Once the new model of MSX comes out...

Hi MSX friends, I normally only read here but I feel the urge to share my thoughts...it's quit some text..sorry :-)

Over the years there where lots of initiatives to create a new MSX compatible computer. But all these initiatives are either dead, not sucessfull or will never be very successful and my fear is this one will share the same fate.

if a new initiative pops up, there is always (of course) a discussion about the specs. Some just look in the past and want just a replacement for their 30 year old MSX maybe with SCC, MSX Music + Audio, OPL 4 etc...and others want at least a Zillion MHz MSX. Requirements regarding specs are important but very narrow minded (nerd perspective). No offense meant!

If we want to rejuvenate or resurect our beloved MSX platform (I have the impression that most of us would like this, or not?), even in a niche market, we should learn from the past.

There is obvious a lot of hard and software talent available in the scene and they are willing to put a lot of effort in creating new things but all attempts to create a new computer had or have the wrong starting point:

They never determine what the market proposition should be and there is no plan other then creating a piece of hardware!!
So basicly they don't look at the 'big picture'.

Some points to think about:

- What would be a possible and realistic (niche) market proposition? Look for example at V-tech or Arduino, raspberry PI, etc.. they created succesfully a new educational 'platform' in the crowded IT market.
- What would be success factors for the chosen (educational) proposition? Example (from my perspective :-) ):
* New cheap basic MSX hardware platform with high(er) performance, USB & network. (not perse a zillion Mhz but
powerful enough for some' modern' software development). SCC and other extensions are not relevant at this stage.
* A ' new' standardized OS (SymbOS for example)
* Support for 1-2 modern and easy Software Development languages. (like LUA (any idea how many apps & games
created in LUA are released each day? ) a new MSX platform could benefit from the vast amount of software
releases and would also be interesting for hobby software developers (please have a look at Coronalabs.com for
example they have a big home-brew scene)
* Digital Software distribution system (app store like..could be less slick). Advantage: Developers are able to
distribute their software cheap and easy and consumers have a lower threshold to buy software.

I can elaborate more but I think you get the point.
It would be fantastic if a scene wide initiative would arise.
This scenario is possible if only the scene would unite and eventually start a kickstart or indiegogo project for the needed cash. :-)

Just my 2 cents..(maybe a few euro )

Ivke

Ps. Of course it's not my intention to offend anyone. All respect to the MSX diehards who are working hard to create new MSX hardware and/or software!

zPasi, I understand what you mean, but fragmentation will always occur when someone creates a piece of software for a particular hardware.
It happens on any retro-compatible new generation console (gb to gb color, gb color to gba, gba to ds, ds to 3ds, 3ds to new3ds... Same with play station 1 and 2, and many others.
It also occurs on the MSX every time someone composes a song for moonsound only, or creates a demo for 2+ or turbo-r, or even for gfx9000.
It also happens when a game has only FM musics, and there are many msx owners who do not have an fm pac...
This fragmentation can also occur if I finally manage to convince a hardware genious to develope new gfx9000, as happened with the opl4 being manufactured today.
The initial fragmentation is unavoidable, but it must be redirected and educated as an incentive to adquiere this new hardware, and once enough (if not all) users have this hardware, it finally becomes an standard.
To put this optional hardware in a new msx prototype (so you can use it or not) is not bad if enough units are sold.
The more units sold, the more developers will risk creating new software for this new hardware, and new software incentivates the purchase of said hardware, so it is a snowball, but the snowball must gain enough initial momentum to start growing.
Anyway, if I ever find myself in the situation of creating an exclusive software for a particular hardware, I would also make a retro-compatible version for the old standard. That would be a good option to avoid fragmentation and also increase sales...
There are many options. We just have to explore them.

Like other old systems, a big part of their charm is in its limitations. Working / playing with, and coding within those limitations. Like for ZX81, Spectrum, C64, Apple II, Atari 2600 etc.

I wouldn't mind an MSX with some limitations removed. Say a high-clocked CPU. Or CPU -> VDP transfer speeds upped. Extreme-speed VDP blitter. Many more sprites on a screen line. Some extra screen modes. Maybe a DMA engine, or some memory -> audio device 'co-processor'. Some re-arranging of the MSX'es complex slot mechanism into an easier to use 'flat' memory space. Things along the lines of ZX Spectrum ULA+ efforts: enhanced features, but still compatible to existing software.

