Is there any future of 6502?

[Icy]

Yes, PIC's processor is lousy. 65x02 is wonderful. Programming 65x02 feels like vacation, you cannot get tired doing that. It is natural, logical, it doesn't give you headache.
So, why couldn't someone, like WDC, come up with a single chip microcontroller that is at par with PIC or Motorola? WDC's microcontrollers aren't microcontollers of today. WDC lives in past. Microcontroller of today is a one chip, and nothing else is needed to attach.
All of that core stuff will disappear in a few years. And, so will WDC, unfortunately. Unless they come up with something up to date. Who wants to tell them?

[kc5tja]

Icy wrote:

Unless they come up with something up to date. Who wants to tell them?

Both Garth and I have been regularly telling them.

However, they are reporting that sales are up (and this is evidenced by the fact that their website actually has content on it now!), so they are obviously doing something right, even if it isn't directly visible to us.

[GARTHWILSON]

Quote:

So, why couldn't someone, like WDC, come up with a single chip microcontroller that is at par with PIC or Motorola? WDC's microcontrollers aren't microcontollers of today. WDC lives in past. Microcontroller of today is a one chip, and nothing else is needed to attach.

We would all like an array of 6502-based µC's made available as standard products. I don't think WDC lives in the past at all; but again, their primary business as I understand it is not selling hardware, but licensing the intellectual property to other companies that do in fact make their own dedicated custom µC's with everything in one IC and the 6502 at the heart of that IC. I wish WDC would license their IP to Microchip. It would be nice if there were a widely available parallel line of PICs built around the 6502 instead of Microchip's little RISC. Everything would still be in one IC, as it is with the currently available PICs.

I suppose the nature of WDC's business and the lack of standard (as opposed to custom) parts is why it appears to most people that the 6502 got relegated to ancient history by industry's negligence to keep a good thing up to date. Actually the 6502's growth just took a less-visible path than the x86 did with desktop computing or Microchip did with standard-part microcontrollers. Surprisingly, more 6502's per year are being sold today than at any time in history. You probably own 6502's in modern consumer electronics products and didn't even know it.

I personally am convinced that what people get familiar with as hobbyists and students gets carried forward into their jobs later when they get into industry and favor the parts they're familiar with to design into their company's products, and I think that is why WDC would do well to support the little customers better. It seems that they are very slowly moving toward that. That's not to say a hobbyist who buys one IC for $5 has permission to take $100 of WDC's time now and then to get help troubleshooting something; but at least they've dropped the $100 minimum order requirement (which I didn't think was that bad anyway) and put their programming manual online instead of insisting that we pay $60 a copy, and they seem to be starting to look more favorably on this website as a supplier of ap notes and things they've been lacking. They're also starting to talk about flash-based µC's that can be programmed on the workbench.

[orac]

Hi Everyone,

Remember... the ST7 from ST micro is a 6502 look alike. It IS a single chip device with Flash, RAM, and IO.

Note, I haven't actually used one of these yet.

Cheers,

Paul

[kc5tja]

Isn't STMicro one of the licensees for WDC's IP?

Also, while I couldn't find any detailed documentation on the ST7, I did find some example code snippets, and its mnemonics are half-68000, half-Z80 (a very interesting combination to say the least). From what I've seen, it does not look anything at all like a 6502. Moreover, I have yet to see them use the Y index register.

[leeeeee]

Quote:

while I couldn't find any detailed documentation on the ST7

http://www.st.com/stonline/products/lit ... m/4020.pdf

Quote:

From what I've seen, it does not look anything at all like a 6502. Moreover, I have yet to see them use the Y index register.

It's very like a 6502, accumulator, X an Y index registers, zero page addressing etc.

Of course there are differences, the decimal mode flag had been replaced by a Z80 like HalfCarry, the stack is smaller - only 64 bytes - and it isn't binary compatible, e.g. LDX #nn codes as $AE $nn not $A2 $nn as it does on the 6502.

Lee.

[Icy]

Hey, how about this:

Quote:

I wish WDC would license their IP to Microchip. It would be nice if there were a widely available parallel line of PICs built around the 6502 instead of Microchip's little RISC. Everything would still be in one IC, as it is with the currently available PICs.

