DT's IQ4 Tech Camera Test Teaser

[Alkibiades]

it seems to be a real Revolution. till now there was always the same story: more pixel-more Problems with the Color cast. now what a surprise.
Other story: I was talking with a Rodenstock leading technicians about this new back. He told me that non Rodenstock lens cant really resolve this 150 MP back.
Could somebody see any differences in the details between the new 150 MP and the older 100 MP?

 

[dougpeterson]

Our testing has been posted. We included the 28HR, 32HR, 43XL, and 70HR.

Phase One IQ4 150mp Tech Camera Testing

Do note that Phase One has requested that we not post the raw files publicly as they were done with a prototype which produced much more noise than the final shipping unit will. We expect to continue tech camera tests in the coming week or two with the shipping units arriving to our LA office on Monday and will certainly share those raws.

 

[dougpeterson]

Other story: I was talking with a Rodenstock leading technicians about this new back. He told me that non Rodenstock lens cant really resolve this 150 MP back.

Rodenstock is notoriously conservative when it comes to technical specifications. While I generally do appreciate the mantra under-promise over-deliver to the opposite. For example, Schneider's tech cam lenses were spec'd based on the area of illumination which led users to expect more movement-flexibility from those lenses than they afforded. But I think Rodenstock takes under-promise over-deliver to an extreme.

In our lenses for 150mp article we list the 32HR, 50HR, 90HR-SW, and 105 HR Macro as A++ Quality lenses and the 23HR, 35HR, 40HR, and 70HR as A Quality lenses.

To make your own evaluation you can reference, for example, the Phase One IQ4 150mp raw files we have posted with the 32HR and 23HR.

 

[Wayne Fox]

Other story: I was talking with a Rodenstock leading technicians about this new back. He told me that non Rodenstock lens cant really resolve this 150 MP back.

So to me this can be viewed from two perspectives. If a lens “doesn’t resolve” to the sensor because the lens sucks, that’s one thing. But if a great lens “doesn’t resolve” to the sensor, then what we have is an optical system that delivers the maximum that glass can deliver, and effectively means the results are as good as the lens can do. Be that the equivalent of a 120mp or 130mp sensor ...

While conventional wisdom maintains the system should be limited by the sensor, at some point if you achieve enough resolution then the system delivers the best the glass can do. There is no negative to this and in fact for some lenses it’s a positive because the sensor can now record all the nuances that the lens might have. I think there are photographers that shoot lower resolution cameras because they feel the results are more “film” like or “better” in some unexplainable way. To me this is the result of the sensor outresolving the glass.

We’ve been hearing about sensors out resolving lenses every since the first 24mp sensors arrived, yet it really has never happened.

 

[f8orbust]

...I was talking with a Rodenstock leading technicians about this new back. He told me that non Rodenstock lens cant really resolve this 150 MP back ...

Don't we hear something like this every time a new high MP sensor comes along ? Ctein wrote a piece for the Online Photographer a few year ago that seems pertinent here (direct link):

... until the lens resolution drops to only half that of the sensor, improving sensor resolution will produce an observable improvement in image resolution.

So I wouldn't worry too much about having to stick that $10k Alpa R/S 32mm up on eBay anytime soon :shocked:

Jim

 

[dougpeterson]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

This morning I added the below image from the Schneider 60 XL Digitar with 20mm rise (from a vertical frame) to the article:

This is with NO color cast correction.

 

[Pemihan]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

This morning I added the below image from the Schneider 60 XL Digitar with 20mm rise (from a vertical frame) to the article:

This is with NO color cast correction.

Dough, you should be banned for showing images like that :banghead::banghead::banghead:

 

[TheDude]

Amazing how much difference backside illumination makes.

Next big change, I assume, "stacked" sensor for a global (electronic) shutter.

 

[dougpeterson]

Next big change, I assume, "stacked" sensor for a global (electronic) shutter.

That seems likely, though probably several years away.

 

[TheDude]

That seems likely, though probably several years away.

It's either frontside and backside illuminated, ie. either pregnant or not, but "stacked" sensor is as much a marketing term as a technical term. It's really an ongoing process.

Would be nice to have a test comparing the rolling shutter effect of IQ3 vs. IQ4.

 

[GrahamWelland]

Would be nice to have a test comparing the rolling shutter effect of IQ3 vs. IQ4.

+1 :thumbup:

 

[dchew]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

This morning I added the below image from the Schneider 60 XL Digitar with 20mm rise (from a vertical frame) to the article:

This is with NO color cast correction.

I give up.
:facesmack:

Dave

 

[TheDude]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

This is with NO color cast correction.

In a few years, we will look back at color cast correction, and wonder how did we ever put up with something like that.

 

[vonalpen]

Hi Doug
I still enjoy using my IQ3 100, but after these news I am eagerly waiting for a call from my dealer here in Switzerland that their test unit has arrived...

