SEO

Sun Apple

architecture 64-bit unix 64-bit unix
CPU 1X1GHz 2X2GHz
cache 1MB on-chip 512k on chip per cpu
mem b/width 4.2GB/s 8GB/s per processor

RAM 1 GB DDR 266 512MB DDR400
(max 4GB 8GB)

Disk 1X80GB 1X160GB
(max 160GB 500GB

graphics Sun XVR-500 ATI Radeon 9600 Pro

DVD 16X read 8X read/4X burn
firewire 400 2 2
firewire 800 - 1
PCI 5 0
PCI-X 0 3
modem - internal 56k
ethernet 10/100/1000 10/100/1000
wireless - 802.11g built-in
bluetooth - optional

OS Solaris OS X
(fully 64-bit) (partially 64-bit

--
Chris Morgan
"Post posting of policy changes by the boss will result in
real rule revisions that are irreversible"

- anonymous correspondent

In article <86pthz8hh9.fsf@elrond.bloomberg.com>,
Chris Morgan <cm@mihalis.net> wrote:

> Sun Apple

> architecture 64-bit unix 64-bit unix
> CPU 1X1GHz 2X2GHz

RAM error ECC NONE AT ALL
correction

--
Göran Larsson http://www.mitt-eget.com/

hoh@invalid.invalid (Goran Larsson) writes:

> In article <86pthz8hh9.fsf@elrond.bloomberg.com>,
> Chris Morgan <cm@mihalis.net> wrote:
>
> > Sun Apple
>
> > architecture 64-bit unix 64-bit unix
> > CPU 1X1GHz 2X2GHz
>
> RAM error ECC NONE AT ALL
> correction

Yes, that is true. Perhaps you would just use the extra RAM, CPU power
and memory bandwidth to double-check really important calculations!

Ok, I'm not serious, but would I buy the Sun machine instead of the
Apple just for that? Probably not.

Chris
--
Chris Morgan
"Post posting of policy changes by the boss will result in
real rule revisions that are irreversible"

- anonymous correspondent

hoh@invalid.invalid (Goran Larsson) writes:

> In article <86pthz8hh9.fsf@elrond.bloomberg.com>,
> Chris Morgan <cm@mihalis.net> wrote:
>
> > Sun Apple
>
> > architecture 64-bit unix 64-bit unix
> > CPU 1X1GHz 2X2GHz
>
> RAM error ECC NONE AT ALL
> correction

Also, it remains to be seen just how 64-bit the next OSX
will be. I haven't heard anyone confirm that there will
be an option for 64-bit userland processes.

-SEan

Sean Burke <burke_sp31415@pacbell.net> writes:

> hoh@invalid.invalid (Goran Larsson) writes:
>
> > In article <86pthz8hh9.fsf@elrond.bloomberg.com>,
> > Chris Morgan <cm@mihalis.net> wrote:
> >
> > > Sun Apple
> >
> > > architecture 64-bit unix 64-bit unix
> > > CPU 1X1GHz 2X2GHz
> >
> > RAM error ECC NONE AT ALL
> > correction
>
> Also, it remains to be seen just how 64-bit the next OSX
> will be. I haven't heard anyone confirm that there will
> be an option for 64-bit userland processes.

True, but I did say this :

OS Solaris OS X
(fully 64-bit) (partially 64-bit

I don't think Apple have a knockout with this comparison, but I give
them the victory on points.

Chris
--
Chris Morgan
"Post posting of policy changes by the boss will result in
real rule revisions that are irreversible"

- anonymous correspondent

In article <86llsn89cj.fsf@elrond.bloomberg.com>,
Chris Morgan <cm@mihalis.net> wrote:

> Yes, that is true. Perhaps you would just use the extra RAM, CPU power
> and memory bandwidth to double-check really important calculations!

Only if it stays up long enough to complete the important calculations.

> Ok, I'm not serious, but would I buy the Sun machine instead of the
> Apple just for that? Probably not.

