We all know by now that Nishi is planning to create a one-chip computer based on the MSX standard. I would like to know what you think about the one-chipper and what functionality it should have to make it a success.



Things nishi had in mind:



- FPGA so you can add functionality to the hardware

- Wireless internet communication

- Translation from a universal language to 189 different languages

- MSX Compatibility??



My suggestion:



Let the computer make use of any kinds of energy available. Sun, wind, kinetic (especially the latter)

heh.. nice suggestion, but the computer was supposed to be $100, so that probably won't happen... But ofcourse a more expensive model could have sun and/or kinetec energy

What I think is most important is MSX compatibility, which can be achieved in two ways.

Either a MSX-mode, or tight integration of an emulator within the OS.



It should be possible to program the system on different layers just like on the current MSX. The hardware should be powerful and simple, just like the current MSX.

Also the video processor should be versatile and simple. Like V9990 but with more GBA-type functionality added to it. But I would be happy with a design like the V9959 as presented on some japanese website. (I think it was part of ESE's FPGA project)



I am glad Nishi suggested the ARM CPU. It is a very efficient RISC chip which in contrast to most other RISC CPU's is easily programmable in assembly. (GBA is also based on an ARM CPU)

Yes, and if they hurry up, the specifications will be out dated. I think when the time comes for actually producing the thing, they have to look again and propably add a new faster ARM or better VDP.



Look at at, it was almost 3 years ago when nishi planned these things. The computer i use, was assembled from brand new parts just 6 months ago. And look what has happened? My computer is already far behind with it's lame 1.2 GHz and just 256 MB DDR. Ok, my 80 GB HDD will do and my mouse and keyboard will also do but the hardware that supplies the mathematical and visual power is already... ancient.



And nishi's plans are muc more out dated then my computer is. I think they will bear that in mind as well. They have to.

Don't ask me to be more specific, coz I can't, but in my opinion, the computer Nishi is trying to build will not be outdated as quickly as 'regular' PC's do nowadays. Why not? Because there's a whole different philosophy behind this new onechip computer. It can't be compared. Different philosophy, different rules.

Two things:

1. it was 2 years ago

2. its specifications are NOT set, they don't have to 'look again'. If you'd have done some research you'd know there were no specifictions of clockspeed and CERTAINLY not of what VDP to use.



And Latok is right. Nishi's ideas are (again) revolutionary and can't be compared.



And LOL, you think you 80GB isn't outdated?! There are harddisks available of more than 3 times that size.

heheh... sure, but as long my 80 GB isn't filled yet, i won't complain. And well, it doesn't matter if they did or didnt exactly specify the hardware power such as CPU speed. They just have to (and i believe they will anyway cuz their not stooopid either) build the computer with the standard hardware of the future, when the start building the thing.



The hardware of today will be much cheaper within 3 years. But maybe not in stock anymore. It just can't be that easy to build a computer based upon specifications that have been predicted. The market is too turbulent like we all know.



I certainly would not believe anyone saying that the computer will be 5 Ghz or whatever .

IMHO Nishi's design offers a lot of flexibility to do some 'last-minute' changes (like a slightly faster ARM core or whatever).

yep! just as it should be

For the people not feeling very fond of reacting to this topic: I'd like to point out that we do have (minimal) contacts with the MSX Association. If interesting ideas or suggestions emerge at this forum, we will - from time to time - summarize them and pass them on to our Japanese contacts.

There is at least one thing I really consider important for myself to make this one chip computer interseting to me: it should be able to run (almost(!); even MSX2 wasn't able to run all MSX1 software) all existing MSX software, independantly of the used medium, so indeed, it should have possibilities of connecting cartridges and floppydrives...



But I also think the device should be able to compete with modern-day existing devices, whether PDA, Smartphone, TabletPC or similar...

This point is also where my doubts are: is it possible to build a 21st century device based on MSX ideals? To have any chance of commercial succes, it would need a unique feature or "killer-app", this compared to other devices already marketed. And I'm afraid even 100% MSX compatibility is not the killer-app the device needs. Maybe for us, old-time MSX enthusiasts, but not for John Doe buying a new PDA or Desktop computer or fridge, for all I care .

That's what I'm afraid of: will the device consist of features that all have been seen or done before...? For instance, I already have a Symbian based phone, if someone would build an MSX emu for it, I wouldn't need a new MSX PDA or phone, would I? (just pretend I'm not an old-time MSX enthusiast who would buy anything with an MSX logo on it )

Just for the record, the things already planned are



- A price of $100

- Powerful ARM CPU (version/speed unknown)

- Translation from a universal language (and back) in 189 languages

- Use of solar and wind enery (tip from snout: what about kinetics? )

- Worldwide permanent radio-internet connection

- Integrated 'modern online payment method'

- Recycleable



As an added bones, the core chip will be licensed to other companies for integration in other products. The main computer can communicate with all these chips individually (Remember Nishi's toilet examples??)



Of course, the MSX compatibility can be done through emulation, or by switching to an MSX mode. Like Grauw said. There are too many legacy limitations to keep the entire machine based on MSX.

Quote:

Of course, the MSX compatibility can be done through emulation, or by switching to an MSX mode. Like Grauw said. There are too many legacy limitations to keep the entire machine based on MSX.

Well, I'd be happy with a USB port with support for cartridge-readers and floppydrives

(or nicer: Firewire...)

well, i think i can conclude that the thing MUST be extremely versatile in every aspect.

What will it look like? Will it have a build-in display? Will it be just another Gameboy with internet? How will it use wind energy? Why should it be recycleable? Will it have a harddisk? Enough memory? What about gfx? What about sound? And why put everthing in just one chip?

sjoerd - I only know the answer to your last question for sure.



The number of components highly affects the price of a computer system. Less components is a lower price. Nishi intended the original MSX computer to be single chips already