gdenby

Let me guess. The surface of the painting is itself shiny, or under glass. Hence, the glare. You moved the light source off to the side to get around that, but now you have glints wherever the paint is raised, and probably shadows of the stroke texture too. You might be better off re-shooting. Make the light source really diffuse. Bounce several sources at equal angles off of semi-reflective surfaces, or shine them thru thin fabric. I'm not a photographer, but worked w. guys who were doing archival shots of oil paintings, etc, and the process was jumping thru hoops, repeatedly trying to find the best angle and distance and timing. If you can get the number of glints down, my suggestion is to try selecting them, feather the selections and use gaussian blur to blend them into the surround. Doing a clone replacement in most situations would be extremely tedious. For clone tool source selection, its alt/option click to select clone source in Mac APhoto.

You have my sympathy. I used Corel Draw extensively some years ago. I tried to ensure my work was portable, and saved it of as .pdf as well as in native format, and .ai when that was available. Unfortunately, this was before the PDF format was defined as it was today. The result, an archive of 4 - 5 years of work completely inaccessible on my current system. Moral of the story: no matter how versatile a proprietary format is, always try to duplicate your work in the most widely used standard formats. Serif is good in that one can save to .svg, which tho' limited in some respects, is a published open standard that anyone may implement w/o charge.

I spent some more time w. doing it all by graphic manipulation, assuming the proportional resizing. I thought I had it last night, but it didn't look quite right. Upon looking closer this morning, it was just that my eyes were just not in focus, and the work was sloppy. Just draw the outer and inner rectangle. Group them. Create a diagonal into the center. W. snapping to object geometry on, select group, copy paste, resize constrained around the center till the copy outer edge meets the 1st inner edge. Repeat, somewhat tedious, but it forms a perspective recession. So either constant or proportional resizing is pretty easy. Now, how to do it with non-quadrilateral shapes...

Yeah, seems I misunderstood the question. I see the request for "the same size padding." I was taking it as "the same (proportional) size padding" seeing that the width of the aqua shape was not an even offset.

Here's my offering. If I understand what FlourescentTurban wants, the procedure goes like this. There is an arbitrary sized rectangle. It is copied and scaled down around the center of the parent rectangle. There is then an apparent boundry "frame." The copy and scale is to be repated with the constraint that the bounrdry has the same proportions. When the boundry is created, if the horizontal and verticle lines of the smaller child rectangle are extended, there are 4 corner rectangles. As the next smaller rectangle is made, those corner rectangles will grow. So the corners of the main rectangles decrease along diagonals leading to the center scaling. The corner rectangles also increase along that diagonal, but their scaling centers are the outer corners of the parent rectangle. The size of the outer rectangle increase as they scale up is limited by diagonals drawn from their corners to the parent scaling center. Where those diagonals intersect the first child rectangle's horizontal and verticle edges is the bound of the next corner expansion. Draw a rectangle within those points. Make the next set of corners extend to where the corner diagonals intersect the 2nd rectangle periphery. Continue as desired. Hopefully, the illustration will make things more evident. The process I used needs a bit of refinement to be accurate, but I did a test by inserting the series of corner expansions on inside another, and visually at least, they seem to decrease in the same proportion as the main rectangles are.

Hi, FlourescentTurban Let me preface this by saying I've been up about 17 hours, my eyes are falling out of my head, and I'm not quite sure I even understand exactly what you are trying to do. Here's an image that is my guess about what you want. It can all be done via copy, group and scale with snapping turned on. No numeric entry. If its sort of what you want, I'll post in the morning w. something that is hopefully coherent. Nitey night.

Hi, ralphonz, In this particular case there's something that might suit your need. Select the star tool, and choose a largish number of points, maybe 36. Draw the star, and then slide the inner radius control outward till it shows the top vertex lining up w. the ones 90 degrees to the side. Convert to curves, select all the nodes as in the example .mov, and change to smooth nodes.

... awaiting with bated breathe. No, with panting.

Humor follows. So the other day I took my old drafting table out of the basement, put it together, trued everything up. Still had some good paper, and one of my Rapidograph nibs wasn't corroded. One of the india ink bottles was still wet. So I proceeded to draw. I placed my triangular metal ruler on the paper with the mm scale at the top edge, and placed the pen end down. Then I moved to approximately 3mm to the right, and put the pen end down again. !!There was no line connecting the entry points. I have to ask, whoever made this "reality," why is this is so-o-o lame!?. Do I actually have to drag the pen tip across the paper, and expect I must manually stop where the line needs to end?

