Re: Disabling memory refresh in UltiMax mode Re: 6510 handling of $00 and $01 registers

From: Gerrit Heitsch <gerrit_at_laosinh.s.bawue.de>
Date: Sat, 10 Dec 2011 19:36:00 +0100
Message-ID: <4EE3A690.2070304@laosinh.s.bawue.de>
On 12/09/2011 03:18 AM, Segher Boessenkool wrote:
>>>> If I remember correctly, I sent the chip to be read. It was from my
>>>> oldest C64, serial number 32xxx, with ceramic DRAM chips (350ns IIRC)
>>>> and 6569R1 and so on.
>>>
>>> An antique!
>>
>> Not only that, but the 6569R1 can be detected by software, at least
>> this program can do it:
>>
>> http://noname.c64.org/csdb/release/?id=89406&show=summary
>>
>> With a 6569R3 or newer, it will just show '6569', but is somehow able
>> to detect the R1 and list it. It is also quite good at detecting the
>> rest of the hardware. With some I have no idea how it does it though.
>
> Excellent fun! It sees if the light pen interrupt can be triggered
> between raster lines x'136 and 2; if so, it's a 6569R1, if not, not.

Did they change the video timing or just the legal lines an LP IRQ can 
be triggered? On my 1084 monitor, the welcome screen looks OK no matter 
what VIC I use.


> The discrete vs. CMOS core logic thing seems to check whether flipping
> the VIC address space (VA15,VA14) glitches (doesn't on the CMOS).

In what way does the old PLA design cause a glitch?


> 6526 vs. 8521 CIA checks exactly how long it takes a timer to trigger
> (8521 is a clock cycle earlier it seems).

Hm? From what I read everywhere, the HMOS-CIA (8521 or 6526 after week 
4x in 1986) takes an extra cycle with the timer IRQ compared to the NMOS 
6526.


>> Demos written for the 6569R3 or higher look quite odd on the R1.
>
> Because of the luminance palette difference, I take it? Or is there
> some more awesome difference :-)

When testing a C64, I use the type check program mentioned above plus a 
RAM test and the 'Edge of Disgrace' demo. The latter has a few 'plasma' 
effects and on those you notice that they were written for a 9-luma VIC. 
On the 6569R1 they look less smooth.

To bad no one has yet been able to find a software trick to distinguish 
between the 6510 and the 8500. :)

  Gerrit


       Message was sent through the cbm-hackers mailing list
Received on 2011-12-10 19:00:03

Archive generated by hypermail 2.2.0.