-
Howdy folks,
Just a little sneak-peak my the latest work with tetsting the MEGA65's built-in ethernet adapter: I managed to port WeeIP, an embedded TCP/IP stack to use it (and to work with CC65's limitations). WeeIP was so small it didn't even have a DHCP or DNS client, so I implemented functions for those, and then hooked it all up together to prove that I can make a TCP connection over the internet.
It still needs some considerable work, but the hard part is done, in that it sends and receives packets etc. I'd love it if someone has interest on working on this further, and adding some of the missing things, like re-transmission of DNS queries if no response etc.
Those interested can see the code a github.com/mega65/mega65-weeip
LG
Paul
-
All sounds great, and looking forward to testing the new high-res text mode.
I also love the IF MEGA65 GOTO hack 
Beautifully simple.
LG
Paul.
-
Hello,
The C65 ROM has IEC code that works at 3.5MHz in C65 mode, I believe. The C64 mode code assumes 1MHz. We could implement something like that, as it does make sense for fastloaders etc. File an issue for me on github, and I'll have a think about how I might best do it. Probably will be based on changing any of the IEC lines, rather than arbitrary writes to the CIA.
LG
Paul.
-
DMA jobs cannot be interrupted on the MEGA65 at present, so you never need to worry about this. Any interrupts triggered during the DMA job will be delayed until the end of the DMA job.
LG
Paul.
-
ok. The MEGA65 book has some documentation on the DMA controller already, but let me know if you need more. It can copy in either direction.
LG
Paul.
-
I think the existing line length limit is fine at 88 chars, as you say. Putting variables and arrays into other memory banks strikes me as being relatively simple, in theory at least, as you can just use 32-bit pointers for their addresses, and use the ZP indirect 32-bit pointer addressing mode.
LG
Paul.
-
Pushing a probable fix now on 138-hdmi-audio-27mhz
LG
Paul.
-
Ah, that actually explains a number of similar problems we have been seeing.
This presents a bit of a problem: Either they should be hot, but then you can't relocate the screen and use disk routines. Or they should not be hot, in which case we have other problems. Presumably they should only be hot when hot regs are enabled.
LG
Paul.
-
This looks fantastic! Are you thinking about being able to put variables etc into another RAM bank?
LG
Paul.
-
So you know there is only 32KB of colour RAM? You can only choose the offset within that 32KB range, not put it anywhere. The colour RAM is its own RAM area, but the first 2KB is mirrored on top of the top of the 128KB at $1F800-$1FFFF. (Note that some models may have 64KB of colour RAM.)
LG
Paul.
-
Yes, I quite agree that it is almost certainly a bug in the VIC-III, that they figured was no point in fixing, since it can be so easily worked around.
LG
Paul.
-
Thank you for your kind words It's great to be part of such a fun community.
I suspect the BASIC READY loop detection magic in Xmega65 might not know about the OpenROMs enough.
LG
Paul.
-
Actually FeralChild wrote the post, I was just the post man
I continue to be impressed and super appreciative at what he is achieving with this, and super glad that he is doing it, as the result is better than I could have done myself, even if I had the time to work on it, which I don't.
LG
Paul.
-
I have started working on 80x25 text mode support, but I ran into a small glitch - the first text column seems to lack some pixels (look at 'R' in 'READY." on the right side):
Anyone knows what could be the problem? Here are the register changes I did after switching to 40 MHz ($41 value to $00) and unlocking VIC-IV:
- $D710 set to 0 to disable badlines and slow interrupt emulation (already checked, that this is not the cause of the problem)
- $D030 set to %01000000 to switch to C65 character set (already checked, that this is not the cause of the problem)
- $D031 set to %10100000 to enable 80x25 mode with extended attributes
- $D058/$D059 set to $50/$00 to set 80 column logical mode (already checked, that this is not the cause of the problem)
- pointers at $D060 (4 bytes) / $D064 (2 bytes) to the start of screen/colour memory
Display More
It's a "feature" of the VIC-III that the 80 column text mode starts a pixel or two to the left of the 40 column text mode. Real C65s do this. C65 ROM does it. You just have to shove it right a little using $D016 smooth scrolling. PEEK $D016 in C65 mode with original ROM and you can get the value that they use.
LG
Paul.
-
Yes, the X16 6502 FAT32 implementation is quite interesting, and quite likely worth taking a look at. I'll likely have a look at it when I get to fixing lots of the problems in the HYPPO FAT32 code.
LG
Paul.
-
Hi Snoopy,
Yes, you understood it correctly, and congratulations on finding the other place to patch
We will give it a go when we get a moment.
LG
Paul.
-
We did discover something funny with this overnight, though, where it seems that the initial transaction to 8 or 9 might still be intercepted by the ROM. We would get DEVICE NOT FOUND error once, and then it would be fine. But if we used a stock C64 ROM, then it would be fine from the first time. Will have to investigate this further.
Paul.
-
Well done!
The two-drive limit is only for direct connect 3.5" drives. It doesn't limit IEC drives on the serial port, of which you can still have a gaggle, until the pull-ups from all the drives results in a non-operable bus.
LG
Paul.
-
I think this legal issue has no chance of being real, but I totally understand your desire to play safe and avoid all chance of trouble.
LG
Paul.
-
A50T is too small. A100T is smallest that will fit. QMTECH Wukong A100T is the cheapest board I know of, that can run a MEGA65 core. Cost is about US$99. Currentlz no keyboard suppor, as I only tested HDMI output and SD card on it.
LG
Paul.
