Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 24 Feb 2006 09:01:26 +0100
From:      Martin Nilsson <martin@gneto.com>
Cc:        freebsd-amd64@freebsd.org
Subject:   Re: AMD 64 stability
Message-ID:  <43FEBD56.4040200@gneto.com>
In-Reply-To: <Pine.SOC.4.61.0602232343410.13647@tea.blinkenlights.nl>
References:  <43FCEF9C.5050308@bluelight.org.uk>	<20060223113840.78b0a1c2.kgunders@teamcool.net>	<43FE048F.7030702@datafarm.de> <200602231550.48331.joao@matik.com.br>	<43FE094A.8030302@datafarm.de> <Pine.SOC.4.61.0602232343410.13647@tea.blinkenlights.nl>

next in thread | previous in thread | raw e-mail | index | archive | help
Sten Spans wrote:
> Opterons don't use chipsets to access main memory ...
> The motherbord just needs enough slots and 
 > good enough traces to support a fully populated memory bus.
   ^^^^^^^^^^^^^^^^^^

And it's here where it usually fails when you fill the board with 2GB 
modules. The Chinese server-wannabee manufacturers products just can't 
handle the load that these modules puts on the system. They are very 
much aware of it, but their tech support never admits that it's their 
crappy motherboard designs that causes the stability problems, instead 
the come up with all sorts of strange excuses and finally closes all 
support requests with: "not able to reproduce here - no fault"


/Martin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?43FEBD56.4040200>