Re: Theorizing: chip select lines as clock speed selectors in a overclocked 6510 on a C64

From: silverdr_at_srebrnysen.com Date: Wed, 29 Jun 2022 15:58:44 +0200 Message-Id: <51B881C4-0F32-4236-84D3-315A7CE20044_at_srebrnysen.com>

> On 23/06/2022 19:24, silverdr_at_srebrnysen.com wrote:
> 
>> I had once a discussion with Jens (yes,*that*  Jens 😉 about relation between DOTCLOCK and "the rest of the chips of the system" and he claimed that using DOTCLOCK for basically anything, trusting it to be in sync with the rest is a no go... Although we haven't experienced any of the issues he said he had when trying to use DOTCLOCK, I have no reasons to distrust his words and take that we were simply lucky.
> 
> On 2022-06-29, at 15:33, smf <smf_at_null.net> wrote:
> 
> Well you certainly can't know for sure that the VIC2 will count the same
> number of dotclock transitions at power up, as you do.

Certainly. The discussion wasn't about the absolute number of cycles/transitions though. It was about relation between DOTCLOCK and "the rest". We wanted to use DOTCLOCK directly in relation to other clocks and everything seemed to work fine. Stiil, his strong advice was not to rely on DOTCLOCK keeping its relation to other ones, which was/is crucial for proper BR operations.

-- 
SD!