Is the difference of DR on MFDB vs 35mm dslr discernible on print?

[rethmeier]

I agree that Australia is very pricey with gear,dat is de reden warom ik mijn spulletje koop in the US.( Note voor Dustbak)

However Sun Studios has a Sinar eMotion 75LV for $8000 and a Hy6 body for $2500 AUD

Two years ago that same eMotion was $47.000 AUD

The P65+ does really cost $55.000 AUD, I just called the dealer.

[douglasf13]

That's weird, never heard that before.  And all my Imatest test were done by taking pictures through the lens and some were higher than 11 stops.  Can you point me to any reference where this is detailed.  
Thanks,
Eric

  Here is a post from Iliah just from today:  link

[EricWHiss]

Here is a post from Iliah just from today:  link

Not sure who Iliah Borg is but he's not exactly definite in his statement where he writes "usually" the scene is compressed.  I'd like to see some real science showing that to be true.     Anyhow were this true that lenses can only reproduce 11 stops DR then how come so many tests including my own show higher?

[Ray]

I would not really mind showing you images that have been captured with either one these lenses, 120 & 90 mm Schneider or any of the HC lenses since I own all of these. However since most of my 'testing' is done on work I have I tend to use only one system and the lens I need. I rarely repeat the same setup with another set because for me that would mean a waste of time. This means I can show you images of all lenses but I don't do tests. I don't have 2 of the same images made with different lenses.

I totally understand that you are sceptical about claims people make about performance. I have the same which means that I try the things I am interested in myself. I don't want to rely upon anyone's claims this includes DxO. My testing parameters are my eyes, my time and the perception of the people I make my images for. Yes, these things are no scientific parameters (mostly) but they do count for me

It is not that I don't want to do the tests people like you ask about but I don't see the point in doing so. They don't mean anything to me. I don't feel the need to proof anything. Whatever I say is the way I think I have experienced something whether someone chooses to believe me or not is up to them. I couldn't care less.

Dustback,
I generally find when I do my own comparisons of equipment, that I'm so thorough and meticulous that my results are already in a form that is close to being suitable for posting on LL. However, I understand there are situations where the differences between lenses or cameras are so obvious that it seems pointless taking the trouble to do a proper comparison.

That's how I feel about my Canon 10-22 with 50D, compared with my D700 with Nikkor 14-24. The 50D has a higher pixel count than the D700. Maybe that factor, I have surmised, will raise the system resolution to an extent that the 50D/10-22 will be very close to that of the D700/14-24 combination. But general shooting indicates the D700 with 14-24 is in a different league. I don't have the time at the moment or the inclination to confirm with meticulous tests what my eyes are revealing through general use of both systems, but perhaps I'll eventually get around to this. Perhaps I'll find that at some apertures the differences are minimal, which would be useful to know.

However, having seen tests of the Canon 10-22 on other websites, such as Photozone, it seems possible I have a rather poor copy of this lens, which leads on to another issue about professional equipment. One expects professional equipment to conform to a more rigorous standard than prosumer or consumer equipment, whether it's lenses, camera bodies or DBs.

But is this the case? Phase One rates the base ISO of the P65+ as 100. This is an international standard with precise specifications. Now I understand that for the sake of simplicity the manufacturer needs to round up, or round down, these ISO figures. The D3X claimed base ISO of 100 is actually ISO 78, according to DXO. Perhaps they should have described it as ISO 80. However, ISO 78 is closer to ISO 100 than ISO 50. But Phase One claims the base ISO of the P65+ is also ISO 100. One has to give DXO credit for consistency of testing procedure, even if you don't agree with their results. According to DXO, the base ISO of the P65+ is only ISO 43.

What the heck's going on here? Describing ISO 43 as ISO 100 is more than mere 'rounding up', wouldn't you say? Would plain lying describe it?

[bjanes]

But is this the case? Phase One rates the base ISO of the P65+ as 100. This is an international standard with precise specifications. Now I understand that for the sake of simplicity the manufacturer needs to round up, or round down, these ISO figures. The D3X claimed base ISO of 100 is actually ISO 78, according to DXO. Perhaps they should have described it as ISO 80. However, ISO 78 is closer to ISO 100 than ISO 50. But Phase One claims the base ISO of the P65+ is also ISO 100. One has to give DXO credit for consistency of testing procedure, even if you don't agree with their results. According to DXO, the base ISO of the P65+ is only ISO 43.

