gdenby
-
Posts
1,887 -
Joined
Posts posted by gdenby
-
-
-
Hi, Mr. Lucky,
Just a note. I've been using image vectorizing apps since at least '96. I've never found one that didn't require lengthly sessions clearing up extra nodes, and/or odd dips/peaks.
The best I could do was pre-processing before sending the bitmap to the vectorizer. Things like finding just the right amount of blurring so that the vectorizer might find a smoother arc at one particular grey level. Scaling the image way up, and then down, so the resampling routines would give a very smooth, if somewhat blurry image. Messing w. the vectorization settings to try and find which one would produce the least number of nodes that then had to be re-positioned to approximate the original pixels.
Designer's smoothing routine sometimes reduces the number of nodes, and retains a close enough match that the adjustments are feasible.
I 'spose that's why spending extra for a nice vector keeps the makers of same going.
-
If you hold down shift, the pixel line will be constrained to vertical and horizontal while drawing.
-
Use the pixel tool. Press "b". Welcome back to CG, c. 1980.
-
I went to you tube to watch thru the full vid. As I mentioned, I'm no longer familiar w. AI, and so I could only guess at most of what he was doing, considering there was no commentary. I did notice he had to fuss w. positioning a number of times.
At any rate, I tried moving circles around in a similar matter. Didn't really have much of a problem. 2 things that happened. Sometimes I had the item snapped, but evidently did not cleanly release the mouse button, because when I clicked back on the item just moved, it was slightly offset. Also, as the numbers of objects built up, I noticed that often I was snapping to a center or periphery of the wrong object. I needed to frequently click on the object I wanted to snap to so that it was at the top of the snapping list.
Here's a file I made doing this. Seems to me most things are positioned properly.
-
Hi Dan,
So how did you make the example?
-
On 10/2/2018 at 8:05 PM, TikiAles said:
Thanks for the repply, i am using that approach, but it isnt exact, even if i snap to object geometry and all of that, it sometimes gets missallign. I´m making circular grid logos, and the fact that there isnt a proper shape builder tool, i am converting to curves, and then adding nodes, then cutting the rest of the nodes.
I'm not understanding what you are doing. I never needed to use the shape builder tool in Ai, so am unfamiliar w. how it works.
However, the built in parametric shapes in AD are resized and positioned by their bounding box. They don't really have a fixed geometry till converted to curves w. nodes at specific places relative to the center of rotation. The snapping options won't work very well based on the bounding boxes.
Where are you adding extra nodes? And why do you need to cut the rest of the nodes? Forgive me if I'm seeming dense, but I can't infer what you are trying to do.
Can you post an image of something that is misaligned, or a file?
-
Hi, Wizaaerd,
Look at the fx options, such as 3d and bevel/emboss, etc. Those apply a bit map overlay in and/or around the vector perimeter.
-
Might be do-able, but I recall a developer saying true 3-D was not considered. At best, 2.5-D via isometric views. Various people have offered work arounds for simple 3-D seeming projections. To date, I haven't come up w. anything with which I've been really satisfied.
I did search around dreamstime a bit, and it confirmed my supposition. The image you showed was close to others that were from 3-D apps. And I did find a 3-D .obj file of celery stalks, which is a standard used by various apps, including the freeware Blender. The image seemed to me to be celeru, w. the root end clipped off.
-
Hi, simonlayfield,
I did download the file, and the problems are really numerous. Attached may be a solution.
Here is the problem as I understand it, hoping my explanation is clear enough. All of the "curves" objects are comprised of multiple joined curve objects. About 25 if I counted right. The way I came at it was to select all, and divide, which made all the curve objects inherit the bottom most object's attributes, which was a black fill. However, once divided, they are pretty easy to re-work. The holes in the letters, which had been in a separate "curves" can now be subtracted one by one from the surrounding letter shapes. Likewise the eye holes from the face, and the interior of the mouth. The only way I could think of to get the cut out along the top was to subtract the area above the letters, but within the outermost shape. Once there is a curve w. just an inside and outside, its perimeters can be broken w. the node tool, cut sections removed, then rejoined, and closed. All the remain shapes just need to have their fill attribute changed to white.
Total time was under 10 min., tho' I did spend more time coming at it from other directions w/o better results.
-
Hi, Artcraft,
While there are quite a few similarities between Illustrator and Designer, basically they work differently. From what very little i know of Illustrator, none of the pathfinder operations has a direct equivalent.
A Designer method that might work for you is to create the three shapes of the top most leaf/seed triad. Two all black on the sides, the center w. a thick white stroke. Select those, and do a power duplicate in an appropriate downward interval. Then, dupe one of the all black leaves, rotate it back to vertical, and align on the stack of cent leaves.
See below. Took about a minute, and most of that was trying to do approximations of the proportions of the parts by eye. The top triad is selected to show what was duplicated.
-
Hey, moksha,
Here's my entirely crude hand drawn approximation. Was messing around trying to figure out the original image geometry, and got frustrated. Figured I could do it faster and near good enough by hand. File saved w. history, but the history is incomplete. Hope it helps.
-
16 minutes ago, Y kay said:
Hi @gdenby , It was so nice of you put your time to create an example to help someone . Shall check the same and revert after trying the same.
But the problem is that the shapes are not closed vector boundaries in Autocad most of the times . And the layout plan example I shared was 1/15th part of the whole floor plan , and making boundaries for each item is difficult in ever evolving layout plan drawings .
Pattern fill shall be hugely useful..whenever it becomes available in Affinity-Photo .
There is an option in Gradient-fill for bitmap ....but that is giving me some un-intended results ....maybe mostly due to my lack of knowledge about Affinity Photo.
Not a problem. Didn't take much time at all.
