Commander X16

8 hours ago, Scott Robison said:

And until the X16 has native internet access, an app store would be limited to sneaker net anyway. I'd love to see the popularity surge enough to justify someone taking that task on (I won't be doing it myself) or someone doing it and thus causing a popularity surge. I would happily admit I was wrong. The purpose of the app store is to make download and install of applications easy. I'm not saying an app store couldn't work for X16, but it would have several strikes going against it from the get go with what is currently known about the design.

If we adopt Bruce's ideas about SPI expansions or something similar, you could produce an ultracheap wifi interface using an ESP32 (don't know the other ESP chip). All you'd have to do would be to wire the ESP32 to the SPI expansion and write a bit of code in it to connect and use it as a bridge.

I think the current Downloads with decent games and utilities and so on would help the hardware in that it would encourage more people to purchase and develop for it, especially if it is cheap. The advantage of the X8 idea, whatever else one may think, is that it's beer money almost, whereas a built X16, Mega65, Foenix256, whatever their merits as platforms, aren't. Even the backwards compatibility doesn't really help. If you actually just want to play C64 or Spectrum games, it's cheaper and easier to use an emulator, or a Mister than the Next or Mega65.

I think this was the rationale behind 8-Bit-Dave's original idea of selling it for £30. It wasn't about the money from the platform so much as widening the uptake.

People do make games for the Next that come in proper boxes (rather like PETSCII robots and so on) and people buy them. I suspect it's not a huge amount, but presumably it's enough to make it worthwhile as opposed to just releasing it as PD.

I think some of y'all misunderstood the thrust of my argument re app store. Actually, I am the one who used that term so it is probably my fault. When I said "app store" I really just meant "a section of the current download page on this website that has content you pay for vs content that us free".

Also someone upstream responded to my last stating that in their opinion a few thousand units sold and a few dozen offerings in the downloads section would be a rousing success. Given my understanding as to the purpose of the project I respectfully disagree.

My understanding of Dave's (ie nortorious 8BG) vision is that he wants a spiritual successor to the 8bits he and a lot of us came up on. That from the spiral-bound lucid manual to the ability to grok and directly access the hardware, the goal is to have the experience of computing in that way is facilitated by the X.

If I am right about that, then there is something else those beloved 8bits have that the X does not and that's developers. Think about it - you may have learned to code on a c64 or atari or whatever, but I will bet you good money that is not the only reason you have such fond memories of the machine. I bet you ALSO have a lot of happy memories gaming, participating in the demo scene, BBSing, etc. In other words there was a scene at that time and that place and the computer wasn't the thing that made it happen- the computer was the thing that got you to the thing.

The point is what matters is not the mechanics of the app store or even if the developers have a way to get paid. What matters is the X will not live up to 8BG's vision if a critical mass of developers doesn't form making funky !@#$ for the platform. Creating the experience of that era of computing us about more than the hardware - it's also about the developers who make the hardware shine.

I could be wrong though. Maybe y'all are on the trajectory you want to be on and are perfectly happy with an end result wherein a very small number of people have an X and an even smaller number of people are actively using it. That's ok too. For what it's worth, I deeply respect the talent and effort that the team has out into this. It is actually that respect driving my words. If i am right about 8BG et al's intentions then it would be tragic if all that passion and hard work resulted in falling short at the finish line.

Regardless, if you guys come out with anything I do not have to solder myself (my hands shake too much not ti make a mess of things) I will buy one. You have won my admiration and earned some of my money ?

2 hours ago, paulscottrobson said:

If we adopt Bruce's ideas about SPI expansions or something similar, you could produce an ultracheap wifi interface using an ESP32 (don't know the other ESP chip). All you'd have to do would be to wire the ESP32 to the SPI expansion and write a bit of code in it to connect and use it as a bridge.



