Jump to content
IGNORED

Supercart I Modification / Daughter Board Question


Recommended Posts

So here is my thinking. I do not like the quality of the sound output that I can get fro the built in 1 bit buzzer on the Aquarius. I realize it is cheating a bit, but since nobody runs my stuff anyway, I figure as long as I can do it, it is still valid. I am thinking that Jay's Supercart board has access to the address bus of the Aquarius. In fact, my guess is that it at least has connections to the entire address bus.

 

How difficult do you think it would be to tap only some leads on the Aquaricart and read the entire address / data bus for either port FC (or some other unused port) and use that as an 8-bit source for a daughter board with an 8-bit DAC attached.

 

http://www.ti.com/li...ink/dac0808.pdf

 

The above chip is a nice CMOS chip that can interface with TTL logic levels and uses a 5 V source. It doesn't draw a lot of power.

 

The question is, where do I start? This one is way out of my capability and element, so be gentle!

Link to comment
Share on other sites

EDIT: Fixed (I think) the NAND/NOR logic below. Double-check my work in any case.

 

-----

 

Ignoring the SuperCart itself for a moment, let's concentrate on what it would take to interface a DAC to the Aquarius, via an IO port address. For reference, here's the Aquarius pinout: http://www.vdsteenoven.com/aquarius/bus.html

 

You can do this pretty easily with standard TTL parts bin components, if you wanted to try breadboarding something up.

 

The Z80 outputs the IO port address on address lines A0 - A7. It signals an IO port access with the line ~IORQ. (The ~ means it's active low, ie. it's 0 during an IO request and 1 otherwise.) It signals a write with ~WR.

 

To drive the DAC, you need to latch an 8-bit value. A 74LS273 or similar should do the trick. Wire its data inputs to D0 - D7, and its data outputs to A8 - A1 on that DAC. Take care: According to that data sheet, the DAC's A8 should hook to the latched D0 and the DAC's A1 should hook to the latched D7. They labeled their bus in the opposite direction from what we're normally accustomed to.

 

Now grab a 74LS273 datasheet. (Here's one: http://www.datasheetcatalog.org/datasheets/90/375556_DS.pdf ). It says here that the latch is positive edge triggered. That means when its clock input goes from low to high, it copies the input to the output. Otherwise, when the clock is low, or if it stays high, the output will remain constant.

 

Looking at my Z80 manual, it says the Z80 puts the data on the bus a little before it drops ~WR from 1 to 0 (about half a clock cycle), and holds the data on the bus a little after it raises ~WR from 0 to 1 (again, about half a clock cycle). This gives you the option of capturing the data at either end of the write.

 

Probably the easiest way to decode the bus is to use some NAND and NOR gates. You want the clock signal to change when A7 - A0 = $FC, ~IORQ = 0 and ~WR = 0, and stay stable the rest of the time. Quite frankly, there are many ways you can do this. Here's one: Get two 74LS00 quad 2-input NAND, and one 74LS02 quad 2-input NOR, and connect as follows:

 

* NAND #1: A7, A6: Output goes 0 when both address bits are 1

* NAND #2: A5, A4: Output goes 0 when both address bits are 1

* NAND #3: A3, A2: Output goes 0 when both address bits are 1

* NOR #1: A1, A0: Output goes 1 when both of these go 0

* NOR #2: ~WR, ~IORQ: Output goes 1 when both of these go 0

* NAND #4: Output of NOR #1 and NOR #2: Output goes 0 when A1, A0, ~IORQ, ~WR are all 0.

* NOR #3: Output of NAND #1 and NAND #2: Output goes 1 when A7 - A4 = 1

* NOR #4: Output of NAND #3 and NAND #4: Output goes 1 when A3, A2 = 1; A1, A0, ~IORQ, ~WR = 0

* NAND #5: Output of NOR #3 and NOR #4: Output goes 0 when A7 - A0 = $FC; ~IORQ, ~WR = 0.

 

You should be able to wire the output of NAND #5 to the "clock" pin on the 74LS273. If, for whatever reason, you need to invert that clock, you can pass it through NAND #6 (feed it to both inputs) to invert it.

 

Now, I picked the 74LS00 and 74LS02 since they're ubiquitous. You may even be able to find them on the shelf at a Radio Shack still. You can do this with fewer components if you got fancier.