I haven't worked w. Autocad output in at least 20 years, and wondered if the vector shapes were closed. Back then, they weren't, and so were useless to me.
I know the tiling fill issue has come up before from people who have been using Photoshop, and there were some work arounds, but I don't recall what they were. I spend 10X more time in Designer than Photo, and the gradient tool in it is very different than the fill tool in Designer.
-
About 10 minutes for this. Hadn't used the fx for 2 years, and had to thrash about a bit. Not quite right, but I've got to go and eats some breakfast.
-
If a .pdf file is created w. the vectors available, both Designer and Photo will do the fills. I don't use Photo much, but at this point, there isn't a pattern flood tool. Both programs doing pixel flood fills must have areas that are already at least at different grey levels, which is the same as any bitmap flood fill I have ever used.
Attached, a very quick Designer plan, using techniques you want. It was really trivial.
-
Hi, Y kay,
As Pšenda said, it is likely Designer is a better choice than Photo. Both do flood fills and and bit-map pattern fills. But both require the areas to be defined, and I know almost nothing about how AutoCAD makes those.
2 hours ago, Y kay said:b) Affinity photos makes some concession in the product development for Architects as our requirements vary fro mthe the phtographers or Ilustrators
I can not speak for Serif at all, but here is my understanding. Serif as a company is at least an order of magnitude smaller than Adobe. And the Affinity suite is very "young" compared to about everything else comparable. There is much to do for the people already using it.
Nevertheless, as already mentioned by Aammppaa, a small sample of what you would like to work on would let the users here what might be done.
I also suppose anyone who is proficient w. AutoCAD could pick up the necessary skills for the fills very quickly.
-
Hi, StephBradley,
The .mov goes by pretty quick, but it seems you have everything selected. Select only the black items and add those. The white items, if placed above the added layer, will continue to show. Those can also be added, and changed as one.
-
Hi, piginhat,
I don't use Photo enough to say for sure, but it seems to me that vector work is displayed w. anti-aliasing. In Designer, the default appearance is vector, and one must switch to a pixel view to see what the screen rendering looks like. Attached, the same vector shape made in Photo, one shot from Photo, the other in Designers default.
-
Yeah, Flame Painter is nice. Shows what one can do if the app has 1 focus only.
One can get some interesting "wavy" things from Designer.
Image attached. 1 zig zag pen line, duplicated. 1 w. rotation center to far left, power duplicate with a tiny bit of resizing & rotation. After a dozen or so, shift rotation center to opposite side. Repeat. Repeat Repeat. Group all.
Move 2nd dupe to close to the 1st, Repeat the above, except shift the rotation center side. Group.
Select all curves in both groups, and change nodes to smooth.
Duplicate a group. Grab vertical columns of nodes across all curves, and rotate, scale, reposition.
Not terribly slow.
-
In my experience, none of the vector shapes created in Designer are automatically rasterized when edited in Photo. Most of what can be done w. vectors in Designer can be done in Photo. There are some exceptions. You can't create or de-synchronize Designer symbols in Photo, but you can still do mass transform among the symbols. Designer has good abilities for adjusting bit-maps (photos, etc) but no ability to develop them from RAW files, or merge various edjusted layers into a single pixel layer.
-
5 hours ago, Pšenda said:
Halo tomgreen,
I apologize for the question - I'm ignorant in this specialization, but why is there a difference between what the laser and the printer see? Is not the problem that the laser machine incorrectly interprets source data, and can not do clipping?
I too know very little about laser cutters. It appears some models have advanced RIP software, others are more rudimentary. It appears the OP needs large numbers of open vector curves. Evidently, the machine he uses does not interpret clipped shapes, and AD's booleans close vectors. I spent a few sessions working on ways to clear away unwanted closes, and found a routine that worked most of the time, but was always fairly tedious. But I was only clearing out several hundred fragments, and the OP talks about thousands. On can use AD for the task, but it is not very efficient.
As a BTW, the attached set of open curves cut into a simple rectangle took about 5 minutes. More complex groups of curves cut around by a curved form took well over 15.
-
The Serif made tutes are on Vimeo. Check this one
A quick .mov file I made showing how the manipulation of a pen line is displayed real time in the transform panel.
Anecdote. I come from a different frame of reference. When I got my 1st personal computer, and it came w. 2500 pages of docs on how to use it, I was thrilled, and couldn't sleep for the next 3 days. And I didn't even have to make punch cards or teletype tapes. Whee!
-
Turn on snapping. Have "snap to spread" enabled. There will be a green vertical line showing when the object is at the V center, red for the H center, both for page center.
-
Hi, TBUA,
For doing traditional bit map painting, Photo is a little better than Designer. Both benefit greatly when used w. a tablet.
I infer from your 1st post you mostly work on paper. In Designer, the "pencil" is a specially designed tool to turn hand motion into a geometric vector representation. At this point, Designer does not have a vector eraser. The best on can do is delete individual nodes of the stroke, or break it into smaller pieces. There is an advantage to this. The strokes can be infinitely adjusted, repositioned, duplicated, made to have different thicknesses, etc. Designer has a very nice tool called the vector brush, which works much the same as the pencil, but stretches or repeats a bit-map along the stroke.
Here is an analogy. Working w. vectors is a bit like making a collage, or setting up a mosaic. Erasing really isn't an option.
Textured Brushes on Strokes- How do I get rid of the gap? (screenshot attached)
in Pre-V2 Archive of Affinity on Desktop Questions (macOS and Windows)
Posted
Hi Johny Johnson,
Since most stroke brushes are designed to have beginning and end tapers, they will have gaps of varying sizes. If you make a brush that has a square end, that can be used for closed shapes.