| 2 Aug 2021 |
 rondnelson99#5674 | Thanks for the clarification.
I think it would be more clear written this way:
"There are no extra clocks as a result of this bug, only those corresponding to the actually executed
instructions."
It doesn't seem grammatically correct to me, and I can't really figure out what it's supposed to mean | 00:13:59 |
 AntonioND#7566 | To be fair, I've been asking people to not use that document for quite some time 😅 | 00:33:14 |
| AntonioND#7566 | It's outdated and not even that complete | 00:33:32 |
| AntonioND#7566 | I've been trying to decide what to do with it. I don't want to remove it completely, but I may add a message in big red letters in the cover | 00:34:21 |
 CasualPokePlayer#8731 | in big red letters THIS IS AN UNMAINTAINED AND INCOMPLETE DOCUMENT! THIS DOCUMENT SHOULD NOT BE RELIED ON! | 01:14:51 |
| AntonioND#7566 | Yeah something like that | 01:15:42 |
 PinoBatch#1742 | I used Crop in Photos to get exactly at the edge of the transparent part of my Game Boy's lens. I meant file size, as I doubt 1.4 MB was necessary. In fact, I remember reading somewhere a message from another user to the effect "I just bought a new phone and can no longer upload photos to Discord because they are larger than Discord's 8 MB limit." | 01:41:30 |
| PinoBatch#1742 | I remember neither the exact words nor which server. | 01:41:59 |
|  Nilin joined the room. | 03:32:22 |
| rondnelson99#5674 | I didn't realize it's use was discouraged, I'll keep that in mind. That said, there are a lot of things that aren't documented anywhere else (besides emulator source code) afaik. A couple examples I ran into yesterday and today are the halt bug and the fact that the memory writes are on the last 2 m-cycles of push. | 03:41:50 |
| rondnelson99#5674 | The halt bug seems pretty well-known though. Surely it's documented somewhere else? | 03:42:21 |
| CasualPokePlayer#8731 | the halt bug where IME is 0 is very well documented | 03:48:38 |
| CasualPokePlayer#8731 | the halt bug where IME is 1, not so much (granted, that is a lot harder to trigger for obvious reasons) | 03:48:44 |
| CasualPokePlayer#8731 | (it's technically the same either way really, but the result is different since one actually fires an interrupt and the other doesnt) | 03:48:46 |
| rondnelson99#5674 | Oh wow yes I didn't know that was a thing | 03:50:32 |
| rondnelson99#5674 | But where is the ime=0 variety documented | 03:50:52 |
| CasualPokePlayer#8731 | pandocs should have ime=0 iirc | 03:51:08 |
| CasualPokePlayer#8731 | rgbds' opcode documentation also has it | 03:51:18 |
| rondnelson99#5674 | I looked for it yesterday, maybe I didn't see it | 03:51:27 |
| CasualPokePlayer#8731 | https://rgbds.gbdev.io/docs/v0.5.1/gbz80.7#HALT | 03:51:51 |
| CasualPokePlayer#8731 | * https://rgbds.gbdev.io/docs/v0.5.1/gbz80.7#HALT | 03:51:54 |
| CasualPokePlayer#8731 | * https://rgbds.gbdev.io/docs/v0.5.1/gbz80.7#HALT | 03:51:57 |
| CasualPokePlayer#8731 | tho it doesnt document what happens when the halt bug occurs with halt | 03:52:19 |
| CasualPokePlayer#8731 | * tho it doesnt document what happens when the halt bug occurs with IME=1 | 03:52:22 |
| rondnelson99#5674 | Yeah it is there | 03:52:27 |
| CasualPokePlayer#8731 | * tho it doesnt document what happens when the halt bug occurs with IME=1 (which is interrupt happens but when it returns it executes the halt again, i.e. the PC with the halt is pushed instead of the PC after the halt) | 03:53:01 |
| CasualPokePlayer#8731 | * tho it doesnt document what happens when the halt bug occurs with IME=1 (which is interrupt happens but when it returns it executes the halt again, i.e. the PC with the halt is pushed instead of the PC after the halt, of course that obviously is very rare to happen for obvious reasons) | 03:54:01 |
| rondnelson99#5674 | The interrupts page in the pandocs seems less in-depth than the rest of it, considering how many nuances interrupts have | 03:54:19 |
| rondnelson99#5674 | I'm not qualified to expand it though | 03:54:32 |
| rondnelson99#5674 | The ime=1 version seems potentially catastrophic. If you instructed the Vblank handler to turn off the screen like the way ISSOtm recommends it, it would cause a deadlock. At least it's suuper unlikely though | 03:58:25 |