 

Tada... now you have your cheap DAC board. :-)

 

If you don't want to spring for the DAC itself, you might try just building an "R-2R" resistor ladder, and buffer it with a simple emitter-follower circuit... but, I'll stop here and let you catch your breath. :-)

Edited by intvnut
Link to comment
Share on other sites

Ok, now warping back into the SuperCart context... The SuperCart already does some bus decoding, and it has a latch for latching the databus when it sees a write inside the cartridge space. If you wanted to expand that logic slightly, say to control two latches depending on the LSB of the address, it seems like that wouldn't be very difficult to do. I'll let Jay comment on what sorts of things you'd need to jerry-rig on there to get that two-latch action going.

 

In that strategy, then you'd write to, say, even addresses for bankswitching, and odd addresses for updating the DAC. Or, if you wanted to minimize the modifications to the SuperCart itself, only address-gate "clock" for the new DAC latch. That would make all addresses continue to work for bankswitching, and, say, odd addresses for updating the DAC. You'd have to follow any write to the DAC with a second write to set the bank back to what you wanted. Sloppy, but less work.

Edited by intvnut
Link to comment
Share on other sites

What I can't figure out on the Supercart is how it knows a bankswitch command has occurred. It doesn't use a port. All you do is this:

 

  ld   de,$c000	  ; load the target address into register DE
  ld   a,$00		 ; load the target bank number into the accumulator
  ld   (de),a	    ; write the bank number into the cartridge space

 

So we load the bank number and then write it to $C000 - no port used at all.

 

So, how does the Supercart latch the incoming data? Is it on the Address bus?

 

Need help from Jay here. I was also thinking based on the above that I could go ahead and just make a DAC cartridge on my own.

 

I wish there was a breadboard with an edge connector that I could play with...

Link to comment
Share on other sites

Sorry for not chiming in sooner ... I've been away from my files for the past day or two, but I'll post some info on the SuperCart later today, along with some pictures to indicate the solder points. Most of the signals you should need for integrating a DAC as Joe described (thanks for the excellent write-up, Joe!) are present on the SuperCart. The only exception is ~IORQ, which the SuperCart does not use, but I can point out the card-edge finger where you can attach a wire to access that signal.

 

(Of course, if the Aquarius's built-in sound is too limiting, you might want to see what you can do with the AY-3-8910 PSG in the Mini Expander instead. The PSG is a widely-used component in classic computers and arcade coin-ops, and there are already trackers and other tools available for it. Might be an easier option than integrating a whole new component.)

Link to comment
Share on other sites

What I can't figure out on the Supercart is how it knows a bankswitch command has occurred. It doesn't use a port. All you do is this:

 

  ld   de,$c000	  ; load the target address into register DE
  ld   a,$00		 ; load the target bank number into the accumulator
  ld   (de),a		; write the bank number into the cartridge space

 

So we load the bank number and then write it to $C000 - no port used at all.

 

So, how does the Supercart latch the incoming data? Is it on the Address bus?

The SuperCart bankswitching logic is an extension of the design that Atari used in their 7800 cartridges. The 6502 CPU uses memory-mapped I/O, which is why bankswitching in the 7800 is controlled through memory addresses and not I/O ports. I kept it that way in the SuperCart design because this made the logic much simpler to implement, since the Aquarius already handles the necessary address decoding.

 

The SuperCart's 74LS02 initiates the bank switches, based on the state of the relevant signals. When you write a bank number to the cartridge, ~MREQ goes low to indicate a memory request, and shortly afterward, ~WR and ~CS also go low. ~CS is a chip select signal that is generated by the Aquarius; it is low when both A14 and A15 are high, which indicates an address inside the cartridge space. When all three signals are low, the 74LS02 sends a clock pulse to the 74LS374, which latches the bank number from the data bus. The ROM's top address lines are driven by the latch outputs, so when the latch's state is changed, the active ROM bank is changed. Voilà ... instant bankswitching!

 

(Actually, at the risk of going into too much detail, ~MREQ is being used to control the 74LS245, which buffers the other signals I mentioned. When ~MREQ goes high, the buffer is tri-stated, and the resistors near the 74LS02 pull the other signals high. This "freezes out" bank switches while ~MREQ is inactive. If you're using the SuperCart's 8K bank mode, the 74LS374 is also tri-stated for addresses $E000 and above, and its outputs are pulled high by the resistors near the ROM. This is why the last 8K ROM bank is always used as the "fixed bank".)

