az2000

Have you looked at fonts that are outline style? Look at some of the font sites. (I've used dafont.com, fontpark.net, acidfonts.com, larabiefonts.com, simplythebest.net, fontspace.com).

It's a good idea to get in the habit of putting your image on at least one layer above the background layer, and saving your original work in PDN's native file format. Use "save as" to save to other formats like .png. That way, if you need to go back and make changes, you won't have the situation you just found yourself in.

It sounds like you want to put the camo on a layer behind the image (person, car), select the image edge with a lasso, fill the image with a transparent color, and let the camo beneath it show through? Then, lasso the edge, copy and paste the selection into the background picture? (If that sounds right, you'd also want to add some feather of the selection so the edges aren't too jagged.).

If it's saved as JPEG, maybe try indexed GIF or PNG. JPEG always looks fuzzy to me when it's something with straight lines and solid colors. Like a screen shot of a desktop, or a scan of a diploma, or transcripts, vehicle registration.

BTW: Regarding the scale which applies to all three images(?), you could include a ruler in each photo. Then resize two photos to so the ruler is the same size as the other photo. Copy the the ruler from one photo and paste it in a 4th layer, positioning the ruler at the bottom of the collage. You can cut out the ruler of all the photos as part of the process of removing their backgrounds. That's just one idea that came to mind, without seeing an example from the antique site. After you do a few of these, it's not as complicated as it sounds.

Just to help untangle this for future readers: Your post appears to renew your original post which was locked by Rick due to a vague title. The second half of this new post contains my reply to your original post. It's not clear if you're asking for additional help. Or, need clarification about anything I suggested. Please pose a new question if you have one.

Look at the "layers" feature. You can create 3 layers in addition to the first layer which should be considered the background layer. Put one image in each layer. You can move each image independently by changing which layer you're working with in the layer's window. (You can hide or show layers with that window too. Useful when laying things out.). You should post an example of the scale. But, that would just go in another layer, and positioned independently just like the 3 images. That was discussed recently here:

Gradient text: Two-mountain fade with transparency Camaro fade with sepia http://forums.getpaint.net/index.php?showtopic=12184

There are a few tutorials for exploding planets, some beginning with images of things like cracked mud. Exploding planets (from cracked mud) - Exploding planets - Exploding earth - Explosion An image of a brick wall in this tutorial: Maybe you could figure something out. (Or, maybe someone else has ideas.).

I'm curious about one thing. The OP said: ==== I saved the .JPG image as a .BMP file then opened it in regular Paint that comes with Windows. From there I saved it as a .JPG and it now works fine with FF & IE. ==== That sounds like a Paint.Net saves .JPG differently than Windows Paint?

I'm not entirely familiar with PDN, but in my old version of Photoshop I would do something like individually select the 3 individual rectangles, apply a "feather" to the selection (which I'm a little confused about how to do in PDN because the feather feature is a plug in), than adjust the brightness/contrast of the selection. The feather'ed selection would prevent the brightness/contrast change from being abrupt at the edge. It would help it blend. (Adjusting all three boxes, bringing them all to a level of brightness/contrast that's halfway between each other would also help the blending illusion, instead of just adjusting the middle, lighter box). (This is one thing I commonly do in my old PS 6 that I haven't translated to how to do in PDN. I'm curious to find out if experienced PDN folks would agree with this approach, and how to feather a selection.).

Post an example. I might not know how. But, someone will.

Is this tutorial (blending or fading two mountain scenes into each other) close to what you'd like to do?

How does the feature you describe differ from Windows 7's native "explorer" feature? When I File->Open in PDN, I can adjust the size of the displayed thumbnail (or, choose "details" displays). It remembers this setting from one session to another. And, it has a tree view on the left. I'm using Win 7's "Classic" theme. Not the "aeroglass" them. I don't know if that may effect what I'm seeing. I've forgotten if I changed any other settings. But, I'm definitely seeing what it sounds like you describe.

