Help collecting pictures of CPU and PPU packages

Discuss hardware-related topics, such as development cartridges, CopyNES, PowerPak, EPROMs, or whatever.

Moderator: Moderators

loglow
Posts: 48
Joined: Mon Sep 19, 2022 1:47 pm

Re: Help collecting pictures of CPU and PPU packages

Post by loglow »

I borrowed a microscope!

Please forgive the slightly scattershot nature of this assortment, since I was hesitant to sacrifice more than a few working chips.

First, the clones...

Unknown CPU with /16 CLK DIV

This was a working CPU with a verified /16 clock divider that had been re-marked via laser as a "UA6527" which it obviously wasn't because it didn't have a /12 clock divider. I had assumed it was an older-style UA6527P, but the die markings make me wonder if it wasn't a UMC chip at all though!? They're unlike any other that I observed, and they were on the very very edge of the die:

UA6527P CLK DIV 16.jpg

Code: Select all

+HN1225002A
*M* HMC
AA AAA
Likely UA6527P with /15 CLK DIV

This was a working CPU with a verified /15 clock divider that had also been re-marked "UA6527" but obviously wasn't for the same reason as above. I suspect that it is actually a later-style UA6527P:

UA6527P CLK DIV 15.jpg

Code: Select all

UMC C M
B6167F
1989.09
UA6540 8833S

This was a working CPU with a verified /16 clock divider and what appeared to be original chip markings that said "UA6540 8833S" on the top. I was curious if this would end up being the same die as a /16 UA6527P. I'll need to crack open a non re-marked UA6527P to check for sure.

UA6540 8833S.jpg

Code: Select all

UMC
1987
C M
TU6142
UA6528 9114-CS 309560

I wanted to try a UA6528 from one of the revisions that has constantly been non-working for me, with all kinds of crazy video artifacts, mostly because I was curious if the dies between working and non-working ones would be different at all. So this was what appeared to be a UA6528 with original chip markings and a "CS" designation:

UA6528 9114-CS 309560.jpg

Code: Select all

UMC
TU6133F
C M
1988
UA6528 9223-CA 221B12

This was a working UA6528 chip with apparently original markings and a "CA" designation. As you can see, it did in fact have exactly the same die as the "CS" version above, so the difference in reliability/functionality is not related to the die itself.

UA6528 9223-CA 221B12.jpg

Code: Select all

UMC
TU6133F
C M
1988
loglow
Posts: 48
Joined: Mon Sep 19, 2022 1:47 pm

Re: Help collecting pictures of CPU and PPU packages

Post by loglow »

Now, the genuine chips...

Re-marked glitchy RP2A03 chips

These three dies were from a batch of sketchy "refurbished" obviously re-marked CPUs that all said "RP2A03 3H4 26" on them. Most of the CPUs in this batch had very similar sprite issues, where there would be no sprites visible at all from the start, or they would all start flickering and then disappear after 10-60 seconds or so. Super weird. One or two didn't work at all, and several appeared to work okay (though I'm obviously still skeptical about them).

RP2A03 3H4 26.jpg

Code: Select all

RP2A03
2 71 1
0 2
D DF F

RP2A03
2 7111
012
D DCCF

RP2A03
2 7111
012
D DCCF
It's interesting how one of them has that middle column of characters blocked off with a single box. Also, there appear to be faint characters before the 7s (looks like a 6) and the Ds (looks like a C or G).

Re-marked faulty RP2A03H chips

These were from a different batch of also obviously re-marked CPUs that all said "RP2A03H 3MM 43" on them in laser-engraved text. Most of them worked fine however, although these two did not. I'm guessing these are not "H" revision chips at all, based on the die markings, and are actually RP2A03G revision chips?

RP2A03H 3MM 43.jpg

Code: Select all

G2A03
2 711
02
D DCF

G2A03
2 711
02
D DCF

Genuine RP2C02G-0

This was a genuine desoldered RP2C02G-0 that did not work at all after desoldering. It may have stopped working before being desoldered also! Unfortunately, I accidentally didn't write down the lot number from the top of this chip. Oops!

RP2C02G-0 LOT UNKNOWN.jpg

Code: Select all

G2C02
11
12 702
3 JCF
Some of the characters were hard to make out, so this is just my best interpretation.