Link to comment
Share on other sites

What I can't figure out on the Supercart is how it knows a bankswitch command has occurred. It doesn't use a port. All you do is this:

 

  ld   de,$c000	  ; load the target address into register DE
  ld   a,$00		 ; load the target bank number into the accumulator
  ld   (de),a		; write the bank number into the cartridge space

 

So we load the bank number and then write it to $C000 - no port used at all.

 

So, how does the Supercart latch the incoming data? Is it on the Address bus?

 

Jay's posted the details of how his SuperCart latches the bank number. For your DAC, you could go memory mapped or I/O port mapped. It's really just about whatever's most convenient. Both are valid approaches. Memory mapped looks slightly faster, if I read the data sheet correctly -- 7 T states for LD (xx), y vs. 11 T states for OUT (x), y.

 

I wish there was a breadboard with an edge connector that I could play with...

 

Ah, but there is something close! http://compare.ebay.com/like/320835739256?var=lv&ltyp=AllFixedPriceItemTypes&var=sbar&_lwgsi=y&cbt=y

 

That one is kinda pricey, but if you look around, I know there are others similar to it. IIRC, Chad Schell used such a prototyping board back when he developed the original Intellicart.

Link to comment
Share on other sites

(Of course, if the Aquarius's built-in sound is too limiting, you might want to see what you can do with the AY-3-8910 PSG in the Mini Expander instead. The PSG is a widely-used component in classic computers and arcade coin-ops, and there are already trackers and other tools available for it. Might be an easier option than integrating a whole new component.)

 

There's a Viterbi-algorithm based encoder out there that will get you roughly 8-bit sound by modulating the 3 PSG volume registers on the PSG. It produces pretty decent audio. Google for it.

 

The PSG audio channels disable to "1", meaning if you just write $3F to the channel enables, you can modulate the output directly by only writing to the three volume registers. No tone or noise will get injected. The viterbi analyzer then computes the best sequence of PSG volume updates to approximate the PCM data you're starting with.

 

When Arnauld Chevallier and I first looked at this on the Intellivision, we assumed the three channels added linearly. However, I've since found documentation online that suggests there's some serious non-linear effects combining the three channels. The attached file (I forget where I found it on the web) has a 3-dimensional table of relative voltage levels produced by combining the three channels. Each sub-table is a 16x16 array of relative voltages modulating two of the channels, holding the third constant. The 16 sub-tables then represent modulating the 3rd channel.

 

Edit: BTW, we did account for the fact that the channels individually have logarithmically-spaced steps. The volume steps are sqrt(2) apart. If you think of the lowest volume as 1, the steps go 1, sqrt(2), 2, 2*sqrt(2), 4, etc. The non-linearity comes when adding them together via a dead short, as the Intellivision and likely the Mini-expander do. If you try to add, say, 2 on channel A and 4 on channel B, you don't quite get 6.

PSG_volume_table.txt

Edited by intvnut
Link to comment
Share on other sites

 

(Of course, if the Aquarius's built-in sound is too limiting, you might want to see what you can do with the AY-3-8910 PSG in the Mini Expander instead. The PSG is a widely-used component in classic computers and arcade coin-ops, and there are already trackers and other tools available for it. Might be an easier option than integrating a whole new component.)

 

There's a Viterbi-algorithm based encoder out there that will get you roughly 8-bit sound by modulating the 3 PSG volume registers on the PSG. It produces pretty decent audio. Google for it.

 

The PSG audio channels disable to "1", meaning if you just write $3F to the channel enables, you can modulate the output directly by only writing to the three volume registers. No tone or noise will get injected. The viterbi analyzer then computes the best sequence of PSG volume updates to approximate the PCM data you're starting with.

 

 

I have seen the Viterbi-algorithm based encoder for the MSX and I realize that I could go ahead and plug in the mini-expander and then pound out sound. The problem that I have is that I just don't have enough cycles to play back decent audio and video at the same time. If I use the 1 bit DAC then I have to oversample to get decent sound and if I use the PSG then I have a similar problem. If I built a simple 8 bit DAC on an unused port then I could simply ld a, (hl) and out $??, a - I'd have plenty of cycles if I were to playback at 22,050.

 

I'm not sure. I don't have to build a DAC, but since Jay tells me that I can mod my Supercart to do it, then that just may be worth it.

 

I may need ANOTHER Supercart! I hope I don't exhaust Jay's supply.

