Jump to content
IGNORED

Pokeymax v2


foft

Recommended Posts

4 hours ago, Gunstar said:

Will the Quad Pokey work in mono, stereo and quad when needed? If it's auto-quad will stereo still sound fine/right if it's in quad mode? Is there some firmware that comes with it to choose the mode or does it auto-change to any mode needed. i.e. mono/stereo/quad?

Quad pokey will work with the majority of stereo software. Basically everything unless its writing to the wrong shadow address.

 

A physical switch can be fitted to select enable/disable 'fancy' mode - i.e. quad or mono in this case. Additionally it is possible to wire A5 with a switch to enable/disable the additional pokeys (must be grounded or connected).

 

There are a few settings on the quad, which may be set temporarily (until power off) or stored to flash. These settings do not include 'quad/stereo/mono'.  They include: mixing curve, auto detect right channel, additional irqs on/off, volume per channel etc.

Link to comment
Share on other sites

I was always more interested in the American market's Covox brand Speech Thing and Voice master. I didn't even know about this the Polish 8-bit DAC Covox until a couple of years ago. I wasn't really all that interested in a separate upgrade for it. But Since it's included in  the Quad Pokeymax for an additional $5, I figured what the hell. I'm sure it will get a lot of use now that I have it. I see an 8-bit musical orchestra in my future...20 voices altogether...is it/will it be possible to have the Quad Pokey playing with/at the same time as the Covox? Or is that a one or the other thing? I'm referring to music that uses both at the same time, for example, is that possible? I have yet to download the manual, it wasn't working when I tried before and after ordering the Pokeymax.

 

Edit: downloading manual now. The problem is that I'm out in the country with few internet choices, I had Hughes-net and recently got rid of them as I get faster and more data just using my smartphone hot-spot that I already pay for. But it is limited too, and I ran out of high-speed, so the manual is downloading at about 15K a second right now...it's going to take about an hour. I'm sticking with the hot-spot until Starlink satellite service becomes available next year.

10 hours ago, Stephen said:

Covox is awesome!

 

Edited by Gunstar
Link to comment
Share on other sites

Can someone attach a PDF of the Pokey Max manual here or send it to me in a PM please? I have been trying to download the manual from the Onedrive link since before I pruchased the Pokey Max, managed to actually start a download once, which I came back and it had failed. Before and since I've tried dozens of time to download it and the Onedrive page just keeps coming up with an OOPs! something went wrong with our servers please try again later...

Link to comment
Share on other sites

  • 4 weeks later...
7 hours ago, foft said:

Warning: If you want to use it with AKI PS/2 it does not work. I don't have an AKI but I think its probably guessing the K0-K5 values.

The AKI should be syncing itself to K0 and K5. I'm doing something similar in one of my latest projects. Can't exactly say how AKI is using the information from those 2 counter bits, but in my implementation I edge trigger off of K0 and use that as the input to an onboard software shadow counter, using the state of K5 to reset it. This effectively syncs the shadow counter with the Pokey counter. Using this method saves 4 I/O bits for other uses, which I've already allocated in the new project.

 

I would think this should still work with the Pokeymax???

 

I guess I should purchase one of them to test with my project.

 

Link to comment
Share on other sites

I use an i2c io expander for the keyboard scanning. I wasn’t expecting expansions not to read k0-k5 properly. I’m looking into if it’s possible to match original timing more closely given the io expander constraints. Currently it scans somewhat slower.

Link to comment
Share on other sites

On 9/21/2020 at 9:05 AM, Muddyfunster said:

My Quad Pokey + Covox + 2x PSG arrived today. 

 

I'd show the picture with the candies, but well, I already ate them :)

 

Really well packaged in a box with lots of protection and the little plastic case for even more protection.

 

 

IMG_E7436.JPG

What is the 2x PSG option? I didn't know about that one. I got the Quad+covox version.

  • Like 1
  • Confused 1
Link to comment
Share on other sites

10 hours ago, foft said:

I use an i2c io expander for the keyboard scanning.

Interesting.

 

10 hours ago, foft said:

I wasn’t expecting expansions not to read k0-k5 properly. I’m looking into if it’s possible to match original timing more closely given the io expander constraints. Currently it scans somewhat slower.

I don't think of it as not reading the Pokey key scan counter properly, but as simply using only part of that counter's output for reference. If you've changed the frequency at which it counts that could be looked at as non-proper and not adhering to spec. However unless you previously knew about the problem with legacy alternative keyboard devices reading Pokey's counter in this way, it would seem innocent to alter the counter's clock frequency since it would still work when reading a stock Atari keyboard.

