PPU variants

Beyond the well-studied 2C02G, we know of the following PPU revisions, both made by Ricoh and other manufacturers:

Official

Chips officially licensed by Nintendo for use in official consoles and arcade systems.

Composite

---

NTSC

Part	Picture	First Seen	Last Seen	Notes
RP2C02		1983-06 3F4 13	1983-08 3H2 10	Extremely rare. Likely only a few thousand made. http://web.archive.org/web/20160315221802/kitayama3800.publog.jp/archives/cat_915765.html (pictures not archived)]
RP2C02A		1983-08 3H1 43	1983-08 3H2 10	Sometimes erroneous sprite pixels appear in X=255. Some modern PCBs generate almost exclusively glitchy pattern fetches. PPUMASK and PPUCTRL seem to be entirely asynchronous, and writes to PPUMASK can disable rendering for one pixel with the commensurate bugs resulting.
RP2C02B		1984-03 4C3 06	1984-04 4D4 60	Erroneous sprite pixels appear at X=255, just like 2C02A. Production seems to have halted to produce the 2C02C, but 2C02C production stopped and 2C02B production resumed, for unknown reasons.
RP2C02C		1983-12 3M1 10	1984-02 4B1 76	Eventually stopped being produced in favor of resuming 2C02B production, for unknown reasons.
RC2C02C		1984-01 4A3 15		Comes in a ceramic package.[1] Currently only found inside serviced Famicoms.
RP2C02D		1984-08 4H2 60	1984-12 4M2 29
RP2C02D-0		1984-10 4K3 63	1984-12 4M1 25
RP2C02E		1984-12 4M3 14	1985-10 5K4 36
RP2C02E-0		1985-03 5C5 46	1986-06 6F4 39	In this and all previous revisions, OAMDATA and palette RAM are not readable. Various OAM evaluation bugs
RP2C02G-0		1987-05 7E2 80	1993-10 3KM 1H	Writes to OAMADDR cause corruption of OAM. Leaving OAMADDR at a value of 8 or greater causes OAM corruption when rendering starts. Particularly susceptible to reflections causing OAM corruption, fixed by putting series resistors between the CPU and the cartridge ROM.
RP2C02H-0		1993-12 3MM 40	1999-05 9EM 5B	Thought have to have fixed some of the glitches in the previous 2C02G-0 revision(which?).
RP2C02H-0 (laser)		2000-10 0KL 40	2003-01 3AL 4B	Reported (along with PAL PPUs) to have some kind of difference that caused problems with address bus filtering in earlier versions of the Hi-Def NES firmware.

PAL

Part	Picture	First Seen	Last Seen	Notes
RP2C07		1985-12 5M4 26		PAL-B PPU. Vblanking is 71 scanlines long. OAM evaluation can never be fully disabled. Red/green color emphasis swapped. OAMADDR works correctly, not corrupting OAM contents]
RP2C07-0		1987-10 7K3 27	1991-10 1KM 13
RP2C07A-0		1992-06 2FM 22		Some subtle differences in PPU that make this work better with Kevtris's HDNES, but otherwise believed identical to 2C07.

RGB

---

Part	Picture	First Seen	Last Seen	Notes
RP2C03B		1984-02 4B3 59	1989-03 9C4 23	RGB PPU. Color emphasis bits set the corresponding channel to full brightness. Otherwise believed identical to 2C02B. OAMDATA and PPU palette are not readable.
RC2C03B		1984-01 4A2 14	1984-01 4A4 30
RP2C03C		1984-03 4C2 63		RGB PPU. Believed to be same core as 2C02C.
RC2C03C		1984-01 4A2 10		Believed identical to RP2C03C.
RP2C04 0001		1984-03 4C2 01	1984-05 4E4 35	RGB PPU with a scrambled palette and some added colors, for copy protection purposes. Otherwise behaves like 2C03C.
RP2C04 0002		1984-07 4G4 19	1984-10 4K2 35	Scrambles the palette uniquely compared to the previous revision.
RP2C04 0003		1984-10 4K5 13	1984-11 4L4 33	Scrambles the palette uniquely compared to the previous revisions.
RP2C04 0004		1984-11 4L3 18	1984-11 4L5 36	Scrambles the palette uniquely compared to the previous revisions.
RC2C05-01		1985-06 5F5 10	1985-06 5F5 11	PPUCTRL and PPUMASK swap locations. Five LSBs of PPUSTATUS return a constant. Otherwise believed to behave like 2C03.
RC2C05-02				Believed to be the same as 2C05-01, with the PPUSTATUS constant changed.
RC2C05-03		1985-07 5G1 12		Believed to be the same as 2C05-01, with the PPUSTATUS constant changed.
RC2C05-04		1987-03 7C3 12		Believed to be the same as 2C05-01, with the PPUSTATUS constant changed.
RC2C05-99		1988-11 8L4 11	1988-12 8M4 14	An RGB PPU with RP2C02E behavior: OAM and palettes are unreadable, but OAM corruption added in revision E is present. PPUSTATUS behavior is normal (the low 5 bits are PPU open bus). Grayscale is normal (palette & $30). Uses 2C02 timing (a dot is skipped every other frame), unlike other RGB PPUs. The palette and emphasis behavior match other RGB PPUs. PPUCTRL and PPUMASK aren't swapped. Used in the Famicom Titler. (picture of Titler PCBs)