What the heck's going on here? Describing ISO 43 as ISO 100 is more than mere 'rounding up', wouldn't you say? Would plain lying describe it?

The ISO 12232:2006 standard is anything but precise, as explained here (see the digital section). The REI (recommended exposure index) specification allows the manufacturer to use any speed rating that they desire, but the rated speed usually is chosen to allow headroom for the highlights. The Saturation and SOS (Standard Output Specification) are more precise, but strictly are applicable only to images rendered into sRGB, but they can be used in a non-standard way with raw images. The Nikon speed rating allows 0.5 EV headroom for the highlights (100/sqrt(4) = 70.71).

[Doug Peterson]

But is this the case? Phase One rates the base ISO of the P65+ as 100. This is an international standard with precise specifications. Now I understand that for the sake of simplicity the manufacturer needs to round up, or round down, these ISO figures. The D3X claimed base ISO of 100 is actually ISO 78, according to DXO. Perhaps they should have described it as ISO 80. However, ISO 78 is closer to ISO 100 than ISO 50. But Phase One claims the base ISO of the P65+ is also ISO 100. One has to give DXO credit for consistency of testing procedure, even if you don't agree with their results. According to DXO, the base ISO of the P65+ is only ISO 43.

What the heck's going on here? Describing ISO 43 as ISO 100 is more than mere 'rounding up', wouldn't you say? Would plain lying describe it?

The P65+ is and has always been marketed as ISO50 for base ISO (base ISO200 with Sensor+ on). If you have a link that shows otherwise please share it so it can be corrected.

The P65+ is ISO 50-800 in full resolution and ISO 200-3200 at 15 megapixels (the files of which compare very well with a 22 megapixel dSLR).

http://phaseone.com/Content/p1digitalbacks...ries/Pplus.aspx
http://www.captureintegration.com/phase-on...one-tech-specs/

Doug Peterson
__________________
Head of Technical Services, Capture Integration
Phase One, Leaf, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870  |  Cell: 740.707.2183
Newsletter: Read Latest or Sign Up
RSS Feed: Subscribe

[bjanes]

Not sure who Iliah Borg is but he's not exactly definite in his statement where he writes "usually" the scene is compressed.  I'd like to see some real science showing that to be true.     Anyhow were this true that lenses can only reproduce 11 stops DR then how come so many tests including my own show higher?

Iliah is a respected digital imaging guru who often posts on DPReview nikon forums. Flare does limit DR in many situations, not by its effect on the noise floor, but by limiting darkest values in the shadows with veiling glare. With Imatest and a Stouffer step wedge, it is important to mask off the target so that none of the light table shows around the edges of the target. Likewise, you should use a lens hood to reduce flare.

For an authoritative statement on flare's effect on DR see this post by David Cardinal here. David has an impressive resume and has published work with the imaging group at Stanford University and appeared in their recent Symposium on HDR Imaging. For more information on lens flare see this lecture on HDR by John McCann--it's an hour long, but well worth the viewing.

[EricWHiss]

Iliah is a respected digital imaging guru who often posts on DPReview nikon forums. Flare does limit DR in many situations, not by its effect on the noise floor, but by limiting darkest values in the shadows with veiling glare. With Imatest and a Stouffer step wedge, it is important to mask off the target so that none of the light table shows around the edges of the target. Likewise, you should use a lens hood to reduce flare.

For an authoritative statement on flare's effect on DR see this post by David Cardinal here. David has an impressive resume and has published work with the imaging group at Stanford University and appeared in their recent Symposium on HDR Imaging. For more information on lens flare see this lecture on HDR by John McCann--it's an hour long, but well worth the viewing.

Well thanks for the extra information. I took a look at the David Cardinal page you linked and its very similar to the DPreview thread.... in that its another voice but neither is substantiated.  Just because its out there on the forums doesn't make it true - I'll wait until I see some real supporting data though your explanation seems entirely plausible.   I don't have an hour to spare watching the symposium, but if I get a chance to watch it I will as it sounds interesting.    In my imatest tests which I posted, I used black matte board around the stouffer step wedge to cover the unused portions of the light table.  It seems clear that more than 11 stops can be had with modern coated lenses if the photographer is careful to use a hood, matte box or flag.

It's probably a waste of time but I want to repeat that the biggest differences you'll find when testing the different cameras will be in 'usable' DR, not the ISO definition of DR.  This usable DR is going to probably be lower or close to 11 stops for the best cameras.  In a way that makes the question of whether the DR of the lenses is a limiting factor or not sort of moot.

