Subject: RE: Network throughput and upgrading kernel questions
To: 'Brian Chase' <>
From: Antonio Carlini <>
List: port-vax
Date: 05/03/2003 09:17:18
Brian Chase wrote:
> consideration.  The VS3100's firmware is limited to booting 
> from < 1GB drives.  So you'd need an older drive for your 
> system disk.

This is true, and for OpenVMS it really matters, but I
thought that with NetBSD it was easy to arrange for
the boot to happen from a partition that lives
entirely within the 1.073GB limit?

> For one, the VAX4000/200 uses the same CPU as the VAXstation 
> 3100/m76, which has twice the performance of what's in your 
> m38.

The-76 does have twice the CPU grunt of the -38 (7.6VUPs
vs 3.8VUPS, which might be where the model numbers came from)
but the CPUs aren't as stated above. The -38 (and the -30
before it) used CVAX chips. The VAX 4000-200 used the SOC
and the VS3100-76 squeezed a Rigel CPU chipset onto the
-38 CVAX design. As for the VAX 4000-200 having twice
the performance of the VS3100-38, it depends. Certainly
it was supposed to have much better performance than the
MicroVAX 3600 series (also a CVAX design from roughly the same
time period) but the different cache design and size led
to *worse* performance under some circumstances.

>  I would guess it also has a superior system bus as well 
> because they were designed for I/O intensive server work.  As 
> for how the LANCE compares to the SGEC, I've no clue.  I've 

SGEC was later and better.

> also no idea how sensible DEC's use of the LANCE is in their 
> design.

My understanding is that LANCE was a DEC design that they
gave to AMD (I don't know why, maybe to widen the market
base and drive costs down). So I'd expect the DEC designers
to be reasonably familiar with LANCe and its limitations.


Antonio Carlini