OT: Simulating a TI calculator with crazy 11-bit opcodes
|
|
« Next Oldest | Next Newest »
|
▼
08-12-2013, 12:36 AM
▼
08-12-2013, 06:07 AM
That's quite awesome, thanks for the link :) ▼
08-12-2013, 02:30 PM
Quote: 1+
Quote: Let's hope that ROM is available as well. That might be nice to have this calculator emulated as well.
Cheers
08-12-2013, 07:52 AM
A TI 57 is a 13 bit opcode machine :) See patents: 4125901 4125867 4107781 4079459 for a rom dump For testing, get go57c on google play or TI-57E from hrastprogrammer for hp 48/49 You can even use TI-57E on go48gx and go49g.
The HP50 version of TI-57E doesn't works well on go49gp (certainly some remaining bugs on go49gp) Edited: 12 Aug 2013, 8:07 a.m. ▼
08-12-2013, 11:03 AM
Hi Olivier
FYI, John McMaster has hi-res photos of the TI-57 chip: see I note that HrastProgrammer fixed a few bytes in his emulator, as compared to the patent. Did you make all the same fixes? Most are OCR errors but at least a couple are not.
Here are the differences between the microcode in the patent and that in the windows simulator: 25: 1abc 1a8c
Cheers ▼
08-12-2013, 12:33 PM
Yes I corrected some words. Some due to OCR, but other to restore the RST behavior of the real 57 (the listing in the patent did not have the same behavior on RST command). A new opcode was also added to allow a light sleep mode for the emulation (to avoid consuming too much batterie)
08-12-2013, 02:08 PM
Quote: This is a very interesting site. Thanks for its citation. After I bought my first calculator in 1972 (Bomar 901B, $130), I was intrigued by this sort of detail hidden within even simple machines. The simulated calculator seems to stop functioning after a division by zero...the display no longer updates.
▼
08-12-2013, 11:34 PM
Quote:Yeah it really is. The 8085 ALU was a very interesting read as well.
Quote:I wonder what the real thing did.
08-13-2013, 12:06 AM
Another bug. Generate an overflow e.g. 1000 * = = and it endlessly loops at the .OVF label. - Pauli |