Fwd: Re: 060 adapter pic

From: Jo Even Skarstein <joska_at_nuts.edu>
Date: 20 Jan 99 23:28:29 +0000

Some of you might find this mail from Michael Grove interesting...


/*
** Jo Even Skarstein http://www.stud.ntnu.no/~josk/
**
** beer - maria mckee - atari falcon - babylon 5
*/
---------- Forwarded message -----------
From: Michael Grove <mgrove_at_iwvisp.com>
To: Jo Even Skarstein <joska_at_nuts.edu>



Jo Even Skarstein wrote:
>
> On Tue, 19 Jan 1999 20:42:46 +0000, M Grove wrote:
>
> > Well, I didn't get a Fuji, but the machine didn't burn-up :(.
> > It apears that there is no system clock. After re-reading the 060
> > application notes, it looks like the adapter doesn't setup the pinout
> > for the 060 to allow the bus to run at 1/2 speed. There also needs to be
>
> This shouldn't be too hard, or what?
>
> > a 5ns delay inserted in the BCLK (clock in) to the 060 as it fires off
> > the leading edge of the clock signal, and the 040 fires off the falling
>
> Wouldn't it be simpler to just invert the clock?

I need to look at this closer to understand what is going on. I
haven't looked at the field service manual in quite some time.
Maybe you can explain it in a nut shell;). Doesn't the Falcon
come up at 16/16Mhz (cpu/bus), then (magic happens) and the A/B
switches to the 32Mhz clock sometime after bootup? The Mighty
Sonic is controlled by software to do this, but I guess the A/B
does this with logic??? Sooo, at bootup the A/B is getting the
16Mhz clock from the expansion bus, then picks off the main
(32Mhz) clock at some startup point. If that is true, I see one
problem. At startup, the 060 would have to be set to run at bus
speed, then when the change over takes place, the 060 would have
to set to run with bus at 1/2 CLKEN, as well as offsetting the
CLK (5ns) for the CLKEN.

Does the A/B get the 32Mhz clock from the ribbon cable (as the
Mighty Sonic). I have never really tried to see how the A/B
works.

Something else I just noticed looking at the 060 application
sheet is I don't think that the CLKEN is connected with the
adapter I bought. That would be a problem:). The unit is at home
and I will have to take a look later.



>
> > edge. There is no floppy access, the IDE drives do not come up. It looks
> > just like a bad buffer installation. I am pretty sure that is the
> > problem, and after the guy posting about the Milan said the adapter
> > takes care of some timing issues, it makes sence. Now to see if I am
>
> What did you pay for the 060+adapter? The Milan 060-card is rumoured to cost
> about 250USD, perhaps it's easier and (almost) as cheap to buy this? And you
> get a L2-cache as well.

The adapter cost $250.00 for a single unit. If the Milan adapter
could be had, it would be simpler if the same changes to get it
working on the Milan are also what is required to adjust the
timing for the A/B. Isn't the Milan a 25Mhz CPU? If so, then they
surely run the bus at the same speed, which might cause problems
with the A/B switching to 1/2 bus after bootup. Not sure here, I
am no expert. I am going to call the manufacture of the adapter I
purchased, and ask a few questions. Once I know for sure what the
adapter does (there was no documentation) I will call Steve at
Wizztronics. They may not ever release anything, but the guy is
pretty swift with this sort of thing.

The L2 cache would be a nice feature. My HBS T-36 36Mhz 68K
really wakes up with it enabled. However I have a DiiMO 50Mhz 040
card for my Mac with L2 cache, and performance increase is
minimal with it enabled. Not sure if it is just because it's a
Mac or what:).
>
If you have some time, please look over the 060 application
notes, specifically section 11 of document 68060UM.PDF (the 68060
User's Manual). It's almost over my head at some points as I am
not an electronics type, just a hacker with good luck so far;).

Mike
Received on to. jan. 21 1999 - 00:28:29 CET

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