I think the current Downloads with decent games and utilities and so on would help the hardware in that it would encourage more people to purchase and develop for it, especially if it is cheap. The advantage of the X8 idea, whatever else one may think, is that it's beer money almost, whereas a built X16, Mega65, Foenix256, whatever their merits as platforms, aren't. Even the backwards compatibility doesn't really help. If you actually just want to play C64 or Spectrum games, it's cheaper and easier to use an emulator, or a Mister than the Next or Mega65.



I think this was the rationale behind 8-Bit-Dave's original idea of selling it for £30. It wasn't about the money from the platform so much as widening the uptake.



People do make games for the Next that come in proper boxes (rather like PETSCII robots and so on) and people buy them. I suspect it's not a huge amount, but presumably it's enough to make it worthwhile as opposed to just releasing it as PD.

I will be doing just these things once there is hardware. I've already done some basic connectivity experiments using two Raspberry Pis connected together, and I have worked with ESP32/NodeMCU in the past as a connectivity tool.

For the record, I strongly support using NodeMCU, as NodeMCU lets you do exactly what you want, and you can always write a simple AT emulator in Lua for legacy terminal programs. 

My plan is to write software for microcontrollers: Arduino, ESP32/NodeMCU, and Rasberry Pi, and let users roll their own cables to connect their favorite dev board to the User port on the Commander. 

 

All this talk about an app store, software development leading sales,  etc., all that...       It  seems almost chimerical.    Especially  in this thread, since the whole point is that the core architectural details are potentially in transition or at least may include an additional platform.  

 

It's a very difficult topic to raise. Dave was right to do so I think, but it can cause people to wonder what hardware is actually coming down the pipe. Which is a shame, because AFAICS it's really about how you get data into the VRAM.

Raises a question. The X8 uses a RAM Window, the X16 the pipe. Is there a reason the X8 couldn't use the port as well as the window ? Does it ? If it does the only issue is the amount of VRAM (and the extra SRAM). It wouldn't be difficult to write code that could use 16 colour graphics or 256 colour graphics accordingly, and recompile (and dither ?) on the gfx conversion. I usually use Pillow for Python which I think does it for you.

I can see a dual port VRAM issue perhaps (?) FPGAs not my thing really, but you could have the use of one disable the other if that was a problem.

4 hours ago, paulscottrobson said:

If we adopt Bruce's ideas about SPI expansions or something similar, you could produce an ultracheap wifi interface using an ESP32 (don't know the other ESP chip). All you'd have to do would be to wire the ESP32 to the SPI expansion and write a bit of code in it to connect and use it as a bridge.

I can see this as a super cheap way to add Internet connectivity, to support software for simple BBS style communities, or even simple FTP and HTTP type features.  I made a suggestion some time back for this kind of thing, but people seemed mostly uninterested.

5 minutes ago, paulscottrobson said:

It's a very difficult topic to raise. Dave was right to do so I think, but it can cause people to wonder what hardware is actually coming down the pipe. Which is a shame, because AFAICS it's really about how you get data into the VRAM.



Raises a question. The X8 uses a RAM Window, the X16 the pipe. Is there a reason the X8 couldn't use the port as well as the window ? Does it ? If it does the only issue is the amount of VRAM (and the extra SRAM). It wouldn't be difficult to write code that could use 16 colour graphics or 256 colour graphics accordingly, and recompile (and dither ?) on the gfx conversion. I usually use Pillow for Python which I think does it for you.



I can see a dual port VRAM issue perhaps (?) FPGAs not my thing really, but you could have the use of one disable the other if that was a problem.

I feel like you're going to get contention problems if you try to do both... it really has to be one or the other, partly because the FPGA resources needed to share the RAM between the CPU and GPU are limited. 

In fact, part of the attraction of the C8 is the improved clock speed and performance. I mostly tool around in BASIC, so the 512K/2M extended RAM gets me no benefit (it's literally faster to read from disk than translate data into banked memory with POKEs), and so the C8 is actually a better BASIC computer than the Commander X16. 

