Page 3 of 7
Addressing logic
Posted: Fri Mar 05, 2021 5:31 pm
by picosecond
16 minutes ago, jbaum81 said:
It's not really necessary
Nope. It's neither necessary nor helpful. Just ground the OE.
Addressing logic
Posted: Fri Mar 05, 2021 5:41 pm
by jbaum81
1 minute ago, picosecond said:
Nope. It's neither necessary nor helpful. Just ground the OE.
The datasheet was a little confusing to me. It did say that OE doesn't affect the flipflops, but it does say that new data can be entered or retained while output is in high impedance state (OE high) . I wasn't sure if the output would remain locked to old values if left low, so I decided to take the safe route and bring it high while I latch in the new data. I'd definitely prefer to just tie it to ground though if the output will literally just echo the state of the flipflops as they change. Here is the datasheet for the latch I chose.
https://www.ti.com/lit/ds/symlink/sn74f574.pdf
Addressing logic
Posted: Fri Mar 05, 2021 6:40 pm
by picosecond
55 minutes ago, jbaum81 said:
It did say that OE doesn't affect the flipflops
Right, CLK and D are the only inputs that affect flop state. OE only connects or disconnects the flops' outputs to the '574 outputs. The logic diagram on page 2 of the data sheet illustrates this pretty well.
Addressing logic
Posted: Fri Mar 05, 2021 8:47 pm
by Wavicle
14 hours ago, Lorin Millsap said:
You are basing that on obsolete information. The X16 uses addresses $00 and $01 for banking control freeing up the VIA for better uses.
Sent from my iPhone using Tapatalk
To be fair, "obsolete information" is the information in the
master branch of the CommanderX16 Documentation GitHub repo. The Readme.md does not hint that the information is obsolete. The zero page RAM/ROM banking documentation is in the
board_r2 branch. The emulator master branch still uses the
VIA for bank control and the
x16_board_r2 branch of the emulator has been dormant for 8 months and is now behind the master (notably the latest
SD card support appears to be missing). From what we've been told something is wrong with the r2 board and an r3 prototype is being planned - it isn't clear that the r2 board documentation or emulator is anything that could be relied upon. By all appearances, the obsolete information is the most up-to-date information one could expect to use when developing for the X16.
Addressing logic
Posted: Fri Mar 05, 2021 9:24 pm
by jbaum81
32 minutes ago, Wavicle said:
To be fair, "obsolete information" is the information in the
master branch of the CommanderX16 Documentation GitHub repo. The Readme.md does not hint that the information is obsolete. The zero page RAM/ROM banking documentation is in the
board_r2 branch. The emulator master branch still uses the
VIA for bank control and the
x16_board_r2 branch of the emulator has been dormant for 8 months and is now behind the master (notably the latest
SD card support appears to be missing). From what we've been told something is wrong with the r2 board and an r3 prototype is being planned - it isn't clear that the r2 board documentation or emulator is anything that could be relied upon. By all appearances, the obsolete information is the most up-to-date information one could expect to use when developing for the X16.
Gets better, I seen a video going over changes in the 2nd revision and it appears they're completely moving around the IO area as well. Looks as if both Via's will now co-exist at the top of the range. I definitely like those changes, it's a lot more comprehensive. None of it is updated in the wiki.
Addressing logic
Posted: Fri Mar 05, 2021 9:30 pm
by SlithyMatt
The README on the master is to support the current emulator. If you use 0/1 it will fail in emulation. So, unless you have a physical board, this doesn't help you. Once the emulator is updated to reflect the latest hardware changes, then we can worry about what's "obsolete".
Addressing logic
Posted: Fri Mar 05, 2021 9:45 pm
by Wavicle
6 minutes ago, SlithyMatt said:
The README on the master is to support the current emulator. If you use 0/1 it will fail in emulation. So, unless you have a physical board, this doesn't help you. Once the emulator is updated to reflect the latest hardware changes, then we can worry about what's "obsolete".
The Readme on the master branch supports the emulator on master branch. If you use 0/1 it should work just fine
if you use the r2 board branch. (See
this line which implements the zero page RAM bank select.)
Addressing logic
Posted: Fri Mar 05, 2021 9:59 pm
by Lorin Millsap
Well to clarify on that, the emulator hasn’t been updated because Micheal hasn’t received an updated board yet. However official announcements to the real hardware have been announced and that is the relevant and current data. Yes the emulator does need to be updated but it’s not getting updated till the person writing the emulator has the real hardware and can make sure that the emulator and real hardware behave the same. Much of this has very much been affected by the global circumstances. We would like to be moving much more quickly but sometimes we have to wait for updated components. Sent from my iPhone using Tapatalk
Addressing logic
Posted: Fri Mar 05, 2021 11:28 pm
by Wavicle
55 minutes ago, Lorin Millsap said:
However official announcements to the real hardware have been announced and that is the relevant and current data.
Is that true? Flipping through the official announcements the closest I see is a short reply on the second page of discussion to the
announcement that the second prototype boards have arrived. Isn't that obfuscated for an official announcement? I'm not trying to just kick dust here; it feels like even highly competent people are going to have difficulty finding non-obsolete information. I'm an engineer who has worked in, around, or close to VLSI and board design at Intel and Microsoft for nearly 20 years so I like to think I'm pretty competent at picking up on these things and I only knew about the change because Kevin mentioned it in the New Year's video.
Addressing logic
Posted: Sat Mar 06, 2021 12:44 am
by Lorin Millsap
Is that true? Flipping through the official announcements the closest I see is a short reply on the second page of discussion to the
announcement that the second prototype boards have arrived. Isn't that obfuscated for an official announcement? I'm not trying to just kick dust here; it feels like even highly competent people are going to have difficulty finding non-obsolete information. I'm an engineer who has worked in, around, or close to VLSI and board design at Intel and Microsoft for nearly 20 years so I like to think I'm pretty competent at picking up on these things and I only knew about the change because Kevin mentioned it in the New Year's video.
Kevin’s video is the one. There was no announce until the board with the changes was actually working.
Sent from my iPhone using Tapatalk