Commander X16

7 minutes ago, BruceMcF said:

Wait, how did you know that I ...



... OH! I get it ...



... make that two.

I'm willing to give it a try, but I suspect I would make three.

Number 4 already making a mess with solder over here ... !

4 minutes ago, James Anders Banks said:

Number 4 already making a mess with solder over here ... !

Five with me, my poor soldering skills will probably make a fried potato instead of a computer !

3 minutes ago, VincentF said:

Five with me, my poor soldering skills will probably make a fried potato instead of a computer !

Mmmm, fried potato!

Don't get me wrong: I've soldered things before.  I've even soldered things within the last three years.

But they were 5-pin VL53L0X units, and a few Raspberry Pi Zero headers.

But... solder an entire X16?  Me?  Fail!

4 hours ago, EMwhite said:

Is there an official or quasi-official place that details the various codename / product names that are being discussed here with a one-line description of what is at play.



For the longest time, forum members have been asked to not talk or ask about release dates or push for features but it appears that the lid has been blown off now and the detail is overwhelming in both good and not-so-good ways.



Between the original proposed list of products and those that I see that appear new, it's a dizzying array.  Forgetting about specific capabilities, a simple list of ranked (least cost to furthest from release/assumed highest cost) would be helpful.



I don't think I'm alone but maybe I am?

I know that references to the Commander X16p, X16c and X16e are scattered through the FAQ, but I don't know which is the announcement or discussions where Perifractic described them more compactly.

Basically, the X16p is the direct end goal of the boards that we have been seeing. All through pin ASIC chips, mostly all of the shelf, except for the Vera board assembly, and one "in stock" old IC, the YM2151 sound chip.

The X16c is the proposed cost-reduced version of the X16p. Still mostly off the shelf ASIC chips, but surface mount where available, for modern automated assembly and oven soldering. The specific features can't be ticked off, because no final feature checklist has been released ... quite possibly, it is not finalized yet ... but it seems like it will have only one slot, and the over 10 or more "glue logic" chips that the CX16p uses similar to the glue logic in late 70s and early 80s designs may be replaced by one Complex Programmable Logic Device (CPLD). The smaller size of the surface mount chips, and possible reduction in addressing logic chip count, means the board can be smaller. And the X16e, the version that has most of the X16 system hardware implemented in an FPGA.

The X8 does not appear in the FAQ (at least, not yet). It seems to have been a "proof of concept" prototype for the X16e, though fitting it into the same FPGA used for the Vera Video/Audio system on the X16 involved some sacrifices and some departures from the X16 system. This is the one that David asks about in the first question, but although it is a completed design, it hasn't been a public part of the current development path, and we have relatively few specifics about it.

2 hours ago, rje said:

I guarantee that at least one person on this forum is a coder who would be terrible at getting an X16 kit working.

 

14 minutes ago, BruceMcF said:

Wait, how did you know that I ...



... OH! I get it ...



... make that two.

 

7 minutes ago, Scott Robison said:

I'm willing to give it a try, but I suspect I would make three.

 

5 minutes ago, James Anders Banks said:

Number 4 already making a mess with solder over here ... !

 

2 minutes ago, VincentF said:

Five with me, my poor soldering skills will probably make a fried potato instead of a computer !

I'm glad to see folks respect the difficulty of the kit, as I recall Dave questioning a year or so ago whether folks would be over-eager to buy a kit and then fail to assemble it -- and need extensive troubleshooting support, assuming they haven't simply toasted various components through their molten metal mis-mangement. As I said earlier, I'm also intimidated. Willing to spend money and try, but still intimidated.

5 minutes ago, StephenHorn said:

 



 



 



 



I'm glad to see folks respect the difficulty of the kit, as I recall Dave questioning a year or so ago whether folks would be over-eager to buy a kit and then fail to assemble it -- and need extensive troubleshooting support, assuming they haven't simply toasted various components through their molten metal mis-mangement. As I said earlier, I'm also intimidated. Willing to spend money and try, but still intimidated.

I'd buy one just to support the project (unless it looked like there was a limited number, I wouldn't want anyone to miss out).

 

Well, I guess I'd have a go at soldering it one night after a few pints!

 

X8 & X16……

if you are of a certain age and from the U.K. this has a feel of BBC micro and Acorn Electron. Everyone who bought an Electron wanted a BBC model B and wished their parents hadn’t got them the cheapo version with missing features. A version that was more expensive to upgrade as well. 

