you should have a recycle bin that let's people/programmers (they are different) delete programs without having manually do it (even if it isn't actually hard)

Hmm, maybe make it available only in 15 mhz processors, to avoid any significant slow-down.

Multitasking?  Well, it might not be too slow for very simple text-based games, but (wouldn't it theoretically make programs twice as slow for two threads/tasks, three times as slow for three threads/tasks, etc., which is pretty slow for a Z80 calculator?).....

A mouse?  That's probably going to complicate program selection and will probably result in the GUI being heavy.

EDIT:  Also, if it uses a mouse, then programs won't be selected like how Flash Applications are selected on the TI-89 Titanium/Voyage 200, which you said you would do, right?  You don't have to do it that way, but I'm just informing you....

* Jon moves his index finger in the direction of interrupts, with a high interrupt frequency, so the split between interrupt execution, and regular code execution is about 50/50.

Multi-thread can be fun, but it won't make work anything faster... (in fact, slower than just doing 1 at a time because of the time needed to save registers and swap places and setting things up.)

Using interrupts can be of great use, but if using interrupts for greyscale, then you're pretty much lost. Maybe like a counter or a call routine that preserves all registers? inserted in the program loop at certain intervals, like inside a GetKey call.

A fast version of GetCSC, without the Bcalls.

c1-->

CODE

ec1 GetCSC:   Logged Jon LV5 Advanced (Next: 300) Posts: 278 Rating: +0/-0 NexGenGS « Reply #41 on: March 02, 2007, 04:15:00 pm » I know this gonna piss you off, but I'll say it anyway, I have a better version of the key input routine.  It isn't buggy, and it takes way less space, and doesn't destroy ix. c1--> CODE ec1 ;RAM needed: ;1 byte for lowercase_ofs ;1 byte for temp_b keyinput:   Logged Halifax LV9 Veteran (Next: 1337) Posts: 1334 Rating: +2/-1 NexGenGS « Reply #42 on: March 22, 2007, 12:57:00 pm » Ok update update: I knocked off 26 lines of the 8xn masked sprite routine. Yeah!!!! And god knows how many bytes(ill check later) Also I am currently working on another 16xn masked sprite routine that will knock off some pounds too. That coming later on tonight. Another update: I just finished my 16xn sprite routine with xor logic so that it is faster than ionlargesprite 8xn masked sprite routine # of lines = 50 lines 16xn sprite routine # of lines = 51 lines I have obviously optimized them to the fullest as ionputsprite takes up 42 lines. Logged There are 10 types of people in this world-- those that can read binary, and those that can't. Fallen Ghost Guest NexGenGS « Reply #43 on: March 22, 2007, 02:12:00 pm » Great! You made them from scratch or one someone's code? The number of lines is not necessarely a good sign, because it may mean a greater number of loops... If they go faster than existing ones, I definitely will need to use them... Logged Liazon Guest NexGenGS « Reply #44 on: March 22, 2007, 02:17:00 pm » if this is going to be an app, size won't be that much of a concern will it?  I'd think with an app, speed over size might be more attractive. so ya, I'd agree with Fallen Ghost. btw, will there be clipped routines?  just wondering since insanely fast masked+clipped 8x( routines have been made, though obviously not app friendly since they were mainly for non flash calcs.  ^^ Logged Print Pages: 1 2 [3] 4 5 ... 7   Go Up « previous next » Omnimaga » Forum » Calculator Community » Other Calc-Related Projects and Ideas » TI Z80 » NexGenGS   SMF 2.0.15 | SMF © 2017, Simple Machines Dream Portal 1.1 © 2009–2024 Dream Portal Team XHTML RSS WAP2 Server load over the past 5, 10 and 15 minutes respectively: 0.42333984375, 0.4599609375, 0.4296875 Page created in 0.15 seconds with 61 queries.