Not quite current CPU sources

Colin McGregor colin.mc151-Re5JQEeQqe8AvxtiuMwx3w at
Tue Jul 8 18:09:08 UTC 2008

On 7/8/08, Lennart Sorensen <lsorense-1wCw9BSqJbv44Nm34jS7GywD8/FfD2ys at> wrote:
> On Wed, Jul 02, 2008 at 03:44:08PM -0400, Colin McGregor wrote:
>> I am looking for a Socket A (aka: Socket 462) AMD CPU chip / heat-sink
>> /fan (i.e.: Athlon XP chips)... Who in the city sells these not-quite
>> current CPU chips?
> I think I have an XP 1700 or 1800 lying around somewhere (I run a
> 2400XP-M in my machine now).

That would be of interest. thanks.

Thanks to the kindness of John McGregor (who is on this list) I got a
Duron 800 and associated regular ATX motherboard. Didn't solve the
original issue, fixing my microATX based MythTV box (turned out the
microATX motherboard I picked-up was dead, sigh...). Still did let me
get a machine fixed up for my brother.

Any event this evening I will be bringing some bits and pieces to give
away at TLUG...

>> Bit of background, last month I bought an ASUS M2N-MX SE PLUS (with an
>> appropriate AM2 CPU / memory / etc), which was shipped defective as
>> far as I was concerned, the MAC address for the built in network port
>> was b5:52:17:C6:1f:00 (for those who don't know why this is defective,
>> look up multicast MAC addresses). This meant this motherboard was
>> TOTALLY useless for the one reason I got this motherboard (a diskless
>> front end box for my MythTV system). I have grumbled loudly at ASUS,
>> got an RMA number, and as I write the motherboard is in transit to
>> ASUS to get fixed.
> It looks to me like the driver didn't deal with it right,  It is well
> known that certain nvidia chips have the MAC address stored in reverse.
> Quite how the driver is supposed to know I am not sure, but it certainly
> has worked on many boards so far and should hopefully be able to work on
> that one too.  Perhaps you just needed a newer kernel.

The driver deals with things in a more-or-less sensible way, if it
detects an invalid MAC address, it will replace that MAC address with
a valid one chosen at random. This is a problem in my sort of set-up
where the server needs to know what the MAC addresses of the diskless
clients is going to be...

>> In the mean time I want a diskless box ASAP, and I am in no mood to
>> spend anything more than pocket change to get said box running
>> (spending bucks to ship a defective motherboard has left me feeling a
>> bit grumpy). I have gotten a used (and long story, but free)
>> motherboard that uses Socket A (aka: Socket 462) CPUs. I hope, but
>> have not tested, the motherboard, so I hope works. My goal is to have
>> the client up and running by the end of this weekend. This way I will
>> have something until ASUS repairs their motherboard...
>> So, back to the start point, suggestions as to who would have Socket A
>> (aka: Socket 462) CPUs for dirt cheap?

Colin McGregor
The Toronto Linux Users Group.      Meetings:
TLUG requests: Linux topics, No HTML, wrap text below 80 columns

More information about the Legacy mailing list