Phase One IQ4-150 RAW files to download

[BobShaw]

On the rumor/announcement thread, I've just linked to the Hartblei adapter for Canon tilt/shift lenses for the Fuji. For me it's the Fuji announcement of the show, and it's already happened: Pros are going to want that for product shots and architecture, and they're all gonna get it; 100MP you don't need but SHIFT, now that is really really useful.
Btw, I'm sure Stefan could be persuaded to make an adapter for the XD as well, but it looks as if the Fuji has substantially more working shooters.
Edmund

There is already a Canon to X1D adaptor from Kipon. I got it a couple of weeks ago and it works well.
I did a post about it in this forum

[ErikKaffehr]

Hi Edmund,

The image was shot at f/11 according to EXIF. It is clear that diffraction reduces sharpness at f/11, at least when using excellent lenses.

What I also saw that that sharpness varies quite a bit across the field. The image is taken at some distance. We do not know about weather, so I felt it was word mentioning.

But, you are right.

This was tested on a 4.77 micron pitch camera and 30 year old Minolta 100/2.8 macro.

f/5.6	f/11

So, diffraction is clearly visible already at f/11 on a 16MP APS-C camera used with good but in no way excellent lens.

Best regards
Erik

Let's not go for complicated first. I suggest diffraction.

Edmund

[paulreiffer]

Hi Slobodan,

Here is file one in C1, with no sharpening, at actual pixels.

Best regards
Erik

Erik,

Could you include your zoom percentage slider on that screenshot?
Given the toolbar sizes, that screenshot would be a 200% crop, not at 100% zoom, showing very different results to a 1:1 pixel comparison?  May be wrong, but here's what a 100% looks like at my end on the same 1920 screen dimension you've shared...

[ErikKaffehr]

Hi,

Screen shot from Photoshop and Capture One. Definitively 100%. Yes, it seems that 100% in your C1 screen dump is different from 100% in my C1 screen dump. I don't understand, would be very interesting to find out.

Best regards
Erik

Erik,

Could you include your zoom percentage slider on that screenshot?
Given the toolbar sizes, that screenshot would be a 200% crop, not at 100% zoom, showing very different results to a 1:1 pixel comparison?  May be wrong, but here's what a 100% looks like at my end on the same 1920 screen dimension you've shared...

[paulreiffer]

Hi,

Screen shot from Photoshop and Capture One.

Best regards
Erik

Thanks - very odd!

[cgarnerhome]

Something is clearly off!  I shoot extensively with the 40-80 zoom (non-BR) on an XF100 and I don't see that much softness at f16.

[ErikKaffehr]

Hi,

This is not f/16, it is f/11!

But, it could be just a couple of bad samples.

Best regards
Erik

Something is clearly off!  I shoot extensively with the 40-80 zoom (non-BR) on an XF100 and I don't see that much softness at f16.

[Steve Hendrix]

If you create a new variant - essentially creating an "as shot" file, and you:

- Pull NR slider all the way to zero (as Paul has done)
- Push Sharpening up a bit
- Push Radius up a bit
- Pull Threshhold down to zero
- Check Diffraction Sharpening box

Then, it looks not bad. This is not a 22mp or 39mp image, it is a 150mp image, it will require adjustments to standard sharpening settings.

I still might expect better from what I see here, but keep in mind this is a 2 minute exposure.

There are lots of elements that could impact sharpness on a 2 minute exposure.

But this is just an initial couple images. There will be plenty more real world sample images coming along from multiple sources, including CI.


Steve Hendrix/CI

[ErikKaffehr]

Hi Steve,

What surprises me, a lot, is that Paul seems to have a different 100% view than I get. Paul's view is like 70% of my view.

I have exported my image to Photoshop, and that image was around 150MP. In Photoshop it looked exactly like in Capture One. I also tested exporting to DNG in C1 and developing in LR.

I checked the image without sharpening. Of course, sharpening is always applied in the process, but I wanted to see what the sensor can deliver.

I would think that some thing is wrong with that camera as sharpness differs over the field. The top of the image is pretty sharp.

Yes, we need more images before jumping conclusions.

Best regards
Erik

If you create a new variant - essentially creating an "as shot" file, and you:

- Pull NR slider all the way to zero (as Paul has done)
- Push Sharpening up a bit
- Push Radius up a bit
- Pull Threshhold down to zero
- Check Diffraction Sharpening box

Then, it looks not bad. This is not a 22mp or 39mp image, it is a 150mp image, it will require adjustments to standard sharpening settings.

I still might expect better from what I see here, but keep in mind this is a 2 minute exposure.

There are lots of elements that could impact sharpness on a 2 minute exposure.