Unofficial

These chips are found exclusively inside of Famiclone systems, made by multiple companies.

Part	Picture	Notes
UA6528		UMC-made clone of 2C02E (or something earlier) [2]
UA6528P		UMC-made variant for PAL-N (Argentina) [3] System crystal is 21492337.5 Hz (exactly 6×229.2516×15625)
UA6538(P??)		UMC-made variant for playing NTSC games in PAL countries. Emits PAL-B video. Vblank interrupt intentionally emitted 50 scanlines later than 2C07. Video DAC is brighter and more saturated (how so?). See also Clock rate.
UA6541		UMC-made clone of 2C07 [4]
UA6548		UMC-made variant for PAL-M (Brazil) System crystal is 21453671… Hz (exactly 6×227.25×4500000÷286)
UM6558		UMC-made variant of UA6538 for SECAM countries. Emits 8-bit "Color Data" digital bus, for conversion into SECAM by UA6559 IC. Color palette noticeably off. System crystal is 21312500 Hz (exactly 4×341×15625). Maybe supports both 50 and 60 Hz operation?
UM6561xx-1	No Picture	UMC-made NES-on-a-chip for NTSC. PPU half believed to be mostly identical to UA6528. Revisions "xx" F, AF, BF, CF known. Emphasis is much stronger than on UA6528 and official PPUs.
UM6561xx-2		UMC-made NES-on-a-chip for PAL-B. PPU half believed to be mostly identical to UA6538. Revisions "xx" F, AF, BF, CF known. AF revision has graphical and timing glitches in Prince of Persia perhaps caused by sprite 0 hit being missed for reasons not yet understood. F and BF revisions are not affected. Emphasis is much stronger than on UA6538 and official PPUs.
TA-02N		??-made die-mask clone of 2C02G, despite the "6528" label.[5] Chip underside also has two codes of currently unknown purpose.
TA-02NP		??-made PPU, Dendy compatible (not a 6538 clone). Pins 14-17 are background color in, like normal 2C03[6]
TA-02NPB		??-made PPU, "NTSC for PAL-B" (NPB) Dendy-compatible. Has a unique video switching capability shared between it and other TA-02NPx chipsets, as well as the WDL chipset.
TA-08		??-made PPU, Second source alternative to UM6558. Color Data bus bit-reversed?
MG-N-502
MG-P-502		Micro Genius / TXC. Die shot matches UA6538.
1818N		??-made NES-on-a-chip, NTSC timing.
1818P		SiS-made[7]] NES-on-a-chip.[8]. Requires external 2KiB RAMs for CPU and PPU. UA6538 timing.
PM02-1		Gradiente-made variant for PAL-M (Brazil). [9]
VT01		V.R.Tech-made clone of UA6561. Only seen as chip-on-board. Supports composite out; RGB out; or 2bpp STN LCD, either greyscale or red/cyan checkerboard, 240 px wide, 80/120/160/240px tall. The chip was extended significantly in VT02 and newer NOACs.
GS87008		(Goldstar??)-made NTSC clone [10]
KC-6078		Found in MT777-DX famiclone, behaves exactly like UA6538
6022		??-made NTSC clone. Details unknown.
2010
2A02E		Dendy Timing. On a large solid-color background (especially yellow, green), every other row has a horizontal stripe.
02
GT-01		In PLCC
HA6538
6538U-8
6538A
6528		WDL made chip. NTSC timing. Palette like RP2C02.
8Z02

If you know of other differences or other revisions, please add them!

See also

• CPU variants
• Clock rate
• NES_2.0#Byte_13_.28Vs._hardware.29
• https://forums.nesdev.org/viewtopic.php?p=150127#p150127