Thanks for featuring my game, Kevin!

@openendedd1008

3 жыл бұрын

That looks awesome Matt! Feels kind of familiar in some way... but in a good way :)

@Dangerousdaze

Жыл бұрын

Looks like a lot of fun! I really like the changes you've added (especially the moving fruit!) But why is it every time I've seen it demoed, no one seems to have the first idea of basic Pac Man strategy? :)

Use a cap (0.1, 0.01 uF) across the power button to clean up glitching as a form of low pass filter. Value would depend on the size of the pullup resistor. Looks good so far! Cheers,

I'm so thankful you guys are keeping the YM2151!!

@m00dawg

3 жыл бұрын

Glad to see this too, although the 1099 was still in production (I thought?) but the YM2151 definitely isn't. From a cost contaiment standpoint, having the VERA manage sound, still offered a lovely landscape of sounds (especially if the DPCM was hardware loopable - though I'm not sure if it is - ala the GB, FDS, and TG16 for "drawable" chiptune waveforms). The FAQ says 16 PSG channels. That's a lot! But of course, 16 PSG, 2 DPCM, and 8 FM is quite a monster!

@calags

3 жыл бұрын

Is that a given though? It seems to me that it makes sense to have the YM2151 on an expansion board for people who want it but not make it a requirement. This is in view of when stocks of YM2151s run out and there are no more to be had and would simplify things for phase 2 and 3 of the design - it'll mean one less chip for the integrated designs and VERA ends up being kinda like a 21st century TED or VIC chip.

@m00dawg

3 жыл бұрын

@@calags yep I'm wondering that too. At first the idea of the VERA taking over sound kinda bummed me out but it will still make the X16 unique, the VERA can still be upgraded (albeit with many caveats mentioned in the video) and external cards can be used. Having 16 PSGs and 2 DACs is still handily above any 8-bit home computer I know of. TG16 had 8 I think - all programmable as I recall though - not sure what waveforms the VERA may have here. I think I read on the forums the VERA doesn't have ADSRs either (whereas the YM2151 does), but even so, the base sound seems like it would be quite impressive without the 2151. Cost containment, I thought, was the highest priority and since we have expansion slots (a killer feature I think) - FM sound can still be supported. And given the "room" available on a card, that also opens up YM2151 compatible solutions that might otherwise not fit in the original chip pinouts (notably a YM-on-FPGA solution). Trouble with cards is compatibility for games and things but I'm rather excited to see how that all might shake out. Sort of a weird retro-soundcard-competition-rennaissasnce?

@brucemcfarling7810

3 жыл бұрын

@@calags AFAIU, unlike the SID, the YM2151 is an all-digital chip, and there are already cores implementing it available. AFAIU, that was in fact part of what swung the decision to the YM2151 -- there are a substantial number available, and if the supply runs out, there's a fall back option. Indeed, it's likely that the "3 or 4 chip" CX16e has it in the same CPLD/FPGA replacing all of that glue logic, and it might be the same for the CX16c. Only the most expensive CX16p would necessarily have to have the chip, and that is likely to be the one that sells the fewest units.

@marioalexanderski9598

3 жыл бұрын

26 channels... And 256 color mode on both text and actual graphics, with smooth scrolling, sprites and whatnot. Why isn't Commander X16 considered 16-bit?!

I kind of figured the SAA's days were numbered once the PSG was included in the VERA core.

One way to solve the “J14” problem is to wire in a small 5v PCB-mount reed relay (SPDT), whose contacts replace J14. One side of the coil (and the cathode of a 1N914) goes to pin 3 (Vcc), and the other side of the coil (and the anode of the same 1N914) goes to pin 26 on the printer/user-port. In normal use, the relay will be off, and the N.C. contacts will connect pin 24 to Gnd. If a user-designed device wants Vcc, they just jumper (any of) pins 8,10,12,14,16, or 18 to pin 26, and this will pull in the relay, connecting Vcc through to pin 24. The other option is to put a resettable picofuse in series with the line from J14:2 to J7:24. This way, if someone does put the jumper on J14 from 2 to 3, then plug in a device that connects J7:24 to ground, the fuse will go high-impedance and limit the current. Once the short (or the jumper) is removed, and the picofuse cools down, everything will return to normal. Finally, you could just connect J7:26 to Vcc. A Centronics parallel port is a DB-25, so pin 26 isn’t used, but a developer could still make a board that plugs into the 13x2 male pin header (J7), and get Vcc from pin 26. HTH! 😎