That's all of my de-capping results for the time being! I hope some of this was helpful or at least mildly interesting to someone out there. I should also be getting some much higher quality photos of the entirety of the recovered dies soon (silicon damage, residual epoxy, and all), so it'll be interesting (at least for me) to compare them and see if I missed anything. I'd be happy to share them here too, if and when I get them.
User avatar
Quietust
Posts: 1918
Joined: Sun Sep 19, 2004 10:59 pm
Contact:

Re: Help collecting pictures of CPU and PPU packages

Post by Quietust »

loglow wrote: Sun Oct 09, 2022 4:19 am It's interesting how one of them has that middle column of characters blocked off with a single box. Also, there appear to be faint characters before the 7s (looks like a 6) and the Ds (looks like a C or G).
Each column represents a layer during the fabrication process - 2 is Diffusion, the invisible one is Depletion Implant, 6 is Buried Contacts, 7 is Polysilicon, 10 is Metal Contacts (i.e. vias), 11 might be a second Metal Contacts layer (e.g. one for connections to Diffusion, another one for connections to Polysilicon, with them being merged together in later revisions), 12 is Metal, and 14 (the solid box on the right - if you look close enough, you'll be able to see letters and numbers underneath it) is overglass.

loglow wrote: Sun Oct 09, 2022 4:19 am Re-marked faulty RP2A03H chips

These were from a different batch of also obviously re-marked CPUs that all said "RP2A03H 3MM 43" on them in laser-engraved text. Most of them worked fine however, although these two did not. I'm guessing these are not "H" revision chips at all, based on the die markings, and are actually RP2A03G revision chips?


RP2A03H 3MM 43.jpg

Code: Select all

G2A03
2 711
02
D DCF

G2A03
2 711
02
D DCF
If they say "G2A03", then they're almost definitely G-revision chips.
Exactly where on the die were those markers located? On the 2A03G it's top-center (immediately to the right of the 2 audio output pins), and from what I've seen on a (very low-resolution) 2A03H die it's definitely not there.
Quietust, QMT Productions
P.S. If you don't get this note, let me know and I'll write you another.
lidnariq
Posts: 11430
Joined: Sun Apr 13, 2008 11:12 am

Re: Help collecting pictures of CPU and PPU packages

Post by lidnariq »

loglow
Posts: 48
Joined: Mon Sep 19, 2022 1:47 pm

Re: Help collecting pictures of CPU and PPU packages

Post by loglow »

Hiya, I added some new photos to the PPU wiki page!

A full chip photo of an RP2C03B. There was only a center-crop image of one before.

PPU=RP2C03B 9C4 23.jpg

A new full chip photo of an RC2C03C. Previously no photo at all.

PPU=RC2C03C 4A2 10.jpg

A new full chip photo of a UA6538. Previously no photo at all.

PPU=UA6538 9214-BA 221530.jpg

The name column next to this last one currently reads "UA6538(P??)" which is kinda weird. I've never run across a UA6538P before. I'm not claiming that it doesn't exist, but I'm not claiming that it does exist either. I didn't change the name because I didn't want to step on anyone else's toes, but I do think it should probably be changed, or at least cleaned up somewhat.
User avatar
krzysiobal
Posts: 1036
Joined: Sun Jun 12, 2011 12:06 pm
Location: Poland
Contact:

Re: Help collecting pictures of CPU and PPU packages

Post by krzysiobal »

In behaviour (M2 divider, M2 duty) identical to regular UA6527P:
seniton 6527UP.jpg
I've got also 3 UA6527P with -BW after date code. None of them works, even no M2 is output.
Either all are broken or what is more probable, there is something subtle with their clocking circuit, as the clock from famicom/famiclones is really very poor quality, sometimes if you replace the big crystal with smaller one, normally working cpu also stops working.
WIll need to investigate it.
ua6527p-bw.jpg
Image My website: http://krzysiobal.com | Image My NES/FC flashcart: http://krzysiocart.com
User avatar
org
Posts: 155
Joined: Tue Aug 07, 2012 12:27 pm

Re: Help collecting pictures of CPU and PPU packages

Post by org »

GM Chipset.