For many users it is important. I can not understand why Apple ignored
error correction. I can accept the lack of ECC in my iBook, but for
my servers (SPARC/Solaris) and workstation (SPARC/Solaris) that has to
be up 24/7 the lack of ECC is completely unacceptable. I gladly trade
CPU speed for stability.

--
Göran Larsson http://www.mitt-eget.com/

Chris Morgan <cm@mihalis.net> writes:

> Sun Apple
>
> architecture 64-bit unix 64-bit unix
> CPU 1X1GHz 2X2GHz
> cache 1MB on-chip 512k on chip per cpu
> mem b/width 4.2GB/s 8GB/s per processor

Further info.

best* SPECint 589 8800
best* SPECfp 884 840

obviously this is per processor, and probably reflects Sun's SPECfp
"hacks"**

Chris

*I had to compare Apple's figures for the G5 from
"G5_Processor_WP_062303.pdf" from
http://www.apple.com/powermac/performance/
vs. official figures for Sun from www.spec.org

**Sun's Forte compiler defeats the cache-busting attempts of one of
the SPECfp benchmarks which inflates their SPECfp scores
disproportionately compared to how often one could expect to enjoy
that kind of analysis and speedup with real code
http://groups.google.com/groups?hl=e...com%26rnum%3D1
--
Chris Morgan
"Post posting of policy changes by the boss will result in
real rule revisions that are irreversible"

- anonymous correspondent

hoh@invalid.invalid (Goran Larsson) writes:

> For many users it is important. I can not understand why Apple ignored
> error correction. I can accept the lack of ECC in my iBook, but for
> my servers (SPARC/Solaris) and workstation (SPARC/Solaris) that has to
> be up 24/7 the lack of ECC is completely unacceptable. I gladly trade
> CPU speed for stability.

I don't see why workstations need to be up 24/7. To me that's what
servers are for. Anyway, you are well within your rights to insist on
ECC, but most of my users choose a PC without ECC over a SPARC
workstation or other machine with that protection more than 99.9% of
the time.

Despite the fact that I would prefer if Apple at least made ECC an
option, memory errors are fairly rare and Apple buyers are more
price-conscious than the remaining SPARC workstation buyers. Perhaps
it's a very deliberate tradeoff.

Chris
--
Chris Morgan
"Post posting of policy changes by the boss will result in
real rule revisions that are irreversible"

- anonymous correspondent

In article <86ad9384wm.fsf@elrond.bloomberg.com>,
Chris Morgan <cm@mihalis.net> wrote:

>*I had to compare Apple's figures for the G5 from
> "G5_Processor_WP_062303.pdf" from
> http://www.apple.com/powermac/performance/
>vs. official figures for Sun from www.spec.org

Oh well.

/wfr
Fredrik


--
Fredrik Lundholm
dol @ ce.chalmers.se

Chris Morgan <cm@mihalis.net> wrote:
: I don't see why workstations need to be up 24/7. To me that's what
: servers are for. Anyway, you are well within your rights to insist on
: ECC, but most of my users choose a PC without ECC over a SPARC

Uh... because one might be often running applications on a workstation that
are kept open or running 24/7/365 in order to get the most efficiency out of
the resource/business-asset called "workstation". When you have thousands of
these critters then any system/application instability from memory errors
carries a high cost to the operation, definitely a higher cost than the
what the incremental cost might have been of having ECC memory in the first
place. No, we aren't YET talking servers here, but workstations.

One must remember, "workstation" does not equal "PC". You are correct
that a PC doesn't need ECC memory, but a workstation does (even workstations
that might run Microsoft OS on x86 hardware). One should not confuse the two
terms.

Don't get me wrong, I love my Apple TiBook and Cube at home, but I also take
care of Sparc day in and day out at work.

However, I do agree with the original implied premise that a high-end Mac G5
qualifies more as a workstation-class system that should have had at least
the option for ECC memory. On the other hand, as a good employee, I am happy
that Apple isn't even trying to compete in the UNIX workstation space, so
that more UltraSparc processors can be sold. Regardless, the opinions
expressed here are my own and are not speaking for any my employer may or may
not have on the subject.

Chris Barrera
cbarrera@t i . c o m