Jump to content
Our response time is longer than usual currently. We're working to answer users as quickly as possible and thank you for your continued patience.

Recommended Posts

I want make puzzle from photography. Before I make something more complicated I create simple image. At the now Knife tool is not exist so then how split / divide photograp based on paths?

 

On the example paradise island photo I want create 5 parts from them in 5 separate files. I'm stuck how divide photograph background when I draw lines. I try select all objects and use binary divide - is not working. I try put photograp inside rectangle then select all curves and use binary operation - nothing change too. I'm looking for way to make custom guide and split based on them too - I can't find this option.

 

I use AD 1.5.3.69. Could you suggest me any solution?

post-50654-0-21034000-1495644254_thumb.png

Link to comment
Share on other sites

i did it this way, create 5 closed shapes than copy the picture and paste inside (ctrl alt v) in each shape.

this is the result.

 

post-35341-0-34114600-1495646596_thumb.png

 

 

intel core i5,  16GB 128Gb ssd win10 Pro Huion new 1060plus.

philips 272p 2560x1440px on intel HD2500 onboard graphics

Razer Tartarus Chroma

Link to comment
Share on other sites

Hi affi.usr,

 

You can convert the photo into curves and draw over to the pen a line that crosses it.

Select the line, "Expand the stroke". Then select the photo and the line and make "Divide" and delete the sections of the cutting line.

 

EDIT

 

Vous pouvez garder la portion centrale de la ligne, la copier et la coller pour l'associer à chaque morceau de la photo.

post-37692-0-32178300-1495646923_thumb.png

post-37692-0-22966900-1495647892_thumb.png

Link to comment
Share on other sites

i have tryed some more, and a puzzle can be made this way.

 

attachicon.gifpuzzle3.png

How hard was it to create the puzzle shapes? Creating each one individually seems tedious.

All 3 1.10.6, & all 3 V2.03 Mac apps; 2020 iMac 27"; 3.8GHz i7, Radeon Pro 5700, 32GB RAM; macOS 10.15.7
Affinity Photo 
1.10.6; Affinity Designer 1.10.6; & all 3 V2 apps for iPad; 6th Generation iPad 32 GB; Apple Pencil; iPadOS 15.7

Link to comment
Share on other sites

  • Moderators

Thank you guys! Yours answers are very helpfull.

 

I'm curious. Is a posibility in AD make custom guides from paths?

 

Hi affi.usr,

No, currently this is not possible. The best you can do is to group the objects you want to use as guides, lock the group and enable snapping to align other objects/elements with them.

Link to comment
Share on other sites

Hi affi.usr,

No, currently this is not possible. The best you can do is to group the objects you want to use as guides, lock the group and enable snapping to align other objects/elements with them.

 

Hi MEB,

 

thank you for your information. Custom guide is basic needed feature fx. distribution text based on the same value degree from center point paths. I hope that will be available soon.

 

*

 

I need some testing. I use reglico technique for test image posted on the first post and I have 4 parts, not 5 after divide. I use power duplicated to create path from center. Right horizontal line not splitting image. When my power supply come back I'll try and I'll write what I observed.

Link to comment
Share on other sites

Hi affi.usr,

 

To get 5 parts, draw a first line (red in my example), divide, move away the left part of the picture, draw a second stroke (blue) with the angle in the white part (out of the picture). Move the right side of the photo and draw the third stroke (green) from the white part between the cut pieces.

post-37692-0-51037300-1495736677_thumb.png

Link to comment
Share on other sites

Hi reglico,

 

I try first another aproach. Create by pen tool line from center to corner, power duplication by 360/5 degree with transformation panel and then use your method. Your way is the easiest. First I need only create paths as guide first and lock them to easier drawing (crucial was 72 degree for path).

