Jump to content

Recommended Posts

Posted

I am reviving this old issue as I've just discovered that it is still the case with the latest version (1.10.5) of Affinity Photo.

I have a scan of a CD cover saved in TIFF format, which I have cleaned up and resized in Affinity Photo. The last step in my workflow is to adjust the levels when needed, and then proceed to flatten the adjustment layer down and to export the image in JPEG 100 quality format. Unfortunately, as other users have reported, doing the flattening creates a banded white/transparent border of 1 pixel all around the picture, ruining the end result.

Due to copyright reasons I can't attach the .afphoto file for public consumption. I can however provide it in a private manner to the support staff for further analysis, if needed.

Thanks.

Posted

Can you reproduce the problem with an image that's not copyrighted? If so, you could share that.afphoto file here, and more of us could help with the analysis.

-- Walt
Designer, Photo, and Publisher V1 and V2 at latest retail and beta releases
PC:
    Desktop:  Windows 11 Pro 23H2, 64GB memory, AMD Ryzen 9 5900 12-Core @ 3.00 GHz, NVIDIA GeForce RTX 3090 

    Laptop:  Windows 11 Pro 23H2, 32GB memory, Intel Core i7-10750H @ 2.60GHz, Intel UHD Graphics Comet Lake GT2 and NVIDIA GeForce RTX 3070 Laptop GPU.
    Laptop 2: Windows 11 Pro 24H2,  16GB memory, Snapdragon(R) X Elite - X1E80100 - Qualcomm(R) Oryon(TM) 12 Core CPU 4.01 GHz, Qualcomm(R) Adreno(TM) X1-85 GPU
iPad:  iPad Pro M1, 12.9": iPadOS 18.5, Apple Pencil 2, Magic Keyboard 
Mac:  2023 M2 MacBook Air 15", 16GB memory, macOS Sequoia 15.5

Posted

Hi Walt,

Unfortunately not. I've just tried to download a couple of royalty free images from the Internet, but with them I cannot reproduce the bug. I could try to scan something different and see if it occurs only with those TIFF images (Epson scanner, with Epson calibrated color profile).

Thanks.

Posted

OK, I have been able to reproduce the issue from another dummy scan, although I think the problem lies on the resize of not merged images. Here you will find attached two .afphoto files: the first come straight from a scanned (TIFF) image. I've cropped the source image, added a levels adjustment layer, and if you merge down the layers the resulting image shows no artifacts, that is no white borders.

The second image is exactly the same as the first one, but I have scaled down the dimensions just before saving it with a different name. I have left the layers not merged on purpose, and here you should be see that when the the layers are merged down the white/transparent border pops up around the merged image. Adding the adjustment layer before or after the rescaling doesn't matter, the white border is always added when merging down.

Hope it helps, and thanks again.

test_scan_no_resize.afphoto test_scan_resized.afphoto

Posted

The 2nd (resize) document has partial pixel values for the X,Y coordinates of the background layer

Change these values to the nearest whole number (or Rasterise & Trim the background layer first) then the Merge Down will work as expected

To save time I am currently using an automated AI to reply to some posts on this forum. If any of "my" posts are wrong or appear to be total b*ll*cks they are the ones generated by the AI. If correct they were probably mine. I apologise for any mistakes made by my AI - I'm sure it will improve with time.

Posted
18 hours ago, carl123 said:

The 2nd (resize) document has partial pixel values for the X,Y coordinates of the background layer

Change these values to the nearest whole number (or Rasterise & Trim the background layer first) then the Merge Down will work as expected

@carl123, thank you for your answer. Where did you see the partial pixel values? How do I change these values?

Second question: as I've explained, I have taken the first image, scaled down with Affinity Photo to 960x960 pixels, and saved as the second image. The first image had no scaling, I did just crop out part of the scanned image. Does this mean that those "partial" (fractional?) pixel values has been created by the resize operation? How do I avoid this?

To be quite frank, I would expect the program to work like any other image processing app by default, and to be more esoteric on demand. You are not expected to hunt partial pixel values upon the first usage, right? That is, it has never occurred to me anything like the above with PhotoShop, paint.net or other apps.

Thank you again for your time.

Posted
5 hours ago, WrongWorld said:

Where did you see the partial pixel values? How do I change these values?

Look in the Transform panel

Make sure you have set your decimal places to 2 or more in Edit > Preferences > User Interface

To save time I am currently using an automated AI to reply to some posts on this forum. If any of "my" posts are wrong or appear to be total b*ll*cks they are the ones generated by the AI. If correct they were probably mine. I apologise for any mistakes made by my AI - I'm sure it will improve with time.

