Matej Posted October 31, 2019 Share Posted October 31, 2019 (edited) Hello friends. I take 2x quad version for TMC2 experiments. Matej Edited October 31, 2019 by Matej 1 Quote Link to comment Share on other sites More sharing options...
enito Posted November 1, 2019 Share Posted November 1, 2019 Hi , I'm interested in one . Thx Enviado desde mi MI 8 mediante Tapatalk Quote Link to comment Share on other sites More sharing options...
foft Posted November 11, 2019 Author Share Posted November 11, 2019 (edited) On 10/31/2019 at 9:02 AM, foft said: Thanks. Do you know how I adjust it? I can search for D5 in the binary but the ones I looked at had many addresses. I guess if there is source code it’d be easier. I think to fix sid and/or quad pokey in the pokeymax it’d be best to get a larger fpga. It’d be possible to fit in the smaller one but honestly it’s more work then I think I’ll actually do! Fortunately they are not very expensive. I've included the sid 8580 core in the latest EclaireXL core (for larger A4 FPGAs) at 0xd240 and 0xd260. xxl kindly modified the slight sid player to access it and it seems, to my ear, to sound good. Though I'm not sure how its meant to sound. I probably shouldn't say that my wife said it sounds better than the pokey! SLIGHTSIDPLAYER-eclair.ATR Edited November 11, 2019 by foft Updated attachment: xxl made a version that silences sid on escape 4 Quote Link to comment Share on other sites More sharing options...
foft Posted November 11, 2019 Author Share Posted November 11, 2019 Err, realized I'm passing the wrong frequency to sid. It should be 1MHz and I'm enabling it at 1.79MHz by mistake! 2 Quote Link to comment Share on other sites More sharing options...
bocianu Posted November 12, 2019 Share Posted November 12, 2019 does it work for EclaireXL mini? rev 3.2 (core 37 - 11.11.2019) because i'm getting only some noises on audio output, or it just hangs... Quote Link to comment Share on other sites More sharing options...
foft Posted November 12, 2019 Author Share Posted November 12, 2019 It works on the Eclaire mini if you have the larger A4 fpga Quote Link to comment Share on other sites More sharing options...
bocianu Posted November 12, 2019 Share Posted November 12, 2019 Uh, so probably i don't have one Mine looks like that: Quote Link to comment Share on other sites More sharing options...
foft Posted November 12, 2019 Author Share Posted November 12, 2019 Yeah that is an A2. It’s not far off fitting into that but for now it doesn’t fit. Quote Link to comment Share on other sites More sharing options...
foft Posted November 12, 2019 Author Share Posted November 12, 2019 The problem is the sid tables each take 14K ram. I can clearly share that between stereo sids, saving 14K (or have mono). Though I wonder if I do can piecewise linear interpolation instead, like on the pokey mixer. That might make it fit. Will pop the data in Octave and have a look what they contain. 2 Quote Link to comment Share on other sites More sharing options...
bocianu Posted November 12, 2019 Share Posted November 12, 2019 Thanks for info foft! It's slightly off topic, but i'm just curious: what else do I miss, by having A2, not A4? Quote Link to comment Share on other sites More sharing options...
Lynxpro Posted November 13, 2019 Share Posted November 13, 2019 On 8/26/2019 at 12:31 PM, Mathy said: Hello foft Quad Pokey, hands down. We wanna show the power of the Pokey/Atari, not SID/C64. Sincerely Mathy PS and we need Quad Pokeys for arcade machines. I just had a thought of using 2 "Quad POKEYs" in a system. Reminds me, Candle said any A8 could address 16 POKEYs... Although it felt like a kick in the teeth when he said a C64 could address 32 SIDs. Quote Link to comment Share on other sites More sharing options...
foft Posted November 13, 2019 Author Share Posted November 13, 2019 A4 va A2: Mostly potential! Two things now: sid and 256k extended ram in block ram (faster than in sdram at higher cpu speeds) Number of pokeys: Well in 0xd200-0xd2ff 16 could be mapped, but they could be mapped elsewhere or bankswitched. Quote Link to comment Share on other sites More sharing options...
Lynxpro Posted November 13, 2019 Share Posted November 13, 2019 5 hours ago, foft said: A4 va A2: Mostly potential! Two things now: sid and 256k extended ram in block ram (faster than in sdram at higher cpu speeds) Number of pokeys: Well in 0xd200-0xd2ff 16 could be mapped, but they could be mapped elsewhere or bankswitched. What is of this sorcery that you speak of? Bankswitching POKEYs? An unlimited number of POKEYs? Quote Link to comment Share on other sites More sharing options...
Chilly Willy Posted November 13, 2019 Share Posted November 13, 2019 8 minutes ago, Lynxpro said: What is of this sorcery that you speak of? Bankswitching POKEYs? An unlimited number of POKEYs? Well, just as an example, you could simply make all POKEYs respond to $D200-$D20F, and writing a byte to $D2FF would set a latch that selected one of 256 POKEYs. That would be super-easy, and also super-silly. 2 Quote Link to comment Share on other sites More sharing options...
xxl Posted November 13, 2019 Share Posted November 13, 2019 or unused $D20C Quote Link to comment Share on other sites More sharing options...
