nesdev.org

As topic said, now I have three UM6558 and UM6559 (russia dendy clone )

But there is no any datasheet or pinout data on UM6558
I found some pinout data on the Russian BBS, but there was not much help. Their data are in conflict......

Are you in SECAM land? https://forums.nesdev.com/viewtopic.php ... 70#p100870 implies that the UM6559 part won't be useful otherwise ...

lidnariq wrote:Are you in SECAM land? https://forums.nesdev.com/viewtopic.php ... 70#p100870 implies that the UM6559 part won't be useful otherwise ...

nope. im chinese , recnetly , i found these chipset in other topics, and mentioned this can be modifyed to rgbs singal output.

but I can't find the UM6558 data only ， UM6559 Already found !

HardWareMan might be able/willing to help you, whenever he stops by again. Maybe.

Otherwise, all I can do is suggest how you would reverse-engineer the pinout using a multimeter. I wouldn't be surprised if the pinout is almost the same as the 2C02's, just with the extra eight pins at the bottom forming the color data bus.

As you've probably already found, it looks like there's a serious design flaw with the colors $1E, $2E, $3E, $1F, $2F, and $3F not being black.

lidnariq wrote:HardWareMan might be able/willing to help you, whenever he stops by again. Maybe.

Otherwise, all I can do is suggest how you would reverse-engineer the pinout using a multimeter. I wouldn't be surprised if the pinout is almost the same as the 2C02's, just with the extra eight pins at the bottom forming the color data bus.

As you've probably already found, it looks like there's a serious design flaw with the colors $1E, $2E, $3E, $1F, $2F, and $3F not being black.

ah , thank you for your reply.

I have already used the resistance value to analyze its pinout,

Combined with the real PCB picture found it really and 2c02 or 6528/6538 differents.

I have noticed its palette problem, but I think this is not a big problem..

The biggest problem is that CPLD PPU is not easy to buy in China, then FPGA PPU does not have a chassis that looks good.

a VS arcade 2C03 and too expensive, only UM6558 / 6559 chipset each set only 50 CNY , It is an almost the cheapest solution of famicom rgbs mod...

chipset x 3 = including demosteic shipping fee = ca.18USD$

bestmmk wrote:I have already used the resistance value to analyze its pinout,

Combined with the real PCB picture found it really and 2c02 or 6528/6538 differents.

I'm not clear: have you figured this out? or do you still need help?

lidnariq wrote:

bestmmk wrote:I have already used the resistance value to analyze its pinout,

Combined with the real PCB picture found it really and 2c02 or 6528/6538 differents.

I'm not clear: have you figured this out? or do you still need help?

NO _(:з)∠)_ i just already found VCC and GND ， and some pin linked with cpu， but i also need help

it's better to find the datasheet for UM6558 ，

If find it can be used directly

in other， I found these in the Russian Forum

1)UM6558 AND UM6538 DIFFERENT ,
I guess it may be similar to UM6538, but I don't understand the text of the picture


2)REAL MACHINE PCB PICTURE FROM DENDY CLONE
I'm still looking at this picture. But there's a lot of doubt...


3)UM6559 Application diagram and pinout
and finally , i find the um6559 in russian forum. it's high-reso scan, total 2pages

In the image with nominal pinout of the 6558, those are the instructions to replace a 6558 with a 6538.

So, undoing it ... the lines about 10, 11, 12, and 13 say that, unlike the 2C02 or UA6538, the pin order is not CPUA2, CPUA1, CPUA0, /PPUCE, but instead /PPUCE, CPUA2, CPUA1, CPUA0.

The little red text similarly says that pins 18, 19, 20 are (in order) +5, the master clock, and /VBL

The only thing that's not clear is what pins 29 (labeled as 21) and 30 (labeled as 22) need to become.

I've seen a vague insinuation somewhere else that some pin controls whether the UM6558 emits 60Hz or 50Hz video.

lidnariq wrote:In the image with nominal pinout of the 6558, those are the instructions to replace a 6558 with a 6538.

So, undoing it ... the lines about 10, 11, 12, and 13 say that, unlike the 2C02 or UA6538, the pin order is not CPUA2, CPUA1, CPUA0, /PPUCE, but instead /PPUCE, CPUA2, CPUA1, CPUA0.

The little red text similarly says that pins 18, 19, 20 are (in order) +5, the master clock, and /VBL

The only thing that's not clear is what pins 29 (labeled as 21) and 30 (labeled as 22) need to become.

I've seen a vague insinuation somewhere else that some pin controls whether the UM6558 emits 60Hz or 50Hz video.

the um6558 may need a separate oscillator ， or common use NTSC oscillator ， i think NTSC (21.4772) oscillator and SCEAM (21.3125) oscillator are closer.


Now I have a HVC-CPU-07 PCB , I will start this test, I have 3 PPU and the 3 chance _(:з)∠)_

dont close the topic ， There will be a final result in recent weeks

Where did they get UM6559 datasheets from? These are unavailable in the whole internet, but looks like official ones.

in a russian forum but I forgot where it was _(:з)∠)_
hereis pinout！ finally i finished！

it tested by 孙大师(a.k.a master) and confirmed it works with paltte problem。

Maybe you can put some CPLD logic between UM6558/UM6559 on CD0-7 lines, which changes problematic dot colours in the fly to the correct ones, that don't cause troubles?

I have to admit I'm really curious what the contents of the ColorData bus are, for no good reason.

lidnariq wrote:I have to admit I'm really curious what the contents of the ColorData bus are, for no good reason.

CD0-CD7 should be color data of graphics. This requires a logical analysis. If the palette problem is corrected, it will be of great value

krzysiobal wrote:Maybe you can put some CPLD logic between UM6558/UM6559 on CD0-7 lines, which changes problematic dot colours in the fly to the correct ones, that don't cause troubles?

Haha, I think so too. In fact, it's too hard for me, I need more help, but now the situation has surprised me . Only a few people in China know the existence of this chipset.