@brucemcfarling7810

3 жыл бұрын

My favorite is actually the last of the three ... with all of those grounds, you can connect to of the DB25 grounds to one ground line and then have a ground and VCC on a separate output ... say a mini-DIN4 or 5 with the ground and VCC where they normally go on common users of that pin count mini-DIN.

To start with the system "off", you need a charge-up delay on the S input to U36A using a capacitor and resistor. Otherwise that chip will power up in unpredictable states. This should also be connected to the 555's reset pin. If adding components is a problem, you already have an RC delay on the CV pin of the 555, so it could also be connected to 555 reset and U36A S.

@TexElec

3 жыл бұрын

Thanks, very helpful. I'm not the best at designing these types of circuits. I must admit, this is where I'd use a microcontroller generally. And honestly, I'm planning to use an ATTINY84. I think I can also drive the reset & power from this IC. It really will be a more stable solution in the long-run. I spoke to David about it, and it's not really part of the core logic and the code will be realy simple to understand if someone wants to see under the hood. I'm even just planning to use the arduino debounce library, so I bet the code won't fill a single screen. Still in the spirit of the system.

@dieSpinnt

3 жыл бұрын

@@TexElec Thanks for the great development work. Just a remark, ATTINY85 is the 8pin one ... for no reason i assume that it may also be 8pin like the NE555. Oh and a good decision. NE555s ( warm memories from the 80ties:) ) aren't known for their deterministic behaviour, hehe.

@theondono

2 жыл бұрын

@@TexElec I wouldn’t jump to a microcontroller right away in any case. There’s simpler chips that will do the trick out of the box (reset monitor/controllers)

Wow that Chase Vault is a lot more polished than the last time I saw it.

One way to limit random trace-frying is to make a trace-fuse. Meaning: make a dedicated section of a trace thinner than the rest. This will ensure which section of the trace that is going to fry, in case of a short. If the thinner section of the trace is fitted with a solder mask, then it is possible to re-solder the thin section of the trace, with a single strand of a wire. So this section will remain the section that will fry, at the next short-circuit condition.

As for your power-on “soft switch”, use a SPDT microswitch; wire the center (common) to +5Vsb, and either side to a pull-down resistor. Connect the N.C. end to the “R” input of an R/S flip-flop, and the N.O. end to the “S” input. Make sure the microswitch is “Break-before-Make”, though! Then hook the “Q” output to the 555 so it provides a single pulse of the desired duration to the PSU. Hook the “Q-bar” output so as to reset the monostable MV (the 555). Of course, the 555 and the TTL chip implementing the R/S flip-flop need to be connected to +5Vsb, also. When the circuit powers up, the N.C. contacts of the microswitch connect +5 (logic high) to the “R (reset)” input of the flip-flop, guaranteeing it starts up with “Q” low, and “Q-bar” high. This also ensures the 555 timer (in monostable mode) is also off (pin 3 low). When you push the button on the microswitch, the pull-down resistor connects “R” logic low. As soon as a pulse from the “N.O.” contacts of sufficient duration is seen, the “S” input of the FF will pull the “Q” output high, which will activate the monostable, and provide your power-on signal to the PSU. Also, you can do away with a lot of the complexity and use a 74LS14 (one section), and a simple resistor/capacitor charging circuit into the input of the inverter, and the output will go to “\PWR_ON”. Wire a 10 uF electrolytic cap and a 47 kilohm resistor (in parallel) from Gnd to the input pin of the inverter. Now, hook the power button in series with a pull-up resistor to +5Vsb, and hook to the input of the inverter section. Now, only when the cap charges up to the point where the “schmidtt-trigger” feature of the 74LS13 inverter activates, will the output go low, turning on the PSU. When you release the power button, the 47k resistor will discharge the 10 uF cap, and the input will go logic low again, and the output will return to logic high. You may have to play with the values of the resistor(s) and the capacitor.

