General Category > Amstrad CPC hardware

Don't you think it's time to stop creating hardware interfaces ?

(1/6) > >>

first I have almost no knowledge in electronic; I barely know how to use a soldering iron properly.Do not see this post as an attack to hardware creator but a friendly discussion for improving the ergonomic of our CPC expansions.

as a demomaker I use lots of interfaces to ease my crossdev works and I have followed the trends to upgrade (ie replace an old version by a newer one from someone else or the same creator) them with time. I have also bought some that I do not use.
It ends with lots of cards more or less incompatible between them that cannot fit all together on a CPC for several reasons: lack of expansion port that can be solved by  a mother x4 and impedance? changes that cannot be solved and break some demos (when I have several cards plugged, at some point there is 1 nop? of delay around the CRTC? and GA? that break demo effects (which is quite contradictory with using extensions to help crossdev).Some people also encounter voltage issues with all these things. It seems some cards are hard to produce because their chips do not exist anymore.

Don't you think it's time to use another approach where you have a single expansion card plugged on expansion port (and eventually printer port and disc port as well as video output to easily allow OSD) that hosts various additional output interfaces (WiFi? USB/serial to plug on the PC?/rj45?/cartridge port). The extension would not be hardwired on the card but provided as a software. To alter/add/remove an expansion would consists in modifying the software or its configuration.
It would be less messy in the CPC desk, maybe it would solve the demo effect issue I was discussing before, maybe it would ease the creation of novel or more powerful interface.
I am pretty sure it is a totally realistic thing to do with a bare-metal software on a raspberry pi (no wait of a Linux machine to start, existing expansion source code already available in several emulators) or a fpga implementation (think about the mister project).
Of course this imply several things: hardware people would need to gather forces to implement the board to be sure nothing is missing for the future, user people should be ready to put all their existing cards to the garbage (or another CPC;)) and by this additional one that would be more expensive. That would not change on the existing CPC software stack as the emulated expansion would be the same that the real hardware ones.
Otherscenes have such integrated complex cards. For example Atari ST has the cosmosex and C64 as the 1541 Ultimate-II. Maybe they are not as easily modifiable in comparison to what I have described, but they are powerful enough to be self sufficient.
So I hope, I'm not dreaming alone in front of my computer and hope one day I could by such thing for my CPCs.

Sorry, but when I read you, the standards meme came to mind. In any case I agree with you, a lot of different hardware is coming out and I don't think it's good.

In any case, I don't know much about hardware, the experts had better say.

The more hardware the better.....BUT!
It needs to be thoroughly reviewed and listed on all the the major CPC sites.
And here is a problem - unlike commercial magazine reviews "back in the day" any review will be made by someone who is already a "CPC scene" member and that's when it gets personal. ::)


The answer is flexibility and choice. If I only want RS232 connectivity or just want to connect a mouse or read a few ROM's, why should I be forced to buy an all singing and dancing massive expansion (with most likely a massive price)?
Also the Rpi/Linux approach is something that the purists will argue against. If the CPC is just a keyboard and display for a much more powerful processor, then is it a CPC expansion at all or did you just create a device that allows you to use a CPC Keyboard and screen on a Pi? And putting it all on one board won't solve the NOP problem either, these are limitations of the CPC, not the expansion hardware.

I own a MX4 and a pile of expansions, but I rarely want them all connected at the same time. I just connect the one(s) I need at the time.


From my point of view (and this is just my opinion), for a new add-on to be "standard" or considered by any new CPC software:

- it must be available, so anyone can get one
- the hardware should be open hardware, and anyone should be able to build it (or even sell it, which goes nicely with previous point)
- it must be supported by a few emulators (ideally supporting major platforms like Windows, Mac and Linux)
- ideally there should be enough units out there, but if something becomes a "de facto" standard, this is likely to be true

Also take into account that add-ons that provide ways of using standard interfaces (e.g. load games from a DSK file) are out of scope here because as long as the CPC software is compatible with whatever the add-on implements, that's it (e.g. my games are M4 compatible, and I don't do anything specific: the same works on any emulator and the real CPC).

Either new add-ons are compatible with existing interfaces (either CPC itself or existing add-ons), or their authors will have a difficult time to get their new hardware adopted because the market is saturated at the moment (e.g. I love my M4!).

Yet supporting this type of add-on will always be controversial, but what can we do? People complain if a game is 64K, if is 128K, if requires Dandanator, etc; but putting that noise aside, this is my opinion at the moment as gamedev and user.


[0] Message Index

[#] Next page

Go to full version
Powered by SMFPacks Reactions Mod
Powered by SMFPacks Alerts Pro Mod
Powered by SMFPacks Mentions Pro Mod