Details are unknown. Going to decap to see what's inside :)
Attachments
IMG_8715.jpg
User avatar
org
Posts: 155
Joined: Tue Aug 07, 2012 12:27 pm

Re: Help collecting pictures of CPU and PPU packages

Post by org »

Ricoh "E" Revision.
Attachments
Ricoh_E_Revision.jpg
User avatar
org
Posts: 155
Joined: Tue Aug 07, 2012 12:27 pm

Re: Help collecting pictures of CPU and PPU packages

Post by org »

UA6548 :shock:
Attachments
IMG_8717.jpg
loglow
Posts: 48
Joined: Mon Sep 19, 2022 1:47 pm

Re: Help collecting pictures of CPU and PPU packages

Post by loglow »

org wrote: Sat Jan 14, 2023 6:57 am GM Chipset.

Details are unknown. Going to decap to see what's inside :)
My wild guess would be:

GM-6827 = UA6527P
GM-6838 = UA6538

...just based on the kinda similar part number scheme. But now I'm curious to see!
loglow
Posts: 48
Joined: Mon Sep 19, 2022 1:47 pm

Re: Help collecting pictures of CPU and PPU packages

Post by loglow »

org wrote: Sat Jan 14, 2023 6:58 am Ricoh "E" Revision.
I've found a non "-0" E revision chip like this under every "2C02E" stamped heat spreader so far. I've pulled off and cleaned up maybe 10 or so. I think I've only ever seen one or two non "-0" E chips without heat spreaders.
loglow
Posts: 48
Joined: Mon Sep 19, 2022 1:47 pm

Re: Help collecting pictures of CPU and PPU packages

Post by loglow »

org wrote: Sat Jan 14, 2023 9:01 amUA6548 :shock:
PAL-M PPU I think
User avatar
Individualised
Posts: 309
Joined: Mon Sep 05, 2022 6:46 am

Re: Help collecting pictures of CPU and PPU packages

Post by Individualised »

org wrote: Sat Jan 14, 2023 6:57 am GM Chipset.

Details are unknown. Going to decap to see what's inside :)
Gradiente?

Are all of these "non-UMC chipsets that act exactly like UMC chipsets" clones or were they officially licenced from UMC?
User avatar
forple
Posts: 29
Joined: Sat Dec 30, 2017 2:09 pm

Re: Help collecting pictures of CPU and PPU packages

Post by forple »

Got better pictures of both the TA-02N and the TA-03N chips (front and back), feel free to use as desired.

TA-02N:
TA-02N
TA-02N
TA-02N Bottom
TA-02N Bottom
TA-03N:
TA-03N
TA-03N
TA-03N Bottom
TA-03N Bottom
With a bit of testing, These chips are likely die-mask clones of G-rev Ricoh chips. This means the 6527/6528 identifier is, well, either purely cosmetic or a theory suggests it's to let clone manufacturers know that these parts are perfectly compatible with the UMC equivalent (the 652x chips).
loglow
Posts: 48
Joined: Mon Sep 19, 2022 1:47 pm

Re: Help collecting pictures of CPU and PPU packages

Post by loglow »

I've found a number of new (to me) chips that I know virtually nothing about.

I don't have any of these myself, I've only found pictures of them. Any info is appreciated. I'm trying to get my hands on a few if possible.

SENiTON 6527AP

All of them appear to have the yin-yang logo and brand name followed by a likely date code (1990 or 1991 plus week) then a C or A. Some have an additional 4 digit code beneath (eg. "CGG4", "CGD6", etc.) The one with "CFAKC" is an outlier as it appears to have the positions of this extra code and its date code reversed.

seniton_6527AP.png

SENiTON 6527UP-8

All appear to have the same markings with no date or fab codes (could be on the bottom, I suppose). I thought the last one might be an S instead of an 8 but after closer inspection I think it's actually an 8 like the others. The top two are two different photos of the same chip.

Edit: Oops! I just noticed that krzysiobal already posted about the 6527UP-8 just a few posts ago.

seniton_6527UP-8.png

SENiTON 6527P-SS-P03

Only came across one of these. Doesn't have the same SENiTON logo, nor any other codes on the top side.

seniton_6527P-SS-P03.png

SENiTON 6538A

Similar marking layout to the 6527AP, date codes (1990 or 1991 plus week) followed by D or C. Some have the 4 digit code beneath, some don't.

seniton_6538A.png

SENiTON 6538U-8

Markings just like the 6527UP-8, no other codes on top.

seniton_6538U-8.png

Continued in next post...
Last edited by loglow on Thu Mar 09, 2023 10:52 pm, edited 1 time in total.
Post Reply