The power on the printer port could be solved by using a PTC in series with the 5V and limit the maximum power/current. The power on problem, is it only a matter of denouncing or something else? If so, I would go for a simple RC circuit and a Schmidt-trigger. Either as a IC or two transistor discrete.

@retrofitter

3 жыл бұрын

PTCs are used on all powered I/O ports on modern motherboards, they are green in colour

@chruder83

3 жыл бұрын

I thought tha same. Why not just use a fuse or a polyfuse to protect against a short circuit. An other option would be to use pin 26 for the 5V. Then there is no compatibility issue with D-Sub or Centronics printer cables. But all of this doesn't help if someone set pin 20 or 22 as an I/O and connects a printer. This could/would still cause some damage. So either way, you need to have some faith in the users. And there will always be someone who still makes some damage, doesn't matter how much effort you put in to prevent this ;-)

Exciting!

ChaseVault is eerily reminiscent of a game I've seen before , but i can't quite put my finger on it 🤣

Can't you simply put in a diode (or some other form of short-protection, like a self-resetting fuse) on the VCC line to User Port pin 24? I would prefer to actually have VCC integrated in the user port connector instead of having to wire it separately

This is so awesome I grew up more with the 386 and 486 and higher era but I would have loved to play with the old commodore 64s and those style of computers and chips. Love retro computers and love their is a community of people that want to bring back that retro feel in a new design that is not just emulation but real hardware.

@Mnnvint

3 жыл бұрын

It's a little bit of each! Chase Vault, for instance, feels a lot like an old shareware DOS game to me, with the FM sound and the richer palette than your typical 8-bit.

Must be so cool to see that running on the real hardware!

Very cool! Do you have at least a loose release date for when the commander will come out this year? I would imagine fall of this year since that would give developers more time to make games for this machine.

@TexElec

3 жыл бұрын

I don't. Honestly, I personally hate committing because I want to make sure it is done. However, we really got over a major hurdle recently, and we don't plan any more major circuit changes so I hope it will around fall of this year. No promises though. :-)

@mattmyers9351

3 жыл бұрын

@@TexElec thanks! I get it, you have a small team of people working on this, and it's a lot of hard work, but also a labor of love. Take your time and make the best product you can make! I love what I'm seeing so far! Good work man!

it would be cool if the commander x16 had a sound chip like what the sid was supposed to be: 32 voices of pure analog synth

@TexElec

2 жыл бұрын

kzread.info/dash/bejne/epmE16d_hrLZZrg.html

7:30 ... why not just use a ptc fuse? Limit to say 2A that way a giant PSU won't be able to pump more through

Why not have the power link as originally planned and add a small fuse?

Use a BJT to buffer one of the printer port output pins; set the pin to high to drive 5v on that pin.

I still don't understand how the /Initialize Output line to the LPT port on the User Port line 6 works. Does it require initializing the CX16 in order to initialize the device attached to the LPT port? In the ECP interface, isn't that the NReverse Request indicating that the bi-directional data lines are inputs into the system rather than output from the system, and the peripheral device reset in the EPP interface?

@TexElec

3 жыл бұрын

I really followed the standard port, as I don't think ECP is possible. Doesn't it require DMA? Maybe EPP, but please, if you have a better suggestion I can certainly change it. That's exactly why I posted it out there. Just kinda making this stuff up as I go along. :-)

@brucemcfarling7810

3 жыл бұрын

@@TexElec On my first question, is that a yes that the /Init out is tied to the reset line of the CX16? It's EPP that I was actually looking at, the ECP pinout just happened to be on the same pinout page. But EPP doesn't use as many pins as LPT does, and the line that connects to PB4 is not used in the EPP. EPP just has one interrupt line input and then the system reads status to find out what the issue is, so the other three status input lines used by LPT are not used by EPP. So a jumper that selected between PB6 and the RESB through the diode on the /Init output on the block header pin 6 would work to allow a system-controlled peripheral /Reset. PB6 being echoed on block header pins 4 and 6 in "EPP mode" wouldn't matter since EPP doesn't use the pin connected to block header pin 4.

