Hi, I have the Sharp EL-9600 graphics calculator and one thing it allows you to do is take screen shots of whats currently on your screen. One thing I have noticed about the screenschot utility however, is that it is highly efficient.
I did a bit of experimenting and here are some results:
Estimated speed by running some simple basic programs through both calculator and Qbasic on P266 laptop gave the calculator a rating of "5 Kilohertz+ (Plus indicates PR rating )"
eg. finding 69! hundreds of times to test the math unit, working moemory storage and retreval etc. multiplying numbers by pi to test fpu.
Screen size is 62x126 pixels black and white
Theoretical image size for uncompressed image = 62x126x2/8 = 1953 bytes
Here are some experimental results
Blank screen - 15 bytes (for file management etc.)"Inkblots" made from a program that:
If random<0.25, go 1 pixel left and change colour of that pixel
If random<0.50, go 1 pixel right and change colour of that pixel
etc. Repeat for 1000 iterations( 5 mins 58 second to run). Most importantly, this was repeated on both all black and all white backgrounds, no change in file size.
This typically gives a very messy area about 1/4 the size of the screen and nothing else.
They usually come in at about 200 bytes.
A random bit of text snipped from a program came in at roughly 700 bytes
A screen filled with the number 8 came in at 1010 bytes.
A screen filled with pseudo random dots came in at 1048 bytes
all these are less
(started off with screen in checkerboard patteren, Do some funky stuff to the program to compensate for Basic's horrible pseudo number generator, changed colour of random pixel. repeat for a LOT of iterations (not all at once so I cant say how many)
The thing is, these things weigh in at less than or around 2:1 compression and I was wondering how such a puny processor could be apable of such compression ratios. (I am particularly impressed with the inkblot at a 10:1 compression ratio working with both Predominantly black and predominantly white squares.
I did a bit of experimenting and here are some results:
Estimated speed by running some simple basic programs through both calculator and Qbasic on P266 laptop gave the calculator a rating of "5 Kilohertz+ (Plus indicates PR rating )"
eg. finding 69! hundreds of times to test the math unit, working moemory storage and retreval etc. multiplying numbers by pi to test fpu.
Screen size is 62x126 pixels black and white
Theoretical image size for uncompressed image = 62x126x2/8 = 1953 bytes
Here are some experimental results
Blank screen - 15 bytes (for file management etc.)"Inkblots" made from a program that:
If random<0.25, go 1 pixel left and change colour of that pixel
If random<0.50, go 1 pixel right and change colour of that pixel
etc. Repeat for 1000 iterations( 5 mins 58 second to run). Most importantly, this was repeated on both all black and all white backgrounds, no change in file size.
This typically gives a very messy area about 1/4 the size of the screen and nothing else.
They usually come in at about 200 bytes.
A random bit of text snipped from a program came in at roughly 700 bytes
A screen filled with the number 8 came in at 1010 bytes.
A screen filled with pseudo random dots came in at 1048 bytes
all these are less
(started off with screen in checkerboard patteren, Do some funky stuff to the program to compensate for Basic's horrible pseudo number generator, changed colour of random pixel. repeat for a LOT of iterations (not all at once so I cant say how many)
The thing is, these things weigh in at less than or around 2:1 compression and I was wondering how such a puny processor could be apable of such compression ratios. (I am particularly impressed with the inkblot at a 10:1 compression ratio working with both Predominantly black and predominantly white squares.