FS: 240MHz cpu module for sgi indy/challenge s
Michael Loeffler
best offer.
Michael Loeffler
spoo...@purplehair.net
Matthew Ferson
based cores.
So rare in fact that there is no such beast!
Cheers,
-Matt
Ian Mapleson
Michael Loeffler wrote:
> I have a 240MHz R4400SC module for a Indy or Challenge S
Impossible. Best R4K for Indy is 200MHz.
Ian.
Michael Loeffler
produced. And this is not a 200 that has been overclocked because I
personally do not like to overclock.
Michael Loeffler
spoo...@purplehair.net
Matthew Ferson <phu...@together.net> wrote in message news:<3B174410...@together.net>...
Gavin Saxby
> produced. And this is not a 200 that has been overclocked because I
> personally do not like to overclock.
>
> Michael Loeffler
> spoo...@purplehair.net
Any chance you could provide us with a hinv output or better still a
photo of this rare beast?
Cheers,
Gavin.
--
Remove the obvious anti-spam (.naespam) measure to reply.
Gavin Saxby
Email: gavin.sax...@uk.sun.com
Central Test Engineering
Sun Microsystems Scotland BV
Michael Loeffler
You can always buy it and see for yourself.
Michael Loeffler
spoo...@purplehair.net
Ian Mapleson <i.map...@salford.ac.uk> wrote in message news:<3B177F3C...@salford.ac.uk>...
The Linux AntiChrist
> That is what I thought at first.
> You can always buy it and see for yourself.
yeah sure. get lost fucknut.
cheers.
Scott Elyard
spoo...@purplehair.net (Michael Loeffler) wrote:
> That is what I thought at first.
> You can always buy it and see for yourself.
How much?
--
.oO=----------------------------------------------------=Oo.
| Comic: http://www.archosaur.org/oscar/ |
| Life (in progress): http://www.archosaur.org/ |
`~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
Tele Monster
Cos, its real. I got the scoop from elsewhere.
Paul Kaloudis
ill buy if you bring it over and let me try it out first.
you also have to agree to buying my time if you waste it.
"Michael Loeffler" <spoo...@purplehair.net> wrote in message
news:3d04dc4.01053...@posting.google.com...
Michael Loeffler
will be tuseday morning (sorry for the delay in time)
Michael Loeffler
spoo...@purplehair.net
Benjamin Gawert
news:3d04dc4.01053...@posting.google.com...
> I have a 240MHz R4400SC module for a Indy or Challenge S
There愀 no 240MHz CPU module, there愀 even no R4400SC 240MHz CPU. The
available clock ratings for the R4400愀 are 150MHz, 175MHz, 200MHz and
250MHz. For the Indy the fastest module was 200MHz.
I sometimes got offers like this, and *always* it was simply a standard
module overclocked with a new oscillator...
Benjamin, who wouldn愒 pay any cent for an overclocked CPU, no matter if PC
or SGI
Penio Penev
> I will get pictures the next time i have access to the machine which
> will be tuseday morning (sorry for the delay in time)
What happened to the pictures?
--Penio.
Greg Douglas
There will be no pictures.
OK, here's the (real) scoop. Using an Indy R4400SC-200 Mhz CPU module,
We replaced the 200 Mhz. R4400 CPU with a NEC R4400-250 MHz. CPU,
and replaced the 50 Mhz. crystal oscillator with a 60 Mhz. oscillator.
The systems seem to run just fine, under both Irix 6.2, and 6.5.12.
So, nothing is being overclocked (the 133 Mhz. 4600 modules use a
66 MHz. oscillator, so we're not even overclocking the module).
hinv from the PROM monitor says 200 Mhz. R4400. Once Irix boots,
hinv says 240 Mhz. R4400
The systems do not run reliably at 266 Mhz (with a 66Mhz. oscillator *
4).
(interestingly enough, PROM monitor hinv says 270 Mhz, Irix hinv
says 300 MHz. :)
We are looking into trying the original 50 Mhz. oscillator, but with
a 5X multiplier, for true 250 Mhz. Indy R4400 CPU. So far we have
been unable to get this to work, by reprogramming the serial EEPROM
that clocks the boot configuration into the CPU on reset.
