Alpha hardware

[Jay Estabrook]

Yup, that would be one of the SIMM platforms that used 128-bit mem bus.
Earlier ones (Jensen, Noname/UDB, at least, IIRC) had only 64-bit
mem bus, and thus required SIMMS in pairs, not quartets... :-)

SIMMS == supplied 32-bits wide each
DIMMS == supplied 64-bits wide each

Some later machines took DIMM pairs (DS10), others quartets (XP1000/DS20).
Memory bandwidth, of course, is better the wider the mem bus; that's why
XP1000 feels faster at 500MHz than DS10 at 667MHz. And DS20/ES40/DS25/ES45
have 2 256-bit wide mem buses, when configured/populated correctly.

IIRC, no DEC Alphas could use single SIMM/DIMM.

But at least API's UP1100 would work with a single DIMM.

--Jay++

Hoover, Tony wrote:
> I know that my EB64+ required SIMMs to be installed in groups of 4.  It
> had 8 slots.
> 
> 
> 
> On Fri, 2008-12-05 at 13:36 -0500, Smith, Todd wrote:
>> Hello Jay,
>>
>> I only tried one module since that is all that I had, but I thought my
>> UDB will boot to SRM with only a single 8MB module.  I just dragged it
>> out and tested just a single SIMM and it never booted.  Maybe another
>> 128MB SIMM module would do the trick!
>>
>> Thanks for remembering some fundamental facts.  :)
>>
>> Todd
>> -----Original Message-----
>> From: axp-list-bounces at redhat.com [mailto:axp-list-bounces at redhat.com]
>> On Behalf Of Jay Estabrook
>> Sent: Friday, December 05, 2008 13:10
>> To: Linux on Alpha processors
>> Subject: Re: Alpha hardware
>>
>> Hi, Todd,
>>
>> Did you try just 1, or a pair?
>>
>> I thought that all the Alphas (other than some API machines that use
>> and AMD chipset) required SIMMS/DIMMS in pairs... ???
>>
>> I know that I've loaded UDB with 64M SIMMS with success, but never had
>> anything larger to attempt. If those SIMMS were really "late", it's
>> possible
>> that the SROM code isn't up to configuring them. I don't know if any of
>> the SIMM-based Alphas were ever tested with 128MB SIMMS...
>>
>> Good luck.
>>
>> --Jay++
>>
>>
>> Confidentiality Note: The information contained in this message
>> may be privileged and confidential. If this e-mail contains
>> protected health information, you are hereby notified that any
>> dissemination, distribution or copying of this communication is
>> strictly prohibited,except as permitted by law. If you have
>> received this communication in error, please notify the sender
>> immediately by replying to this message and deleting it from your
>> computer.  Thank you.
>>
>> _______________________________________________
>> axp-list mailing list
>> axp-list at redhat.com
>> https://www.redhat.com/mailman/listinfo/axp-list
> 
> _______________________________________________
> axp-list mailing list
> axp-list at redhat.com
> https://www.redhat.com/mailman/listinfo/axp-list
>