Resolution Independent Mac OS X Leopard?

[Catfish_Man]

It will be implemented anyway.

If by "it" you mean a vector based UI, then no, it won't. Vector graphics have some interesting limitations.

 

[dejo]

Vector graphics have some interesting limitations.

Care to elaborate?

 

[savar]

Savar you should read that Ars link I posted above. There's no good reason why the entire desktop should be vectors and there are many reasons why it shouldn't be. High enough resolution bitmaps can be used to elements that it doesn't matter that they become blocky. Its not likely for example that someone would ever need a 1 foot wide "close" button on their window title, so making a 100 pixel high version and scaling according to the DPI of the monitor is good enough.

I did read that link and I think you (and Siracusa) are totally correct...I mean I don't see any desktop system being 100% vector for years, but there's no reason why you can't render vectors once and cache the rasterized images. Essentially thats what you're talking about, except you're saying that the graphic designer will "cache" the vector images, whereas I was implying the OS would do it -- I was intentionally oversimplifying because a lot of people are confused about this issue. In reality it will be a mix of the two strategies that shifts towards increased vectorization as time passes.

 

[dr_lha]

Care to elaborate?

For one, we have gotten used to nice photo-realistic icons in Mac OS X, e.g. a Hard Drive actually looks like a hard drive. Reducing this down into Vectors is basically impossible, because the definition of it in vectors would take up more memory and CPU time to render than simply having a large bitmap that you scale up and down.

 

[macgeek2005]

What exactly does resolution independent mean?

 

[mkrishnan]

What exactly does resolution independent mean?

Hehehe, without doing a here-we-go-again here...

Resolution Independent means that software specifies physically (in the real world -- in inches or cm or what have you) how big an element should be, rather than specifying the element's size in pixels. The element is then adjusted or rendered based on whatever is stored to be as sharp as possible given the number of pixels the display you are using has in the given physical space.

 

[jsw]

FWIW, it also means objects can be defined with respect to proportional size on a display (i.e., 1/10 the width, 1/8 the height) without regard to how many pixels make up the display. Many resolution-independent mechanisms use this definition, so all screens look "alike" regardless of the size or pixel density of the given display upon which they are rendered.

 

[mkrishnan]

I'm sorry, thank you for the catch. That's a really good point.

 

[Catfish_Man]

Care to elaborate?

Consider the problem of representing TV static as vectors. I believe you'll find that it essentially turns into a really inefficient bitmap format (because each 'object' described by the vectors is the size of a pixel). Obviously that's an extreme case, but it shows up in less severe forms in many real world things.