2:40 when things are mapped to $01 and $00 what does that mean? aren't those just two separate bytes? do they act like pointers controlled by the bank switching apparatus?

@TexElec

3 жыл бұрын

It's memory location $0000 & $0001. I just used the zeropage reference in 8-bit.

Performing arcwelding on a pcb sounds both fun and scary.

For the power supply protection issue: why not use a polyfuse or other self resetting semiconductor fuse?

@TexElec

3 жыл бұрын

I believe this is the approach I am going to take. I wasn't sure if this would be adequate, but I've been reading into it more and this is what I'm adding to the next proto. Unfortunately, to make things safe, looks like I should add a dummy load. Not welcome news to have to eat around 15watts...

@ReneKnuvers74rk

3 жыл бұрын

@@TexElec I wonder why you would need to add a load? Overvoltage protection maybe, but a load? To tame the switching power supply? That will most likely not be necessary. The multiple output voltages are well regulated with often one being the master and the others derived from that one supply. All loads of the XC16 will eventually accumulate and help the PSU to stabilize. Be sure to use wide traces to the short circuit protection. And some extra caps over each supply. Select caps with a much higher rated voltage as the actual voltage. This helps extending the life span.

@WarrenGarabrandt

3 жыл бұрын

@@ReneKnuvers74rk Most PC power supplies require a minimum load on them to turn on safely, otherwise they become unstable and the ringing can cause high voltage spikes. This is why it's a VERY bad idea to just jumper a power supply on when it's not connected to anything to "test" it.

Why did you move from barrel jack to ATX supply? Just curious, but I think it just makes things more expensive and complicated for little to no benefit. Also, I actually like the drop of the SAA, and I think nobody will want to drop the VERA chip, It’s plenty powerful and I Can’t see a possible better alternative for graphics

@TexElec

3 жыл бұрын

There are pros and cons to any approach. While ATX does complicate the power input a bit, it is very flexible. You can use pretty much any PC case made in the last 20 years, and likewise, you should be able to use about any ATX power supply. I would still look for a lower wattage supply, but I'm shooting for it working with a 500w or less supply. Ultimately, cases are an issue, getting a case molded is seriously in the $100k+ range. We're not antcipating C64 mini numbers, so we have to scale the system to be built within the means of the project.

I used to buy a lot of stuff form you on Ebay im just surprised that you made it here

i really see someone plugin in a 1200W psu and make some arc welding with that... please help these people

Why do you need to use TTL chips for a part of the machine that isn't (and shouldn't be, probably) hackable?

What about a reset supervisor IC for the ATX power on? If you use one of those, you don't have the complications of a CPLD or microcontroller or violate the "no controller faster than the CPU" issue.

@TexElec

3 жыл бұрын

I'm actualy sort of cheating and using a reset IC. The Dallas DS1233A. It's the debouncing which isn't as stable as I like. No through-hole debouncer ICs I can find.

@PHamster

3 жыл бұрын

@@TexElec It’s not cheating persay. Instead of using resistors, comparitors and open collector. :). Couldn’t the debounced be solved with a capacitor and a Schmitt trigger like the TI SNx414?

@KonjonoAwesome

3 жыл бұрын

@@TexElec According to the DS1233A datasheet, it does switch debounce. Is that not working for you properly?

@TexElec

3 жыл бұрын

​@@KonjonoAwesome I suppose I could... But, I'm pretty commited to the microcontroller now. I still need a bit more monitoring capability and I sort of have one more plan now. :-)

@KonjonoAwesome

3 жыл бұрын

@@TexElec I totally get it. I think the microcontroller is the better option myself.

I'm a bit confused about the power supply issues. Coming from 8 bit ataris all thats needed there is 5v dc, which pretty much any usb wallwart rated for 2.5 amps would be able to do these days. So basically like a Pi. Surely the solution wouldn't need to be much more complicated than that here with some additional switch? Sounds a little overkill to actually use atx supplies.

@TexElec

3 жыл бұрын

It's an ATX form factor board meant to go in an ATX case, so it stands to reason the best supply to use would be an ATX power supply. Honestly, I've done a bit more research, and I feel like it's going to work out fine.

