Doug's test of IQ 250, first observations

[ErikKaffehr]

Hi,

I would like Doug very much for posting these images.

I looked at two of them IQ250 to the left and IQ260 to the right. Some shadow detail pulled 4 stops, with noise reduction set to zero in Phase One.

After that I looked at center, adjusted exposure so they looked similar and removed sharpening (as the IQ 250 image looked quite a bit oversharpened to me). Both produce quite a lot of moiré and In my view the IQ 260 shows less aliasing. To me the IQ 250 seems in some sense more aggressive.Oh, just found out, IQ 260 has 67% single pixel noise reduction as defult causing the softness!

There is about 1.3 stop exposure difference, it seems.

Personally, I am not into technical cameras right now, so I just looked at shadow detail and sharpness. Regarding colour rendition I don't feel I have much to say, except I don't think the images are a world apart. I set WB on one of the books. I presume that they are yellowish, so the image would come out a bit cold. A bit to much moiré for my taste, but it was expected with a subject having much fine detail.

Best regards
Erik

[Paul2660]

Shadow details pushed 4 stops on the 250 file is impressive.  Reminds me of what you can do the the 36mp 35mmCMOS.  Can't get 4stops but often will get 3 to 3.25.


Paul C

[ErikKaffehr]

Hi,

Yes, I agree. But this corresponds to about 30 MP on 35, so they should be in similar league.

I would also add that the sharpness really impresses on those combinations of lenses and sensors. I get nothing like that sharpness with my Zeiss lenses of yore on the P45+. But the very sharp lenses also exaggerate the issues with aliasing.

Best regards
Erik

Shadow details pushed 4 stops on the 250 file is impressive.  Reminds me of what you can do the the 36mp 35mmCMOS.  Can't get 4stops but often will get 3 to 3.25.


Paul C

[torger]

I would also add that the sharpness really impresses on those combinations of lenses and sensors. I get nothing like that sharpness with my Zeiss lenses of yore on the P45+. But the very sharp lenses also exaggerate the issues with aliasing.

Doug's shots are made at f/9. As many like the no-aa-filter-crispiness-wow-effect f/9 is a quite good demo aperture for lenses that can handle it, but if I would shoot with this camera for production I'd probably use f/11, to get more depth of field and less aliasing. I think they would sharpen up really well anyway.

I have noted though that among many/most tech camera users there's a preference towards aliasing and false colors rather than diffraction, so I guess most will actually shoot at f/9 rather than f/11 even if I wouldn't 

[Ken R]

Doug's shots are made at f/9. As many like the no-aa-filter-crispiness-wow-effect f/9 is a quite good demo aperture for lenses that can handle it, but if I would shoot with this camera for production I'd probably use f/11, to get more depth of field and less aliasing. I think they would sharpen up really well anyway.

I have noted though that among many/most tech camera users there's a preference towards aliasing and false colors rather than diffraction, so I guess most will actually shoot at f/9 rather than f/11 even if I wouldn't 

Rodenstock lenses are optimized to be used at wider apertures and from what I have seen with my own Rodenstock HR lenses the optimum aperture seems to be about f8 (best combination of resolution and depth of field). The Schneider lenses are usually best at about f11 so Doug probably chose f9 as a good compromised since he used both Rodenstock and Schneider glass. Just a guess.

I a lot of times use the Rodenstock lenses at f11 and while "acuity" is reduced a hair the resolution is so high to begin with that it can easily be sharpened if one wants but it still looks great.

Also remember (I'm sure most of you here know this) when looking at images at 100% that the IQ250 is a larger file (pixels) than an A7R/D800E so it is like looking at a larger print at the same close distance. Given equal print sizes the IQ250 will magnify less any imperfections you might see.

[Doug Peterson]

I looked at two of them IQ250 to the left and IQ260 to the right. Some shadow detail pulled 4 stops, with noise reduction set to zero in Phase One.

[...]
There is about 1.3 stop exposure difference, it seems.

Note that for some of the combinations I shot both an OE (Over Exposed) and Normal exposure. You're coming an OE from one back to the Normal of another.

I know - a lot of files right? 

[Doug Peterson]

Rodenstock lenses are optimized to be used at wider apertures and from what I have seen with my own Rodenstock HR lenses the optimum aperture seems to be about f8 (best combination of resolution and depth of field). The Schneider lenses are usually best at about f11 so Doug probably chose f9 as a good compromised since he used both Rodenstock and Schneider glass. Just a guess.

Actually an even more practical reason - In this case I was already projecting it would take me most of the day to shoot the combinations I had planned, so going to even longer exposures to use f/11 was not in the cards.

I do hope to do some series of images with Schneiders at different apertures, but likely not for another month or two. It's very likely the Schneiders would have benefited from stopping down a bit further.

[ErikKaffehr]

Hi Doug,

I interpreted it as Zero Oh, my fault! The enclosed images are hopefully the correct ones!

Anyway, my take is right now that:

- The IQ 250 has a significant advantage in shadow noise over the IQ260. This is pretty much expected.
- The all backs perform well with no shift withe the 32 HR lens.
- The lens has very high MTF at pixel pitch, leading to excessive aliasing (in my humble opinion).

Essentially I would say that the IQ 250, unshifted,  behaves as expected. With shifting I don't know the expectations.

My impression? The IQ 250 is an impressive product, that I will not buy, for the simple reason I cannot afford it.

Best regards
Erik

Note that for some of the combinations I shot both an OE (Over Exposed) and Normal exposure. You're coming an OE from one back to the Normal of another.

I know - a lot of files right?