New WDC W65C268QBX Board

[floobydust]

While the new board seems interesting, it still lacks a user-friendly approach (out of the box). While it has a built-in monitor, there's nothing provided to get somebody up and running quickly. It should have included a serial-to-USB interface (FTDI) so you can simply plug it into a USB port (which would also provide power) and interface it with a terminal program quickly.

Not knowing the price, that might also be an inhibitor to being accepted. In short, as it's nothing beyond the '265 controller, some leds and a few miscellaneous parts, the pricing should only be another $10 or so above the price of the controller itself. Basically, it needs to sell for $25 or less and include a quick connect (FTDI) and provide a link to a downloadable toolset to allow the end-user to get something done quickly. Some sample programs which can be assembled and downloaded would be a plus (I think the monitor supports S-record download).

As noted, the competition (arduino, Pi, etc.) are providing something for a low cost and making it easy to get started quickly without adding any other hardware beyond a USB cable.

[Arlet]

Quote:

the pricing should only be another $10 or so above the price of the controller itself

For comparison, STM sells the Nucleo series of eval boards for less than the total price of the components if you were to buy them individually.

[rwiker]

It appears to be $29.95: https://www.tindie.com/products/dcwdc/m ... ocomputer/.

[Alarm Siren]

I just got a promotional email about this board, apparently its officially titled the "MENSCH Microcomputer"
and is available on amazon now for $29.

https://www.amazon.com/WDC-65xx-6502-ME ... rds=MENSCH

Quote:

The MENSCH™ microcomputer W65C265QBX is a small single board computer (SBC), 1.25” x 2.25”, using the System on a Chip (SoC) W65C265S MCU. The W65C265S was invented and designed by Bill Mensch WDC’s founder after he invented and designed the 65816 microprocessor. As a recognized pioneer of the microprocessor industry. Mr. Mensch’s twenty two patented inventions were/are used on the original Motorola 6800 Microprocessor Family, MOS Technology 6502 Microprocessor Family, and WDC’s 65C02 and 65816 Microprocessor Families. As an electrical engineer, 1971 BSEE from The University of Arizona (UA), having taught graduate level classes at Arizona State University (ASU) on SOC design for five years, advisor to four university colleges of engineering for over two decades, the MENSCH™ was designed in hopes that it will attract learners to consider a career based on an Electrical and Computer Engineering (ECE) education, hardware and software engineering. The MENSCH™ can be used to learn about all kinds of electronic components and all kinds of state-of-the-art programmable logic. Operating at 3.3 volts provides for easy connectivity with various state-of-the-art electronic components and modules such as sensors and actuators. Operating at 5.0 volts provides easy connectivity to 5.0 volts electronic components and modules.

Quote:

W65C265S MCU operating at 3.6864 MHz
Works at 3.3V or 5V (Cable not provided), giving access to a wider range of sensors, memories, FPGAs, etc.
Serial Terminal Programming Interface Serial Port #3 (P66/P67/P56/P57) NOTE: Requires external FTDI FT232 or similar cable/board (not included)
8 LEDs connected to P7x
2 – 18 pin connectors plug into standard breadboards for quick prototyping. Signals accessed are Ports P0x (A0-A7), P1x(A8-A15), P3x (D0-D7), Serial Ports S0 and S1; 1 x 4 pin Connector with PHI2, WEB, NMIB and IRQB signals for expansion.; 1 x 4 pin Twin Tone Generator Connector (Driven by TG0/1)

[BigEd]

$29 is a much more realistic price (previously £89 from Mouser?) - and what you get is a surface-mount SoC on a breadboard-friendly carrier PCB.

[Oneironaut]

Now if only they would sell their C compiler at the hobby level, this could go someplace.
Last time I tried to puchase a license, I was hit with a "not for hobbyists" brick wall.
I found that extremely odd considering the efforts to drive interest at the hobby level.

Brad

[BitWise]

BigEd wrote:

$29 is a much more realistic price (previously £89 from Mouser?) - and what you get is a surface-mount SoC on a breadboard-friendly carrier PCB.

No RAM, only the ~580 bytes in the chip.

I'm not sure if you can add external RAM. The address bus, data bus and control signals are routed to the pin connectors but the qualifying chip selects are not -- they are being used to control the LEDs.

You might be able to get away connecting an external RAM to the buses, tying the RAM's /OE to ground and its /WE to the processors R/W signal and having PHI2 as CS. Depends if the processor minds having the SRAM output to data bus when its accessing the internal mask ROM image and I/O registers.

Otherwise you'd need a patch wire to the LEDs to get a clean chip select signal. Shame they didn't route them to pins, There is space at the top of the board by the other control signals.

[Arlet]

Quote:

Otherwise you'd need a patch wire to the LEDs to get a clean chip select signal

And maybe remove the LED to remove some of the signal loading.

[GaBuZoMeu]

Oneironaut wrote:

Now if only they would sell their C compiler at the hobby level, this could go someplace.
Last time I tried to puchase a license, I was hit with a "not for hobbyists" brick wall.
I found that extremely odd considering the efforts to drive interest at the hobby level.