I was looking at the emu code the other day... i noticed how much it flips banks when at the basic prompt... result of the keybd check code being in a different bank than the basic code i guess... right!?? .... also getting ready to test dual video... will see.

@TexElec

3 жыл бұрын

That's definitely a question for Michael Steil, but you are correct, I know it's banking something when in BASIC, I've seen it on the analyzer many times, and I wasn't able to get the system to come up until I worked out the banking. Fortunately, it is more simple now and should be pretty quick. Michael has been able to do some pretty cool stuff with banking and there is now 512k which will leave plenty of space to add other code. The system ROM is really flash, so it can be written to. I put a jumper on the write line (don't want bricking programs out there!) and it will likely be off by default. We also have yet to test the write routine, it requires a preamble sequence to write, but I suspect it will work fine.

@Alperic27

3 жыл бұрын

@@TexElec banking is a trivial thing to do once u get the right addr decode logic in place (disclaimer: which i did screw up a number of times...at the start). actually, i was wondering about the ROM, so u just answered my question (today i bought some flash to replace the eeprom in my current proto). I will also soon be able to test vera.. but more on that later, if you are interested. you guys are doing an awesome job, and I am thankful for inspiring it has been in these otherwise dreary times... the kid i was with my apple iie never thought that one day i would make one so easily, thank to what you guys are doing. 🙏🏻

I was expecting to see how the 16 bit CPU was dropped back to a 6502. I guess that was already decided previously? I'm curious to what reasoning was used.

@TexElec

2 жыл бұрын

Our kernal developer is not a fan of the 816 and keeping full compatibility with the 6502 was deemed more important. I did however setup the motherboard to where a 816 could be installed, but there will likely need to be kernal mods and there will be no easy way to access beyond 64k.

@puzzud

2 жыл бұрын

@@TexElec So what doesn't make this project the CX8 at the moment? The bus? The VDP?

Whatever they end up doing with the power button, I hope they keep it like the C64 in that you can turn the system off at any time without having to worry about a "proper" shutdown* (one of things I don't like about the raspberry pi). * Obviously taking care to not turn the system off when it's writing data to the SD card or other storage devices.

@amarioguy

3 жыл бұрын

The only reason “proper shutdowns” would matter is if a cold shutdown can result in data corruption, something that only matters when dealing with multi user, incredibly complex operating systems This is so much simpler than that, and as a result acpi shutdowns shouldn’t be necessary nor would they even work.

@JustWasted3HoursHere

3 жыл бұрын

@@amarioguy Awesome. Sure was nice to have a computer that can "boot up" in about 1 second and be shut off at any time (except when writing data to a device).

can make it so can power the board from battery´s

FTDI sells usb controllers in 28-TSSOP packaging. If I happened to make an expansion card, then you'd have kb/mouse/sd control. Maybe ethernet along there if it's not in already.

@WarrenGarabrandt

3 жыл бұрын

Adding the physical port would be trivial compared to the complexity of implementing the device drivers in 6502 assembly, I think.

The VERA gets a little too many SYSTEM functions, instead of being a tool aka "VGA mod" just for graphics, imo.

J14 should be left in place since this is supposed to be an enthusiasts computer. If you really must exert such control over everyone else you could just solder a wire link on the board instead of the header, owners could remove that link and add the header if they wanted to. Have you considered making it a closed system and then encapsulating it in potting compound?

@TexElec

Жыл бұрын

The design has changed a bit since this video, and I now added +5v to pin 26. This will not expose it on the LPT port, but it will expose it on a 26-pin ribbon cable. We have had interest in the past from schools and other technical institutions who want to use this system as a learning aid. With that it mind, some level of safety must be considered in the design. I don’t think I’ve risen to the level of potting compound; this system is VERY open and there will be no mystery.

@johnafish

Жыл бұрын

@@TexElec Good to know it's still progressing.

