At 6:08 pm +0100 30/3/99, Ian Collier wrote: >On Thu, Mar 04, 1999 at 12:17:46AM +0000, Andrew Collier wrote:
Have you been asleep for a month? >> >The images seem quite >> >dark, btw. This may be a function of display gamma, since I've speculated >> >before that your display is gamma corrected while most other kinds of >> >display (except SGIs) are not. > >> Yes it is - and will you stop trying to tell me that this is a fault at my >> end!! > >I am not saying that it is a fault. I am saying that it is a difference. >However, if I may make an analogy, if you were to put up a program in >MacBinary format you should be aware that it isn't much use since most >people around here seem to use Windows. I'm afraid it's a fact of life >that you are the odd one out. No, I'm sorry but that's just a silly analogy. The situation is nothing like that. With file formats, I can choose either a proprietry format (eg MacBinary) which only a privelidged few can read, or a standard file format (eg gzip) that everyone is going to be able to use. I have the choice of providing something which will be correct for everybody, wheras with the demobase images I have to publish something which looks right on one monitor, even if it might be entirely different on someone elses *and I can have no control over that variation*. By the nature of publishing images on the web, I have to choose *something* to be "right". If I adjust the pictures to look correct on your monitor, then they'll look far too bright on somebody else's. An uncalibrated monitor can display just about anything. Yours is too dark. Someone else's might be too light. If *something* has to be defined as being correct, don't you think it is reasonable that that particular something should be a gamma corrected monitor? I'm not going to change my pictures for your dark monitor, unless you can prove to me that there aren't an equal number of monitors on which the picture already looks too bright. (See Bob, I don't only argue with _you_) >> It is totally predictable, and I should imagine it is not recoverable. It >> throws away every alternate frame, and averages the pixels in each pair of >> adjacent horizontal lines. It then scales whatever picture from the signal >> is left, to the size of the window which, by default, is 320x240. It can go >> up to 640x480. > >Tricky. But if you can get it so that one line equals one pixel and if it is >mathematically averaging pairs of lines together then (assuming you can guess >one line - say, a line of constant colour in the border) the process ought to >be reversible. I'll wait for you to demonstrate that... but I rather think that TV input card increases infomational entropy. How, for example, are you supposed to tell the difference between a set of horizontal single-pixel black and white lines, compared to a filled grey rectangle? Andrew -- | Andrew Collier | email [EMAIL PROTECTED] | Talk sense to a | Part 2 NatSci | http://carou.sel.cam.ac.uk/ | fool and he +----------------+-----------------------------+ calls you foolish | Selwyn College Student Computer Support Team | -- Euripides
