need sharp edges PDF line art :(

[lepr]

1 hour ago, Lagarto said:

I think this is a bit disturbing, this kind of a change should not happen when there is a good ground to assume that the included images are basically just "saved", when downsampling is disallowed. I think there is a similar issue in a thread related to exporting TIFF images: users are confused when seeing that exporting to TIFF requires selection of "resample" method when they basically just want to save, that is, keep the original image without losing a pixel.

Exporting a standard TIFF (i.e. without Affinity extensions such as layers and vector objects) results in non-destructively transformed raster-based objects (Image, Pixel and Mask objects) of the Affinity document being resampled to the export pixel density for flattening to the final image, and so the resampling method is in effect there even when the export is at full scale.

[thomaso]

52 minutes ago, Lagarto said:

I think this is a bit disturbing, this kind of a change should not happen

I actually don't trust anymore what APub or my (old) Acrobat do tell me about specific data.
I just experienced another odd export compression result: a JPG being ZIP compressed while another still JPG as expected. Both differing in their original DPI only (same px dimensions):
https://forum.affinity.serif.com/uploads/monthly_2020_06/gabe's_afpub_and_my_pdfs_-_compression.m4v.933d1575634b83cf24e48a7f4ab2ed55.m4v

So I better quit now with testing this aspects any further, to avoid useless results or confusion, possibly caused just by wrong meta data interpretation within the apps.

[thomaso]

1 minute ago, anon2 said:

Exporting a standard TIFF (i.e. without Affinity extensions such as layers and vector objects) results in non-destructively transformed raster-based objects (Image and Pixel objects) of the Affinity document being resampled to the document pixel density for flattening to the final image, and so the resampling method is in effect there.

It appears to be similar for PNG, which also appears to get ZIP compressed on PDF export and not being converted to a JPG inside the PDF.

However, as Lagarto mentioned, it would be helpful to know a documented info about how it should work 'by design' in APub, to be able at all to judge whether this 'design' is also the intended behavior or possibly an issue.

[lepr]

1 minute ago, thomaso said:

It appears to be similar for PNG, which also appears to get ZIP compressed on PDF export and not being converted to a JPG inside the PDF.

All flat raster file exports are (by necessity) processed as I described for a flat TIFF file. A non-flattened PDF export, depending on settings, is capable of containing non-destructively transformed raster-based objects and, in that case, the resampling method would best be greyed-out when resampling is disabled.

[thomaso]

5 minutes ago, anon2 said:

A non-flattened PDF export, depending on settings, is capable of containing non-destructively transformed raster-based objects

Does it mean in a workflow with ZIP-compressed resources (like TIF), an exported PDF must be flattened to force a re- and/or JPG-compression, possibly conscious with less quality in the sake to get a smaller PDF file size? – If yes, then on the other hand the PDF file size might also increase in case flattening causes any transparency in the .apub being rasterized.

An unavoidable win-loose situation this way?
And what would happen if I rasterize before export certain objects (e.g. TIF)? How would that influence their compression on export without flattening?

That just makes me wonder if the command "Rasterise..." isn't meant to open an options window because of its three dots (which, at least on mac, is an indicator for 'harmless use' because of a following necessary confirmation by the user):

I assume instead it rasterises directly on click and according to the current document resolution, right? So to get it rasterised in higher DPI it would help to temporarily increase the document dpi?

[lacerto]

(...)

[lepr]

2 hours ago, thomaso said:

Does it mean in a workflow with ZIP-compressed resources (like TIF), an exported PDF must be flattened to force a re- and/or JPG-compression, possibly conscious with less quality in the sake to get a smaller PDF file size? – If yes, then on the other hand the PDF file size might also increase in case flattening causes any transparency in the .apub being rasterized.

An unavoidable win-loose situation this way?
And what would happen if I rasterize before export certain objects (e.g. TIF)? How would that influence their compression on export without flattening?

Make of this what you will.

I placed a Grey 8 BPC 1200 PPI TIFF (ZIP compression) with Greyscale D50 profile in a RGB 8 BPC 300 PPI Publisher document with sRGB profile. Then I made several duplicates of the placed Image, scaling and/or rotating them.

Using the PDF "for print" preset as a starting point, various combinations of resampling and JPEG compression being enabled or disabled resulted in various PDFs containing RGB TIFFs and/or Grey PNGs with PPI of 1200, 96 or 72. (PNG is not a typo by me.) None contained a JPEG. At least, that's what Publisher's Resource Manager says is in the PDFs. I am at home at the moment, where I choose to use no Adobe app, so I am unable to check the PDFs in Acrobat until back in the workplace (whenever that will be).