Posted
15 hours ago, carl123 said:

Look in the Transform panel

Make sure you have set your decimal places to 2 or more in Edit > Preferences > User Interface

Weird place to check, to say the least.

In the meantime I did some more search in this forum, and I've discovered that my issue is a recurring one for which users have asked since years to have a fix, to no avail. 'cause to me all this story about blurring when merging down (but not when merging visible layers!), fractional pixels, white borders, banding borders (yes, I had also these ones), etc. are a bunch of unneeded annoyances for the user, been (s)he well versed toward digital photo editing or less.

It really amazes me that Serif pretends to be the only one on the right track, and all the other competitors (Adobe included) not having a grasp on how to deal properly with layers merging. Except, all the others get it right, and Serif doesn't.

I am also still wondering how, from a scanned TIFF image (using integer pixels, and not even multi-layered) that has been cropped, it is possible with a resize operation (with integer values) to end up with an image having partial pixels which needs "black magic" to properly merge down. Not that I expect some official answer, well to me is a (big) bug.

Apologize for the rant, and thank you very much for your time.

Posted
On 4/16/2022 at 1:35 PM, WrongWorld said:

I am also still wondering how, from a scanned TIFF image (using integer pixels, and not even multi-layered) that has been cropped, it is possible with a resize operation (with integer values) to end up with an image having partial pixels which needs "black magic" to properly merge down. Not that I expect some official answer, well to me is a (big) bug.

It remains unclear to me what sequence of operations you did, exactly, to get the first file. Here is what you said:

On 4/14/2022 at 6:55 PM, WrongWorld said:

the first come straight from a scanned (TIFF) image. I've cropped the source image, added a levels adjustment layer,

But here is what I see:

image.png.ff11ee2147a9ee6165eb617c4dced44f.png

Note that the Background layer is selected, but the visible content it is much smaller than the blue bounding box that shows the layer size. Also note that the bounding box starts at negative X/Y coordinates as shown in the Transform panel, and that the width/height of that layer are bigger than the document size (which is 1700 x 1700px.

From all that, it would appear that indeed you have cropped, but:

  1. How did the background layer get positioned at negative coordinates; and
  2. Why is it bigger than the document size? Did you resize the document without mentioning it? Or did you perhaps start with an empty document that was 1700x1700 px and then Place the scanned image into it at those negative coordinates, and then crop it?

In any case, if you've Cropped it you might want to "complete" the cropping and use Layer > Rasterize & Trim (or the same option is available if you right-click on the layer in the Layers panel).

-- Walt
Designer, Photo, and Publisher V1 and V2 at latest retail and beta releases
PC:
    Desktop:  Windows 11 Pro 23H2, 64GB memory, AMD Ryzen 9 5900 12-Core @ 3.00 GHz, NVIDIA GeForce RTX 3090 

    Laptop:  Windows 11 Pro 23H2, 32GB memory, Intel Core i7-10750H @ 2.60GHz, Intel UHD Graphics Comet Lake GT2 and NVIDIA GeForce RTX 3070 Laptop GPU.
    Laptop 2: Windows 11 Pro 24H2,  16GB memory, Snapdragon(R) X Elite - X1E80100 - Qualcomm(R) Oryon(TM) 12 Core CPU 4.01 GHz, Qualcomm(R) Adreno(TM) X1-85 GPU
iPad:  iPad Pro M1, 12.9": iPadOS 18.5, Apple Pencil 2, Magic Keyboard 
Mac:  2023 M2 MacBook Air 15", 16GB memory, macOS Sequoia 15.5

Posted
28 minutes ago, walt.farrell said:

Note that the Background layer is selected, but the visible content it is much smaller than the blue bounding box that shows the layer size. Also note that the bounding box starts at negative X/Y coordinates as shown in the Transform panel, and that the width/height of that layer are bigger than the document size (which is 1700 x 1700px.

From all that, it would appear that indeed you have cropped, but:

  1. How did the background layer get positioned at negative coordinates; and
  2. Why is it bigger than the document size? Did you resize the document without mentioning it? Or did you perhaps start with an empty document that was 1700x1700 px and then Place the scanned image into it at those negative coordinates, and then crop it?

In any case, if you've Cropped it you might want to "complete" the cropping and use Layer > Rasterize & Trim (or the same option is available if you right-click on the layer in the Layers panel).

Hi Walt, thank you for your answer. This is what I did, and I've just repeated the operation on the original TIFF image just to be sure I didn't goofed up involuntarily.