You have to set the image's DPI to equal your printer's DPI. It's possible your printer's DPI isn't exact (to scientific tolerances). If you set the DPI to the printer's DPI, but don't get the exact size you expect, you may have to experiment with the image's DPI to get the printed result (size) you expect. If you're measuring it on the screen, I don't believe you can force an image to be an exact measurement. I believe different displays will display an image differently. For example, I had two laptops with the same LCD screen size. One was a low-resolution (WXGA at 1200 x 800) that would display a 1"x2" image pretty large. (I never actually measured it with a ruler.). The other was extremely high resolution (WUXGA at 1920 x 1200) and that image was much smaller (if I were to measure it with a ruler.). You'll always have variations like this because even the larger, standalone monitors will vary slightly in size while using the same resolution. (A 19", 20" and 22" monitor might all use 1920 x 1200).

I can't answer the first question. But, your image looks pretty good. If you read the following tutorial, it shows how you could make that ring around the shield more continuous and 3D: - And then, instead of the gassy halo around that semi-circle band, you could put some highlights on the shield and the ring, like they're glossy and reflecting light. These tutorials show how to do that: - - - -

You happen to be in luck. I just went through the tutorials and have a fresh idea of what matches the image you posted. These two match because they bring together different images and effects: - http://forums.getpai...rmers-tutorial/ - http://forums.getpai...ial-render-sig/ Maybe this one too (just for the elaborateness): - http://forums.getpai...make-my-dj-sig/ And this one for the glowy line around the object in your image: - http://forums.getpai...m-my-joker-sig/ The best thing you can do is spend a day or two reading all the newbie, creation and distortion tutorials. There's no single way to create a banner. It's a combination of overlay techniques and special effects. You have to develop an eye for the different components you might see in a banner. (Like yours, even though I don't have that developed eye, I see the female zoomed out and her head spun a little to create the illusion of movement. You might find that technique in another tutorial.).

2600 x 2200 is the pixel diemisions (height x width). That's the recommended minimum size for a 20"x24" print. But, they don't say what the DPI should be. Usually for printing, the the DPI is 300 to 1200. It depends on the printer. Changing DPI will change the number of pixels necessary to create a size of print. For example, at 110 DPI, Walmart's suggested minimum pixel dimension produces 20"x24". But, that's an oddball DPI. At 300 DPI, it requires a 7200 x 6000 pixel dimension to produce that same 20"x24" Unfortunately, Walmart isn't giving you enough information. An example of how you get what you pay for at Walmart. You might try one of the other internet-based canvas print businesses. (Google for "photo canvas prints."). Or, if Walmart allows returns, just try one and see what it looks like. The problem with not knowing the DPI is that, if you use a smaller DPI, it won't look as good as it could. The information Walmart is proving doesn't make sense. The pixel width and height should be a constant given the DPI. The way they say it should be a minimum height x width makes it sound strange to me. Maybe others have some thoughts on this. There is a tutorial about DPI which might help. You can play with DPI, pixel width x height and printed dimensions (in inches) using Image -> Resize. See how changing one affects the others. To me, DPI is the key. You know the printed size (inches). The pixel dimensions (2600 x 2200 "minimum") are meaningless (IMO). They're a result of DPI and printed size. You could try uploading images with different DPI sizes (72, 96, 120, 150, 300, 600) to Walmart. See what it accepts. Then experiment with different pixel dimensions. But, larger pixels dimensions doesn't equate to higher quality. If Walmart shrinks the image to fit the printed size, it could get unsharp.

I found this tutorial the other day to do those pastel cartoons like we used to see on the Charles Schwab commercials. (Scroll down that page to see finished examples). And, this tutorial shows how to create an oil painting appearance, including the canvas texture (if you wanted to save money). Walmart's page says the size should be a minimum 2600x2200. Nothing about DPI. I didn't understand what you recounted from the Walmart page's error. Maybe you could reproduce the error message exactly?