[Ray]

The P65+ is and has always been marketed as ISO50 for base ISO (base ISO200 with Sensor+ on). If you have a link that shows otherwise please share it so it can be corrected.

The P65+ is ISO 50-800 in full resolution and ISO 200-3200 at 15 megapixels (the files of which compare very well with a 22 megapixel dSLR).

Doug,
The Canon 5D and Nikon D3x also have an ISO 50 option, but the sensitivity at ISO 50 is exacly the same as it is at ISO 100 which appears to be the manufacturers' base ISO for all three cameras. In fact, according to DXOMark, the P65+ at ISO 50 is actually ISO 45 whereas at ISO 100 it's actually ISO 44. Here's a link to the DXO sensitivity tests.

http://www.dxomark.com/index.php/eng/Image.../(brand3)/Canon

[Ray]

The ISO 12232:2006 standard is anything but precise, as explained here (see the digital section). The REI (recommended exposure index) specification allows the manufacturer to use any speed rating that they desire, but the rated speed usually is chosen to allow headroom for the highlights. The Saturation and SOS (Standard Output Specification) are more precise, but strictly are applicable only to images rendered into sRGB, but they can be used in a non-standard way with raw images. The Nikon speed rating allows 0.5 EV headroom for the highlights (100/sqrt(4) = 70.71).

I see. I didn't realise there was so much flexibility in the standard. It certainly explains the frequent discrepancy between the manufacturer's claimed sensitivity and the tested sensitivity.

I suppose the point here is that, whatever methodology DXO has used to measure sensitivity, it's a standard procedure they apply to all cameras they test, which makes their results valid for comparison purposes.

Considering that the D3X has at least equal DR to the P65+ at equal image size, and greater DR at some image sizes (such as 8"x12"), the fact it can achieve such a DR rating with a good 2/3rds of a stop less exposure is even more remarkable.

In other words, if the SBR of the subject being photographed is very high and I need a full ETTR exposure to capture as much detail as possible in the shadows without blowing the highlights, then in circumstances where I would use a 1/160th sec exposure with the D3X at ISO 100, I would need to use 1/100th sec with the P65+ at the manufacturer's rating of ISO 100. If in addition we then adjust F stop for equal DoF, the shutter speed for the P65+ decreases to 1/40th sec. This seems quite a big disadvantage to me. If one needs to use a slow shutter speed with the D3X one can always fix a neutral density filter to the lens.

[BJL]

A P65+ w/ 28mm = 18.0mm w/ FF dSLR
Film w/ 28mm with film = 17.4mm w/ FF dSLR

Both still less wide than the 14-16mm examples I offered for 4/3, EF-S, DX and 35mm DSLRs, confirming the refutation of the comment about MF offering greater wide angle coverage than smaller formats like "APS-C". (100% coverage VF's are available in certain 4/3, EF-S, DX and 35mm models too, so I do not understand your other comment.)

[Edited to add EF-S to the list of DSLR formats offering 100% coverage VF's, after reading Michael's 7D first look!]

[bjanes]

Well thanks for the extra information. I took a look at the David Cardinal page you linked and its very similar to the DPreview thread.... in that its another voice but neither is substantiated.  Just because its out there on the forums doesn't make it true - I'll wait until I see some real supporting data though your explanation seems entirely plausible.   I don't have an hour to spare watching the symposium, but if I get a chance to watch it I will as it sounds interesting.    In my imatest tests which I posted, I used black matte board around the stouffer step wedge to cover the unused portions of the light table.  It seems clear that more than 11 stops can be had with modern coated lenses if the photographer is careful to use a hood, matte box or flag.

It's probably a waste of time but I want to repeat that the biggest differences you'll find when testing the different cameras will be in 'usable' DR, not the ISO definition of DR.  This usable DR is going to probably be lower or close to 11 stops for the best cameras.  In a way that makes the question of whether the DR of the lenses is a limiting factor or not sort of moot.

Those sources are not the average Joe Blow forum member, but highly regarded experts. David Cardinal works with the Stanford imaging group, which is about as high powered as you can get in this area. Perhaps MIT or Harvard would be comparable.