  1. Open the TIFF image from Windows Explorer, choosing to open it with Affinity Photo (via Open with)
  2. Select the crop tool
  3. Use 1700x1700 for the cropping area (typed directly in the text boxes), and then align with the mouse the cropping area to the bottom right corner of the image

The image is then (apparently?) cropped, however now that I know where to look for I've selected the layer with the pen tool, and effectively the "blue" border is still stuck at the original image size. This is counterintuitive:

  • If the idea was to create a sort of cropped "viewport" on the image (as the transformation panel hints after the crop operation), I don't see a simple way to reposition such cropped "viewport" on the whole image (PhotoShop can, for instance, albeit I seem to remember you have limitations on the background layer).
  • If indeed the intention was to crop the document, than I fail to see why I have to right click on the background layer, already a pixel layer itself, and ask for Rasterize and Trim... to complete the operation. How am I visually hinted that there is still a hidden portion of my original image after the crop, and that I should really trim it out with a different operation if I want to get rid of it?
  • Whatever is the decision behind this uncropped part being left hidden within the image, I still fail to see how it has to affect negatively a subsequent down-merging of adjustment layers if I decide to further scale down the image's dimensions.

I am attaching the original dummy TIFF image, just in case someone wants to reproduce the issue.

Thanks again.

img_671.tif

Posted
1 hour ago, WrongWorld said:

If the idea was to create a sort of cropped "viewport" on the image (as the transformation panel hints after the crop operation), I don't see a simple way to reposition such cropped "viewport" on the whole image

You can reposition while cropping, if you want. After, you can drag an edge of the layer to uncrop, but to do that you'd have to unlock the layer first.

1 hour ago, WrongWorld said:

If indeed the intention was to crop the document, than I fail to see why I have to right click on the background layer, already a pixel layer itself, and ask for Rasterize and Trim... to complete the operation.

Cropping in Photo is non-destructive. But sometimes you need to make it destructive.

1 hour ago, WrongWorld said:

How am I visually hinted that there is still a hidden portion of my original image after the crop, and that I should really trim it out with a different operation if I want to get rid of it?

The bounding box is one visual hint. Mostly, as you learn how the application works, you learn when you need to rasterize and when you don't.

2 hours ago, WrongWorld said:

Whatever is the decision behind this uncropped part being left hidden within the image, I still fail to see how it has to affect negatively a subsequent down-merging of adjustment layers if I decide to further scale down the image's dimensions.

That will, I think, be a function of when you applied the adjustment and whether you Rasterized & Trimmed. 

The way you did it, you probably have an adjustment layer that is larger than the cropped layer content, which may be why you had the problem. (I have not tried to confirm that.)

-- Walt
Designer, Photo, and Publisher V1 and V2 at latest retail and beta releases
PC:
    Desktop:  Windows 11 Pro 23H2, 64GB memory, AMD Ryzen 9 5900 12-Core @ 3.00 GHz, NVIDIA GeForce RTX 3090 

    Laptop:  Windows 11 Pro 23H2, 32GB memory, Intel Core i7-10750H @ 2.60GHz, Intel UHD Graphics Comet Lake GT2 and NVIDIA GeForce RTX 3070 Laptop GPU.
    Laptop 2: Windows 11 Pro 24H2,  16GB memory, Snapdragon(R) X Elite - X1E80100 - Qualcomm(R) Oryon(TM) 12 Core CPU 4.01 GHz, Qualcomm(R) Adreno(TM) X1-85 GPU
iPad:  iPad Pro M1, 12.9": iPadOS 18.5, Apple Pencil 2, Magic Keyboard 
Mac:  2023 M2 MacBook Air 15", 16GB memory, macOS Sequoia 15.5

Posted

Walt, thanks for the time spent on giving me some details on how AP works. I still stand that in certain aspect the implementation of cropping is somewhat convoluted, and granted: once you know you can understand (still...), yet it should be more straightforward and without needing to look at the bounding box (which is not shown immediately after the image is cropped). So, I agree it's not ideal (IMHO), but the logic works.

Where I do not agree at all (not with you, clearly) is that there should be a "right" moment to use an adjustment layer. The cropped image is going to be down-sampled from 1700x1700 to 960x960 pixels, and this is all what I want to know. I don't want to care about the fact the document size is still larger, because even then let's have the adjustment layer aligned and sized properly, and leave all this partial pixel nonsense out of the equation (and I am not the only customer complaining for this behavior). I repeat myself: no other program I know, with destructive or non-destructive editing, acts like this, and for a reason. And just to be sure, I have just edited the same TIFF with ON1 Photo RAW (also not destructive) and it produced a down-merged, scaled (exported) image with not borders.

Thanks.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...

Important Information

Terms of Use | Privacy Policy | Guidelines | 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.