Using the IQ4 150 on a tech cam, is there finally a possibility to add more info to the exif data than just the f-stop? Like focal lenght, amount of shifts and tilts? I like many others have been asking for this for a long time....

Appreciate all your information, thank you.

Jost

 

[0beone]

Thats is just sick... I loved (and still do) my Trichromatic but this seems to be in altogether another ball park...
I see the poor house in the distance!

 

[TimoK]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

This is with NO color cast correction.

Too bad! Even worse You if did dare to take same kind of picture as this with clear blue sky in both upper corners.
If it were as clear blue in your image, You should know, what it does mean to people in Dantes hell:banghead:

 

[dougpeterson]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

Too bad! Even worse You if did dare to take same kind of picture as this with clear blue sky in both upper corners.
If it were as clear blue in your image, You should know, what it does mean to people in Dantes hell:banghead:

I assume you're referencing the fact that with previous backs like the IQ3 100mp the use of a "problematic" lens+movement combo like the Schneider 60XL with 20mm rise (on a vertical frame) would result it the subtle (or in extreme cases, not subtle) loss of saturation in colors in the outer image circle area. This was because there was so much color cast that C1 was not able to fully correct it. This would mean that with an IQ3 100mp, in some cases, clear blue skies would fade in richness toward the edge of the frame.

The image I posed, in my opinion, has more than enough blue sky throughout the image to evaluate whether this loss of saturation occurs on the IQ4 150mp, and the answer seems quite resoundingly "no" - at least for a 60XL with 20mm of rise on a vertical frame (not the absolute worst-case-scenario, but one that would have thrown an IQ3 100mp into kanipshins). Even without color cast correction the sky color (hue and saturation) is visually very good throughout the frame, meaning that the LCC, if you did one, would be doing very light work to bring it back to being perfect.

I suppose another example with blue skies throughout the entire frame would illustrate this even more clearly. We will be bringing the IQ4 to San Diego, LA, San Francisco, Seattle, Portland, Phoenix, and Denver over the next two weeks (you can register for those events here: Phase One IQ4 tour). I hope to get outside to do more testing (and just plain shooting for fun) during that time; the West Coast may provide me some blue skies, but no promises – of all the new features on the IQ4 "weather control" is not one of them .

Or am I misunderstanding your post?

 

[TimoK]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

Am I missing something?

The image I posed, in my opinion, has more than enough blue sky throughout the image to evaluate whether this loss of saturation occurs on the IQ4 150mp, and the answer seems quite resoundingly "no" - at least for a 60XL with 20mm of rise on a vertical frame (not the absolute worst-case-scenario, but one that would have thrown an IQ3 100mp into kanipshins). Even without color cast correction the sky is very nearly correct, meaning that the LCC, if you did one, would be doing very light work to bring it back to being perfectly correct.

Or am I misunderstanding your post?

Maybe misunderstanding, maybe it's my English. I mean that I most often see the color cast problems in the most shifted part of picture. And most often in flat blue sky. You did mask those extreme corners with the trees.
If You did make a less artistic composition with blue sky from upper left corner to right corner it was a more challenging case.
And, Yes no loss of saturation in the corners (= crosstalk?) wich is a miracle.
If all this true, I'm in Dantes hell too.

 

[dougpeterson]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

Maybe misunderstanding, maybe it's my English. I mean that I most often see the color cast problems in the most shifted part of picture. And most often in flat blue sky. You did mask those extreme corners with the trees.
If You did make a less artistic composition with blue sky from upper left corner to right corner it was a more challenging case.
And, Yes no loss of saturation in the corners (= crosstalk?) witch is a miracle.
If all this true, I'm in Dantes hell too.

Here is a markedly less-interesting image with the 60XL and 15mm rise on a horizontal frame which has blue to the corner. Again: Uniform Light (which would be less required if the 60XL center filter was used, which it was not) has been applied to this image, but Color Cast reduction has NOT been applied to the image.



It's not perfect, and some users may still choose to do an LCC in this case to get it perfect. But even without an LCC, at least to my eye, it's really quite good.

I'll see if I can oblige an image on a 43XL or 60XL with blue to the absolute corners and 20mm of movement (beyond what I'd recommend for sharpness, but for shots where the corners are sky this matters less). Of course the sky has to agree to help make such a shot.

Here is the 10mm fall from the same scene. The light changed pretty significantly in the short time between the two shots, so they don't qualify for a nice stitch, but you can see the color in the corners continuing nearly unabated. (the bottom half of the frame is correctly much cooler, as it's in deep shadow from a low-angle morning sun so is in mostly blue-dome light, whereas the rest of the frame is receiving direct warm-morning-light). As with the previous shot (and all shots in the article), no Color Cast correction applied to the image below.

 

[TimoK]

Re: DT's IQ4 Tech Camera Test Teaser (full article now available)

But even without an LCC, at least to my eye, it's really quite good.

To my eye, it's better than my clients ever have seen.

Of course the sky has to agree to help make such a shot.

So, we are living in the Hell under the Sky.:facesmack: