sack-c0s

The idea is fine - the big pain in the ass is that it's only a single buffer so you can double-buffer the character map but when it comes to time to flip the buffers you still need to race the beam to update the colourRAM. You were so close, commodore....

I really don't know what to say at this point, or what the protocol is, but as this is public knowledge I'll repeat the message I received: 'I am absolutely devastated here. Jason Kelk passed away at 12:40pm today. He had been seriously ill again for the past 7 weeks. He requested to end treatment for long covid last weekend.' I really am lost for words...

Something else for people to have a go at if they want to... C64 version: Original for reference: The original GoatTracker 2.75 file is attached as a .zip file so that notation can be extracted. I've been wondering about getting to a Pokey version for a while, as I'm not 100% happy with the way it is around the 1:50 mark/position 18 in the tracker file - I was really feeling the need for a 4th channel at that point and didn't really find a 3 channel compromise I was completely happy with. I haven't really got the time at the moment so I figured I'd throw the source file out into the world and let other people play. Channel 1 is the bassline, Channel 2 is the melody, channel 3 is chords + snare drum. here comes your man mk2.zip

To be fair - i think it's the 1541 ultimate playing the mod, and not really the C64

I think there was some interesting discussion to be had, but none of it was really about the video.

That would explain why people aren't talking about it - i appreciate the feature being there for compatibility but the original feature itself does feel a bit useless. To be fair banking out 256 bytes could be useful if you did it on pages 0 and 1 because it would help with task switching...

I think that it would match the timings of an ideal 1541 for compatibility purposes (although after 30 years of use i don't think a real one would meet 'ideal' any more)

Not sure how I missed that feature. I should look into it

The 1541 Ultimate isn't a 'surrogate processor'. It emulates a disk drive and a tape recorder, which is accessed in the same way that the real devices would be in a cycle-exact simulation, which means that they are compatible with turbo loaders. As for the extra memory - that acts like a Ram Expansion Unit (REU) that was produced during the commercial life of the machine, only expanded to 16MB in the same way that other clones were. Honestly it's more like a RAM disk than a memory expansion, so not even as flexible as an Atari memory expansion. You need to copy in and out of the expansion RAM through a port and it stops the CPU whilst this is happening. You do get the advantage of a 1 byte per cycle transfer (which is impossible using a 6502 loop) but the downside is that it has to be a linear transfer so it's not easily abusable as a blitter. You can stop destination address incrementing however so it will hammer a single address with a sequence of bytes once per cpu cycle. It adds nothing to the CPU power of the machine. That remains the same as ever. The one awesome thing it does do that wasn't generally possible back in the day (unless you were a studio with an expensive PDS setup)is allow you to inject and run code into RAM over ethernet for cross-compiling and hardware testing, and if somebody ever points me towards an Atari 8-bit device that does the same thing you can be damn sure I'm having one of those!

that's not how you spell RISC OS

Way I see it: Atari folks try to beat the C64 by making something better: Cool. I'll download that. C64 folks fight back by trying to make something better: I have both side-by-side on a shelf so I'm going to download that too. So keep on fighting... I win either way

so Pacman is a kaaskop?

With this playback technique there's no need for blood though - we have a shared playroutine and potentially we can all go do our thing in terms of preferred editor.

nice work! I suppose the next step is to investigate writing an editor. If we're outputting a register stream then we're not necessarily tied to the whole concept of tracker patterns and instrument editing, so that should give a lot of freedom to lay out an editor in a new way... tdk_2.obx

If you are looking for a way to extract notation from C64 stuff this looks interesting....

I was waiting for the 'inevitable banhammer for posting fake news', but alas it wasn't to be... Thanks for everything, RIP

Don't know why I didn't notice it before, but now I see which part you mean. I just lowered the volume on that part and it does sound better as a result. The slides and effects are taking a bit of experimentation so I'm still working on those

Everything is fine. The short version is: *Just released a game in my dayjob, and everything was crazy busy getting things working, and then along come a few thousand players who find new ways to break stuff, so more work to fix things. *I Posted this as a bit of a break, literally just before leaving for the airport because I had some free time. *Then I went on a long overdue holiday (Which wasn't cancelled this time), which is where I am now. Also trying to buy a house, so that's also chaos. Once things calm down some more normal service will be resumed

It's been a while (it turns out releasing a game for modern platforms is actually hard work), but I finally have time to catch my breath, go on holiday, and go back to playing with 8/16-bit stuff. The bassline may or may not be accurate for C64 fans - that depends on if you listened to the original on a 6581 or 8580. It's closer to the way it sounds on my 8580 and personally I prefer it that way, but YMMV.

I actually used to be an Acorn Archimedes coder. After that I went back to the C64 (before branching out to just about anything that would let me run code on it). That thing had a nice, fast, simple CPU, linear chunky screen memory, 256 colours, it slaughtered the Amiga in terms of 3D power, and could give it something to think about in terms of 2D too. So... why would I code an Amiga game instead? *because the people who would download and try the game aren't a pain in the arse about it*. Endless amounts of crap about 'not showing the power of the machine enough', 'not being an Amiga beater', 'Just not being the game I'd have written so you shouldn't have either'. It was a complete and total buzzkill on a machine I actually liked. It just seemed completely and totally pointless. I'm seeing some of that here too. It's not technical, it's not a question of cycle counts, register usage or anything like that. Sometimes it's just that people just make it outright unsatisfying to code for some platforms. If you want an honest answer about why people may choose 'the weaker platform' then I'm afraid that's it. Compared to some 8-bit machines writing a Sinclair Spectrum game would be like walking both ways uphill in winter whilst carrying a rock on your back, but given that there would be a certain amount of *appreciation* for having made the effort...

I would bitch and moan about type-ins, but here's me personal relationship with the damn things: 1) tried typing them in 2) They didn't work. 3) saw correction 1-2 months later, but that was an age as a kid 4) gave up, and figured out enough to just write what the description said the game did, using the screenshot as guidance 5) Ended up on the less popular platforms for a while (Acorn Archimedes anyone?), so more of the same.... 6) ended up porting games for a living so I pretty much built a career out of typos in type-ins I suppose

The C64 doesn't have a monitor, but if you have a cartridge (such as the Action Replay) then you have a monitor and freezer available. don't get me wrong - I love my C64 - but once you try an Action Replay it doesn't feel like a full machine when it's disconnected. If you have actual hardware then I really recommend getting one

The z80 is clocked slow (2mhz i think) on the 128 to get it to work with the system bus - People seem to be of the opinion that the 6502 would just outperform it. I think the 16-bit registers and the stack handling is better on Z80, so it is sometimes nicer to code for, but for me it's the custom chips that make the machine - and that almost always leads back to the 6502. That said - there's always the Sega Mastersystem....

if your use-case is right for the job, couldn't you use an EOR filler instead?

I noticed the other day that you can kick the 6510 off the bus from the cartridge port so there's an interesting potential to just turn it into a zombie shell of a C64 with the custom chips driven by whatever you want on the cartridge. Sort of like a SNES SuperFX cartridge on steroids. Part of me wants to wire up a FPGA board to a connector to try that out....