To those designing replacement Gottlieb boards and are using a 6502 based design. If you get tired of using Chinese retreads for 6502s then you are in luck. There is a still manufactured alternative out there -- W65C02S6TPG-14 made by Western Design Center.
BUT - your design must accommodate the WDC 65C02 'Byte Enable' function. To do this - your design MUST include a 3K pullup resistor between VCC and pin 36

If you are doing another spin on your bare boards - it is recommended to add a place to install this resistor in your design so you can use the W65C02 in future builds.

So far, not yet.
The reason it isn't a drop in replacement is due to the "BE" or "Byte Enable" input. It must be pulled high before the IC will work properly.
The BE input does not have internal pullups (too bad) so if unconnected - it floats. If the BE isn't pulled high, it turns off some of the busses ... which is never a good thing.
The main reason I brought this up --> when the wonderful world of Chinese 6502s dries up then these W65C02s can be forced to work assuming people start planning now for them.

Based on what I have read, I think it will be fine in most regards but need to try it. Biggest risk I see is a tight timing loop might complete a clock cycle per loop cycle sooner. It all depends on exactly what the code is doing. If this happens, we may need to tweak the U2/U3 system ROMs but unlikely need to tweak the Game ROMs.

Other differences between the W65C02 and the old R6502:
W65C02 (like the R65C02) introduced a few new instructions. This shouldn't be an issue as the existing code should not have been using these previously undocumented opcodes.
Few flag issues with IRQ and NMI (mostly decimal flag) but if 6502 code was originally written correctly (big IF) then this isn't an issue. System 80 series didn't use NMI but frequently used IRQ.
And there are a few instructions that may take 1 clock less to execute. Actual difference depends on whether data crosses a page boundary. In tight timing loops where no page boundary is crossed, this can cause issues if loops complete faster. It all depends on what the loop is doing.

I have a Gottlieb System 80A board that I will be trying soon. The existing 6502 is soldered in so just gotta make time to pull the old one, install a new socket and try it - put it through some test code and put it through several tests and then run the game code.
But as real 6502 supplies start drying up - this may need in depth research. Does make me wonder - with all of these 6502 remarks that are coming in - how many started off life as 65C02s before being remarked?

I do know Boston Pinball uses these W65C02 CPUs on their System 3 boards. The original R65C02 CPUs did not have this BE pin. Boston Pinball must have tied off the BE pin to get it to work properly on their board. I doubt they have done any coding changes to facilitate timing changes so we know they are System 3 code compatible.

So far, I have a case of W65C21N6TPG-14 PIAs on order. Still need to order some W65C22s but just didn't get to that point yet. And potentially order W65C02s but the jury is still out on this one.

No luck on a 6532 replacement. After repeated attempts, they finally flat out told me 'no, we will not be creating a 6532 replacement'. Damn...
I can easily mimic the 6532 function in a programmable device but this programmable device still wouldn't be a drop in anyways.

Ed

** >> ME << **
https://www.greatplainselectronics.com/proddetail.asp?prod=W65C21N6TPG-14

I have 300 at this time.
Unlike Mouser - I buy them a full case at time.
Mouser does violate manufacturer MAP pricing so, unfortunately, I must do that as well. For the 250+ pricing, they must be getting better pricing than me.

I don't have Fedex on website but I can do that. With USPS, the shipping would be unrealistic so Fedex is required. You have to do 'request for quote' on website to get Fedex.

Qty 1000 would be difficult at this time. Last time I checked at WDC, they didn't have that many in stock at manufacturer.