 

[lacerto]

(...)

[thomaso]

20 minutes ago, Lagarto said:

: I had used the Vector Crop Tool on my image, and that causes the image color space to turn from gray to CMYK,

Oh, the vector crop tool is known and logged for causing unexpected things.
Its opacity issue was listed to be fixed in a recent beta, but reported again, and still happens in v.1.8.3:

 

[lacerto]

(...)

[lacerto]

(...)

[Wosven]

21 minutes ago, Lagarto said:

why did you feel confused about this post?

Not confused, but sad about the side effect of crop tool!

But usually I use form like rectangle and modified, or some made with the pen tool, and I need to export the result to PDF or a file format using a profile, so I don't think it's causing problems with my files.

I can image receiving a file and searching why some colours get ugly (like red) once exported to CMYK…

[lepr]

2 minutes ago, Wosven said:

Not confused, but sad about the side effect of crop tool!

People ambiguously use the sad emoticon for two very different reasons here:

1. as you did, to indicate that you feel sadness about a situation, or

2. to indicate that they think a person who posted is "sad", as in a "loser" or, in other words, someone to be despised. (I got slapped with one of these today, as it happens, lol.)

 

[lacerto]

(...)

[Wosven]

1 minute ago, anon2 said:

People ambiguously use the sad emoticon for two very different reasons here

The second option is a subtlety I'm not aware of, and I can't remember reading it in a book with this meaning, but if so, I certainly was confused and didn't understand.

There wasn't any opinion but facts in your post, @Lagarto , don't worry.

[lacerto]

(...)

[MikeW]

Thoroughly testing options/results and reporting same isn't being pedantic. 

At least not to me. I enjoy reading your thoroughness even if I don't respond. Well, I guess like here I might respond in an offhand, off topic manner...

[thomaso]

3 hours ago, anon2 said:

People ambiguously use the sad emoticon for two very different reasons here:

1. as you did, to indicate that you feel sadness about a situation, or

2. to indicate that they think a person who posted is "sad", as in a "loser" or, in other words, someone to be despised. (I got slapped with one of these today, as it happens, lol.)

Isn't it actually that the two negative emoticons in the forum are more used to "judge" the author (in particular her/his mood in the eyes of the reader) than to judge in fact the content, info or opinion? Otherwise any bug report would deserve the sad icon, because its info always is sad – even though the fact that it got reported rather is positiv. Whereas if positive emoticons would judge an author, too, a bug report could always deserve a heart or smiley. 😉

In that way I also first was confused by Wosven's emotion but it appeared clear from Lagarto's text that it was related to the info, not the author. (maybe it would have been less confusing if it was given to the post above, the one with the bug info.)

Sometimes, when I feel a sense of hidden, subtle humor in a post, I would like to press the laughing icon. But I avoid it because it might get read as if making fun of the post or author, in case the subtle isn't felt by others, or even by the author her/himself.

[lepr]

2 hours ago, thomaso said:

Isn't it actually that the two negative emoticons in the forum are more used to "judge" the author (in particular her/his mood in the eyes of the reader) than to judge in fact the content, info or opinion?

I've certainly interpreted the confused emoticon as being used by a person to tell the author that the author appears to be confused or has written nonsense.

2 hours ago, thomaso said:

Sometimes, when I feel a sense of hidden, subtle humor in a post, I would like to press the laughing icon. But I avoid it because it might get read as if making fun of the post or author, in case the subtle isn't felt by others, or even by the author her/himself.

Yes, same for me.

[v_kyr]

5 minutes ago, anon2 said:

I've certainly interpreted the confused emoticon as being used by a person to tell the author that the author appears to be confused or has written nonsense.

Yip, that's how I would rate that emoticon too in such circumstances. - Personally, I thus often prefer to express certain things in text instead, where things may be then less unmistakable.

[lepr]

3 minutes ago, v_kyr said:

Yip, that's how I would rate that emoticon too in such circumstances. - Personally, I thus often prefer to express certain things in text instead, where things may be then less unmistakable.

In my opinion, the forum would be better with only the "like" and "thanks" emoticons.

[v_kyr]

24 minutes ago, anon2 said:

In my opinion, the forum would be better with only the "like" and "thanks" emoticons.

Basically yes.

Also I for my part can't imagine to ever would have a need to express in comments something from of all those here then ...

[thomaso]