When creating a thumbnail of an image (for preview on a website), I do not get the same result when saving direcly from and AI document and saving from Photoshop. Here is how it goes:
1. In AI, I set the crop marks and go to 'Save for web and devices'. I then change the pixel dimensions from 250x250 to 150x150...ah ha, no matter what I do the image is now fuzzy (whether I save to a jpeg, gif, png, etc.).
2. In Photoshop, I open the file from the pdf I created in AI. I then 'Save for Web and Devices', change the pixel dimensions from 250x250 to 150x150...and voila it is much shaper.
So I guess it has something to do with downsizing a PDF and downsizing the raw image in AI. But why? I know that AI and Photoshop have two different ways of compressing images, but is there a way to adjust the illustrator settings (besides the standards such as dithering) so that you can create a thumbnail directly from illustrator as opposed to having to go through photoshop?
I haven't used Illustrator much for creating web images. Photoshop is the tool to do that sort of thing, it has more interpolation methods; Nearest Neighbor, Bicubic, etc.
If this is going to be a typical stage in your workflow, consider creating an Action or Droplet in Photoshop to perform the task.
You could create a droplet that you just: drag the AI file onto the Droplet, and it will automatically open and rasterize the file, resize it, save it to wherever you want, than close the file.
With Actions, you could do a batch of AI files, in one shot.