Link to comment
Share on other sites

 

(Of course, if the Aquarius's built-in sound is too limiting, you might want to see what you can do with the AY-3-8910 PSG in the Mini Expander instead. The PSG is a widely-used component in classic computers and arcade coin-ops, and there are already trackers and other tools available for it. Might be an easier option than integrating a whole new component.)

 

There's a Viterbi-algorithm based encoder out there that will get you roughly 8-bit sound by modulating the 3 PSG volume registers on the PSG. It produces pretty decent audio. Google for it.

 

The PSG audio channels disable to "1", meaning if you just write $3F to the channel enables, you can modulate the output directly by only writing to the three volume registers. No tone or noise will get injected. The viterbi analyzer then computes the best sequence of PSG volume updates to approximate the PCM data you're starting with.

 

 

I have seen the Viterbi-algorithm based encoder for the MSX and I realize that I could go ahead and plug in the mini-expander and then pound out sound. The problem that I have is that I just don't have enough cycles to play back decent audio and video at the same time. If I use the 1 bit DAC then I have to oversample to get decent sound and if I use the PSG then I have a similar problem. If I built a simple 8 bit DAC on an unused port then I could simply ld a, (hl) and out $??, a - I'd have plenty of cycles if I were to playback at 22,050.

 

I'm not sure. I don't have to build a DAC, but since Jay tells me that I can mod my Supercart to do it, then that just may be worth it.

 

I may need ANOTHER Supercart! I hope I don't exhaust Jay's supply.

 

I'm not sure why you think you don't have the cycles to get good quality out of the PSG. Are you basing it on the complexity of the MSX decoder? You could always make your own decoder, and make different tradeoffs. You're already outputting the 1 bit data at a fairly high rate. You won't need to oversample so much with the PSG and its 4-bit volume registers.

 

A different question: Exactly which PSG variant does the Mini-Expander use? If it uses a variant that still lets you turn the I/O ports to "output mode", then you could just wire the DAC to the hand controller port...

 

Link to comment
Share on other sites

Speaking of the Viterbi encoder (which I found here: http://www.msx.org/downloads/related/sound/pcm-encoder-001 ), it looks like it uses a linear summing of the volumes for its model of the PSG.

 

   double vol[16];
   vol[0] = 0;
   for (int i = 1; i < 16; i++) {
       vol[i] = pow(2., i / 2.) / pow(2., 7.5);
   }

//... and a little later:

   double curV[256][16];
   for (int i = 0; i < 16; i++) {
       for (int j = 0; j < 16; j++) {
           for (int in = 0; in < 16; in++) {
               curV[i * 16 + j][in] = vol[i] + vol[j] + vol[in];
           }
       }
   }

 

I wonder if it would get better results if we load that non-linear table I shared earlier into curV. Seems like it ought to improve the SNR further.

Link to comment
Share on other sites

Speaking of the Viterbi encoder (which I found here: http://www.msx.org/d...pcm-encoder-001 ), it looks like it uses a linear summing of the volumes for its model of the PSG.

 

double vol[16];
vol[0] = 0;
for (int i = 1; i < 16; i++) {
	vol[i] = pow(2., i / 2.) / pow(2., 7.5);
}

//... and a little later:

double curV[256][16];
for (int i = 0; i < 16; i++) {
	for (int j = 0; j < 16; j++) {
		for (int in = 0; in < 16; in++) {
			curV[i * 16 + j][in] = vol[i] + vol[j] + vol[in];
		}
	}
}

 

I wonder if it would get better results if we load that non-linear table I shared earlier into curV. Seems like it ought to improve the SNR further.

 

Ehhh.... I don't know. I fought so long with the DAC and I am intimidated by the PSG, but, you're right, I hadn't considered using the hand controllers as the DAC output and I really didn't look at the viterbi encoder very closely. Because I don't use C++ (I took it in college and I am just not comfortable living in it and I don't know what free windows based variant to use) I would have to recode the encoder in FreeBasic to understand it. I also have to take the MSX decoder and then convert it to the Aquarius. Having said that, now that I went through the 1 BIT DAC, I have to imagine that I would be able to get a handle on it pretty quickly.

 

I also wanted to stay focused on a stock system - but if I were willing to build a custom cartridge, and nobody would have that, then I guess I should be willing to use the PSG. We've drifted a bit off topic for this thread.

 

Maybe it is time to bite the bullet and join the rest of the free world and just go ahead and use C++ like everyone else (arg, it will be painful.)

