TNI announces 萌ＳＸ

WOW ... looks like a giant project ...

Interesting... (But) how about a discussion forum/platform for this? Or a collab environment by means of e.g. Phabricator?

Other possible way is to define a sequence of opcodes that switches the processor between "Legacy" and "Native" modes. This is how Hitachi did on 6309, Intel did on x68 family (when changing from Real mode) and lately AMD did on x68-64 (32-64 bit modes)

A small nitpick about this newpost: They are looking for developers/support in general (for any generation MSX/different types of projects), not for this new system exclusively.

You need this: https://www.youtube.com/watch?v=Vhh_GeBPOhs

As in: people who create content. So, question is: what would such people actually want? What can they manage? You can give a Cray supercomputer to the FlierpyFlierpy Indian tribe from the rainforest, and that'll make as much sense as giving a modern PC to someone who prefers to make MSX1 games (and not MSX2, 2+, tR, G9k, Moonsound, 1cM). See, there's a whole flock o' people who like primitive sprites, tiles and noises. For them it's retro, style, and it reminds them of when they were young and got their first MSX. You could add all kinda of videochips, audio, 32 bit processors and heaven knows what, but eventually you'll end up with a PC. I wonder whether these people want that. Maybe they do, I simply wouldn't know.

So, what would people want to do with such a system? Games? Possibly. Demos? Possibly. In such cases (full-screen games/demos) you don't need multi-tasking anyway. But, word processors? E-Mail tools? Spreadsheets? Databases? Internet browsers? I dunno, probably not.

Something to compare it with. I wonder, how many people use SymbOS? I mean, I'll be the first one to say it's the best OS/GUI-system on MSX ever, and truly a milestone for an 8-bit system, engineered by a single person. But how do people use it? If not, you end up with something done merely to prove it can be done, and little more than that.

If anyone's about to make a new MSX, my tips are:

• A range of Z80 CPUs, so that you can dedicate audio, gfx etc. to them. See, our current ultimate chips (G9k, Moonsound) are just a tad more meaty in their possibilities, so they could need more CPU to get every last drop of potential from it.
• An interrupt generator that ticks consistently anywhere on this planet. No tempo crap with music @ 50/60 Hz anymore.
• What annoys me is the PSG vs SCC/FM balance that's a bit off with certain models, so have that fixed too.
• Playing music during disk loading.

See, it's not even all that much, but these have been kind of the 'bugs' in the MSX.

note:
The TNIASM by 0.x version does not support macros!
In this case, I do not really care about this compiler for MSX.
I greeting and SJASM is supported with macro.

that is because 0.x is no longer being developed.

TNIASM v1.0 support macros and its pretty good

https://www.msx.org/forum/msx-talk/development/looking-for-t...

Also this news post is not about tniASM, its about a new msx standard

Do tell! Really, what do people do with it? (this is a serious question)

Regardless of its Win95-looks, in use it reminds me of my Windows 3.11 days. It just sat there on my HD while everything I ever used back then (486DX2 days) was DOS, and Norton Commander was my file manager. To me it was a tad silly to first boot Win3.11 from DOS, only to click a DOS-icon and go to DOS again. So, Win3.11 ended up largely unused.

While you may think this belongs into another thread, it does belong here because of its relation to multi-tasking. Typically, someone uses an MSX for one task at a time. A game, or a demo. So, you'd have to wonder whether you need an OS that supports multi-tasking. So, again: prior to designing all this new stuff, perhaps one should ask MSX-users what they intend to use it for.

I agree on the faster CPU (or multiple cores/co-processors), the 9958/9990 combi. The BASIC would be nice for the BASIC fanatics, although fast could refer to things being compiled, or to the way it works (or perhaps both). If the line numbers could be wiped, block-constructions added, local variables, and the whole thing running with a proper IDE, then it'd be something.

switching tasks without the help of the app itself (like having a render event) is not possible, but the hardware can has "surfaces", where all the vdp i/o impacts, so virtualizing the memory and registers pool for the VDP into a major memory bank, the MSXBIOS can switch the vdp to show one bank or another, in real msx it is not possible because we haven't access to switch the vdp's register pool. But it is doable.

Ofcourse also is the software solution with extra feature help, like do the vdp registers really readable, and let the faster cpu backup/restore all the vram and register bank to the ram. Which is the same effect.

But, it is not just about VDPs, it is about sound, cartridges, everything, it will be hard to implement it, except if any single chip that we use often is virtualized in the circuit behaviour description (what coding in VHDL is , describing the behaviour of the machine, then the compiler create the related circuit from that behaviour description).

So, if everything fits in some large fpga or serveral of them, no needs to plugs carts, except for .... ROMs games and file I/O.

Just when switching from one task to another it will glitch the video/sound just a click, but that is a minor thing, todays monitors resync very fast the vertical, it will be like changing the channel in a TV.

I do I use it almost daily And i love to support it! The multitasking is very usefully. Especially when it becomes to networking support. Next to that it is very easy to create a application for it. It's like rapid development. In a couple of hours you can make a chat program which is fully working. That's just crazy.