Oooh, how about *several* USB cables for power? XD

arrow max1000 fpga (max10 with 8kLE) is very cheap, around 20 bucks, usb/gpio/memory 8mbytes

@Jkauppa

2 жыл бұрын

palette animation, assuming palette has 256 entries, for around 60x40 screen sizes, for fast "rendering"

It seems like the game you played was just Pac-Man with different levels

ARMSID wasn't an option? Maybe too pricy, right? - ...asking for a friend. ^_^ xD

Interesting that the X16 is a 8-Bit Computer but for me it is like a 16-Bit Computer. Edit: The PC Engine/Turbografx-16 is marketed as a 16-Bit Console but in there is a 8-Bit CPU.

@shamringo7438

3 жыл бұрын

That is because the 6502 has a 16 bit address bus. It still handles data in 8 bit. That's why it's called an 8 bit CPU. The PC engine/Turbografx16 being marketed as 16 bit console is because its graphics processor (Huc6270) is 16 bit.

sad thay you dropped SAA1099 :( chipstunes really make retrocomputer :( now it sounds like PC on 90's

Just go MCU. You can buy them quite cheaply already programmed with whatever you want for a very small fee. The dinosaurs are taking the "must be ttl" a tad too far. It will NEVER be anything like what they used back in the day any way so it's just silly to go overkill

@PHamster

3 жыл бұрын

I don’t understand the need for “TTL”. Is the need for it the current drive, the “low” logic high at 2.4V vs 75% of VDD.

@TexElec

3 жыл бұрын

@@PHamster It's David's requirement. He wants no CPLDs, etc. Also no chip more powerful than the CPU, IE, no microcontrollers. I may have to talk him out of it because I really don't think I'm ever going to be happy with driving this the way it is. I could also drive the reset with the microcontroller and save a few cents more on parts. I can use CMOS or whatever, I just meant it needs to be traditional parts.

@Willborb

3 жыл бұрын

@@TexElec Compromising the safety of the rest of the system's power regulation, simply to adhere to TTL sounds flawed. It makes sense in the logic side of things, but not for the board supply. Designing a compromised circuit that could have critical flaws in situations you guys didn't plan for is worse then adding a single IC that will "do it safely, but isn't TTL."

@ScottWalde

3 жыл бұрын

Could you convince him to let you use something like an ATTiny85? They might be close, but arguably not more powerful than the CPU. (And available in DIP-8) If accessibility to modification is part of the requirement, you could even use the Arduino environment to program it... barrier to entry is pretty low then. I do have a soft power switch circuit that works that I'd be happy to share if you really want, but I'd have to install Eagle on a computer to take a look. On the "power" issue... I'm not sure I understand the problem. If you hook up a 900W power supply, your circuit isn't going to magically start drawing more power. If you're worried about someone doing something dumb and then melting traces, just add a properly sized polyfuse where the power enters your board. Again, they're easily available as through hole parts.

@PHamster

3 жыл бұрын

@@ScottWalde I wonder if they’re dealing with in-rush from all the capacitors used? The in-rush of the PICO ATX Supply won't be the same amount of as a 900W Corsair PSU. Since it's suppose to be simple with "TTL" mentality, the polyfuse is your best bet.

Why a barrell jack? You're already using an ATX power supply, you have a bunch of molex connectors inside that can be passed through if need be. Also, this is entirely opinion but I absolutely hate the FM sound.

@brucemcfarling7810

3 жыл бұрын

If you are asking why they are using a barrel jack with an ATX power supply, did you miss that they said the barrel jack was what they USED to have, before they transitioned to an ATX power supply?

@papeleradereciclaje4375

3 жыл бұрын

"I absolutely hate the FM sound" I did too, but nowadays I can appreciate it more. A lot of games did not make good use of it and ended up sounding like robot farts. But, there were also a good a amount of games that sounded extremely good FM music, especially in the arcades and computers like the PC-98 and the Sharp 68k. There were also some Adlib games that had good music, like Dune, MegaRace and KGB/Conspiracy. Let's not forget Tim Follin's work on Time Trax for the Sega Genesis, which is absolutely outstanding.

@Mnnvint

3 жыл бұрын

@@papeleradereciclaje4375 Yeah, I think a lot of DOS era adlib music was just composed as midi, on a system with much better midi instruments than the adlib defaults. But if it's a composer who knows what they're doing, and treats it as an instrument in its own right, it can be great. I'd add the very first game with FM sound, Marble Madness, to the list of gorgeous FM tracks.