Bang for the Buck, comparing Celerons to Xeons (and PIIIs?)

[George Bonser]

On 11-Sep-99 Alan DuBoff wrote:
> I was comparing 2 different SMP machines just now.
> 
> On both machines I'm running Debian potato which was updated, dist-upgrade'd,
> and upgraded before starting the test. Each machine is running the latest
> compilers, 2.95.1 from unstable.

I have a couple of such systems. One is a dual Celeron 366 at 550 on a BP6 mobo
and the other is a dual 300 at 450 on a supermicro slot-1 mobo using the MSI
slotkets. Out of curiosity I just compiled a 2.2.13pre kernel on the dual
300 Celeron @450 box with:

time make -j4 bzImage

With the following result:

real    2m57.413s
user    4m38.690s
sys     0m21.870s

The dual 366 Celeron @5550 was:

real    1m57.180s
user    3m29.450s
sys     0m16.220s

Dual P-III 500:

real    2m10.220s
user    3m47.150s
sys     0m19.000s


But actual compile times will depend on configuration of kernel options (these
are not standard Debian kernel .config files)

I have beaten these
systems to DEATH and have not had a problem with overheating, sig 11 faults or
anything. On the other hand, I have a pair of Celeron 366's that will not
overclock at ALL. I have these systems running the distributed-net client to
keep the CPU's at 100% useage so they stay about as hot as they can get. 

Oh, and if you should decide to run that distributed-net client on those dual
Celerons, you might join the Debian Linux Power Users (2.2.12 SMP) team. We
were rated the #597 team yesterday ... with only three people.

Note the dual 450 box is without air conditioning of any sort in the South Bay.