But this is just an initial couple images. There will be plenty more real world sample images coming along from multiple sources, including CI.


Steve Hendrix/CI

[Jack Hogan]

One of the benefits of having more pixels is that we get a better resolution of fine detail. But, increasing resolution would also make the picture look softer. If we look at an image at 1:1 pixels, it will be more magnified on a 150 MP back than on a 100 mp back.

In the enclosed sample I show a small crop from the harbour image. On the top it is the original image, without sharpening in C1. Left below is same crop, but from the image downsized to 100 MP. Just downsizing the image makes it sharper. The image on the right size is the "100  MP" crop sharpened with focus magic. That is pretty sharp.

I agree Erik and Ronald, that seems pretty good for a ~3.8um pixel at f/11.  Let's not forget that almost four of them would fit inside the diameter of the green Airy Disk at that f-number.  Plus if it was a 2min shot there could have been a little vibration in there somewhere.

Jack

[vjbelle]

Actually a two minute exposure should be almost immune to vibration unless the vibration is constant.  Remember the days of shutter vibration?  Images were especially sensitive for exposures made at 1/15 - 1/100.  Above or below and it was mitigated.  And for really long exposures (greater than 30 seconds) it was invisible. 

Victor

[Jack Hogan]

Actually a two minute exposure should be almost immune to vibration unless the vibration is constant.  Remember the days of shutter vibration?  Images were especially sensitive for exposures made at 1/15 - 1/100.  Above or below and it was mitigated.  And for really long exposures (greater than 30 seconds) it was invisible. 

That would be true of mirror slap and/or shutter shock.  I was thinking of actual ground or wind induced vibration transmitted through the tripod.  It would be easy to rule out if the photographer were to comment on the setup.

Jack

[Doug Peterson]

That would be true of mirror slap and/or shutter shock.  I was thinking of actual ground or wind induced vibration transmitted through the tripod.  It would be easy to rule out if the photographer were to comment on the setup.

Notably, the XF is the only camera I’m aware of that tracks vibration *during* the exposure (from both internal and external sources) and presents it (optionally) to the user a time chart of that vibration after the exposure is complete.

However, having tested prototype cameras many time, I sympathize with the surprisingly difficult nature of it. That tool may or may not have been operable and may or may have been noticed alongside the bevvybof other considerations during such tests.

[narikin]

Come on guys, you're making excuses for dreadful soft images that should never have made the cut!

It's like you're trying to find ways to justify them. They are simply unacceptable, and as the first images released promoting their new hi-end pro back, a disaster! 

[Slobodan Blagojevic]

What are the file dimensions for the 150Mb back?

[E.J. Peiker]

What are the file dimensions for the 150Mb back?

14204 x 10652

Here's the specs - just scroll down a bit
https://www.phaseone.com/en/Camera-Systems/XF-Camera-System/IQ4/XF-IQ4-150MP-Camera-System.aspx

[Slobodan Blagojevic]

14204 x 10652...

Thanks.

So, when looking at a 100% section of a 150Mb file on my screen (109 ppi resolution), it is like looking at a print 130" or 3.3m long, printed at 109 ppi, right?

Looking at my 50Mb files on the same screen is like looking at a print 50" or 1.3m long. I am looking at both from the same distance.

Does it come as a huge surprise then that a print 3.3m long, viewed from the same distance (arm's length or less) as the 1.3m long print would look less sharp?

[BernardLanguillier]

Come on guys, you're making excuses for dreadful soft images that should never have made the cut!

It's like you're trying to find ways to justify them. They are simply unacceptable, and as the first images released promoting their new hi-end pro back, a disaster!

Yes.

DSLRs companies get killed for samples 10 times better than this.

Again, this is not about the camera, I am sure it can do much better.

But it may reveal the fact that extracting the image quality potential of these 150mp backs is becoming real hard.

Cheers,
Bernard

[dchew]

Phase One posted a raw file of the Rodi 23mm along with an LCC file. Pretty impressive, although in my opinion C1 default sharpening is still too strong.

Almost no color cast too. Good sign.
IQ4 w 23hr

Dave

[DaveRosenthal]

Phase One posted a raw file of the Rodi 23mm along with an LCC file. Pretty impressive, although in my opinion C1 default sharpening is still too strong.

Almost no color cast too. Good sign.
IQ4 w 23hr

Dave

For anyone who judged the camera by the original two samples, you must to download these samples instead--they are much, much better. I think the 32mm example (file in the same link) is even more impressive than the 23. These samples show the per-pixel sharpness that you should expect when you pair a good lens with good technique. I'd expect similar results from the 35 BR, etc. for us XF users but we'll have to wait and see.