This would be ideal situation for WDC. Just think of all that huge market suddenly open for them.
This could blow my mind. No more headaches programming crappy, hostile PIC's RISC!

[kc5tja]

Icy wrote:

This could blow my mind. No more headaches programming crappy, hostile PIC's RISC!

Although still not as convenient as a 6502, I do find AVR microcontrollers to be MUCH better than the PICs from a programming point of view.

[8BIT]

kc5tja wrote:

Although still not as convenient as a 6502, I do find AVR microcontrollers to be MUCH better than the PICs from a programming point of view.

I agree. My video project was my introduction to the AVR family. I found the opcodes to be very similar to the 65C02. The last 6 of the 32 gerneral purpose registers work as three - 16 bit index registers (X,Y,Z). These allow for easy access to SRAM and FLASH memories. The stack is only limited by the available SRAM. They also run at 16 MIPS which is faster than the 14MHz 65C02. The AVR's use dedicated IRQ vectors for a multitude of functions, making IRQ decoding simple. I have not experimented with multiple-IRQ handling, so cannot say they have better response than the 65C02. The biggest plus to me is the onboard I/O (SPI, UART, timers, etc). The biggest minus is the limited SRAM onboard. Some versions do support external SRAM, but access speed to this is cut in half.

My two cents... Daryl

[Nightmaretony]

Trying out other CPUs is a pain in the rear. Am deisnign a testing system for work, wanted a seperate sub computer to handle some switching. In the interest of time, wanted an off the shelf SBC with certain sets of IO and USB. Called WDC, no dice. And bummer, my fave person Ralph is no longer there Picked up an SBC using an 8051 with basic. The assembly is as you said, hostile on that unit. Basic is a tad strange format to one used to Applesoft. Had to sstruggle a bit with it.

Cant wait to get back to doing the pinball project. Got too many new projects, including a movie soundtrack....

[kc5tja]

What, if I may ask, is your set of requirements that you opted not to go with a 65xx solution? I'm curious, because it seems to me that something like the Kestrel-1 or the SBC-1/2 probably would have fit your needs except for a few key details. What are those key details?

Thanks.

[Nightmaretony]

RS232 communication with routines

basic or machine hooks, but the communication. Originally usb but opted out of it.

several i/o ports to work with

available finished and ready for action.


Already got this one based on the 8051, using basic for it. I still prefer 6502instead. Sigh. Little late now to not use it, already got thingtws going in there... refunding isnt an option since I already began modding into the board...

At least for the pinball and haunted house projects, it will remain 6502 all the way.

[Nightmaretony]

Actually, one other thing it will ened to do is IR command lines sending and recieving. If I cant get the other one to do it, I will switch it over pronto and go back to what I know and love best.

So what it would be:

RS232 i/o

8 or 16 i/o pins
the ability to recieve and send IR command sets such as RC5 or custom sets definable.

The trick was time, so I opted for the off the shelf. I dont know how it will do for the IR. If those iexist for 6502, let me get back to it



NOTHER fun project to keep me busy...
http://www.imdb.com/name/nm2158451/

[kc5tja]

Nightmaretony wrote:

RS232 i/o
8 or 16 i/o pins
the ability to recieve and send IR command sets such as RC5 or custom sets definable.

I remember one time reading about how someone was using a VIA chip for a UART. But their approach was quite unique! They used a timer to determine how long to keep the TXD line high or low. So, if you wanted to send 01001100 binary, and u is the bit duration, they'd first program the TXD line low, then tell the timer to expire after 2u. Then they'd simply EOR the TXD bit to toggle it, and tell the timer to expire after 2u again. Toggle, time, toggle, time, etc.

I'm sure that receiving was done a bit more traditionally, but I thought it was kind of interesting, especially since the baud rate was fully programmable within a certain range. However, I'm sure the baud rate wasn't all that great. But it should be able to handle at least 20kbps though, I'd think.

[Nightmaretony]

High baud rate isnt important. The host ocmputer will send out commands to the sub cpu and the sub may return values or send back IR messages. So in a sense, the RS232 monitoring would need to be kept going on a continious basis as an event loop....