Re: MACH chip question

From: Magnus Kollberg <>
Date: Tue, 23 Jun 1998 09:58:03 +0200


> >You said that the Centurbo 2 accesed the motherboard in 4 cycles and that
> it should
> >be faster. The difference is not very big if you are not only reading/writing
> >16-bit words. A blocktransfer should be much the same.
> >
> Whatever the size of the transfert on the 040 bus is, it reads (or write) a
> 32 bits word. So if you make a move.b (a0)+,(a1)+ you've got 2 32 bits
> accesses.

My assembler isn't what it should be, but is the above the same as 2 16 bits
access on the 030 or 4 16 bits access? Are you moving 32 bits or 64 bits?
> >Also you said that the Cent2 could handle a transfer at 14MB/s to ST-RAM. How
> >would that be possible when a normal Falcon can only do 5MB? It does also
> >differn depending on which operation you do.
> The CT2 boost the falcon bus speed to 25 Mhz so the bus transfert speed
> grow exponentialy.

Nice and this works on all Falcons?

> >Have you tried Nembench on the Cent2?
> Yes (not me, but Centek does) and it geave 14 Mb/s for STRam transfert.

Not to bad.

> Yes there are less cache troubles on a 030 than on the 040 with the TOS
> programs.

But I havn't realy found many programs which have these problems so for me it's
not a valid argument. The 16-bit bus is a bigger one.

> >Have you looked at this?
> Not yet. but i will take one week of holidays in about 3 week and take
> some time to look for this and continue my exploration of the AB040
> expansion bus pinout :-)

And optimize memory and ST-RAM access. ;-)
> >It would be interresting. It would also be interresting to know what is
> >limiting the AB040 from using more than 64MB RAM. Maybe they didn't add logic
> >for decoding all address lines?
> I think that is the problem : no more free I/O pin on the Mach

To bad.

> >If you need an idea for a great project, it would be very nice with a PCI
> >interface connected to this 32-bit bus.... ;-)
> Yes that's why i want the AB040 bus pinout :-))))

Yummy. A 060 and 32-bits PCI! and we wont need to access the ST-RAM anymore.


//Magnus Kollberg
Received on ti. juni 23 1998 - 20:41:00 CEST

This archive was generated by hypermail 2.3.0 : ti. nov. 03 2015 - 20:07:54 CET