I mean the code to send something is just like this:

Function SendSomething(mystring)
	If mystring>"" Then 
		Net.TCP.Send(mystring + "|")
	EndIf

EndF

At this stage it is not. But when the graphic library is done for SymbOS it would be possible to make fullscreen games or programs. Multi-tasking is not only about running 2 applications but also for game development. I did battleship as a network game and the Multi-Tasking was very usefully for that. As i didn't care about the network traffic while doing other things. Of course in the end the z80 can only do one task at a time but you can use all the free CPU time to do other thing.

So i can see the benefit for a multitasking dos system

true

and maybe symbios got the possible start for some parts of next genmsx

But that is the problem of SymbOS, it is not MSX.

What TNI team mean with DOS3 is that you can program following the actual MSX standard hardware/software and have multitasking at the same time, so it will be MSX with multitasking.

That is the main cause why the results of symbOS is limited about adoption by the comunity.

I can't help thinking that I already own a computer that can execute multiple msx programmes at the same time: my Linux machine. But if they come up with a better BASIC, A cross between MSX-BASIC and de ZX Spectrum's Beta BASiC would be awesome. Beta BASIC had local variables, functions (no return statement, but you could pass variables by reference if you wished), advanced tracing, and much more modern stuff.

There's nothing such as the sweet smell of vapour in a monday morning.

you seem to have a very sensitive nose then

To call something vaporware merely a few days after its announcement is just not right....

1) Because I see an MSX more as a one-task machine. A game, a demo etc. fullscreen, and not both at the same time.

2) Because I don't see an MSX as an office machine anymore, and that's about the only purpose I see for real multi-tasking, where it may be practical to have two windowed text editors etc.

Anyway, what I wanted to point out is to find out what people would be expecting to do with a new MSX. I'm merely stating that a term like multi-tasking may sound great 'n cool, but that in my opinion only few would really do something useful with it. Often the idea is nicer than what it works like in practice.

Actually, perhaps it'd be handy to define what we're talking about. Are we talking task-switching (stuff remains idle until you switch to it) or multi-tasking (multiple processes run at the same time)?

If you use your machine mainly for playing games and watching demos multitasking isn't really needed.
But for "working" with it it is really useful already on systems like the MSX. Syn already gave a good example. Only because you didn't use it before on the MSX doesn't mean that you wouldn't like and need it.
Other typical examples:
- running a music player in the background
- unZipping files
- writing some texts in a texteditor
- still have the possibilities for doing other stuff inbetween without the need to stop any of these tasks
As EdoZ mentioned, when it comes to networking not having multitasking is really boring. Somehow multitasking is a requirement of networking. You can...
- run your Samba server in the background
- download stuff with WGET or whatever
- have your messenger/chat(s) open
- and do anything else you want
All these tasks can already be done fine with an MSX today, and I am sure people would appreciate it if they can run several of them at the same time.

That's what's interesting to find out first then. Do we run games/demos, or do we actually work (office-type work) with an MSX, anno 2016? And your other typical examples would fit office-type - for lack of a better term. I would initiate an MRC-poll for that!

Do note though, contrary to what you may think, I'm not against such a system at all. It's just that it'd be a shame if large chunks of horsepower would go unused because it simply doesn't interest the majority of people. Horses for courses. And I think MSX is a game/demo horse, not an office horse anymore. But hey...

yeah i took a too quick look at symbos, I read MSX dos so I thought it runs under dos but its just the installer/setup.

Next Vaporware

I think I understand that the word vapoware is being trown alot against this post. This project is quite something big we are talking about. I do however have trust in some of the people behind this project. As far as I know there is a lot of skill doing stuff at TNI, something that should not be underestimated.

I seriously wish the best to the people at TNI for taking this huge effort. If it completes: awesome! If it turns out to be vaporware: at least someone seriously tried.

Big projects are hard projects but I love big projects and they all have my moral support and in the far future my time and (crappy) coding skills.

We must not forget that big projects eventually spawn smaller projects. Even if they fail to complete it is never in vein.

Aim lower and crowfund a minor upgraded machine like this: http://www.specnext.com/

Well said John.

Maybe it is not really clear but the goal of this project is not to release new hardware. Instead it aims to create a new MSX standard. Ofcourse new hardware (in the shape of a new msx computer or 1ChipMSX/Zemmix Neo type device) could be released in due time in cooperation with a MSX hardware manufacturer or with the help of crowdfunding.

Regarding the progress: The focus right now (the thing the developers are working on atm) is to improve the current system software (for example the Basic or BIOS). Once that is done, the new hardware functionalities will be implemented (in FPGA).

@Syn: Thanks for the update! We'll have to be patience a bit more

and?
anything happen?

They are still working on it. I'll ask them about it later on for more info

Really nice interview, thanks Manuel.

It is a really nice project, and the fact that is is designed to be used in currently FPGA MSX computers (even if it needs bigger fpga like MSXSM) makes it adoption a lot easier.

I'm just concerned that it seems to be a lot of work for just one guy (even if this one is a talented guy like Lesparre). Enhancing BIOS, DOS and Basic while designing a new computer is too much work. I hope some other talented people can join him.

and?
Anything happen?

Cool, I’d love to hear more about this!