 

my choice would be a DIP DIY version of the full X16

At the risk of throwing a monkey wrench into the discussion... I was thinking (always dangerous) that if one of the issues is to help recoup costs, then why not release the VERA module?  I'd love to have a video interface like the VERA for some other projects.  This would let me connect something like the STM32 to a monitor or TV, or add some cool features to Ben Eater's 6502 kit.  The VERA could become a whole product on it's own with higher priced versions that could go up to 720 or 1080i/p HDMI.  I'd be happy with a version that might still have some bugs in it if there were a simple way to "Field Program" it.  (get it, get it, eh, eh... sorry)  Since it's an FPGA the feature set could evolve over time or allow for some trade offs that might not make sense for the X16 but might for other applications.  Just thinking out loud.

6 hours ago, BruceMcF said:

 



For the core markets for the CX16p and CX16e, there isn't going to be any appreciable cannibalization. For the more fringe markets, as you describe above there might be some, but I think it is less than you are describing above, and as implied by what you describe above, what cannibalization there might be can be avoided by simply launching the crowdfunding campaigns in parallel.

The only problem is that we may have something of an externality akin to the "New Coke Problem".   

The CX16 has been actively touted (and yes, promoted) in the form of its core architecture (banked RAM, VERA working the way VERA docs say VERA works) for quite a long time.   And not just with the glossy mock ups or purely 'paper spec.'   No, there's been detailed docs with a programmer's reference guide for both the system and the VERA with revision histories going back to 2019.  There's a  frequently updated emulator that has had 38 releases (and with r39, we must assume, being close to ready).  The X16 emulator (just from the time this site came into being and started tracking its downloads ( and not counting any other download locations or people who built their own from the repo) has something approach 9000 downloads.   

There's an ecosystem of software going back years.    There are literally HOURS of tutorial videos (and I'm just counting hours of stuff where the creator walked through code that teaches people about VERA according to the current official VERA documents and not the alternative VERA of the X8.   

There are folks who are working on programming languages (e.g., PROG8), compilers, environments, text editors, other dev tools and plenty of demos games.  As I look at the listings, it appears to me that most of those among the top downloads (especially with the vast majority being written in assembly, C, or other non-BASIC langauges etc., ) will currently use RAM banks,  VRAM access through the published data ports specifications, etc.    Heck, even many of the BASIC programs are specific to the X16 and use banks, the YM chip (there's one that demos a hell of a good song and sounds amazing on the X16 emulator), sprites, tiles and PSG stuff. 

Now personally, I didn't have any big problem with New Coke versus Coke "Classic", but market sentiment is not necessarily rational.  There's a potential for  something akin to a sunk opportunity cost value that people will subconsciously assign to their own time that has been spent working up stuff on the X16 platform, and there may be sour grapes if there's release of a drastically altered version and any questions at all on the timeframe for the 'original' conception of the project to see a real release.    There's also the thing that's caused 8BitGuy so much obvious frustration these last many months... the hurdles and pitfalls they've run into getting the X16 protos to a release point without timing issues, board issues, etc., and  that 'uncertainty and doubt' can also have an effect.   I don't blame him, he's had a string of total rocking successes with Planet X series of games, PETSCI Robots and pretty good channel growth.   But those gremlins that the team is having to battle,.. could also cause some question marks about the timing and assuredness of the X16 release even if  (as I agree they should) the Kickstarter goes up tomorrow.    I can't quantify the impact of any of this, and maybe it won't alter things things at all on balance. (After all, geeky enthusiasm is an externality in the other direction). 

Some of these concerns can be at least partially ameliorated with release of some specific details about the X8 nuts and bolts -- especially the specifics in a level of detail similar to what we have for VERA now -- regarding how that 256 byte window works, and what the story is with the address range from $A000 to $BFFF on the X8.    Info on whether there's RAM that can be accessed under the ROMs, for example.   And finally, getting that X8 emulator in people's hands will let those interested take a shot at porting over their existing X16 programs to the X8.   Those smarter than I have mentioned potential upsides to the 256 byte window for VRAM access depending on the setup and details of moving the window, so having the emulator in folks' hands will cancel out the uncertainty factor. 

Bottom line, as a practical matter I'm not sure the 'parallel' crowd-fund launch would really behave like a parallel launch of product tiers unless, on the one side there's video of a functioning X16 with the keyboard and SD card timing issues squashed , AND on the other side, there's been some hands-on access to the emulator and detailed specs for the X8 so that it can be judged according to the same level of immersion the X16 platform enjoys even now at this moment.