[rescue] Re: Ross Hypersparc

Fri Jan 10 11:09:33 CST 2003

Well, I don't have any hard and fast details.  The extent of my comparison
was between an HS166/512K and an HS180/512K on a SS20.

The extent of the anlysis was in boot/xwindows load time, and in overall feel.

The time to boot and launch xwindows for login was measurably smaller on the
HS166 by (going by memory) 10-20% faster than the HS180.

The rest is impressions of using.  This difference is more suttle and I never
made any realy measurements of anything, however the HS166 generally fealt a
bit faster.... but I never did measure the difference.

Around the time of those tests I learned more about the architecture of the
Ross modules.... and while the 180 and 200 modules architecually are an
improvement (for the most part) over preceeding modules, the half speed cache
is their weak point.  For cache heavy stuff, a 180 is only faster than a 90
due to it's architecture improvements.

I've never been a fond proponent of half speed caches (read: hs180, hs200,
pentium II, and pentium III (non E).  It's like putting cheap tires on a
Ferrari.... (note, I've never done that... and I don't own a Ferrari :-)
although dreams of a 308GT linger in my mind).

I wish I had something more statistical to give you all, but it's only
impressions and boot/OS/X load time testing.

I estimate that on average a HS180 falls in performance roughly between
an HS150 and an HS166 and probably closer to the HS150 side... I personally
would take an HS150 over a HS180 any day.  As for the HS200, I've never had
the opportunity to even put my eyes on one, never mind try one.  I'd estimate
that a 200 falls close to a HS166 in performance.

Now, with ideal situations for those processors, they probably do run faster
than what I claim... however the real world usually falls far short of ideal.

As always YMMV,

  -- Curt

>X-Original-To: rescue at sunhelp.org
>Delivered-To: rescue at sunhelp.org
>From: <wrmartin at verizon.net>
>To: The Rescue List <rescue at sunhelp.org>
>Date: Thu, 9 Jan 2003 23:49:00 -0500
>MIME-Version: 1.0
>Content-Transfer-Encoding: 7bit
>X-Authentication-Info: Submitted using SMTP AUTH LOGIN at out001.verizon.net 
from [192.168.129.213] at Thu, 9 Jan 2003 22:49:00 -0600
>Subject: [rescue] Re: Ross Hypersparc
>X-BeenThere: rescue at sunhelp.org
>X-Mailman-Version: 2.1
>List-Id: The Rescue List <rescue.sunhelp.org>
>List-Unsubscribe: <http://www.sunhelp.org/mailman/listinfo/rescue>, 
<mailto:rescue-request at sunhelp.org?subject=unsubscribe>
>List-Archive: </pipermail>
>List-Post: <mailto:rescue at sunhelp.org>
>List-Help: <mailto:rescue-request at sunhelp.org?subject=help>
>List-Subscribe: <http://www.sunhelp.org/mailman/listinfo/rescue>, 
<mailto:rescue-request at sunhelp.org?subject=subscribe>
>
>Having just gotten a 180, I'll be interested in this also.  Hopefully (when 
things calm down) I'll have a chance to compare it to my 150.  Should be 
interesting.
>
>Bill Martin
>> 
>> From: nick at snowman.net
>> Date: 2003/01/09 Thu PM 11:42:41 EST
>> To: "Curtis H. Wilbar Jr." <rescue at hawkmountain.net>, 
>> 	The Rescue List <rescue at sunhelp.org>
>> Subject: Re: [rescue] Re: Ross Hypersparc
>> 
>> I've never actually played with anything faster than 150mhz.. can you
>> provide details?
>> 	Thanks
>> 		Nick
>> 
>> On Thu, 9 Jan 2003, Curtis H. Wilbar Jr. wrote:
>> 
>> > Regarding the hypersparc post...
>> > 
>> > My experience is that the HS166/512K and HS142/1024K are better procs than
>> > the HS180 and HS200.  The half speed cache on the HS180 and HS200 incurs
>> > a performance penalty for many applications.  
>> > 
>> > Any others confirm/refute this ?
>> > 
>> > -- Curt
>> > _______________________________________________
>> > rescue list - http://www.sunhelp.org/mailman/listinfo/rescue
>> _______________________________________________
>> rescue list - http://www.sunhelp.org/mailman/listinfo/rescue
>_______________________________________________
>rescue list - http://www.sunhelp.org/mailman/listinfo/rescue