This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Messages - DrDnar
Pages: 1 ... 27 28 [29] 30 31 ... 38
421
« on: June 17, 2011, 12:58:11 am »
That's a good question. The Desolate guy remarked to me that experimentation showed that the LCD driver only updated the screen at 60 FPS. I've been meaning to write a program for precisely checking this.
422
« on: June 17, 2011, 12:48:36 am »
Obviously, TI is using slightly different LCD driver models. As long as they're not using the super-slow ones, I don't think we should really care. There's little point to screwing around with the test mode anyway. Incidentally, the blue lines of death only seem to work on the TI-83+.
By the way, I thought that the maximum theoretical bandwidth for the LCD drivers was more like 120 FPS. Do the new ones require less delay?
423
« on: June 17, 2011, 12:41:24 am »
Yeah, I first addressed, that, and the I addressed the BCD in Axe. I'm assuming that Scout is coding for a PC, and I thought that it might be worth pointing out that there are performance considerations in which you choose.
424
« on: June 17, 2011, 12:38:38 am »
That may or may not be faster depending on CPU design, memory bandwidth, caching, integer size, et cetera. If you have high-latency memory, BCD may well make things much faster, as the code for screwing with the individual nibbles could be faster than the penalty for a cache miss.
The Z80 has instructions for helping with BCD a little, namely the DAA instruction. You might consider making an Axiom of a generic open source BCD library.
425
« on: June 17, 2011, 12:11:56 am »
Is there a converter between zStart fonts and Omnicalc fonts? It looks like Omnicalc uses a different header and stores 7 bytes per character, whereas zStart stores 8 bytes per character.
426
« on: June 16, 2011, 11:24:24 pm »
Try using doubles. Integers will definitely not go up to the factorial of 100, which has 157 digits. You'll lose precision in the lower digits, but that may not be not important.
427
« on: June 16, 2011, 09:27:30 pm »
Maybe you can convince Michael to GPL Omnicalc. Then you could absorb Omnicalc into zStart. I figure that Michael might as well, considering that he isn't active anymore, and Omnicalc needs more work for 2.55 compatibility.
428
« on: June 16, 2011, 03:51:42 am »
Sure. You could then read the major version in {L1+1} and the minor version number in {L1}. In fact, Asm(EF6F4C6768)->{L1}r will work.
429
« on: June 16, 2011, 03:44:40 am »
OS version check? Really? Molest me not with this pocket calculator---er . . . well anyway, EF6F4C6768 will return the major version number in the high byte and the minor version number in the low byte. So divide by 256 to get the high byte and do "and 255" to get the low byte. In theory, "or 0" should also work, assuming that doing an 8-bit logical operation implicitly strips the high octet.
430
« on: June 16, 2011, 03:32:53 am »
IIRC, BrandonW has stated that TI cannot fix the universal flash unlock exploit without removing the ability to send variables directly to the archive.
431
« on: June 15, 2011, 03:46:19 pm »
That's a good point. Someone could embed a call to the Clear All Memory OS routine. Or even the boot code self test.
432
« on: June 15, 2011, 03:42:52 pm »
Make that C7.
433
« on: June 15, 2011, 03:32:48 pm »
:AsmProg
:C7No Axe needed.
Edit: C7
434
« on: June 15, 2011, 02:51:55 am »
Correct. Omnicalc, for example, has problems. Although, I released a version which fixed some of the worst problems. I also found 2.53 to be generally quite unstable.
435
« on: June 15, 2011, 02:39:12 am »
It's bad if you've been trying to avoid MathPrint so that third-party software, you know, actually works.
Pages: 1 ... 27 28 [29] 30 31 ... 38
|
Show Posts