 

The fact is changing that clock frequency creates a  problem if AKI will no longer work because of it. Making it something that really should be corrected on Pokeymax, and not on the legacy keyboard translator. Speaking of which, I now strongly suspect that my method of creating a software shadow counter will also fail because of my use of a hardware timer interrupt in the PIC chip as a pacemaker that assumes a certain Pokey key scan frequency is in play. Probably AKI's code is doing something very similar.

 

A comment from my shadow counter routine...

Quote

This interrupt occurs at a 31.250 Hz interval, twice the speed of Pokey's keyscan rate. At each IRQ,
Pokey Counter bit-0 is checked for state change. Upon change, an internal Shadow Register for Pokey is incremented.
Also bit-5 is checked for sync.

My shadow counter is also running at twice the rate of Pokey's key scan due to the PIC timer's clock frequency, and then later divided down for actual use.

 

So in essence the PIC hardware timer's overflow sets the pace at which the shadow counter is incremented. Change the Pokey's key scan frequency, and you'll start to see the shadow and Pokey counters getting out of sync between corrections. Not a good thing :) .

 

I already have the shadow counter code implemented in my V2.5j firmware, which allowed the unused I/O bits to be allocated for an arrow key to joystick translation in the latest TK-II implementation, both as a stand-alone upgrade for stock Atari 8-bit machines, and also as an upgrade for the 1088XLD. I am also using this for a variant in the upcoming 576NUC+ product where the extra I/O is used for power control, ROM banking, and extended memory configuration. BTW, the 576NUC+ also has an expansion header for Pokeymax extended feature support.

 

So between AKI (now a Lotharek product) and the TK-II variants, there would certainly be a far ranging issue if Pokeymax would not work for them.

 

 

  • Like 1
Link to comment
Share on other sites

@Mr Robot, yes, but it is not produced yet. Everything depends on demand. Now we have Quad plus 2xPSG plus Covox (4 channels) or Quad plus 2xSID. We have some players able to play mods for SID or PSG. But nobody made any music for Quad pokey or Pokey plus SID/PSG. No demos or games. Therefore demand is near zero. 

Production of v3 PokeyMAX having all chips on board (plus maybe Paula?) is expensive. I do not want to freeze money, we must wait for some soft productions increasing demand...

Edited by Duddie
  • Like 1
Link to comment
Share on other sites

With Midi

28 minutes ago, Duddie said:

But nobody made any music for Quad pokey or Pokey plus SID/PSG. No demos or games. Therefore demand is near zero. 

 

I hope this will change once MidiJoy is allowed to be sold after the ABBUC hardware contest. Given the demand for "normal" MidiJoy, I assume that chiptunes musicians will be quite interested...

  • Like 2
Link to comment
Share on other sites

2 hours ago, foft said:

@mytek: I’m seeing what I can do to match original scan. Another option is to modify firmware on the pic to measure the toggle rate then set the timer.

I don't really think that'll be an option for the AKI, unless the person who took over and modified the source several years ago can be persuaded.

 

And I'm afraid it's also not going to be an option for me either, since I have very little code space left in which to add such a routine.

 

Would it help if I were to send you a TransKey with the V2.5j firmware? The Pokey piggyback will work in any A8. Oh wait a minute you are Europe... is there someone here in the states I could send it to instead that could pass it on to the right person. I'll give you one for free, but just don't want to hassle with shipping overseas.

 

Link to comment
Share on other sites

6 hours ago, candle said:

why are you relaying on scanning frequency?

set interrupt on both edges for k0, sync up with k0 and k5 states and you're done

I don't have that option for that I/O pin. Keep in mind the TK-II started out looking at the entire 6-bit Pokey key scan counter. Only recently did I change to a synced shadow counter based on only 2-bits of the Pokey counter. Because this also needed to work with previous revisions of the hardware, the option to move I/O pin assignments was not a possibility.

 

And then there is the AKI based keyboard adapter that Lotharek sells, which is also tied into frequency. Don't have that source code so I'm not privy to what would be required to change its method of synchronization, but perhaps you do know since you used to incorporate that into your simple stereo board. So why is that relying on frequency? ;)

 

Link to comment
Share on other sites

well, i don't know much about AKI interfaces, except I made one with atmega for atari ST keyboard, and another one for usb keyboards

i didn't have any preconceptions on how to approach this, just one observation that AKI that was at some point available in Poland was using only K0 and K5 for counting

this lead me to the conclusion it must be syncing up to the overflow, and then use monothonic property of this counter

whenever there is a diffrence in internal counter value vs monitored pins, state machine is reset and resynced, so frequency doesn't matter

 

Link to comment
Share on other sites

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.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...