Of course, I submitted a bunch of ideas for improving BASIC to deal with banked memory (a BANK command, along with special PEEK and POKE commands for floats, 16-bit integers, and strings), and if those make their way in to the final product, then the CX16 becomes much more attractive for BASIC.

I've watched the CX16 slowly become less attractive over time, with features disappearing: the clock speed got lower, the UART went away, and the BASIC interpreter still has some really stupid shortcomings. I actually welcome the C8 specifically because it's actually faster, apparently uses a boostrapped OS (meaning I can fix BASIC myself), and it has a UART I should be able to tap into.

 

I'm software developer so less hardware, the better. That is why FPGA seems much closer to perfection - emulator ?

Now on topic X16 vs X8. TBH X8 seems much better from software perspective  - 12Mhz, a window in VRAM (no registers), USB. So

On 8/20/2021 at 4:00 PM, ppescher said:

1. 
   Should we release the Commander X8?
   
   Yes, but only if it's made compatible with the X16 from a programmer's perspective.
   

But that compatibility to be reversed - x16 to be compatible with x8 - window in VERA, 12mhz, USB.



Wishful thinking:

The best will be x16 with 65816 that directly address 24bits of address space (no need for windows in RAM/ROM), VERA with window (as X8) or even directly mapped in address space (it will take only 2 banks), RGB to be at least 666 why not even 888, more VRAM so actually 640x480 could be used (now is only a feature on a paper)

6 hours ago, paulscottrobson said:

If we adopt Bruce's ideas about SPI expansions or something similar, you could produce an ultracheap wifi interface using an ESP32 (don't know the other ESP chip). All you'd have to do would be to wire the ESP32 to the SPI expansion and write a bit of code in it to connect and use it as a bridge.



I think the current Downloads with decent games and utilities and so on would help the hardware in that it would encourage more people to purchase and develop for it, especially if it is cheap. The advantage of the X8 idea, whatever else one may think, is that it's beer money almost, whereas a built X16, Mega65, Foenix256, whatever their merits as platforms, aren't. Even the backwards compatibility doesn't really help. If you actually just want to play C64 or Spectrum games, it's cheaper and easier to use an emulator, or a Mister than the Next or Mega65.



I think this was the rationale behind 8-Bit-Dave's original idea of selling it for £30. It wasn't about the money from the platform so much as widening the uptake.



People do make games for the Next that come in proper boxes (rather like PETSCII robots and so on) and people buy them. I suspect it's not a huge amount, but presumably it's enough to make it worthwhile as opposed to just releasing it as PD.

Agreed. It's just that the group of people most likely to find an app store appealing will be the least likely (or so I think) to go out of their way to enhance their hardware to be able to utilize it.

1 hour ago, Squall_FF8 said:

I'm software developer so less hardware, the better. That is why FPGA seems much closer to perfection - emulator ?



Now on topic X16 vs X8. TBH X8 seems much better from software perspective  - 12Mhz, a window in VRAM (no registers), USB. So



But that compatibility to be reversed - x16 to be compatible with x8 - window in VERA, 12mhz, USB.





Wishful thinking:

The best will be x16 with 65816 that directly address 24bits of address space (no need for windows in RAM/ROM), VERA with window (as X8) or even directly mapped in address space (it will take only 2 banks), RGB to be at least 666 why not even 888, more VRAM so actually 640x480 could be used (now is only a feature on a paper)

The nature of the VERA hardware is what prevents the window into RAM in X16 (regardless of CPU model or speed). X8 puts a soft CPU core and VERA on the same FPGA so they can share resources internally without having to use limited physical IO pins. Thus why the design goals of x16 (separate components) are incompatible. It would be possible if there were more pins on the VERA that would allow decoding of more address lines, but that would drive up the price of VERA, and people are already showing resistance to most things that would involve an increased price on the hardware.