 

Decisions, decisions. 1.) 4096 color 60 fps video player, 2.) work out the display timing to chase the beam and get higher resolution, 3.) turn the PSG into a better DAC, 4.) Port Lode Runner, 5.) convert more movies, 6.) Go back and update Flicker mode converter to super high speed [recently discovered new method], 7.) build a DAC add on, 8.) build a "hard drive" cartridge so we can actually have very large programs [1 meg just isn't enough], 9.) build an ethernet embedded cartridge (for real time data transfer, etc.), 10.) Move on to a different machine, like the Intellivision or the Atari 2600...

 

<sigh> So many ideas...

Link to comment
Share on other sites

A different question: Exactly which PSG variant does the Mini-Expander use? If it uses a variant that still lets you turn the I/O ports to "output mode", then you could just wire the DAC to the hand controller port...

It uses an ordinary AY-3-8910, not an 8914 or an 8917 like the Intellivision and ECS Computer Adaptor (respectively). I haven't tried it myself yet, but I'm pretty sure the hand controller ports can be used in "output mode"; I've seen mention of it in some of the Aquarius documentation.

 

Speaking of the Mini Expander ... I don't think anyone has ever photographed the inside of it before, so here's a quick pic:

 

miniexp.JPG

 

Besides the PSG itself and the cartridge/controller ports, there's pretty much nothing in there except the address bus drivers and PSG glue logic.

Link to comment
Share on other sites

I may need ANOTHER Supercart! I hope I don't exhaust Jay's supply.

Hey, no worries there ... I haven't yet opened half of my SuperCart board packs, so there's plenty to go around.

 

I wish there was a breadboard with an edge connector that I could play with...

So did I, especially when I was designing the SuperCart! I ended up cannibalizing an ordinary 16K cartridge board instead (from a FinForm cartridge, so it was no great loss). Here is my first functional 64K SuperCart prototype:

 

proto.JPG

 

Believe it or not, that mess actually worked the first time I tried it, and still works today. The little switches you see at the top later became solder jumpers in the final design.

Link to comment
Share on other sites

Okay, here's a labeled picture of the SuperCart board which indicates the solder points for the various signals:

 

aqcboard_labeled.jpg

 

A0 through A12 and D0 through D7 are in their usual places in the EPROM footprint, of course, and GND and VCC are available at multiple points on the board. If you wanted to experiment, you could even use a blank SuperCart board as a "breadboard"; it would certainly be easier than my messy prototype!

Link to comment
Share on other sites

A different question: Exactly which PSG variant does the Mini-Expander use? If it uses a variant that still lets you turn the I/O ports to "output mode", then you could just wire the DAC to the hand controller port...

It uses an ordinary AY-3-8910, not an 8914 or an 8917 like the Intellivision and ECS Computer Adaptor (respectively). I haven't tried it myself yet, but I'm pretty sure the hand controller ports can be used in "output mode"; I've seen mention of it in some of the Aquarius documentation.

 

Speaking of the Mini Expander ... I don't think anyone has ever photographed the inside of it before, so here's a quick pic:

 

post-3819-0-00877300-1331689009_thumb.jpg

 

Besides the PSG itself and the cartridge/controller ports, there's pretty much nothing in there except the address bus drivers and PSG glue logic.

 

Then yep, the I/O port can easily be set to output on those.

 

So, chjmartin2: You've got two 8 bit I/O ports. You know what that means, right? STEREO OUTPUT! :D Then the only trick is finding a +5v supply, though I imagine that isn't that difficult.

 

If you just want to try something quick & dirty as a proof of concept, build a simple "R-2R" resistor ladder that plugs into that I/O port. Here's the Wikipedia article: https://en.wikipedia...Resistor_ladder

 

Note that unless you use precision resistors, the DAC won't be very linear and may have monotonicity issues. But for a Q&D proof of concept before you spend $ on a DAC chip (or while you wait for the DAC chip to arrive), it can give reasonably satisfying results easily.

 

You'll want to buffer the output. A 741 opamp would do the job nicely, although those have the disadvantage of needing a wider power supply range. Again, Wikipedia is useful here: https://en.wikipedia...rting_amplifier

 

Or, if you want to try your hand at an emitter-follower circuit, check out this page: http://hyperphysics....ic/emitfol.html You can also google around for more info on how to make these circuits. They tend to be pretty simple, once you get past the scary looking DC bias computation.