 

I can't uderstand why in some occasion divide not working. I can't find what it should be first made for binary operation to work with them. I think that is working in that way: I create object and to start dealing with Intersect, Divide etc., but I need select minimum 2 object and object above affect object below when I divide. For now if I draw path with 3 point in V-shape like style and I select them is a posibility divide if I hide all layers and it simply connect to point on border of artboard (see atached picture). I don't understand this behaviour. For me if I've few layers and one curve visible it should be not posible to divide them, but I can be wrong.

post-50654-0-16371700-1495907278_thumb.png

post-50654-0-81633400-1495907282_thumb.png

Link to comment
Share on other sites

Hi affi.usr,

 

Your method with "Power duplicate" is perfect since you want to have the same angle between the paths. If you use it directly in the photo, it will be cut all at once by the operation "Divide". (Do not forget to convert the photo into curves and make "Expand the stroke" for the 5 lines before selecting the set for the division).

For your second question, I do not have the answer because here I tried with a closed form as suggested by dutcshader, and with an open line and the operation "Divide" works normally: see screenshot. I may be missing something.

post-37692-0-05213100-1495913393_thumb.png

post-37692-0-89039400-1495913393_thumb.png

Link to comment
Share on other sites

Hello reglico,

 

you are absolutly right. I want achive the same angle from center, so I use power duplicate. I tried some other aproaches to find out what makes difference between binary operation.

 

*

 

Closed shape

All working as I expected. Simply select all closed shape objects and use what you want. Very usefull aproach is press Alt (I'm Windows user) and create non destructive boolean operation as we can see on this movie:

https://vimeo.com/110131937

 

Effect attached as "closed shape" image. If you don't need non destructive you can make this without Alt:

https://vimeo.com/110161212

 

Open shape

 

Open shape is another story. Is a difference when line is straight or curvy.

 

Curvy line

With expand stroke path split object directly as curve is drawn. Difference is when expand stroke is not applied. At this time Affinity Designer connect begin and end point and make closed shape before operation (in my example divide) and line behave like arc.

 

Straight line

Without expand stroke on object  is simply remove line and make one object - in my test it was rectangle. So when you create rectangle and draw line by pen tool and apply divide nothing is change. Difference makes expand stroke. Then when I make use divide is working as expected.

 

Summary

Details which I wrote above makes my first troubles.

post-50654-0-31692900-1495954787_thumb.jpg

post-50654-0-95518700-1495955063_thumb.jpg

post-50654-0-07152500-1495955698_thumb.jpg

post-50654-0-72648600-1495956272_thumb.jpg

Link to comment
Share on other sites

Hi affi.user,

 

basically, the explanation of what you found is simple. In order to apply a Boolean operation to a pair of objects, be it in a desctructive or a non-destructive way, these objects must enclose a area on the canvas. And though a line can be assigned a stroke width, the stroke width is immaterial with respect to the Boolean operations. These operations will only take the paths, regarded as mathematical objects, in consideration. Now there are precisely two situations you will have to discern:

  • Your object can be a closed path having a shape like (C.) in my example below. It is geometrically evident, that this path will enclose a non-zero area.
  • But your object can also be an open path, such as (A.) or (B.) in my example. In this case, Affinity Designer will virtually join the end nodes of your curve with a straight line, and on the basis of this virtual connection you will have to answer the question whether the curve encloses a non-zero area.

So when your open path is just a straight line (A.) and you virtually connect the end nodes of this curve with a straigh line, the path and the virtual line will be coincident on the canvas. And it is geometrically sound that your object will enclose the null area then. But the Boolean operations are calculated with respect to this area. So when you subtract an object having the null area from any other object, the result will be just that other object.

 

On the other hand, when your open path is not just a straight line (B.), it will enclose a non-zero area, as soon as you imagine its end nodes to be joined by a straight line. And therefore, it will behave under Boolean operations just as if that virtual straight line was drawn. Compare the result of the subtraction.

