bitmap copy of clip very blurry

This weekend while trying to make an animated gif, I used the mask and clip features to use only a portion of an image in my project. I copied the clip and put it into the file I was working on.

Later I wanted to make the file a bit smaller, so I tried the create a bitmap copy of the clipped item. The bitmap is very blurry, unusable.
Is there anywhere I can check to see settings that might cause this? Using version 5.

(I posted a version of the xar file as part of another post I made today http://www.talkgraphics.com/showthread.php?p=319820)

Ramp up the DPI for the bitmap copy, 300dpi ought to do it ;)

Where would I make that change?

In the Bitmap Copy dialog, go to the Bitmap Size tab and change (only) the DPI to 300

Attachment 62907

Ah, Yes. How silly I feel now. Several times I looked right at it and did not even see the dropdown option. :p
Worked perfectly. Much appreciated.

You can also input via the keyboard if you want a higher dpi.

LOL I never even thought about trying to change the dpi with typing. My mind was on the other issue (This weekend, I was banging my head about a non-transparent gif that would only export solidly if I checked the option for transparency :confused:).

I can't grasp the value of using a 300 ppi resolution here as it's for web use where it will be rendered at 96 ppi on the vast majority of computers.

In the attached png, one of the images is at 300 ppi and the other at 96 ppi. Can you tell the difference?

The difference only becomes apparent within Xtreme when zoomed in, as per the image supplied above by Samantha, but at 100% the two should render almost identically, or am I missing something?

Egg, you're correct that for web graphics, you don't want any more than 96dpi. However, if you create the bitmap copy at a high res (at least 300dpi), then export it at 96, you can get a much sharper image with better detail.

In this second generation export, it's a little harder to see the differences between the exports from the 300 ppi and 600 ppi bitmap copies. They were very noticeable in the final output.

BTW, the image with the $50 bills was ultimately used at 300dpi with a big fade from right to left so it could go under text in a brochure.

For our webpage, the pink in the $50s didn't work and the shadows on the $100s were unacceptable. I was able to hide the shadows very satisfactorily using only X5. (Very proud of myself.:D) And I used the same bitmap copy trick on the bills to get good detail in the final output.

Quote:

---

if you create the bitmap copy at a high res (at least 300dpi), then export it at 96, you can get a much sharper image with better detail.

---

This is only really true when making a bitmap from an image perspectivised by the mould tool. It's an issue here because the mould tool renders images without proper anti-aliasing (even in VHQ mode), so by making a higher-res bitmap from it and then scaling it down you are effectively anti-aliasing it manually.

There are some other cases where it can pay to do this kind of ‘super-sampling’ manually, but it's pretty rare. The most common for me is when you've got two paths on top of each other with exactly the same shape; this can make the edges look wrong when rendered using Xtreme's implementation of anti-aliasing, but the effect can be vastly reduced using super-sampling.

When you're super-sampling for web images, it's generally best to use a multiple of 96dpi such as 192 or 384 rather than 300. This lines up the pixel boundaries of the output image to the source image, which results in a slight improvement in sharpness.

Allison/Bob, thanks for the input. That's what I thought, I just couldn't follow Steves reasoning for creating a 300 ppi image. The only reason Samanthas image looked blurry was that she was comparing a 200ppi image alongside a 96ppi image whilst zoomed in.

Quote:

---

Originally Posted by BobInce

This is only really true when making a bitmap from an image perspectivised by the mould tool.

---

It's true that I'm most likely to use it on a mould or something I've squished or skewed in some way. I don't do it unless it looks like I need to. But then, I use a lot of photos as objects, so I may run into it more often than most people.

Quote:

---

TWhen you're super-sampling for web images, it's generally best to use a multiple of 96dpi such as 192 or 384 rather than 300. This lines up the pixel boundaries of the output image to the source image, which results in a slight improvement in sharpness.

---

That's good to know. Thanks!