Request: keystone malfunction (#15943)

Discussions, questions, comments and suggestions regarding Capture One PRO, Capture One PRO For Sony / Fujifilm, Capture One for Phase One and Capture One Express For Sony / For Fujifilm 20.x for Mac

Request: keystone malfunction (#15943)

Postby thomaskyhn » Thu Jan 16, 2020 1:19 pm

I don't have my initial request as I didn't save it. The issue is the regular malfunction of the keystone adjustment tool when making simultaneous vertical and horizontal adjustments using the four adjustment points. The expected result of such adjustment is a rectangle with two horizontal and two vertical sides, but instead of this you often get a parallelogram in which the sides that were supposed to be vertical aren't actually vertical.

Below is a part of what I've sent to support.

. . .

What I expect the keystone correction tool to do is the following:

Regardless of the content of the photo, and regardless of where or how you place the four adjustment points, as long as you have a four-sided figure, the result of applying corrections should be a rectangle with two completely horizontal and two completely vertical sides. Again, the content of the photo is irrelevant here, and so is the distortion that may result from the adjustment (extreme settings will produce extreme results with most other adjustment parameters as well). If for some reason C1's keystone correction tool doesn't work in this way, it should at least be stated clearly in the user guide along with a description of what it can and what it can't to do.

That you may be able to produce the desired result using the adjustment sliders is another matter. The main point here is that keystone correction using the four adjustment points produces unexpected results. If this is due to a limitation in what the keystone tool is able correct, it should be stated in the user guide, so users would know what results to expect – preferably with an explanation of why the tool is limited in this way, when corresponding tools in other RAW editors do not have such limitations. If this limitation isn't intended, I assume the tool isn't working as it should, and I hope that it will be fixed in the next update.
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK

Re: Request: keystone malfunction (#15943)

Postby thomaskyhn » Thu Jan 16, 2020 4:10 pm

Just to give an example of what this is about so that it's absolutely clear.

Here's a model of something you would potentially want to correct in an actual photo, a four-sided figure that should have had 90° angles:

Image

Here it is before applying corrections in Capture One:

Image

And here it is after pressing Apply – obviously not the result you would expect:

Image

And this is how it looks when adjusted in Lightroom – the rectangle you would have expected from Capture One too:

Image

Here's the example file in case anyone wants to try for themselves: https://i.imgur.com/GW5thjc.jpg
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK

Re: Request: keystone malfunction (#15943)

Postby Keith Reeder » Thu Jan 16, 2020 4:29 pm

Who are you talking to, Thomas? This is a user-to-user forum.
Keith Reeder
 
Posts: 1364
Joined: Mon Sep 29, 2008 7:43 pm
Location: Blyth, NE England

Re: Request: keystone malfunction (#15943)

Postby thomaskyhn » Thu Jan 16, 2020 4:33 pm

Keith Reeder wrote:Who are you talking to, Thomas? This is a user-to-user forum.

I know. I suggested elsewhere (https://forum.phaseone.com/En/viewtopic.php?f=78&t=37622&start=105#p176641) that since users don't have access to support requests, feature requests and bug reports, not even their own, it would perhaps be a good idea to post them here, just for reference, so other users know what has been reported/requested and to make request numbers available for reference too.

This way other users would also be able to refer to reports/requests made by other users when contacting support themselves.
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK

Re: Request: keystone malfunction (#15943)

Postby NN635680879799322049UL » Thu Jan 16, 2020 11:17 pm

Regarding the unexpected 4-point keystone "correction" producing a parallelogram where a rectangle would be expected:

I can see this, but not always. When it happens, I can get the correction to work by separately correcting the horizontal and then the vertical. If anything, I would expect the simultaneous 4-point correction to work better. A 4-point correction should have the same result as two 2-point corrections. Perhaps if you report this to support, they will accept that it is a bug and not a user error since it is completely reproducible, cannot be construed as a feature, and needs to be fixed if the 4-point keystone correction is to work reliably.

Jerry C
NN635680879799322049UL
 
Posts: 285
Joined: Sun May 24, 2015 7:12 pm

Re: Request: keystone malfunction (#15943)

Postby thomaskyhn » Thu Jan 16, 2020 11:54 pm

NN635680879799322049UL wrote:Regarding the unexpected 4-point keystone "correction" producing a parallelogram where a rectangle would be expected:

I can see this, but not always. When it happens, I can get the correction to work by separately correcting the horizontal and then the vertical. If anything, I would expect the simultaneous 4-point correction to work better. A 4-point correction should have the same result as two 2-point corrections. Perhaps if you report this to support, they will accept that it is a bug and not a user error since it is completely reproducible, cannot be construed as a feature, and needs to be fixed if the 4-point keystone correction is to work reliably.

Jerry C

Yes, it's definitely reproducible.

I just tried correcting the file in the example above in two separate stages, vertically and then horizontally, and the other way around too, and this way I get two different results, none of them correct though.

In the last email I got from the guy who responded to my support request, he said that he'd forward the request and info to "the R&D and Product Management team as something to review for future releases". Hopefully they'll do something about it soon.
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK

Re: Request: keystone malfunction (#15943)

Postby photoGrant » Fri Jan 17, 2020 4:21 am

Keith Reeder wrote:Who are you talking to, Thomas? This is a user-to-user forum.