 

Finally, when you expand a non-zero stroke, you will actually create a closed path that encompasses a non-zero area, and your object will behave just like the object in case (C.). And as a side note, you will have to understand, that the Divide algorithm is still buggy in Affinity Designer, or rather, it has unexpected results, when an object lies entirely within the boundaries of another object:

 

https://forum.affinity.serif.com/index.php?/topic/12629-affinity-designer-customer-beta-1351/?p=56676

https://forum.affinity.serif.com/index.php?/topic/1319-unexpected-results-using-divide/

 

Hope that makes sense …  :)

Alex

post-1198-0-94507800-1495960816_thumb.png

Link to comment
Share on other sites

Hi A_B_C,

 

thank you for your explenantion! It's make more sense for me!

 

From algorithm perspective and mathematic point of view is more clear for me why boolean operation works in that way. I'm not fully understand (A) situation. When I simply draw it on the paper straitght line and circle has commons points (in geometry titled chord). I think earlier from geometry perspective that common area of points is used to make operation.  From other hand when stroke is 0 in stroke size it is a non zero area as you mentioned (it is like move pencil above paper without any sign of drawind). At this point behaviour is not clear, because on your example all curves has 0 size of stroke, but when line is curvy we have operation and when is straight - is not.

 

Good side of this situation is when we keep this fact in mind we can make design with less troubles.

Link to comment
Share on other sites

Exactly, affi.user, when you look at the Booleans from a set-theoretic view, you are perfectly right. A circle and a straight line intersecting that circle do have points in common and we will call the set of these points a “chord.”

However, the Boolean operations in Affinity Designer do not precisely behave according to the Boolean operations on point sets. Actually, they are no set-theoretic operations in the mathematical sense, witnessed by the fact that you mentioned. So perhaps I should have avoided speaking of “geometry” or “geometric,” since we are inclined to think of geometric objects as point sets nowadays. Unfortunately, I have no better explanation for the current behavior of the Boolean operations. Maybe they were designed according to an older model of geometry, where lines are thought to have a length but no width (Euclid, Elements, Book 1, def. 2: “A line is a breadthless lenghth”), therefore enclosing no area. And when an object that has no area is subtracted from an object, subtraction being strictly area-related, the latter one will remain intact. Honestly, I have no better explanation …

But you must have in mind, that assigning a stroke width to a path simply means to add some sort of decoration to that path. Think of it like an ornament that is mathematically irrelevant. But as soon as you expand the stroke, you will essentially create a new shape from that ornament, and this shape will enclose a non-zero area, if the stroke had a non-zero width. It is essential to have in mind that strokes are just ornaments, nothing more …

Does that make sense?
Alex

Link to comment
Share on other sites

Alex you are great! I really understand now what is expand stroke is. Yes, what you wrote is make sense for me. Now I more oriented how place my design thoughts in Affinity Designer application enviroment than before this topic. Thank you for all answers!

 

From practicall point of view which model are used in programming is less importand from understand how apply our plans in graphics results.

 

At the end of topic I would resolve one issue. When I want boolean operation in picture I have to convert it to curves. Why? Can it be simplier select picture, curves to dealing with on the process and makes operations without that step (convert picture to curves)?

Link to comment
Share on other sites

I assume you are talking about your first screen shot. To be honest, I would do that differently:

  • Grab the Cog Tool and create a cog shape (A.)
  • Adjust control points as seen below (B.)
  • Divide cog shape (C.)
  • Superimpose cog shape on rectangle as seen below, rotate as desired (D.)
  • Select all and divide (E.)
  • EDIT Use the parts of the rectangle as masks or layers to clip to.

I added the cog shape I used for my model. In general, when you are attempting geometric constructions, it is always good to look at the smart shapes. They offer a world of possibilities.

 

Cheers, Alex  :)

post-1198-0-60636400-1495970552_thumb.png

Cog.afdesign

Link to comment
Share on other sites

However, the Boolean operations in Affinity Designer do not precisely behave according to the Boolean operations on point sets. Actually, they are no set-theoretic operations in the mathematical sense, witnessed by the fact that you mentioned...

