Jump to content

Capture 1 Pro and Fuji X-T1


Recommended Posts

I am only getting as little as 15.2 Mpixel from my X-T1 with Capture 1 Pro 7.2.1 (4777 x 3185). I get up to around 4873 x 3249 with X-E1 images - 15.83Mpixel. Strangely it is variable from shot to shot - I believe it is a change to the algorithm in Capture 1 as I switched to from 7.2.0 to 7.2.1 to get X-T1 capability. On 7.2.0 I also had a margin outside the default crop which allowed me to get up to around 17Mpixel, something around 5000+ px on longest side.

 

Any body else noticed it? Does the same happen with LR esp 5.4? All in all it is slightly irritating as 16Mpixel is a bit tight for some purpose; some libraries want 4900px on the shortest side so under 4800px is a bit off. The X-T1 manual quotes 4896 x 3264 - almost 100px more on each dimension.

 

I will raise a support ticket with Phase One when I get the time.

Link to comment
Share on other sites

Interesting, I was using the 55-200mm. I wonder if there is a configuration issue on either the camera or C1.

 

Duncan, when you go into crop mode in C1 on that image do you see additional pixels under the mask? That is what I used to see with the 7.2.0 and the X-E1. I could extend the image crop significantly into that area to well over the stated resolution.

Link to comment
Share on other sites

Interesting, I was using the 55-200mm. I wonder if there is a configuration issue on either the camera or C1.

 

Duncan, when you go into crop mode in C1 on that image do you see additional pixels under the mask? That is what I used to see with the 7.2.0 and the X-E1. I could extend the image crop significantly into that area to well over the stated resolution.

 

I'll be taking some shots tomorrow with the 18-55 & 55-200 so I'll be able to check that out. There were extra pixels over the 4896 at the wide end of the 18-55. I believe these are to cove for any optical corrections required. 

Link to comment
Share on other sites

I read something like this with another X-T1 owner on the Fuji X forum. It turned out that C1 was reading the size of the jpeg that the camera records to show you.  Can't remember if there was a setting or what to change, but you are getting the full size if you are shooting RAW.  

 

I find it incredibly difficult to find previous threads in the Fuji X forum, so I'm not sure if I could find it again.  I think the "variable" bit may come from the jpeg also, we know jpegs sizes vary according to how much information is in the image.

Link to comment
Share on other sites

I did a test this morning.

  • With the 18-55mm I get full resolution with more outside so I can extend the crop beyond spec. But at 55mm I get less than full resolution.
  • With the 55-200mm It gives less than full res, no margin at 55 and reduces significantly at 200mm where I am effectively getting under 15Mpixel instead of the 16Mpx of the spec

C1 Pro was using 100% lens distortion correction and hiding the distortion it was cropping. However it seems to be doing it within the specification image size rather than using the spare pixels outside the image spec.

 

So Duncan, you were right about it applying lens correction (from the file, presumably info embedded by the camera) but apparently not using the spare pixels we thought were for that purpose.

 

In camera RAW conversion produces a full size 16Mpx image where C1 produces one at 15Mpx or even less. That suggests the C1 algorithm needs correcting.

Link to comment
Share on other sites

Hi Martin

 

I've just taken a few shots with the 55-200 and even at 200mm they are showing as being taken at 4896 in C1 (file information screen) but in the Crop Tab, it shows as 4790 @200mm. Exported the file and PS confirms it as 4790px. Not sure what C1 is doing but yes, it does seem to remove 100px from the width with the 55-200 lens. 

Link to comment
Share on other sites

Duncan, check the lens correction tab in C1 - my guess it is set to 100% and probably don't show distortion is ticked. I also got a little loss with the 18-55 at at 55mm for the same reason.

 

As I say it looks as though C1 is doing the lens correction within the image as specified rather than making use of the "spare" pixels. In camera raw conversion comes out with a jpg at the full 4896px.

Link to comment
Share on other sites

Duncan, check the lens correction tab in C1 - my guess it is set to 100% and probably don't show distortion is ticked. I also got a little loss with the 18-55 at at 55mm for the same reason.

 

As I say it looks as though C1 is doing the lens correction within the image as specified rather than making use of the "spare" pixels. In camera raw conversion comes out with a jpg with the full 4896px.

 

I would agree. Convincing them of that (c1) is another thing though!

Link to comment
Share on other sites

I have raised a support ticket and we will see what they say, probably after Easter. Hopefully we will see C1Pro 7.2.2 before too long!

 

When I am next out with the camera I will photograph some brick walls to see what distortion is like, before and after correction.