Link to comment
Share on other sites

http://www.vdsteenoven.com/aquarius/psgprog.html

 

Sweet! I just have to set the PSG register to Port A and then do an OUT command to send the 8 bits to the controller port. There are 9 pins - is it that 8 are data lines and one is ground? Could it really be that simple?

 

Here's something awesome:

 

main:
   ld hl, ROMLOC			   ; load the ROM location to audio data
   ld bc, SAMPLES		    ; load the sample counter
loop:
 ld (a), hl					    ; grab an 8 bit sample
 OUT 247, 14				 ; set to right output
 OUT 246, a				   ; send the sample
 inc hl							 ; move up the sample
 ld (a), hl					    ; grab an 8 bit sample
 OUT 247, 15				 ; send to left output
 OUT 246, a				   ; send the sample
 inc hl							 ; move to next sample
 dec bc						  ; counter down
; do cool stuff here like a real time oscilliscope
 ld a, b						   ; check for bc =0
 or c
 jp nz, loop				    ; rinse repeat

 

I have so many extra cycles even at 44,100 Hz. (88,100 Out's per second, Out = 11 t-states I am swimming in T-states. .21875 * 10^-6 seconds per T-state. the Out is 11 t-states. 22 t-states for outs, 20 t-states for loads, another 20 for clean up and 10 for the loop = 72 t-states per sample or 15.75 * 10^-6 seconds per stereo sample. I've got 6.9 mircosecond left over per sample, or another 32 t-states per sample. SWEET!

 

If I balance out the playback somehow and unroll the loop then I could easily update the screen every 1,000 samples or so and get nice visual effects.

 

Even better, using this method, you'd be able to really have nice real-time sampled sound effects in a game.

 

Ok - let's start finding the parts. First thing I need are old 9 pin joystick connectors without having to destroy good old stuff...

Link to comment
Share on other sites

http://www.analog.com/static/imported-files/data_sheets/AD7224.pdf

 

All I need is this device. Everything is built in - I've got the DAC and the amp buffering. Two of these puppies wired to the controller outputs, with adequate power (I'll have to figure out how to steal it.)

 

My vision is a little black box with two controller wires coming out of it that plug into the miniexpander, plus I'll have a pass through port (why wouldn't I?) with two plugs for the controllers.

 

Since the DAC will be only using the Output side, the input side will still be available. I'll just wire up some appropriate sized diodes to protect the controllers. (Wouldn't be pretty to send volts out to them I think?)

Link to comment
Share on other sites

Somebody has to make a flash / sd card drive for the Aquarius. I just realized that with a stereo 8 bit dac off the mini-expander that I could write an Amiga MOD player! Well.... only if the samples were smaller than the available memory (48k or so right?) We need more memory...

Link to comment
Share on other sites

http://www.analog.co...eets/AD7224.pdf

 

All I need is this device. Everything is built in - I've got the DAC and the amp buffering. Two of these puppies wired to the controller outputs, with adequate power (I'll have to figure out how to steal it.)

 

You'll also need a charge pumps to get a +15v rail for Vdd and a +10v rail for Vref for single supply operation.

Link to comment
Share on other sites

Somebody has to make a flash / sd card drive for the Aquarius. I just realized that with a stereo 8 bit dac off the mini-expander that I could write an Amiga MOD player! Well.... only if the samples were smaller than the available memory (48k or so right?) We need more memory...

You can have up to 32K of extended RAM in addition to the built-in 4K. You can also put 16K of extra RAM into the cartridge space, but that doesn't leave you anyplace to put your program.

 

Some sort of flash storage cartridge is definitely on my long-term "to-do" list, and the "Memory" slot in the Mini Expander would be perfect for it because it would leave the "Program" slot available for software. I've seen several IDE controllers for other Z80-based computers, and it seems like it should be simple (in principle) to start by adapting one of them for CompactFlash. I'm focused more on getting some games written first than on data-intensive stuff like multimedia, but even for some game projects I have in mind, it would be great to have some sort of persistent storage.

 

The nice thing about the hand controller ports in the Mini Expander is that they are flush with the surface, unlike the ones in the Intellivision's ECS Computer Adaptor. That makes it relatively easy to use regular DB9 female connectors mounted on a board. Another project of mine, an Atari joystick/paddle adapter for the Aquarius, will probably make use of the Mini Expander ports also.

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...