Brad

I don´t know the height of your "brick wall" - perhaps this http://www2.mouser.com/ProductDetail/We ... XjHCtOEwM= is within the limits. You get the WDC65C256, 32K RAM, a socket for Flash EPROM (4x 32K), USB <=> TTL-RS232 adapter, and you get an WDC address where you can register and download a C-Compiler/Assembler package (I took both 8 bit only for 65C02 and 16 bit version for 65C816 etc.). It´s looks a bit like old Win98 / XP style software, but it didn´t crash on Win 8.1. I cannot say anything in terms of quality/speed/etc as I didn´t use it so far.

Arne

[whartung]

BitWise wrote:

I'm not sure if you can add external RAM. The address bus, data bus and control signals are routed to the pin connectors but the qualifying chip selects are not -- they are being used to control the LEDs.

Can you explain this? I'm looking at the W65C265S data sheet, and it's not clear to me what signals that you need are coming off of the Port 7.

Oh, I see, this is what you are talking about?

Quote:

Port 7 has a Chip Select register (PCS) that is used to enable Chip Selects (CSxB). A "1" in bit x of PCSx enables Chip Select CSx- to be output over P7x while a "0" in PCSx specifies the value in the output data register is to be output on P7x. Port 7 data register is set to all "1's" after Reset, and PCS is cleared to all "0's" after Reset.

I guess I don't understand how the Chip Selects are used in this case by the CPU.

Am I reading the memory map properly:

Quote:

CS7B -- 4M -- (C0-FF) -- User Memory

So, if I were to try and access memory in the C0xxxx to FFxxxx range, then the CPU would automagically (assuming proper configuration) trigger CS7B?

It says "8 Programmable chip select outputs", so not sure exactly what that means. I guess it means they can be enabled/disabled through software, but it's not like you can create your own memory map, like, say, I want the respective CS to trigger properly for each, say, MB of memory, rather than 4M or 8M blocks, or whatever.

So, that the memory map in Table 1-5, that's a static map?

Can you simply mimic the proper CS signaling through glue logic off of the Address bus? Of course for this thing, they only expose the lower 16 address lines. Maybe, because of that, the lack of CS is moot?

[GaBuZoMeu]

whartung wrote:

So, if I were to try and access memory in the C0xxxx to FFxxxx range, then the CPU would automagically (assuming proper configuration) trigger CS7B?

YES.

Quote:

So, that the memory map in Table 1-5, that's a static map?

YES.

Quote:

Can you simply mimic the proper CS signaling through glue logic off of the Address bus? Of course for this thing, they only expose the lower 16 address lines. Maybe, because of that, the lack of CS is moot?

Without the upper address lines there is no much fun. And you have to pay attention that your external devices don´t collide with the internal ones. All of this has been done within the 265 and you could use it if they have provided the /CS signals to some pins

At least I hope that this may led them produce some 265 in QFP which may then appear @ Mouser some day...

[DerTrueForce]

Does it have to be QFP? I mean, will PLCC do? Because they have that here: Link.

[whartung]

GaBuZoMeu wrote:

Without the upper address lines there is no much fun. And you have to pay attention that your external devices don´t collide with the internal ones. All of this has been done within the 265 and you could use it if they have provided the /CS signals to some pins

As far as I can tell, the chip on the board is just a normal, surface mount chip. The signals are available, but they're not available on the headers (or through the LEDs).

I think not exposing the CS lines is a real mistake. At a minimum they could have exposed CS3, which would have made it trivial to add a 32K RAM (000200-007FFF) to this thing. Exposing the extra upper 8 bits, I guess thats just asking a lot. I guess they feel any potential applications for this have minimal RAM requirements, but it's even more a shame that half of the RAM available is the stack.

[Dr Jefyll]

DerTrueForce wrote:

Does it have to be QFP? I mean, will PLCC do? Because they have that here: Link.

Depending on your preferences, the PLCC may well be a better choice. It's pretty easy to hack a PLCC onto a protoboard, as a PLCC socket fits onto the usual .1" grid. Or, forgoing the socket, you could uncurl the PLCC's J-leads and create a staggered pattern that would allow the PLCC leads (with .050" spacing) to be soldered directly to the .1"-grid. Or get a board that has holes on a .050" grid.

The PLCC avoids the problem whartung mentioned about certain signals being unavailable. Also by creating your own board you get to choose a physical configuration that suits you. There are some additional parts you'd have to buy, such as the two crystals, but it certainly won't take much to equal what the W65C268QBX Board has! (See the schematic on page 7 of the PDF attached.)

That said, I can see the W65C268QBX Board as a reasonable choice if you intend to use a solderless breadboard and/or your primary interest is just to tinker with some '816 assembly language. BTW notice there's no USB interface.

Quote:

The MENSCH™ has a built in operating system called a “Monitor”. The Mensch Monitor™ interfaces through a USB cable providing power and development through all kinds of “Terminal Emulators” found in most computers, PCs, Macs, tablets, and Smartphones.

[GaBuZoMeu]

THX

I have a more or less permanent link to Mouser's W65C265 offer. So I can easily check whether the QFP version - which I prefer - may become available.

I am somewhat old fashioned . I would try to create a small pcb with the 265, some RAM, some RS232 driver, a terminal-emulation, perhaps an additional VIA, etc. Just to create a small standalone 65 system to play with.