Link to comment
Share on other sites

I have raised a support ticket and we will see what they say, probably after Easter. Hopefully we will see C1Pro 7.2.2 before too long!

 

When I am next out with the camera I will photograph some brick walls to see what distortion is like, before and after correction.

 

Good man. What I'd also like to see them do is introduce the Auto Mask feature for Fuji raw files. Would make layer masks a lot easier!

Link to comment
Share on other sites

I had a play with some brickwork and the degree of correction that is being applied by C1 would hardly appear to be noticeable in most subjects. Only those with strong lines parallel to the frame edges. Even then it seems possible that there is overcompensation from slight pin cushion to slight barrel or vice versa. It would not be demonstrable in small images from the pictures (so not posting them) I took of the brickwork of my house - close-up the brickwork is sufficiently uneven to mask the distortion (with or without correction). It would need a more formal test to show it properly.

 

Mind you that does not explain why in camera raw conversion produces full pixel count and from C1 it can lose over 1.22Mpx in some case (down to 14.77Mpx from 16).

 

For now I have switched off the automatic distortion correction and will only use it when the nature of the picture demands it. I will check again when next release of C1Pro comes out Thanks for working with me to pin it down.

Link to comment
Share on other sites

I had a play with some brickwork and the degree of correction that is being applied by C1 would hardly appear to be noticeable in most subjects. Only those with strong lines parallel to the frame edges. Even then it seems possible that there is overcompensation from slight pin cushion to slight barrel or vice versa. It would not be demonstrable in small images from the pictures (so not posting them) I took of the brickwork of my house - close-up the brickwork is sufficiently uneven to mask the distortion (with or without correction). It would need a more formal test to show it properly.

 

Mind you that does not explain why in camera raw conversion produces full pixel count and from C1 it can lose over 1.22Mpx in some case (down to 14.77Mpx from 16).

 

For now I have switched off the automatic distortion correction and will only use it when the nature of the picture demands it. I will check again when next release of C1Pro comes out Thanks for working with me to pin it down.

 

No worries, and thanks for raising it. I was uploading a batch to Corbis and noticed that a few were falling a bit shot of what I would expect in file size (tiffs) and spotted that they were down to 4700px with the distortion correction on. When switched off it gave me 4896px. Doesn't seem much but it can make a difference.... especially if I want to up-scale them to 5000 or 5100px.

 

Like yourself, it is hard to see what they are correcting as when it is off there is very little sign of any distortion. I'd rather have the extra pixels!

Link to comment
Share on other sites

Phase One's response to this issue:

Hello,

Your raw files still have the original size available, however what you are mostlikley seeing is that the images have a lens correction applied to them. Any distortion corrections will limit the size of the image becuase a crop is needed to acount for the curved shape of the corrected file. You can change your correction settings int he Lens correction tool, and even disable automatic corrections if you prefer. To disable automatic corrections go to the Lens correction tool and click on the triangle in the upper right corner. Choose "Disable default lens corrections." Then any new images will not have the correction applied to them. Existing images would need to be reset manually.

Best regards,
Team Phase One Support

 

That is pretty much what I had told them in my support ticket (and a couple of updates). My reply:

 

That is what I determined and have switched off the automatic lens correction. But the raw conversion in camera presumably applies the lens correction as well and it produces full size images without cropping. I assume the camera is using the "spare" pixels on the sensor outside the specified size for the correction rather than doing it within the image area and ignoring those spare pixels. Otherwise what are the spare pixels for?

Sorry I am not at all happy with the obvious answer that you have provided. I believe the Capture 1 lens correction processing is incorrect.

 

I dont' expect to get any further with Phase One. Although I prefer Capture 1 I may still see about upgrading my LR to 5.4. Anyone know what result LR gives after lens correction? With capture 1 lens correction is costing more than 6% of overall resolution on the long end of the 55-200mm..

Link to comment
Share on other sites

Phase One's further reply:

 

Hi Martin,

What is more likely is that in-camera lens corrections simply enlarges the final jpeg to match the size of the original raw. You could achieve the same thing in Capture One by setting your output size to a specific pixel dimension rather than 100%.

Spare pixels on the outside of the sensor are typically used for calibration and ensureing the color and detail at the very edge of the frame is accurate. There are not enough pixels at the outside of a sensor that are typically ignored to fill in the amount of cropped data from a lens correction.

Best regards,
Team Phase One Support
 
So now we know what those "spare" pixels are for, maybe! Actually I can appreciate that some would be needed for demosaicing at the edges of the frame.
Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.