What a useless comment.

Hi Thomas, I tried reproducing this in the latest release with the same results. Thanks for sharing. I'll submit a report to concur. Wouldn't bet on a fix within a reasonable/professional time frame however.
photoGrant
 
Posts: 250
Joined: Thu Sep 18, 2014 5:16 pm
Location: Midbest

Re: Request: keystone malfunction (#15943)

Postby NN635680879799322049UL » Fri Jan 17, 2020 7:07 am

thomaskyhn wrote:
NN635680879799322049UL wrote:Regarding the unexpected 4-point keystone "correction" producing a parallelogram where a rectangle would be expected:

I can see this, but not always. When it happens, I can get the correction to work by separately correcting the horizontal and then the vertical. If anything, I would expect the simultaneous 4-point correction to work better. A 4-point correction should have the same result as two 2-point corrections. Perhaps if you report this to support, they will accept that it is a bug and not a user error since it is completely reproducible, cannot be construed as a feature, and needs to be fixed if the 4-point keystone correction is to work reliably.

Jerry C

Yes, it's definitely reproducible.

I just tried correcting the file in the example above in two separate stages, vertically and then horizontally, and the other way around too, and this way I get two different results, none of them correct though.

In the last email I got from the guy who responded to my support request, he said that he'd forward the request and info to "the R&D and Product Management team as something to review for future releases". Hopefully they'll do something about it soon.


I thought the 2-step correction looked rectangular compared with the single 4-point correction, but I did not look at it except to eyeball it.

When support tells you they are forwarding it to R&D, sometimes it gets fixed in the next iteration, especially when it is reproducible, as this one is.

Jerry C
NN635680879799322049UL
 
Posts: 285
Joined: Sun May 24, 2015 7:12 pm

Re: Request: keystone malfunction (#15943)

Postby thomaskyhn » Fri Jan 17, 2020 10:47 am

photoGrant wrote:Hi Thomas, I tried reproducing this in the latest release with the same results. Thanks for sharing. I'll submit a report to concur. Wouldn't bet on a fix within a reasonable/professional time frame however.

Hi Grant. Thanks for your comment – and for confirming. I'm sure you're right, this will require a great deal of patience.
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK

Re: Request: keystone malfunction (#15943)

Postby photoGrant » Tue Jan 28, 2020 1:17 am

Hi Thomas,

This is still prevalent in 13.0.2.19 (latest build release Jan 27 2020).

I submitted a report.
photoGrant
 
Posts: 250
Joined: Thu Sep 18, 2014 5:16 pm
Location: Midbest

Re: Request: keystone malfunction (#15943)

Postby thomaskyhn » Tue Jan 28, 2020 11:20 am

photoGrant wrote:Hi Thomas,

This is still prevalent in 13.0.2.19 (latest build release Jan 27 2020).

I submitted a report.

Hi Grant,
Yes, it was the first thing I checked after updating. It will probably take a while before this will be fixed. If it will be fixed. At least they're aware of it now.
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK

Re: Request: keystone malfunction (#15943)

Postby thomaskyhn » Tue Jan 28, 2020 12:56 pm

I asked about a fix on Facebook and got the reply that the "Keystone tool is not intended to work in the way that you are suggesting". If this is the case, does anyone know how it is intended to work?
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK

Re: Request: keystone malfunction (#15943)

Postby thomaskyhn » Tue Jan 28, 2020 3:18 pm

thomaskyhn wrote:I asked about a fix on Facebook and got the reply that the "Keystone tool is not intended to work in the way that you are suggesting". If this is the case, does anyone know how it is intended to work?

I just received another message from support about this. It seems they're looking into it.
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK

Re: Request: keystone malfunction (#15943)

Postby photoGrant » Tue Jan 28, 2020 9:03 pm

thomaskyhn wrote:
thomaskyhn wrote:I asked about a fix on Facebook and got the reply that the "Keystone tool is not intended to work in the way that you are suggesting". If this is the case, does anyone know how it is intended to work?

I just received another message from support about this. It seems they're looking into it.


Yeah if this working as designed... I'm not sure what its objective is.
photoGrant
 
Posts: 250
Joined: Thu Sep 18, 2014 5:16 pm
Location: Midbest

Re: Request: keystone malfunction (#15943)

Postby thomaskyhn » Wed Jan 29, 2020 11:24 am

photoGrant wrote:
thomaskyhn wrote:
thomaskyhn wrote:I asked about a fix on Facebook and got the reply that the "Keystone tool is not intended to work in the way that you are suggesting". If this is the case, does anyone know how it is intended to work?

I just received another message from support about this. It seems they're looking into it.


Yeah if this working as designed... I'm not sure what its objective is.

I just received another reply from support saying the ticket has "highest priority". So there's no doubt that they recognize the issue now. A very positive development.
thomaskyhn
 
Posts: 166
Joined: Wed Oct 24, 2018 5:50 pm
Location: CPH, DK


Return to Capture One 20.x Software for Mac



Who is online

Users browsing this forum: No registered users and 3 guests