On Thu, 18 Jul 2002 14:48:27 +0100, Smith, Bradley wrote:
>
>I disagree. If it only took a little thought then almost all programs =
>would
>be able to do it.
>
>It in fact requires a lot of thought and the problem has already been =
>solved
>i.e. virtual memory. There's no point in solving the problem again. =
>Virtual
>memory of some sort is the only solution to not having enough RAM to =
>store
>the whole image and as we all know that makes things chug along slowly.
>Nothing you can do about it. If you compress the image in memory to fit =
>it
>in then you lose quality and surely quality is the whole point of =
>having an
>80 meg TIFF. If you were prepared to give up quality then you'd convert =
>it
>to some detail losing format like JPEG wouldn't you?
Ok. That was simply not my point. What I thought of is this:
You ONLY read from the disk, what you need to display.
That means, if you are showing the whole image at once you have to lose
a lot of information anyway,
unless you put up the image on Time Square.
If you show the image pixel by pixel, you can only show a small
rectangle from the image, so only read this.
And virtual memory is not the cure for poor program design. BTW virtual
memory is NOT equivalent
with swap file. Virtual memory is there to ease some programming tasks.
E.g. you don't have to program
against some particular memory layout, since virtual memory addresses
can be mapped around physical memory.
Granted you can prevent failure of your program under low memory
condition with swapping memory,
but that doesn't release you from your duty as a software designer to
THINK about the problem.
Anyway, reinventing the wheel is nothing bad if the wheel you're
supposed to be using is square.
Jochen
-- Read the List FAQ/Etiquette: http://www.newtontalk.net/faq.html Read the Newton FAQ: http://www.guns-media.com/mirrors/newton/faq/ This is the NewtonTalk mailing list - http://www.newtontalk.net
This archive was generated by hypermail 2.1.2 : Thu Aug 01 2002 - 06:02:51 EDT