DK'tronics Upgrades

[Bryce]

As part of my investigations into new RAM expansion possibilities, I did some playing about with a standard DK'tronics 64K expansion, first converting it to 256K and then converting it to a 256K silicon disc. I know some of this information was probably available somewhere else, but I thought I'd document it properly with pictures in case anyone was interested.

I will probably follow up later with a 512K conversion of the same device when I have time.

You can find it all here: http://www.cpcwiki.eu/index.php/Upgrading_a_Dk%27tronics_RAM_Expansion

Bryce.

[Zetr0]

Now this is VERY nice indeed +1 ^_^

[Cholo]

Niiice! the silicon disc mod looks simple enough that even i could do it  Still, do you need a additional silicon box (i think ive seen pics of silicon disc as a box set of 2) or is it all "onboard"?

[Bryce]

The second Silicon Box was a simple ROMCard for just one ROM. The ROM Image is available on the Wiki page, so you just need to install that on your MegaFlash or any other ROMBoard and you'll have everything. The ROM adds some RSXs to allow you to Load/Save from the RAM as if it's a disc.

Bryce.

[Gryzor]

Niiiice! Bonus points for creating a nice page AND linking to it properly Thanks!

[Bryce]

Well it's kind of pointless writing a page and then nobody finding it because it's not linked to

Bryce.

[OCT]

The second Silicon Box was a simple ROMCard for just one ROM. The ROM Image is available on the Wiki page, so you just need to install that on your MegaFlash or any other ROMBoard and you'll have everything. The ROM adds some RSXs to allow you to Load/Save from the RAM as if it's a disc.

Now I wonder if a second-gen "MegaFlash Mk II" shouldn't integrate dk'tronics-compatible 4MB circuitry like Yarek's all on one board...

[Bryce]

Great idea. Let me know when you've got it finished, I'll buy one

Bryce.

[TFM]

The second Silicon Box was a simple ROMCard for just one ROM. The ROM Image is available on the Wiki page, so you just need to install that on your MegaFlash or any other ROMBoard and you'll have everything. The ROM adds some RSXs to allow you to Load/Save from the RAM as if it's a disc.

Bryce.

Well, I would suggest to use Dobbertins RDOS, because if provide a RAM DRIVE C, even compatible to games :-) And it has 444 KB free.

In addition it works with both CP/Ms :-)

@Bryce: Great piece of work!!! I can see the art of making in it :-DDD

[Gryzor]

Well it's kind of pointless writing a page and then nobody finding it because it's not linked to

Bryce.

Precisely my thinking, but it's not that obvious to many users. Many editors think that once they hit Save it goes into the collective subconscious

[Cholo]

The second Silicon Box was a simple ROMCard for just one ROM. The ROM Image is available on the Wiki page, so you just need to install that on your MegaFlash or any other ROMBoard and you'll have everything. The ROM adds some RSXs to allow you to Load/Save from the RAM as if it's a disc.

Bryce.

That makes sense, thanks!

[OCT]

[MegaFlash (internal if it has to be) with Yarek-style 4MB RAM] Great idea. Let me know when you've got it finished, I'll buy one

I trust that'll be much better if you buy the parts and make them fit together nicely.

By the way, as you were hinting at forthcoming cartridge work in your NoRecess interview, DangerousPrototypes.com have been busy discussing Bluetooth integration on microcontrollers lately (as in "remote-flashable EEPROMs").

[Bryce]

Interesting. I think I would still go for a USB solution, due to the extra cost of adding Bluetooth, and all the handshaking/setting up involved with BT. USB is Plug it in, transfer. BT is Pairing/pin code etc. BT makes sense for portable devices, but does it really make sense for two devices that have to be plugged in to the power and are most likely in the same room?

The hints I made in the interview were more aimed at an ACID replacement and possibly something close to the C64 EasyFlash.

Bryce.

[TFM]

Blue teeth? Well, if you like Wii controllers ;-)

[Gryzor]

I agree with Bryce. BT has been a pain in the arse even for devices (like mobiles) that greatly benefit from it. I wouldn't touch it with a ten-foot joystick cable.

[TFM]

Oh, USB isn't better. It's a pain in the back to program proper USB protocols. Sure the user will never feel it, it just works. Dr. Zed talked to me about this in more detail, so I'm not jealous to people who like to program hardware for USB (doesn't matter if a microcontroller of VHDL).