Since you work with Imatest, why don't you repeat the DR measurement without masking off the light table, allowing a large amount of flare? Your DR would be considerably less. Or you could look at the veiling glare measurement method used by Imatest. Look at the test target just below the snow scene. Assume the luminance of the white is 1000 and that of the black is 1, giving a range of 10 stops. If your lens has 0.25% veiling glare (which would be very good), that black target area will have a luminance of 0.25% of 1000, or 2.5 plus the original 1 for a total luminance of 3.5. The contrast ratio is now 286:1 or 8.2 stops.

Assuming a perfect capture device, the brightest area of the image will be 1000 and the darkest 3.5, giving a DR of 286:1 or 8.2 stops. In this case the DR is limited by flare rather than noise.

(revised 13:15 GMT 24 Sept to include original luminance of 1)

[EricWHiss]

Those sources are not the average Joe Blow forum member, but highly regarded experts. David Cardinal works with the Stanford imaging group, which is about as high powered as you can get in this area. Perhaps MIT or Harvard would be comparable.

Since you work with Imatest, why don't you repeat the DR measurement without masking off the light table, allowing a large amount of flare? Your DR would be considerably less. Or you could look at the veiling glare measurement method used by Imatest. Look at the test target just below the snow scene. Assume the luminance of the white is 1000 and that of the black is 1, giving a range of 10 stops. If your lens has 0.25% veiling glare (which would be very good), that black target area will have a luminance of 0.25% of 1000, or 2.5. The contrast ratio is now 400:1 or 8.6 stops.

Assuming a perfect capture device, the brightest area of the image will be 1000 and the darkest 2.5, giving a DR of 400:1 or 8.6 stops. In this case the DR is limited by flare rather than noise.

Sorry i don't have a current Imatest license.  But that would be a good test.   No doubt shooting into backlight will reduce the contrast but what experienced photographer would do that (unless intended for purpose)?.  
Regarding the posts - I'm not trying to knock these guys but understand I'm not interested in taking that or anything on faith.  No guy from Berkeley, Harvard would make a post with conviction without referencing a study or linking to data.

[Ray]

Both still less wide than the 14-16mm examples I offered for 4/3, EF-S, DX and 35mm DSLRs, confirming the refutation of the comment about MF offering greater wide angle coverage than smaller formats like "APS-C". (100% coverage VF's are available in certain 4/3, DX and 35mm models too, so I do not understand your other comment.)

However, it does seem BJL, that the new Rodenstock 23mm (HR Digaron-S) when used with the slightly larger sensor of the P65+, almost matches the wide-angle capability of the smaller formats.

Depending on whether one crops the 35mm format to a 4:3 aspect ratio, or crops the P65 to a 3:2 aspect ratio, the 23mm Rodenstock produces an equivalent FL of 13.8mm or 15.3mm, which is slightly wider than the Canon EF-S 10-22 on cropped format, but not quite as wide as the Sigma 12-24 on full frame and not quite as wide as the Zuiko 7-14 on the 4/3rds format, but pretty close.

[evgeny]

Is anyone has D3x and MFDB? Where are the images?

[Ray]

Is anyone has D3x and MFDB? Where are the images?

Looks like those who own a D3X either have no need of an MFDB or have sold their MF equipment to buy their D3X plus a few Nikkor lenses and a new car   .

[billthecat]

This person on Flickr seems to have a decent comparison:
http://www.flickr.com/photos/dos-chin/sets...6120567/detail/

Bill

[Christopher]

Looks like those who own a D3X either have no need of an MFDB or have sold their MF equipment to buy their D3X plus a few Nikkor lenses and a new car   .

And those using a P65 don't need a toy like a d3x :-P.

Ok back to reality. I will have some time next week for some testing. However so far I probably will only have a 5dII and Leicas. Still missing a cheap rental option or owner of a d3x in Sydney.

[TMARK]

This person on Flickr seems to have a decent comparison:
http://www.flickr.com/photos/dos-chin/sets...6120567/detail/

Bill

S5 Pro holds its own.

[douglasf13]

Sorry i don't have a current Imatest license.  But that would be a good test.   No doubt shooting into backlight will reduce the contrast but what experienced photographer would do that (unless intended for purpose)?.  
Regarding the posts - I'm not trying to knock these guys but understand I'm not interested in taking that or anything on faith.  No guy from Berkeley, Harvard would make a post with conviction without referencing a study or linking to data.

Iliah is the co-developer of RPP and Libraw. You can email him directly from his dpreview account, and I'm sure he'd be willing to walk you through it.