We have been unable to obtain a 62.5MHz 3.3V SMT crystal oscillator
(it's a non standard frequency).
So yes, we do have R4400SC-240 Mhz Indys working (at least 4 so far).
They are not yet for sale. We are still working on getting them up
to 250 Mhz., the speed of the actual NEC CPU chips that we have.
And yes, they are noticeably faster than the 200 Mhz. R4400s.
Greg Douglas
Reputable Systems
http://www.reputable.com
Pete Plank
> OK, here's the (real) scoop. Using an Indy R4400SC-200 Mhz CPU module,
> We replaced the 200 Mhz. R4400 CPU with a NEC R4400-250 MHz. CPU,
> and replaced the 50 Mhz. crystal oscillator with a 60 Mhz. oscillator.
Interesting! I wonder if it would be possible to do the same with an
Indigo PM2 module to get it past the 150MHz barrier?
Pete
Michael Loeffler
modules are 5v and the 100 and 150 R4400 are also 5v. All the processors
that are over 150MHz that are made by MIPS are 3.3 volt (yes some of the
150s are 5v)
--
Michael Loeffler
spoo...@purplehair.net
AIM: spoonman52
Yahoo: spoonman527
msn: mloef...@hotmail.com
ICQ: 95889001
"Pete Plank" <ne...@nekochan.net> wrote in message
news:8c608e21.01061...@posting.google.com...
Steve Allen
"Michael Loeffler" <spoo...@purplehair.net> writes:
> Sorry but it is not possible with out overclocking which is bad, the PM2
> modules are 5v and the 100 and 150 R4400 are also 5v. All the processors
> that are over 150MHz that are made by MIPS are 3.3 volt (yes some of the
> 150s are 5v)
Am I wrong in thinking, then, that I can plunk a 250MHz R4400 into my
175MHz Indy, and a 60MHz crystal, and it'll work OK?
~Steve
--
Steven R. Allen - SGI Admin Weenie
http://www.eskimo.com/~wormey/
Contrary to popular belief, Unix is user friendly.
It just happens to be selective about who it makes friends with.
The charity that is a trifle to us can be precious to others. -- Homer
Greg Douglas
Steve Allen wrote:
>
> In article <k5zX6.314894$oc7.28...@news2.rdc2.tx.home.com>,
> "Michael Loeffler" <spoo...@purplehair.net> writes:
> > Sorry but it is not possible with out overclocking which is bad, the PM2
> > modules are 5v and the 100 and 150 R4400 are also 5v. All the processors
> > that are over 150MHz that are made by MIPS are 3.3 volt (yes some of the
> > 150s are 5v)
>
> Am I wrong in thinking, then, that I can plunk a 250MHz R4400 into my
> 175MHz Indy, and a 60MHz crystal, and it'll work OK?
Yes.
There may be more to it than that:
You might also need to change the serial EEPROM (National Semiconductor
93C56) boot configuration. I don't know what speed oscillator the
4400-175 has, but I bet it's not 50Mhz. like the R4400-200. Therefor,
the timings would be all wrong. Probably.
The 4400 CPUs get all their configuration information, including
secondary cache timing, master clock frequency, and all kinds of
nasty bits (tricky way of saying "I have no idea what this particular
parameter is supposed to do" - there are 64 of them) from this chip,
whose data is clocked in serially on reset of the CPU.
--
Greg Douglas
Reputable Systems
Steve Allen
Greg Douglas <gdou...@reputable.com> writes:
> Steve Allen wrote:
>> Am I wrong in thinking, then, that I can plunk a 250MHz R4400 into my
>> 175MHz Indy, and a 60MHz crystal, and it'll work OK?
>
> 93C56) boot configuration. I don't know what speed oscillator the
> 4400-175 has, but I bet it's not 50Mhz. like the R4400-200. Therefor,
> the timings would be all wrong. Probably.
Ugh. Well, since I don't have the means to play with that chip, I guess
I'll leave well enough alone...
Thanks,