Looking for TA-03NP/1 clone CPU

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

Moderator: Moderators

Post Reply
User avatar
ImATrackMan
Posts: 37
Joined: Thu Mar 26, 2015 6:34 pm
Contact:

Looking for TA-03NP/1 clone CPU

Post by ImATrackMan »

Recently, a buddy in Poland sold me a Pegasus IQ-502 and whatever other common famiclone was available sround Poland (the 777 one I think). I was really only interested in the Pegasus, so the 777 is only conna be around for parts. Both have clone parts from 1992, so both have the bug where the APU1 duty cycles are swapped.

The Pegasus, strangely, uses a transistor to amplify the audio going to the cart slot, and has incorrect mixing restors on the APU outputs causing massive distortion. I modified the audio circuit by removing the transistor and wiring the audio path to the 40H368 (U7) used by the FC to amplify audio and a PPU signal (this was also faulty due to video signal leakage likely through the ground shared by the gates)

Since the audio circuit is now exactly the same as the original FC one (and actually better, there’s NO power hum coming through the cart slot), I’m trying to track down a TA-03NP family CPU. I hear pretty consistently that these are totally separate Korean clones, not based on the 6527/P and don’t have the duty cycle error. I was able to find one on ebay, but it was only available attached to a very strange clone on a daughterboard for an unknown Goldstar product in Australia, and it would have cost me $60. I’m not really interested in paying $60 for a clone CPU so I figured I’d ask you all.
User avatar
Eugene.S
Posts: 317
Joined: Sat Apr 18, 2009 4:36 am
Location: UTC+3
Contact:

Re: Looking for TA-03NP/1 clone CPU

Post by Eugene.S »

these are totally separate Korean clones, not based on the 6527/P and don’t have the duty cycle error.
If you have 6538 ppu, you need to find TA-03NP1 6527P for correct timing
It doesnt have duty cycles bug

http://hwm.us.to/famimusic/Forum/pictures/img6211t.jpg
User avatar
ImATrackMan
Posts: 37
Joined: Thu Mar 26, 2015 6:34 pm
Contact:

Re: Looking for TA-03NP/1 clone CPU

Post by ImATrackMan »

Ultimately, revision doesn’t matter. I’d obviously prefer to get my hands on an NP1-6527P, but any other version would help me get the timeline down for when UMC got the clone formula right. The reason I think they’re totally separate clones is, despite them being occasionally used with UMC parts, they’re MOSTLY used with a TA PPU on boards populated entirely by Goldstar parts. That, and the fact that the two clones I have sound completely different despite having UMC chips made within 7 weeks of the TA-03NP1/6527P in that picture.

Ebay UA6527P - 8848S, actually just a re-marked 6527 NTSC CPU, bad duty, bad DMC, distorted output
MG-P-101 and UA6527P - 9221, bad duty, good DMC, nearly identical to 2A03 output amplitude

Ultimately, anything helps.
User avatar
krzysiobal
Posts: 1037
Joined: Sun Jun 12, 2011 12:06 pm
Location: Poland
Contact:

Re: Looking for TA-03NP/1 clone CPU

Post by krzysiobal »

I really dont understand your excitement about TA-**** chips.
I have one CPU+PPU of them for research purposes but every time I was fixing DIL40 famiclones, I preferred to use UA6527P/UA6528.
TA CPU (even after fixing the issues with incorrect resistor values) outputs still quite weird (metallic-alike) sound. It must work in pair with TA PPU, otherwise some games (like Micro Machines) glitches and hangs in a few seconds after run.
But TA PPU has real shit (faded) colours.

Maybe I am deaf but I don't hear the difference between sound from normal and swapped duty cycle CPU.
User avatar
ImATrackMan
Posts: 37
Joined: Thu Mar 26, 2015 6:34 pm
Contact:

Re: Looking for TA-03NP/1 clone CPU

Post by ImATrackMan »

My reason for wanting one is NSF playback. Since the pich is so similar to NTSC chips, all it takes in software like Famitracker is to set the clock speed to 50 Hz. I also just enjoy the novelty of owning high quality fakes, and since I essentially collect sound chips, a proper hybrid clone of the 2A03 is of big interest to me.
User avatar
krzysiobal
Posts: 1037
Joined: Sun Jun 12, 2011 12:06 pm
Location: Poland
Contact:

Re: Looking for TA-03NP/1 clone CPU

Post by krzysiobal »

You can make a combinatoric circuit and put it between CPU and memory which restores original duty cycles.
tepples
Posts: 22708
Joined: Sun Sep 19, 2004 11:12 pm
Location: NE Indiana, USA (NTSC)
Contact:

Re: Looking for TA-03NP/1 clone CPU

Post by tepples »

I have no idea how "between CPU and memory" would work. In a write to the affected ports ($4000 and $4004), the 6502 core's accumulator output is connected within the chip to the APU register that holds the duty. Unlike with PPU interposers, such as NESRGB, Hi-Def NES, and the unswap circuit to use a 2C05 with NES software, there's no place to insert your own hardware to swap bits 7 and 6. The one exception is if you reimplement the whole pulse 1/2 side of the APU on a separate chip, in much the same way kevtris did for Hi-Def NES.
Post Reply