Without getting too deep into theory, if the set of all points (nodes) of a line are collinear, it is a one dimensional object. Since it exists in only one dimension, it must have zero area, right?

 

Also, I am confused about the several posts in this topic about converting a picture to curves. As far as I know, there is no way to do that in Affinity -- it would require the missing & often requested auto-trace function, correct? dutchshader's method is to copy the picture (presumably a pixel layer) & paste it inside each curve shape. That does not convert the image to curves; it is just an example of vector masking, is it not?

All 3 1.10.6, & all 3 V2.03 Mac apps; 2020 iMac 27"; 3.8GHz i7, Radeon Pro 5700, 32GB RAM; macOS 10.15.7
Affinity Photo 
1.10.6; Affinity Designer 1.10.6; & all 3 V2 apps for iPad; 6th Generation iPad 32 GB; Apple Pencil; iPadOS 15.7

Link to comment
Share on other sites

I could not find the convert to curves on a image i opend. But when placing a image you can convert to curves

intel core i5,  16GB 128Gb ssd win10 Pro Huion new 1060plus.

philips 272p 2560x1440px on intel HD2500 onboard graphics

Razer Tartarus Chroma

Link to comment
Share on other sites

I could not find the convert to curves on a image i opend. But when placing a image you can convert to curves

True, but the placed image is just converted to a single curve, not a "(Curves)" object, & it doesn't help much as far as 'cutting' it into pieces -- the only way I found to do that other than with vector clipping is to duplicate the image multiple times & divide one with each of the 'cutline' curves, & then delete all the extra pieces.

All 3 1.10.6, & all 3 V2.03 Mac apps; 2020 iMac 27"; 3.8GHz i7, Radeon Pro 5700, 32GB RAM; macOS 10.15.7
Affinity Photo 
1.10.6; Affinity Designer 1.10.6; & all 3 V2 apps for iPad; 6th Generation iPad 32 GB; Apple Pencil; iPadOS 15.7

Link to comment
Share on other sites

Without getting too deep into theory, if the set of all points (nodes) of a line are collinear, it is a one dimensional object. Since it exists in only one dimension, it must have zero area, right?

 

Well, there are infinite lines that actually fill up a space, such as the Peano curve or the Hilbert curve …  ;)

 

But seriously, there might be an implementation of the Boolean operations, where you could subtract a line from a shape, in effect splitting the shape along the line. In a certain sense (but surely not a mathematically, or rather topologically, consistent one), this could be regarded as a somewhat more set-theoretic approach. But this is not how the Booleans work in Affinity Designer.

Link to comment
Share on other sites

That does not convert the image to curves; it is just an example of vector masking, is it not?

 

Yes, you are right. The method I suggested is incomplete. You cannot divide an image object or a pixel layer without further ado. You will have to use a clipping or masking method in order to fill the puzzle pieces. Sorry for that omission. I corrected my previous post …   :(

 

… the only way I found to do that other than with vector clipping is to duplicate the image multiple times & divide one with each of the 'cutline' curves, & then delete all the extra pieces.

 

Sounds interesting. But how does that work in detail, R C-R?  :unsure:

Link to comment
Share on other sites

i did this one, just converted the image to curves. created some shapes and divided.

i did not use any clipping or masking.

 

 

post-35341-0-91695900-1495978025_thumb.png

 

intel core i5,  16GB 128Gb ssd win10 Pro Huion new 1060plus.

philips 272p 2560x1440px on intel HD2500 onboard graphics

Razer Tartarus Chroma

Link to comment
Share on other sites

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...
 Share

×
×
  • Create New...

Important Information

Please note there is currently a delay in replying to some post. See pinned thread in the Questions forum. These are the Terms of Use you will be asked to agree to if you join the forum. | 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.