I noticed that tutorial recommends using brush/line tools to create a border for the Magic Wand tool to latch onto. That must be tedious. Do most people do it that way? Or, do they use the Lasso (select) and Magic Wand technique described here and here? (search for the word "lasso" on those pages.). In my old Photoshop v6 I use a Magnetic Lasso. Is there a plugin to do that with PDN?

To open a file, click "File" in the the menu bar at the top of Paint.Net, and "Open." (Sorry if I misunderstood your question.). If you go to youtube.com and search for "paint.net" you'll find a lot of video tutorials. Watching over someone's virtual shoulder might give you the orientation you're looking for. looks like a good basic video.EDIT: In case it's not obvious when watching that video, there are three parts. The one linked above is Part 1. You should see , and . There may be better videos. This series stood out to me when I searched.

If someone absolutely needs to automate Paint.Net processing, it's possible to use AutoIT v3 (google for it). I've done some insane things with it. It lets you automate mouse/keyboard interaction with a Windows program (file open, apply different effects, save, close... file open... etc.). There are command line tools like ImageMagic and GD. Also IrfanView has some batch capability and plug ins that might do something that could replace a desired function within Paint.Net. But, really, after you learn how to use AutoIT... you can do automate anything in Paint.Net. And, the nice thing is, once you make that investment to learn it, you can use it with other Windows GUI applications. You'd be surprised how often that comes in handy. I'm sure if someone was really into it, they could write a front-end application to simplify various automated tasks (like batch "sharpen"). Then it could be a distributable tool.

I found the answer in the tutorial Beginner Image Merging which demonstrates the gradient tool's transparency mode. Now I see how to create a variably transparent sky like the example .PNG linked in my first post above. I'm still not understanding how to manipulate transparency around the image of objects. Such as a smooth, transparent transition around a font's edge.

I found SH's tutorial easier to understand as a total newbie to Paint.Net. The Camaro example is a bit more advanced. But, just browsing it I didn't get what's happening. (For example, it refers the student to download the faded/manipulated image and use it as the starting point?). However, I wouldn't have found the Camaro example if you hadn't posted a reference to it. It would be good if related tutorials linked to each other so a reader could cross reference them. IMO, tutorials would evolve better in a wiki environment where users could collaborate on development, cross-referencing, and aggregation under hierarchies that would evolve naturally (possibly derived from keyword and feature terminology). In that case, 20 tutorials on a related subject wouldn't be bad since they would complement each other and be easily discovered through each other. Similarly, the ability to contribute to an existing tutorial might reduce the desire to create a variation of it. Just my thoughts. Forums are great for fostering discussion. But, not so good for knowledge warehousing. Likewise, wikis aren't good for discussions. (IMO).

Background ======== I have a lot of experience with Photoshop (going back to the days of Aldus Photostyler). But, I stopped upgrading a 3-4 versions ago, just when they added PNG support. I haven't gained much PNG experience because that initial support was limited. I've searched the forum and tutorials for my question. (Found a lot of useful info that I wasn't looking for, and made notes for future reference.). Question ======= I know PNG-32 can have a range of transparency for each pixel. What I don't understand is how to manipulate that "alpha" value. For example, how to create the kind of variable PNGs seen here (variable transparency sky). I know how to create a layer with 50% opacity, save the image as PNG-32, and when I display it on a web page, it lets the web page's background color show through. The same level of transparency across the entire image. But, what I don't understand is how to get a variable transparency like the sky example linked above. What tool do you use to create more transparency at the top, and less at the bottom? Or, feather the edge of a font to be less transparent near the font, and more transparent a few pixels away (so it's not jagged)? I guess what I'm getting at is: With my older versions of Photoshop I know how to blend a transparent GIF into a color that matches the web page's background color. What I'm not understanding is how to do the same blending with the "a" channel of PNG's RGBA. I'm lost on how to translate my old GIF techniques using variable color to variable transparency. If there is a tutorial that shows how to manipulate the transparency of various parts of an image, please refer me. Thanks