New Member for my collection, SC720

[Derek Cooper]

Arrived at 11:35 and finished by 14:35 even had time for a coffee half way. Very nice kit, easy to assemble -  nice instructions on the website. I only did one extra step and that was to check the 7mhz clk was ok before I added all the chips, just insert the 74HCT02 and check the clock on the CPU pin.

Very please with this.

Thanks for all the hard work.

 

[Chris Brunner]

Oooooooohhhhh that's pretty. I haven't seen the 720 yet.

*runs to SCC*

--
You received this message because you are subscribed to the Google Groups "RC2014-Z80" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rc2014-z80+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/rc2014-z80/9e9c396f-4a22-43a5-abaf-6aa47929bab4n%40googlegroups.com.

[Steve Cousins]

Hi Derek,

Good to know you are happy with your new toy. I'm curious as to why you have not fitted the angled header pins in the holes provided for the serial ports, especially as you appear to have what looks like male to male pins as an adapter. I find the combination of the 6-pin socket plus the 6-pin angled header pins to work really well, allowing the choice of directly fitting the USB adapter or using the 6-way cable as you have done.

Steve

[Derek Cooper]

Hi Steve, it's simple enough. I'm deciding if I want vertical or horizontal pins (considering a case for it) so I left them out and just used an adaptor for now.

Derek

[Randall Crook]

Finished mine on the weekend. 

Have setup Compact flash cards for both SCM and RomWBW, been doing some quick tests and it’s a really nice machine. 

Does the Mandelbrot set nicely. 

Randall. 

--
You received this message because you are subscribed to the Google Groups "RC2014-Z80" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rc2014-z80+...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/rc2014-z80/509c1ef2-7d85-4f2a-89e3-abcaa6c811c1n%40googlegroups.com.

[John Rose]

Oh my goodness, that is a thing of beauty.  The wife's going to kill me, but I'm heading over to Tindie right now...

[Mark T]

I like the replacement of the 74hct670s. I remember there was a thread discussing that approach some time ago and Bill Shen confirmed it worked with one of his cpld boards. I think U5 could be removed on the SC720 and take an enable from U15 instead, but this was probably to save some time on the design. Did anyone notice just how many new boards Steve designed recently?

[Bill Shen]

The original 512K RAM/ROM has 16K banks but RomWBW uses 32K banks so a new design specifically for RomWBW such as SC720 can have significantly simpler banking logic.  In fact, most of my banked memory designs are 32K banks.

  Bill

[John Scherer]

Ok, might as well post that I received my boards (SC719, SC715, SC111, SC119, and SC809 backplane) from Steve on July 24.  I would have had them all assembled that night but got a late start.  There is a LOT of soldering on that backplane.

[Alan Cox]

On Thu, 27 Jul 2023 at 23:14, Bill Shen <coinst...@gmail.com> wrote:
>
> The original 512K RAM/ROM has 16K banks but RomWBW uses 32K banks so a new design specifically for RomWBW such as SC720 can have significantly simpler banking logic. In fact, most of my banked memory designs are 32K banks.

Although it does rule out running MP/M and a lot of other stuff. MP/M
really wants 56K/8K so even then there are better ways of getting
32/32 or 56/8 options than the now hideously expensive 670's

[Alan Cox]

As I was looking at it I added the sc720 to the emulator. I'll take a
look at adding the 128K ROM/RAM board when I have a bit of time

[Steve Cousins]

Hi John, you may need to get a new wife LOL

[Steve Cousins]

Hi Mark,

You are right that U15 hs spare I/O enables, so it would seem possible to replace U5. BUT. I didn't have a spare gate to qualify WR. Also, I couldn't generate all the required addresses with the one 74HCT138. I wanted LEDs at 0x00, SIO at 0x80, and paging register at 0x78. I wanted to maintain software compatibility so those addresses were considered fixed, I know RomWBW is easy to configure for other addresses but there is still an advantage in consistency. There is one further problem. My paging design is only software compatible with the existing drivers if I only enable it for addresses 0x78 and 0x79. With the existing 74HCT138 the address range covers 4 addresses per device. This would cause any initialisation code in ROM to paged out when it tries to program the registers at 0x7A and 0x7B. In short it was a case of compatibility over chip count.

As for the number of boards I've produced recently, well I'm an addict - what more can I say. I just had a quick count up. Looks like I now offer over 90 kits. Its all getting rather silly!

Steve

[Steve Cousins]

Hi John, I expect that system will look very nice when finished. Thanks for sharing.

Steve

[Steve Cousins]

I did sketch out a design for a banking scheme where the common area could be set with jumpers from 1k up to 32k, I think it was, in sensible increments. It was actually quite a simple circuit. The compromise was that the banks are all 64k so the shadow RAM (hidden by the common area) could not be used. Seemed wasteful, but simple to implement. I may make one one day.

Steve 

[Steve Cousins]

Hi Alan

Nice to know the emulator supports SC720. Thanks.

Steve

[Bill Shen]

Alan has mentioned the 8/56 split for MPM before, but this time it has finally sunk in.  Its off topic for SC720, so I'll talked about it in a separate thread in retro-comp.
  Bill

[Alan Cox]

On Fri, 28 Jul 2023 at 21:12, Steve Cousins <steve...@gmail.com> wrote:
>
> I did sketch out a design for a banking scheme where the common area could be set with jumpers from 1k up to 32k, I think it was, in sensible increments. It was actually quite a simple circuit. The compromise was that the banks are all 64k so the shadow RAM (hidden by the common area) could not be used. Seemed wasteful, but simple to implement. I may make one one day.

I did one using a latch and mux. The latch provides the upper bits for
the 64K bank and the mux decides which set of 4 latch bits to use.
It's then simply driven by an AND gate which would be easy to add
jumpers or a bit of glue for software select to select 32/32 48/16
56/8. As the upper and lower is switchable you can use all the memory
if needed, although with 56K/8K you don't lose that much and I'm not
sure what I'd do with all the 8K blocks scraped together in MP/M. With
32/32 obviously it becomes a different game. The big limit is that you
can't have a block appear low and then make it appear high although I
don't think ROMWBW cares.

Alan