Chilly Willy Posted November 13, 2019 Share Posted November 13, 2019 3 hours ago, xxl said: or unused $D20C Yeah, $D20B and C are unused, and that would make the logic decoding the bank register slightly less complex. You wouldn't need to decode for $D2xx since the POKEY chip select would handle that. You'd only need to look for B or C, and C is probably easier of the two. Quote Link to comment Share on other sites More sharing options...
Lynxpro Posted November 19, 2019 Share Posted November 19, 2019 On 11/13/2019 at 9:31 AM, Chilly Willy said: Yeah, $D20B and C are unused, and that would make the logic decoding the bank register slightly less complex. You wouldn't need to decode for $D2xx since the POKEY chip select would handle that. You'd only need to look for B or C, and C is probably easier of the two. Quick....get the over to the 5200 and implement this sorcery... Actually, Team 52 also needs an easy RAM upgrade so the console can shine even more against its contemporaries. After all, the Colecovision has the Super Game Module expansion which provides it with extra RAM and another sound chip [the RAM allowing more MSX titles to be ported]. And some way to implement SIO on it. I'd love to see the 5200 connected to A8s and STs in a MIDI Maze ring in the future...but I digress. Quote Link to comment Share on other sites More sharing options...
mono Posted November 19, 2019 Share Posted November 19, 2019 (edited) On 11/13/2019 at 6:31 PM, Chilly Willy said: Yeah, $D20B and C are unused, and that would make the logic decoding the bank register slightly less complex. You wouldn't need to decode for $D2xx since the POKEY chip select would handle that. You'd only need to look for B or C, and C is probably easier of the two. $D20B is POTGO (W) register. $D20C is used by SIO FIFO: Edited November 19, 2019 by mono Quote Link to comment Share on other sites More sharing options...
Chilly Willy Posted November 19, 2019 Share Posted November 19, 2019 (edited) 3 hours ago, mono said: $D20B is POTGO (W) register. $D20C is used by SIO FIFO: Funny - that POTGO register isn't in my Altirra Hardware Manual pdf... very strange, and I didn't even think about the fact that it was missing. Need to find a different copy of the pdf if the one I have has such a basic mistake in it. And that idea of putting using a FIFO for the serial is nice for high baud rates. It's something a number of old systems needed, like the Amiga. Even for not-so-high baud rates - DMA contention on the Amiga meant that you could lose data over MIDI because there's no FIFO for serial data. Anywho, probably a better idea for compatibility with existing software would be to make the first POKEY respond to $D200-F, a second one to respond to $D210-F, and any others to be bank selected to $D220-F with the bank select at $D23x... or something akin to that. Edited November 19, 2019 by Chilly Willy more info Quote Link to comment Share on other sites More sharing options...
xxl Posted November 19, 2019 Share Posted November 19, 2019 0xxxxxxx - initialization (FIFO cache cleaning) 10aaaaaa - set temporary border value (as AUDF3 content, $ 00 ... $ 3F) 11aaaaaa - set default border value it could have been better thought through: 00nnnnnn - nnnnnn - POKEY NO. 01xxxxxx - initialization (FIFO cache cleaning) 10aaaaaa - set temporary border value (as AUDF3 content, $ 00 ... $ 3F) 11aaaaaa - set default border value Quote Link to comment Share on other sites More sharing options...
xxl Posted November 20, 2019 Share Posted November 20, 2019 if SIO-FIFO is under D20C then just use D21C for bank selected POKEY ? Problem solved. Quote Link to comment Share on other sites More sharing options...
Apostrophe Posted November 25, 2019 Share Posted November 25, 2019 Hi foft, the pokey in my old Atari 400 is defect now. Is the Pokeymax v2 also compatible with the Atari 400? What about the space requirement? Sincerely Martin 1 Quote Link to comment Share on other sites More sharing options...
foft Posted November 26, 2019 Author Share Posted November 26, 2019 I think it will work fine in the 400 but don’t have one to check myself. The v2 is pretty narrow, I guess the (optional) header for the audio out adds a bit of height. Does anyone have a 400 they can measure please? Quote Link to comment Share on other sites More sharing options...
andymanone Posted November 26, 2019 Share Posted November 26, 2019 46 minutes ago, foft said: I think it will work fine in the 400 but don’t have one to check myself. The v2 is pretty narrow, I guess the (optional) header for the audio out adds a bit of height. Does anyone have a 400 they can measure please? Yes, I guess, that should fit. Above the POKEY (between Pokey and the aluminium cage) should be also enough space. But I have a couple of 400's too and I can measure it, when I'm home again, to be absolutely sure about... Gtx., andY Quote Link to comment Share on other sites More sharing options...
andymanone Posted December 12, 2019 Share Posted December 12, 2019 On 10/31/2019 at 8:59 AM, foft said: A little! Finally got the prototype boards to @santosplast week. I was quite busy and it took a few weeks to send! When he has time he will try to build 1-3 of them which we can then test/use to verify/fix issues. It uses very small parts so I’m not sure how building them will go. If they work and it turns out to be ok to make them, then it might be possible to assemble the other boards from this run. If they don’t then ... it depends. If the fix is clear then I can fix the design and then order new boards, probably fully assembled. @foft ...any progress here ?? Thanks, andY Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.