If you haven't heard of this program:
http://www.ticalc.org/pub/84plusce/asm/programs/faster.zip
It'll actually make any 84 PCE up to 30% faster, and yes, it does work, with up to a 31% speed gain in my testing against a stock 84 PCE.
Since even still, the TI-84 Plus CE does graph slower then an 83/84 Plus monochrome calc, I'm wondering how I can make the program "permanently" run, so whenever I turn on my calculator, it auto starts, since this program does make it around the same speed of a default 83/84 Plus graphing calc.
Here's what the readme says about the program:
The speed boost comes from changing the wait-states setting for the flash.
The calculator's CPU actually runs fairly fast for a Z80-compatible CPU, but the
OS is located in flash memory (which also contains your archive space), and all
code executing there is severely hobbled by a 9 wait-state penalty, reducing the
effective clock speed by nearly a factor of 10. This program decreases the
wait-states to 6. (The hardware allows a minimum of 5, but that crashes my
calculator. 4 should work according to the flash chip's datasheet, but only if
you make some assumptions about signal hold times and TI's designed bus topology
which are apparently not true.)
Anyways, suggestions anyone?
http://www.ticalc.org/pub/84plusce/asm/programs/faster.zip
It'll actually make any 84 PCE up to 30% faster, and yes, it does work, with up to a 31% speed gain in my testing against a stock 84 PCE.
Since even still, the TI-84 Plus CE does graph slower then an 83/84 Plus monochrome calc, I'm wondering how I can make the program "permanently" run, so whenever I turn on my calculator, it auto starts, since this program does make it around the same speed of a default 83/84 Plus graphing calc.
Here's what the readme says about the program:
The speed boost comes from changing the wait-states setting for the flash.
The calculator's CPU actually runs fairly fast for a Z80-compatible CPU, but the
OS is located in flash memory (which also contains your archive space), and all
code executing there is severely hobbled by a 9 wait-state penalty, reducing the
effective clock speed by nearly a factor of 10. This program decreases the
wait-states to 6. (The hardware allows a minimum of 5, but that crashes my
calculator. 4 should work according to the flash chip's datasheet, but only if
you make some assumptions about signal hold times and TI's designed bus topology
which are apparently not true.)
Anyways, suggestions anyone?