Phase One IQ4-150 RAW files to download

[dchew]

Yes, the 32 is well-known as the better lens of the two. For those of you “worried” about sharpness, look at the 32 image.

I was most interested in the cast on the wider lens. Looks like most, if not all of the color cast is gone with this sensor. I took the lcc image and white balanced in the center to 148/148/148. The far edges are ~ 39/40/41. Still dark, but color change is almost non-existent. Very encouraging.

Dave

[ben730]

The LCC is very impressive. I hope Phase will also make an IQ4-100 (44x33) back.

Regards,
Ben

[dchew]

Crop of the 32 hr image. Aliasing @ 150 mp. Not a bad lens.



Dave

[vjbelle]

Dave..... I see a shiny new IQ4-150 with your name on it.

Victor

[dchew]

Dave..... I see a shiny new IQ4-150 with your name on it.

Victor

Ha. I don't know. Here's the thing: That 32hr image was shot at f/8. I haven't done the airy disc math, but I bet diffraction is creeping in even there. As others have pointed out, everything needs to be perfect in order to get this resolution. No wind, f/8 or wider, no atmospheric mumbo jumbo, etc. No way I'm going to limit myself to those constraints, so how often will I really see the benefits?

But now with this much-improved LCC, our favorite SK's may have new life. That's a whole different story, even at f/11.3...

Dave

[JoeKitchen]

OMG, there is almost no lens cast in those images.  I really want to see how the Back handles the SK 35mm and 28mm now. 

On a different note, Phase One does need to find someone who can take good release images.  Although these are fairly nice in terms of no camera shake and being sharp, I am putting them both about 1.33 stops under exposed. 

[JoeKitchen]

On a different note, this kid needs to get off his cell phone and enjoy what is around him. 

[Doug Peterson]

The LCC is very impressive. I hope Phase will also make an IQ4-100 (44x33) back.

As I understand it, there are no plans for any future new crop-sensor backs. Phase One is very much a "full frame 645" company now as far as its photographic division goes (the industrial division does use a 44x33 sensor as, in many aerial applications, the speed of readout is the most important thing).

[vjbelle]

Dave.... from my experience most of my Schneider Digitars have to be shot at f10 or f11.  The 35XL must be shot at f11, the 60XL must be shot at f10, even the 120 Asph must be shot at f10.  I'm not buying any of the Rody wides due to their excessive distortion which is not so easy to fix in post.  So.... a lot of what you mention is very true.  The full resolution of the new back is beyond most of my lenses.

Victor

[dchew]

Joe, about the under exposure: Did you create and apply the LCC? Once you do, the edges get pretty bright. At first I thought the LCC was over-correcting for falloff. Then I thought, on an angle of view that wide I bet it is a pretty realistic rendering of the sky based on the sun direction. Even without a polarizer the sky is brighter along the sun's axis.

Dave

Edit: If you switch to linear response there is indeed some headroom.

[JoeKitchen]

Joe, about the under exposure: Did you create and apply the LCC? Once you do, the edges get pretty bright. At first I thought the LCC was over-correcting for falloff. Then I thought, on an angle of view that wide I bet it is a pretty realistic rendering of the sky based on the sun direction. Even without a polarizer the sky is brighter along the sun's axis.

Dave

Edit: If you switch to linear response there is indeed some headroom.

I have always found the light fall off correction in the LCC fix to be a bit overzealous. 

I would assume this is because C1 increases the output of each pixel in the correction by the amount with which it is off in the lens cast sample shot.  However, in reality, through the lens, this is not how it works.  Highlights burn in faster then shadows, so if you correct the fall off optically, with a CF, the light fall off is corrected more naturally and does not look overly bright on the edges. 

It would be impossible to really create a light falloff correction that knows how much to increase the highlights by, and then the mid-tones, and then the shadows (an amount that would be ever decreasing depending on the DMax of each pixel) to simulate a real CF, so this is what we are stuck with. 

So usually I am setting the light fall at 72% to 86%, depending on the lens, and using the CFs when I can. 

Editing the image like this, I got a fairly nice sky and light fall off in C1 while pushing the image 1.5 stops.  (Rather quick edit attached.).

[Doug Peterson]

I'm not buying any of the Rody wides due to their excessive distortion which is not so easy to fix in post.  So.... a lot of what you mention is very true.  The full resolution of the new back is beyond most of my lenses.

You may have missed that Capture One has full lens corrections for the 23, 32, 35, 40, 50, 70, and 90 HR-SW. That includes distortion correction, including with movement (though you do have to manually enter the movement amount).

[Altdo]

On a different note, this kid needs to get off his cell phone and enjoy what is around him.

He and his father were hunting for Pokemon at 8:30 on a Sunday morning... It seemed like it was quality Father/Son time.

[vjbelle]

You may have missed that Capture One has full lens corrections for the 23, 32, 35, 40, 50, 70, and 90 HR-SW. That includes distortion correction, including with movement (though you do have to manually enter the movement amount).

Yes Doug..... but all of that correction comes at a price of loss of 'some' resolution.  Regardless of where that correction is made, either at the raw stage or beyond in post it 'does' come at a price.

Victor

[Doug Peterson]

Yes Doug..... but all of that correction comes at a price of loss of 'some' resolution.  Regardless of where that correction is made, either at the raw stage or beyond in post it 'does' come at a price.

Absolutely. Any change to the pixels costs some resolution. The salient questions are how much is lost, and how does it compare to the alternative? I'd suggest "not much" and "quite well" are the answers in this case.

That is, my experience very strongly indicates that a [32HR + 10mm shift + lens corrections] is significantly sharper than a [35XL + 10mm shift].

[Chris Barrett]

The sharpness is lovely, but it would have been great to see some shift movement.  The images appear to have nothing but rise, and having additional sky is useless for judging sharpness.

I agree, Joe, they should have a working architectural photographer produce a set of images if they really want to speak to tech cam usability.

IMHO,
CB

[JoeKitchen]

The sharpness is lovely, but it would have been great to see some shift movement.  The images appear to have nothing but rise, and having additional sky is useless for judging sharpness.

I agree, Joe, they should have a working architectural photographer produce a set of images if they really want to speak to tech cam usability.

IMHO,
CB

I agree.  I have always found that vertical images with upward shift have put a great amount of stress on the files with the LCC due to how sensors are designed.  That being, they are typically designed to be more forgiving with upward shift in horizontal orientation. 

A much better test would have been a vertical two-point composition near the right corner of the building utilizing 16mm of shift composing the image so nearly the entire image is the building.  This would have not only given us a better sense of sharpness, but also DoF. 

[ErikKaffehr]

Hi,

Clearly, there is a price, but the question is if it is a low price or a high price.

As far as I recall, lens corrections may have started with panotools and they had very good algorithms for correcting distortion.

Best regards
Erik

Yes Doug..... but all of that correction comes at a price of loss of 'some' resolution.  Regardless of where that correction is made, either at the raw stage or beyond in post it 'does' come at a price.

Victor

[ErikKaffehr]

Hi Chris,

I can see your point, but still, this looks very good compared to the first sample. The LCC images seem to hold great promise.

Best regards
Erik

The sharpness is lovely, but it would have been great to see some shift movement.  The images appear to have nothing but rise, and having additional sky is useless for judging sharpness.

I agree, Joe, they should have a working architectural photographer produce a set of images if they really want to speak to tech cam usability.

IMHO,
CB