[Bryce]

USB integration was a lot of hard work on the old µPs where you had to do everything by hand, but new decent µPs have most of the work done for you and even supply sample drivers. Eitherway, what's important is the amount of work/hassle that the USER has, not the developer! ie: The developer does lots of work so that the user doesn't have to. If you save on the development work, you have a product that's really difficult for the user, what use is that!?

Besides, BT is way more complicated to program than USB.

Bryce.

[TFM]

Sure, it must be easy to use and nobody can expect the user to complete the developpers work, no doubt.

But in past this was not the case for all CPC expansions, especially non-commercial stuff. However latest releases of hardware had a high level of the "user-babysitting-facter" or "convenience-factor". YKWIM

[Bryce]

I didn't mean that the user should finish the development, I was more talking about the developer taking the easy option at the expense of the user experience: ie:
A PS/2 mouse interface is easier to develop than a touchscreen, but an iPhone that needs a PS/2 mouse attached would not be as user-friendly as the touchscreen solution.

The same can be said for USB on a CPC device. USB: plug and play, user plugs it in an it all works. RS232: User needs to mess about with baud-rate settings and comport numbers. BT: User needs to pair devices and enter pin codes. So the USB solution might be more difficult to develop than say RS232 communications, but the user experience is better because the developer put in some more work.

Of course when the booster was designed USB wasn't an option, but new expansions should reflect current standards to improve the user experience.

Bryce.

[TFM]

The same can be said for USB on a CPC device. USB: plug and play, user plugs it in an it all works. RS232: User needs to mess about with baud-rate settings and comport numbers. BT: User needs to pair devices and enter pin codes. So the USB solution might be more difficult to develop than say RS232 communications, but the user experience is better because the developer put in some more work.

For the CPC (and I'm NOT talking about the PC!) this all depends on software. One device can be as easy as the others to be used. For CPC imagine an BT device: You just bring it into the room with the CPC and you do nothing, because the software recognices it and pairs with it. The same for RS232 devices: if all soft for the CPC (not PC!) uses the same parameters, it's plug & play

[OCT]

For CPC imagine an BT device: You just bring it into the room with the CPC and you do nothing, because the software recognices it and pairs with it.

That was the whole point of my proposal to use Bluetooth (or whatever RF technology comes to mind) on a remote-reflashable cartridge:
No wired connections from which a CPC might fry&die, no hassle unplugging everything time and again on aging connectors.
Of course this implies the protocol stack is readily implemented (rather than having to be developed from scratch), but that's what it looks like from the sites inclined to such tinkering, e.g. DangerousPrototypes.com and Mikrocontroller.net even for Bluetooth (then again, mentioning the latter was by no means meant to exclude something as simple as e.g. I²C on the cartridge edge for in-circuit reprogramming).

[Bryce]

The ideas I had been working on didn't involve having any communications software or drivers on the CPC or PC side. The hardware would look after all the comms and protocols and it would be complicated to allow pin codes to be entered. This was my idea:

The CPC: Would "think" that a RAM expansion (probably 512K) is connected and can access/manipulate/copy/save or whatever using standard routines or even BASIC. When not being used to transfer data, the device would work as a standard RAM expansion.
The PC: would think a 512K Memory Stick is connected and be able to drag and drop files in the normal way no matter what OS you happen to prefer.
That way neither side needs special drivers or software and the hardware is compatible with all PCs and CPCs.

Now you all know why I have been experimenting with RAM expansions for the last few days

One of the problems with using BT on such a device is that it usually gets powered down quite often, so the pin code/pairing would have to be performed every time you wanted to use the device.

Bryce.

[SyX]

Can i reserve one already??? Please, pretty please, with sugar on top!!!

I love your idea Bryce, and we only need to install the RDOS rom in the MegaFlash to use it as a ram disk perfectly compatible with the basic/firmware. 

In this moment, i'm following with a commented disassembly of it that i began a few months ago, when TFM discovered for me this little wonderful, i'm very grateful to him by that, you are the best my friend

[Bryce]

It's still only an idea at the moment, I've no idea when I will get time to make or test a prototype

Bryce.

[SyX]

Je je je, i know, but i want to give you positive feedback, because i think it's a GREAT IDEA!!!

When you have time, take a look to the RDOS rom, if you don't know, you will see that using the ram disk feel very natural, you only have to use |C (or úC ) to select the ram disk and voila, you can use load, save, ... totally transparent to the user