Re: [NTLK] Changing the SA-110 Core Clock Frequency

From: Chuck England III (cengland10_at_cox.net)
Date: Mon May 03 2004 - 10:39:05 PDT


I would be happy if I could get the thing up to 200 Mhz.
 without adding any components.

Does anyone know what the exact model is, of the SA-110
 in the Newton 2Kx machines? I still cant seem to find
 a Pin out paper for it, or the programming matrix for
 changing the clock speed. A complete data sheet on the
 model of processor in the 2kx machines would be helpful.

P.S. Thanks for replying Julian, I was beginning to think
 no one had read my original message, what with all the
 OT messages recently.

Chuck E.

On 5/3/04 10:56 AM, "Julian Wright" <julianwr_at_paradise.net.nz> wrote:

> Chuck England III wrote:
>
>> I was looking through some of Apples docs and some docs on the Intel
>> website for the SA-110 processor, The information indicates that
>> there is a 4 line matrix that controls the Core clock frequency.
>> In the Newton, they have two of those lines already set and a
>> surface mount resistor at either of the other two lines sets
>> the core clock at either 100 or 160 MHZ. From what I was reading
>> other combinations could produce speeds up to 233Mhz
>>
>>
> You're quite right about the intention of these lines. I looked into it
> a while ago, and the real problem is that while the interface for the
> SA110 was well defined right from the start, it wasn't always fully
> implemented. Specifically there have been several different revisions of
> the SA110, some of which don't support higher clock multipliers than the
> one that produces 160MHz. This seems to be the variation used in
> MP2x00s. What you will need to do is remove your SA110 and replace it
> with a later revision, which supports speeds up to 287MHz or more.
>
> Though it has to be noted that even if you find the correct revision of
> SA110, that particular chip itself may not be capable of operating
> reliably at top speed. It may be necessary to experiment with slightly
> lower than maximum multipliers to get a reliable result. You should be
> able to get better than 233MHz though. But for most people, replacing
> the SA110 may be the insurmountable stumbling block.
>
> Somebody please feel free to prove me wrong about the revision of SA110
> in the Newt!

-- 
This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries
Official Newton FAQ: http://www.chuma.org/newton/faq/
WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/


This archive was generated by hypermail 2.1.5 : Thu May 06 2004 - 13:39:32 PDT