Source cursor display in "copy source" with HF4

Announcement of new releases, bugs, support, suggestions
Post Reply
fredkahan
Posts: 14
Joined: 19.07.2006 21:04

Source cursor display in "copy source" with HF4

Post by fredkahan »

Dan
I had earlier complained that the new "retouching" in HF4 did not take into account the internal corrections made to source slice for inevitable displacement and magnification. You insisted that it does. I WAS WRONG!

I was misled by the cursor display in the source window. It was so far off the composite that I never tried the "copy source" routine. It in fact does sample the intended region of the source slice .. often far from the cursor display .. and does just what I want (usually elimination of edge echoes in the background).

BRAVO Helicon Focus Version 4!!!
Dan Kozub
Posts: 355
Joined: 24.03.2004 18:14

Post by Dan Kozub »

Actually it should copy the area under cursor, so it looks like to be a bug.

I would appreciate the print screen with this situation. And maybe a couple of images from your stack to recreate the problem here.

Danylo
fredkahan
Posts: 14
Joined: 19.07.2006 21:04

"Copy Source" cursor offset

Post by fredkahan »

Dan
My workflow involves 10 or more slices from a D200 as 35 Mb 48bit Tiffs, obviously too gargantuan to upload.

I therefore set up a demo with mid-quality 24bit JPGs of the original .. 200kb each. Imagine my surprise: NO PROBLEM. Perfect registration up and down the stack of JPG slices!!!

So I decided to give you the screen shot from the original Tiffs (where the problem still persists). A new problem: the output cursor does not record with PrintScreen (nor does the mouse usually). This failure to record on the PrintScreen also occurred when I used an auxiliary Wacom pad.

BTW, my system is a dual monitor: ATI Radeon X300 digital to LCD, analog to a NEC monitor with the LCD as primary on XP Pro.

Is their a trick you can think of to preserve display of both cursors during PrintScreen.

At least we are making some progress in narrowing down the bug. I would be glad to scale the TIFFs to see if it is a TIFF specific problem.

Fred
Dan Kozub
Posts: 355
Joined: 24.03.2004 18:14

Post by Dan Kozub »

Fred,

The only way to grab cusor as I know is to use Corel Capture. But it is not the problem.

I think it depends on image resolution. I will check it with 12mp 16 images.

Danylo
TomMeeks
Posts: 45
Joined: 05.09.2006 00:10
Contact:

I also see misaligned cursors in Copy source and processed

Post by TomMeeks »

I also see that the cursors are very misaligned between the source and processed images when trying to copy a section from one of my source images to the processed one.

This makes one, or the other, cursors useless. I suppose from reading this thread that I should trust the cursor in the processed image pane rather than the one in the source image pane.
Dan Kozub
Posts: 355
Joined: 24.03.2004 18:14

Post by Dan Kozub »

Tom,

Have you updated to version 4.03 of Helicon Focus. We made some changes in retouching fucntions.

Please let me know and send me a print screen.

Regards,

Danylo
TomMeeks
Posts: 45
Joined: 05.09.2006 00:10
Contact:

Post by TomMeeks »

Dan Kozub wrote:Tom,

Have you updated to version 4.03 of Helicon Focus. We made some changes in retouching fucntions.

Please let me know and send me a print screen.

Regards,

Danylo
Yes, I have updated. It appears that it's dependent on the magnification used. I was quite surprised to see some of the magnified images had perfectly lined up cursors. I'll play with it more and get back to you as I gather more experience.
fredkahan
Posts: 14
Joined: 19.07.2006 21:04

"Copy Source" cursor display problems in HF 4.03

Post by fredkahan »

Tom & Dan

I have just retested image stacks in the latest HF version and find that the misalignment of the "copy Source" cursor on the input frame persists on 48 bit Tifs but not on JPG's generated from the D200 NEF originals. I used 100% magnification for the evaluations.

Note that the misalignment is an annoyance rather than a hindrance to obtaining the desired effect; i.e. you can ignore the input cursor and pay attention to the output screen.
TomMeeks
Posts: 45
Joined: 05.09.2006 00:10
Contact:

Re: "Copy Source" cursor display problems in HF 4.

Post by TomMeeks »

fredkahan wrote:Tom & Dan

Note that the misalignment is an annoyance rather than a hindrance to obtaining the desired effect; i.e. you can ignore the input cursor and pay attention to the output screen.
I was concentrating on the source image cursor since it had both circle and crosshairs and, more importantly, it has the information that was missing from the processed image. If we rely on the using the cursor in the processed image completely, we might pick up unwaited out of focus information from the source image.

By being careful, it can be done even when there is a mismatch. But, it's a lot easier when they line up exactly so we can go back and forth as best fits the situation.

This isn't a 'show stopper' for me by any means. It's just an effort to make a great tool even better. I LOVE this product.
Dan Kozub
Posts: 355
Joined: 24.03.2004 18:14

Post by Dan Kozub »

Thanks for the tests! I will check it once more with 16 bit images. I am surprised to hear that color depth has any effect on cursors.
I would rather suspect that RAW conversion can introduce some minor shift in the images.
TomMeeks
Posts: 45
Joined: 05.09.2006 00:10
Contact:

Coming directly from Canon 400XTi JPG

Post by TomMeeks »

The images that I'm working with are being saved by the Canon 400Xti camera as JPGs for now. While I normally prefer RAW, the fact that my ultimate goal is Object VR pushes me to skip the step the RAW to TIFF step. (36 views * 10-20 focus planes is a LOT of RAW conversion!)

I understand many programs are having a rough time recognizing the 400XTi's new RAW format. So, I hadn't even tried that directly in Helicon Focus to see if it worked. If I don't have to do the processing for 360-700 photos I'd be glad to work in RAW. :)
Post Reply