Jump to content

gdenby

Members
  • Content count

    1,739
  • Joined

  • Last visited

4 Followers

About gdenby

  • Rank
    Dedicated User
  • Birthday 08/17/1950

Profile Information

  • Gender
    Male
  • Location
    "Michiana," USA
  • Interests
    CG, obviously. Traditional visual 2d and 3d art. History. Music, piano emphasis. Nature conservancy, gardening, cooking.

    Too old for martial arts or treking.

Recent Profile Visitors

1,557 profile views
  1. Thanks heaps for your interest in my Castrol logo GT question gdenby. Very much appreciated. :)

  2. I found a pic of an old sign, and ran it thru a tracer. Ended up w. a file above 50K. Spent a good bit of time deleting nodes, and adjusting curves. Eventually managed to get an .svg that was 14.6 Kb. But it had almost none of the lumpy character that the trace had from the original image. To get the file down from 16 Kb to under 15 was a matter of deleting every excess node I could find. "Oh, look, there are 2 nodes almost on top of one another. And there's three. delete, delete." And I noticed that the export for the .svg had to be for web use. Other types made bigger files. My guess is that even a manual trace of an irregular form is going to make something too big. I've looked at various versions of what I worked on, and it really comes down to the number of nodes defined. Smooth curves don't need near as many points to record.
  3. Hi, iamwoger, Did you use an autotrace utility. The "Casterol" has a really irregular edge, which can happen w. tracers. They can make about a bazillion nodes. If the tracer doesn't have a node reduction or optimization routine, there may be hundreds of nodes that must be cleaned away manually. Or, is it a bitmap image enclosed in a vector shape, w. the .svg format allows. If so, there may be data for every pixel, which would bulk the file up. AD exports .svg 1.1
  4. Hi, Digas, While "My Craft Studio Elite" has lots of convenience built in, Designer has far more versatility. Attached, a quickie that while it did take a minute or so, shows how much variation can be made using the power duplicate function and layering blend modes.
  5. I've been fussing around in APhoto, working on emulating hatched drawing routines. I've had some luck getting a more hand made look for the bitmap. Here's a sample of a sort of wood cut look that is the results of 3 .svgs from Image Vectorizer, a product from the potrace creator's commercial venture. Put them together in AD. Still a little too busy, and I didn't bother filtering out enough of the noise. But I'm getting some decent results.
  6. Considering that potrace has been in development for 15+ years, and is available free in Inkscape, or on the web at Vectorization.org (and in other forms), seems to me like there are other features that should be built into Affinity well before a bit map tracer,
  7. Hi, Kev74, The learning curve is there, but it is not very steep. The UI has a ton of stuff in it, and it will take awhile to dig thru it. Also, the Affinity software is fairly "young." There is a lot of stuff that is available in older apps that isn't in Affinity yet. And features that are Adobe proprietary you will not find. Nevertheless, what is available works well. Its quick and reliable. In my experience, very few program failures compared to others. Some rough corners for some operations, but mostly just irritations.
  8. gdenby

    Pattern fill

    Hi, Angelos58, It appears the problem above is caused by the diagonal lines being slightly offset from the square they fill. See attached:
  9. The original file could contain bitmap images. I just d0loaded the youtube logo as an .eps, and it included a badly pixelated image for the shiny screen glint. It needed to be blurred. I suspect that the pixelization problem may be because the .png file is not large enough to start, or is being exported at a smaller size. As I mentioned above, all bitmaps will have shapes w. pixelated edges.
  10. What format are you exporting? All pixel formats will, by definition, be somewhat pixellated. Of the various formats, .gif is usually the worst (its the oldest, back when everything was pixelated), .jpg not compressed or only slightly will be good, as will .png. Only .svg can be made close to no pixels/dots, because it is not a pixel image till rendered at whatever the highest resolution of the rendering device is. How are you making the art? Are you adding images from other sources into ones made w. Affinity apps? If those are pixellated, you might be able to process them into something useful. As R C-R mentions resampling, which is often helpful. Again, give an example.
  11. Some odd shapes snapped to the periphery of another odd shape. 1.7.0.11
  12. Ah, just checked back after a long day entertaining family. Yes, that does it. Your method selects the parts that are needed, much easier than selecting all the un-needed ones.
  13. I tried to make it easy, but this is really a difficult file. As I mentioned, I had a lot of trouble figuring out what white parts were supposed to be part of the back ground, and which were tube high lites. Check out the attached. The top layer, colored orange, needs to be duplicated. The glints, selected as nodes, need to be wipes out. Then that layer gets subtracted from its self, leaving the original whit layer part.
  14. Hi, LeeScoresby, (Edit: I thought I posted this a couple of hours ago, but here it is anyway. What happens, I 'spose, when working on a morning coffee and vector work.) I gave it a try. Not a fun set of objects to work w. In the end, I made a couple of mistakes, because I was having troubles telling which white shapes were glints on the surface, and which were white areas that needed to de transparent. Attached is my attempt, w. flaws. Here's what I did to get the results. I duplicated the top layer till I had a copy for itself and each layer below. I then began doing a boolean subtraction w. a copy on each of the lower layers. Because of the way the .svg stack was made, that subtraction cut away any part of the under layer that was no the layers color. This is easy enough to do. The hard part was at the end. I had to figure out how to cut away the open areas within the tuba body outline, but leave the white glints. In other words, I had to cut a portion of the topmost image away from its duplicate. I see from the result that I messed up somewhere, and subtracted a few glints, and not the background, so there are a couple of transparent areas where there should white. I put in a patch white rectangle in the layer stack to fill that. The .svg structure was unusual ti me. More often, I find the layers stacked light to dark on top, which means the background is transparent. tuba_silber_vectorized_grau.afdesign
×