I don't think snapping can be made compulsory. The feature just facilitates positioning. For certain kinds of grids, such as the basic square, holding down the shift key will constrain the pen lines to horizontal, diagonal and vertical. That will double your chances of placing nodes only at the vertices.

This is my speculation, nothing more. The work space zero position is the upper left hand corner. All the layers' bounding boxes have their zero position defined as an offset to that. This gives a default location for the objects within the work space, and the base transform is position relative to the zero point. The scaling transform is secondary, and for Affinity at this point, a user defined option from moment to moment. Considering the current beta ability to scale and rotate around any arbitrarily placed center point, the transform panel box diagram is kind of irrelevant.

From what little I know of the .svg format, it would be very difficult, it not impossible to do .svg exports, because an .svg shape with a fill is surrounded by a series of points that start at one node, and end by either connecting back to the first one, of drawing the fill from the 1st to last. With branching networks, one node leading to many others, and those to others, I don't know how the list could be made a continuous series. So while it is no big deal to draw a bunch of connected lines, I'm thinking writing a routine to turn the network into a group of defined areas might be difficult. I have a recollection of a piece of software I used decades ago that did allow such an operation, but it required the user to follow certain steps. My very vague recollection is that the added shapes had to be made only by clockwise node addition. Eventually, there was a utility routine added to fix user error, but it didn't work 100%.

From what I can tell, both Designer and Graphic turn pressure shaped paths to outlines when converting to .svg. I fear what you want to do is not possible. Looks like most of the other export options just rasterize the vector drawing.

Note, the image directorfilms posted is not proper perspective. Looks like what a Quantel Paintbox might have made using just straight line tools. The horizontal lines don't seem to get closer together approaching the horizon. My 1st problem was that I'd mostly forgotten the rules for drawing perspective that I'd learned when I was 16. After thrashing around for awhile, I decided to fake it. Made a right leaning trapezoid that seemed close to desirable. Set the rotation/scaling center at the vertical-horizontal lower corner. Power duplicated vertically. Then manually scaled the copies to reduce in both axes to make a receding series. kinda PITA. Grouped, copied, pasted, moved to the side, and then sheered. Again, again again. Grouped, flipped, moved to mirror position. Did a rotate and scale to make the side walls. Then realized that the manual sizing was not quite accurate, and a combine boolean didn't work well. So I had to open up all nearly 700 trapezoids to apply the colored, blurred style, instead of the B&W lines I had. Note to self, add the style and fx to the beginning element. Made some back ground shapes for color fills and gradients. Still trying to find a tute that shows the easy-peesy perspective grid method I learned w. just a pencil and ruler.

Not so quick, maybe 45 min, but doable.

Maybe this will help: Motion manual page

Even w. your shot with the layer panel, I'm mostly clueless about most of what you did. Looks like there were maybe 40 layers w. masks, etc. But, am I right, the last image was the foundation for all the enhancements?

Tried last night, and yes, the line pressure can be saved as a style, so I'll be putting together a new category of those.

I made a thick stroke. and copied it so I could see how the different pressure graphs were working. After making and saving a bunch of profiles, I deleted the line layers and saved the file. The pressure curves are associated w. the file, I suppose I should check and see if they can be saved as a style. I know dashed lines can. When I re-open the blank file, all the pressures are available in the stroke dialogue, ready for application to the newly drawn vectors. They can be tweaked w/o saving for 1 off variations that might be hard to tell apart as a tiny thumbnail.

I'm gonna guess this is really simple. There is no layer selected, so the brush can't work, because nothing has been selected for work. Logical, because there may be many levels. Gotta pick one. Then, the brush is likely to work.

I'd suggest more physical memory. I'm a casual user, and I've hit my 8 Gb limit once, and so obtained more.

A nice demonstration of a very handy feature. I have a file I keep on hand called "line pressures" that has 8 different pressure curves saved so I can pick and choose as I like, or tweak as I go.

When I d-loaded the file, it was in excess of 30 Mb. I deleted a duplicate pixel layer, and got it down to 15+ Mb. But never mind that. A technical issue. You need to study perspective. Let me suggest this page. (BTW, the resources Mr. McEvoy provide are among the best ever). The problem w. your picture is that the implied horizon line is way up off the top of the page. Simple answer, take the 2 layer groups and squish them down by at least 60%, and stretch them horizontally by about 50%. It'll look like a piece of lawn viewed from a front door.

Using the file you provided, I selected the bear family group, and set the fill to none. Then I selected the base layer, and went to the menu "Document/(select)Transparent background." Export as .png.

See this topic