"65C02: Tear down, A look at the CMOS version of the 6502"

Let's talk about anything related to the 6502 microprocessor.
Post Reply
User avatar
BigEd
Posts: 11463
Joined: 11 Dec 2008
Location: England
Contact:

"65C02: Tear down, A look at the CMOS version of the 6502"

Post by BigEd »

Via Jac Goudsmit on G+, a short video and matching blog entry, from "electronupdate".
Quote:
This particular version was plucked off of an embedded industrial controller... a place where this processor still finds design wins. The die has patent numbers which were very helpful in sorting down the details of this design.
Video:
Image

Blog:
Image

I think some of the commentary on the chip is not quite right. I would say the floorplan is very like the NMOS 6502 - the area below the decode array will be the so-called random logic, and the more organised area in the bottom half of the core will be the datapath - that's where you'll find the registers, the ALU, and the busses. Note that the decode array doesn't have fuses - it is mask-programmed - but the variable features are indeed the programming of the array.

Lots of white space in the chip, as noted - looks pad-limited, which is not great for price, as the core is such a small part of the silicon.
User avatar
jac_goudsmit
Posts: 229
Joined: 23 Jun 2011
Location: Rancho Cucamonga, California
Contact:

Re: "65C02: Tear down, A look at the CMOS version of the 650

Post by jac_goudsmit »

I got the link from TheAmpHour. https://www.reddit.com/r/TheAmpHour/com ... _6502_cpu/

===Jac
calculi
Posts: 68
Joined: 19 Oct 2015
Location: France

Re: "65C02: Tear down, A look at the CMOS version of the 650

Post by calculi »

Very interesting ! Finally we've got better pictures than the tiny ones to be found on WDC's web pages. Maybe some day we'll enjoy even higher resolution since the video clearly shows some zooms on the PLA and the patent numbers. Their CPU seems to come from Sanyo, before WDC's switch to TSMC foundry, it is already a 2-metal process. Too bad that the commentary isn't accurate.
User avatar
BigDumbDinosaur
Posts: 9426
Joined: 28 May 2009
Location: Midwestern USA (JB Pritzker’s dystopia)
Contact:

Re: "65C02: Tear down, A look at the CMOS version of the 650

Post by BigDumbDinosaur »

jac_goudsmit wrote:
I got the link from TheAmpHour. https://www.reddit.com/r/TheAmpHour/com ... _6502_cpu/

===Jac
Just got around to looking at this. Good find, Jac.
x86?  We ain't got no x86.  We don't NEED no stinking x86!
calculi
Posts: 68
Joined: 19 Oct 2015
Location: France

After the 65C02, the 65C22 ^^

Post by calculi »

A new entry recently showed up on ElectronUpdate blog : our well-known VIA !

https://electronupdate.blogspot.fr/2017 ... -past.html

An unusual fact : the bonding pads are labeled with the pin number and the signal name. The third picture matches WDC's datasheets : 1 polysilicon and 2 metal levels. The floorplan seems to share common points with the CPU : the horizontal middle section looks like a datapath (8 horizontal slices) ; it probably hosts the various registers (control registers, timers, I/O)... and the BIG output transistors are easily noticeable on the peripheral ring.

I'd really like to know if they plan to publish super-high resolution pictures... ? imagine a "visual 65C02" :D
calculi
Posts: 68
Joined: 19 Oct 2015
Location: France

Re: "65C02: Tear down, A look at the CMOS version of the 650

Post by calculi »

Bump

For those who are interested, please have a look at an older entry on "Elecronupdate" blog :
https://electronupdate.blogspot.fr/2017 ... -down.html
and click the 1st image ; the upper left part of the chip clearly looks like a CPU, doesn't it ?
If you search the internet for "GHH393" you'll find a chip from SunPlus/GeneralPlus, with a very similar looking CPU core used in the Furby toy -and 65C02-based.
Our preferred processor has a tendency to hide everywhere, just as told by WDC (and Garth ;-) )
User avatar
BigEd
Posts: 11463
Joined: 11 Dec 2008
Location: England
Contact:

Re: "65C02: Tear down, A look at the CMOS version of the 650

Post by BigEd »

Good spot- that's certainly a 6502 of some sort!
6502-spotted.png
6502-spotted.png (95.51 KiB) Viewed 2104 times
Post Reply