Replacing a 6502 with 65816

[GARTHWILSON]

BigEd wrote:

Looking back to Daryl's note, he says the information came from Garth. I wonder: Garth, could you confirm or correct?

8BIT wrote:

Garth told me that WDC has started producing the 65816 in TQFP again

I'm pretty sure I got that info from someone else on the forum. I just left my second message on WDC's answering machine and don't have the answer yet.

[Windfall]

GARTHWILSON wrote:

I'm pretty sure I got that info from someone else on the forum. I just left my second message on WDC's answering machine and don't have the answer yet.

Well, WDC came back to me by email, and said :

WDC wrote:

We do manufacture the W65C816S6TQG-14 QFP part. We currently only have these parts in small quantities, but can ramp up quantities as needed.

Many of our distributors do not have these parts yet, so if you would like to sample or acquire these parts now we could help you with that request direct.

So, rumour confirmed.

[GARTHWILSON]

Good to hear. I still have not gotten a return call from them, and they still are not answering the phone, and I did not leave a third message.

[cbscpe]

I'd like to use the QFP variant. Which distributer has it? Or where can I order it? As an SMD package QFP is much easier to solder than PLCC and would even save more PCB space.

[Windfall]

Did anyone get hold of one or more W65C816S6TQG ? I was waiting for them to appear in the catalog of e.g. Mouser, but it seems nothing at all has happened for over a year (!), and WDC is not responding to my related query this time.

[BigDumbDinosaur]

Windfall wrote:

Did anyone get hold of one or more W65C816S6TQG ? I was waiting for them to appear in the catalog of e.g. Mouser, but it seems nothing at all has happened for over a year (!), and WDC is not responding to my related query this time.

It doesn't appear that the '816 is being produced in QFP at this time. It's either DIP40 or PLCC44.

[Windfall]

BigDumbDinosaur wrote:

Windfall wrote:

Did anyone get hold of one or more W65C816S6TQG ? I was waiting for them to appear in the catalog of e.g. Mouser, but it seems nothing at all has happened for over a year (!), and WDC is not responding to my related query this time.

It doesn't appear that the '816 is being produced in QFP at this time. It's either DIP40 or PLCC44.

Look back in this thread a bit. WDC confirmed that it was being produced. They even offered me some at that time, but I said I was in no hurry and could wait until it was available retail. I'm still not in a hurry, but nevertheless disappointed that they still don't seem to be available retail.

[BigDumbDinosaur]

Windfall wrote:

BigDumbDinosaur wrote:

Windfall wrote:

Did anyone get hold of one or more W65C816S6TQG ? I was waiting for them to appear in the catalog of e.g. Mouser, but it seems nothing at all has happened for over a year (!), and WDC is not responding to my related query this time.

It doesn't appear that the '816 is being produced in QFP at this time. It's either DIP40 or PLCC44.

Look back in this thread a bit. WDC confirmed that it was being produced. They even offered me some at that time, but I said I was in no hurry and could wait until it was available retail. I'm still not in a hurry, but nevertheless disappointed that they still don't seem to be available retail.

What I'd like to see is the 65C816 in a package in which the A16-A23 outputs are brought out on separate pins and not multiplexed with the data lines. It can be done in a PLCC52, since all that is needed is an additional eight pins. I understand why WDC MUXed A16-A23 onto D0-D7 back in the day, but the reason for doing it disappeared with the demise of the the Apple ][gs.

[Rob Finch]

Quote:

What I'd like to see is the 65C816 in a package in which the A16-A23 outputs are brought out on separate pins and not multiplexed with the data lines. It can be done in a PLCC52, since all that is needed is an additional eight pins. I understand why WDC MUXed A16-A23 onto D0-D7 back in the day, but the reason for doing it disappeared with the demise of the the Apple ][gs

What about something like the 65C265 microcontroller ? I think it has all the address lines brought out, and has a wack of built in peripherals. Some of the chip select decoding already done.

[BigDumbDinosaur]

Rob Finch wrote:

What about something like the 65C265 microcontroller ? I think it has all the address lines brought out, and has a wack of built in peripherals. Some of the chip select decoding already done.

My opinion is the design of the WDC microcontrollers is poorly thought out. The ROM has to be masked programmed and the built-in peripherals are something in which I have no interest. If it is desired to use an external ROM then some of the connectivity goes out the window. If I were going to design around a microcontroller I'd go with something that is much more flexible, and more powerful.

However, for hobby purposes, I prefer to stick with discrete components and an actual 65C816.

[Dr Jefyll]

BigDumbDinosaur wrote:

If it is desired to use an external ROM [with the 65C265 microcontroller] then some of the connectivity goes out the window.

That's true. You sacrifice about 45% of your IO pins when an external bus is used. OTOH, that still leaves 40 IO pins free, which is considerable.

In the '265 block diagram below I've highlighted the IO ports that are NOT available when an external bus is used.

Another potential limitation is the 265's 8 MHz speed rating. But if it's expected that code will execute directly from EPROM then the 8 MHz limit may not matter much.

[BigDumbDinosaur]

Dr Jefyll wrote:

Another potential limitation is the 265's 8 MHz speed rating. But if it's expected that code will execute directly from EPROM then the 8 MHz limit may not matter much.

Quite true.

POC V2 (when built) will be able to shadow ROM in RAM and thus avoid the performance limitations of running the firmware from ROM. I'm also cogitating on how to rig up the logic so the block of RAM that shadows the ROM can be write-protected. On paper, this mess should be able to run at 20 MHz. with only I/O hardware accesses being wait-stated.

[GARTHWILSON]

I just got word from WDC, in response to an email I sent last night:

Quote:

The 816 does come in a 44 Pin QFP package. [...] To date these products have not been in our general distribution however we are working to get them into distribution and they should be available from Mouser Electronics by next week.

I hope to make and supply this '802 module, unless someone else would like to do it. (If someone else would like to do it, there are other ones I would like to make and supply.) I still need to figure out how to handle the matter of the pins. Interplex, the manufacturer of pins whose diagrams I posted earlier, has been broken up into two or three separate companies apparently, and I need to remember to call the one on the east coast before their closing time, since they don't use distributors. From talking to another part of the company, it sounds like the minimum order on pins may be prohibitively large (like 30,000 pieces on a reel) unless they have a portion of a reel of the right pins sitting around; so I might need to figure out another solution.

[Tor]

Pins? Something like this? http://oshchip.org/products/Flip-Pins_Product.html
(unfortunately only up to 20)

[KC9UDX]

What I've done is use headers, soldered to a sacrificial socket. It makes for a lot of extra height though.