But as soon as you bring in entirely new IC's, and try to build a 'standard' with that, you get a new machine. And then you have to wonder what place that machine is supposed to take among existing gear.

And that's where ivke2006 hits the nail on the head. Low-end Arduino boards can be had for what, $3~4 a pop? And you have the RPi 2, a quad core 32-bit cpu, built in the millions with a large community & lots of software. You know how many 'RPi-wannabee' single board computers are out there today? Just a few years back perhaps a dozen. Today: many dozens. RPi-sized 64-bit ARM boards coming up around the $50 mark or less. In a few years from now: hundreds of boards like that. What purpose ON EARTH would an über-MSX serve among all that? NONE. No matter how nice a machine it would be, or how well we'd know / love / program it.

A modern replacement for our aging MSX machines: yes please! Pick up a few enhancements along the way: why not! (if done cheap enough). But outside of that: dead on arrival.

later, this message of "if I'm crazy," I'll be a little more serious
MSX is true that the community is small, but so is that is very committed to the machine.
My idea for a MSX future ?, which was developed with the MSXVR a hard to "virtualize" all the capabilities of our system, but maintaining the "feeling of using a real MSX"
with respect to the processor, a Z80 (or compatible) is cheap nowadays, but back to get the stick in the wheel ... can put a "turbo z80" to 20MHz for example, but can switch to 3.57 in any moment.
separate issue is the "cpu century", if it is possible to maintain compatibility would make the curve progression was much nicer, but if it is not and high-level languages ​​are usasen, the progression may be even better.
RAM? minimum 4MB, but at least (in fact there is proof of a virtual Nestor Soriano MSX with 52MB !!!), if we take our limitations, one of the largest is the amount of RAM. (And we must recognize that today is almost more expensive 1MB to 1GB)
VDP? the 9958 is mandatory, but if possible than "the old 9958" a v9958 with 4MB open many doors for the use of the system as a primary platform * (I would explain this later)
the V9990? also if it is virtual or extended VRAM, but it would be even more interesting a "new" concept, as the V9978 being developed for MSXVR, although I think that is already developing a new specification must be ambitious (for example a VRAM addressing above 16MB) *

in sound no problems since there is almost unanimous: PSG + MUSIC (AUDIO close to) the OPL4? It's a great idea, but it would have to consider whether such a complex chip would cost to put (besides that could keep making runs)

SD storage ?, no doubt at least, with the possibility of external USB storage

and of course a powerful BASIC (but with certain limitations), and controllable (I do not want to imagine a basic list 1MB)

If quedase site and budget, a separate slot for future expansion, it would be nice ..
and of course ... quiet, no fans to make noise.
* Primary platform:
My idea for a new MSX is to replace as far as possible the PC, if a new MSX can watch Netflix, navigate albeit crudely, see and answer my emails, the truth is that only go on the PC for some video editing, on work, or some PC games (sorry Starcraft II love me)

I'm sorry I was so heavy

I think, what MSX was in the '80's, it was mainly a games console, which with you could also code and do some other "computing". For the high-end models, there was even professional use, but we've got other equipment for that now.

Nowadays we have more powerful games consoles, but usually they are one-purpose devices. You can't code with them. So, in my mind, the new MSX could be a retro-games console, but open for coding and hacking. It should be able to run classic MSX games, MSX-DOS, BASIC and all that familiar stuff. It could also have a "new mode" that gives more power and everything, but that should come cheap so the console could sell thousands. If it only would sell 100 or so, it just would fragment the MSX base.

If I should design such a device, it would contain three main parts:

1. An FPGA
2. A small ARM-based Linux computer, probably RasPI
3. The familiar MSX connectors (like module slot) and then some (USB etc)

The FPGA would serve as MSX hardware (with Z80, mappers and everything). The ARM computer would be the *new* turbo mode. Four cores, OpenGL ES, a lot to play with. The rest is just software, open source. Most exists already.

I know the RasPI alone is powerful enough for the whole thing, but to make sure all the MSX hardware would be implemented without compromises the FPGA would also be there. It doesn't cost much anyway. And if you still want gfx9000, OPL4, whatever, it all could be arranged, no problem.

Seriously! Wouldn't that be Uber enough?