NIck-T
-
The medium format digital back marketplace can be quite confusing, even for the people who are supposed to be the experts.
I think this comparison is like every other one I see on the interweb It's ok, but for some reason it's either a photo of a brick wall or a tree.
Maybe that's the perceived market, brick wall and tree photographers.
The test I would love to see is skin tones and how they react under different lighting and ambient color.
The Leaf backs I owned were less succeptable to ambient color bounce, but had an awful time with direct hard light on light caucasian skin. One hard Frensel or profoto head on those pasty white models threw up a magenta cast in the arms and legs from highlight to shadow. Maybe because the transition is so abrupt
Then again the Leaf back loved soft key like window light and would produce a beautiful look.
The P30+ I use Works well under direct lighting, really well under direct tungsten light, as long as you have a lot of watts, but it also picks up every single skin blemish, or change in color like nothing I've ever seen. I'm sure this sensitivity is great for shooting food, but sometimes it means a lot of extra post production work to paint out the the over sensitivity to color. Great news if your a retoucher, bad news if your paying for the retouching.
The Hasselblad 39 mpx back I very briefly tested probably produced the best skin tones of any medium format back I've tried. More of a Canon like skin tone, which I find nice.
The thing about medium format is none of the makers really show the true value of their files, which is working a file deep in post. Even a 18mpx p21 file you can beat the hell out of in post and get that deep uber sharp look, especially with hard light.
The p30+ takes more abuse still and that is the one area where medium format squashes the dslrs with their AA filters.
(http://russellrutherford.com/sports/pictures/020rr_sports_june_08.jpg)
I know dealers don't have the resources, but some db maker should show an image from start of capture though all the retouching steps to finish to illustrate how well their files hold up.
JR
-
James, I agree with you all. Thank you for sharing us.
Just want to add P30+ files develop by C1 3.7?
With the same digitalback P30+ develop by C1 v.4. Skin is totally difference.
Skin texture is much more like Phocus or leaf.
kitty
-
beside the fact that other comparisions probably would be nice to see as well I appreciate this comparision (whatever should make someone at Phase One to do comparisons with a DSLR??? This is up to people who have both systems. Or maybe we will see another "measuring megabytes" article on LuLa in the future).
Looking at the images the first impression is that they are... close. The most obvious difference is the crop of the P40+.
Given the change of light I think it's hard to tell differences in DR. Too, if you set the film curve for the P45+ to "extra shadow" it's getting closer to the P40+/P65+ files. So maybe the film curve presets for the new backs are a bit different to those of the old backs.
Colour reproduction is a question of the camera profiles. I like the blues and cyans shifted towards yellow in the P65+/P40+ files. But this is very easy to adjust for the P45+ file as well with the Color Editor.
Some moirée in the P45+ file whereas the P40+ and P65+ are almost clean.
A few days ago I "compared" the Sony A900 of a friend with the P45 (non plus) and the P21+.
Far from scientific or precise... just a quick and dirty side by side setup.
The DB files look "richer" while the A900 files look somehow "airy" (again - to a certain extend - a question of camera profile).
The DB files look more "3D" while the A900 files look more "flat" - detail reproduction in dark tonal values is much better with the DB files.
Though overall the detail reproduction of the A900 with good glass is very impressive.
Info for Doug and Dave: I've downloaded the EIP files. In the P45+ LCC is okay (can turn in on/off). In the P40+/P65+ the LCC is set in the respective menu but has no effect (so the EIPs seem to contain the settings but not the actual LCC file).
edit: after unpacking the EIPs the LCC works.
-
I appreciate these comparison files and especially the inclusion of the RAWs, which is vital for real comparison of backs. With this said, I don't know why Phase One has refused to include download of sample RAW files on their site from ALL their backs. I would like to see RAW files done with each segment of the market in mind....ie: landscape (me), fashion, architecture, long exposure samples and so on. They could take their time to do controlled consistent tests with say a "normal" lens on the Phase camera and make the best shots that the back/camera is capable of making. This shouldn't be hard to do. All Phase has even put on their site are these tiny jpgs., mostly fashion, some architecture, cars, but not really much landscape....... the only way to really evaluate a back is to see the RAWs.
Hope Phase is listening. Eleanor
-
Skin Tone
skin tones
in Capture One V3 there were 4 Skintone Profiles for the H25... I think they were only available in the Windows version for some reason (but I'm not quite sure). So maybe you don't know them:
http://tinyurl.com/mmmhan (http://tinyurl.com/mmmhan) (308KB)
-
I know dealers don't have the resources, but some db maker should show an image from start of capture though all the retouching steps to finish to illustrate how well their files hold up.
JR
Actually, more value, and objectivity, would be had if someone like yourself produced this exercise.
In your spare time, of course :>))
Mark
-
The DB files look "richer" while the A900 files look somehow "airy" (again - to a certain extend - a question of camera profile).
The DB files look more "3D" while the A900 files look more "flat" - detail reproduction in dark tonal values is much better with the DB files.
Though overall the detail reproduction of the A900 with good glass is very impressive.
What do you mean with airy? I did some tests of the A900 (alone), and found that in general the files had a lot of detail, more than Canons, but the darkest shadows, even in properly exposed files, were showing some posterization. This is the type of thing I expect the DBs to handle better.
-
What do you mean with airy?
thin, lean, skinny compared to the DB files ... (auf Deutsch würde ich sagen: "dünn", während die DBs "satte" Bilder produzieren).
I did some tests of the A900 (alone), and found that in general the files had a lot of detail, more than Canons, but the darkest shadows, even in properly exposed files, were showing some posterization. This is the type of thing I expect the DBs to handle better.
yes, I agree. And this is especially noticable when uprezing images (and in this particular use the AA filter is limiting again).
However the A900 is a great camera (above all regarding the price) and there are some very nice lenses in the Minolta/Sony line.
-
thin, lean, skinny compared to the DB files ... (auf Deutsch würde ich sagen: "dünn", während die DBs "satte" Bilder produzieren).
Ah, okay. Perhaps "weak" or "unsaturated". I don't have this impression of the A900 photos, but I have not yet been able to compare all colours. Blues look okay.
-
Perhaps "weak"
yes, thanks :-)
I don't have this impression of the A900 photos, but I have not yet been able to compare all colours. Blues look okay.
it's not just about the colours but the overall "look"...
As to the colours the comparission is unfair anyway - viewing/processing with C1 the colour profiles for the DBs are specific while the A900 profile is a generic profile (that tends to turn the reds to magenta and oversaturate magenta in daylight). With a few adjustments in the Color Editor this can be solved.