Nekochan Net

Hi All,

Yes, I'm at it again, I modded a 250MHz R10K cpu module to a 300MHz R12K and then overclocked it to 350MHz. Pretty much all it takes is changing the cpu from a 250MHz chip to a 300MHz R12K chip and changing 2 of the tiny resistors on the bottom of the cpu module. It runs much warmer than the stock module so I changed the fan on the top of the cpu module to one with about twice the air flow.

o2 1% hinv
CPU: MIPS R12000 Processor Chip Revision: 2.3
FPU: MIPS R12010 Floating Point Chip Revision: 0.0
1 350 MHZ IP32 Processor
Main memory size: 448 Mbytes
Secondary unified instruction/data cache size: 1 Mbyte on Processor 0
Instruction cache size: 32 Kbytes
Data cache size: 32 Kbytes
FLASH PROM version 4.18
Integral SCSI controller 0: Version ADAPTEC 7880
Disk drive: unit 2 on SCSI controller 0
CDROM: unit 4 on SCSI controller 0
Integral SCSI controller 1: Version ADAPTEC 7880
On-board serial ports: tty1
On-board serial ports: tty2
On-board EPP/ECP parallel port
CRM graphics installed
Integral Ethernet: ec0, version 1
Iris Audio Processor: version A3 revision 0
Video: MVP unit 0 version 1.4
AV: AV1 Card version 1, O2Cam type 1 version 0 connected.
Vice: TRE
o2

Some quick numbers:

Media Convert Test (all R10/12K cpus)
250MHz - stock - 2.80min
300MHz - modded - 2.58min
350MHz -mod & OC - 2.12min
400MHz - stock -1.82min

mahjongg / gmake (from SGI Freeware site)
250MHz - stock - 2.63min
300MHz - modded - 2.32min
350MHz -mod & OC - 2.14min
400MHz - stock -1.73min

It's been running all day now, I've done two 3 hour media convert tests for stress and the system has been 100% stable.

This module has 6nS cache chips (as does my 270MHz that I also modded to a 300MHz) and I would guess that a 225MHz module could be modded/OC to a certain point (don't know how far), the limiting factor will probably be the speed of the cache chips. 195MHz and below modules will require adjusting the cpu chip core voltage (not to hard) and mapping the resistors from the VSLI controller chip. I don't have a 195MHz cpu module, but the work should be able to be done in a few hours.

If anyone want to try this, let me know and I will post pictures showing which resistors to change.

Joe

hum, nice indeed, what about 'upgrade' a 150MHz R10K O2?

_________________
We are here on Earth to do good to others. What the others are here for, I don't know.

you need to have at least a 200mhz r10k o2 to do the 300mhz chip swap.

it might have something to do with the cache chips or the clock.

Posting the pictures would be great My R10K 250 O2 is crying out for some speed

there's pictures on the 600mhz thread for moving the resistors to get to 300mhz.

acronym wrote


It's not the same resistors as the 270 to 300MHz mod, I'll post a picture this weekend of the modded / overclocked module and the resistors to change.

Joe

I know this is a little late, but I have the information (pictures) on which resistors to change on a 250MHz O2 CPU module to mod it to 300MHz and/or overclock it to 350MHz.

This is standard NEC VR10K/12K CPU chip control, so bits 9 through 12 set the chip's SysClk multiplier of the module's oscillator (100MHz) to get the Pclock frequency for the CPU chip core. please review this thread for the basics viewtopic.php?t=1230&start=0

http://users.rcn.com/joepage/o2r10k.jpg This is the bottom of a 250MHz CPU module. The two bits we need to change are bit 11 and/or 12.

http://users.rcn.com/joepage/o2r10kzm1.jpg bit 12 - Changing bit 12 will clock the module to 300MHz, it is the 4th resistor up from the bottom of this group, move it from the right set of contacts to the left set of contacts.

http://users.rcn.com/joepage/o2r10kzm2.jpg bit 11 - Changing bit 11 will overclock the module to 350MHz, move it from the left set of contacts to the right set of contacts.

please note: the 250MHz module running at 300MHz (with a 300MHz CPU chip and the stock fan) was not much warmer than the stock 250MHz module, the overclocked module was very warm (way to warm) so I changed the stock 60x60x10mm fan (on the module) to a 60x60x20mm fan that had about twice the air flow and it dropped the module temperature quite a bit. I have had this one running for about 3 weeks now without any problems.

Joe

so for the 300MHz 350MHz R12K, I'd get one of these,

and then do a little soldering.

could I try overclocking the 250 to 300?

are there any other mods for a R10K 250MHz?

also, I now have a 200, and a 250 R10k, and there have been quite a few modications made between the 200, and the 250.

for instance, that place stocks
30710AF5360VR1200
I'm guessing that it's the wrong package, or it somehow wouldn't be compatible with the O2 board.

or http://www.google.com/search?hl=en&lr=& ... tnG=Search

can't get much out of this, but it might have a little info about the chip (my adobe doesn't support unicode or something)
http://www.google.com/url?sa=U&start=9& ... JJ*&e=9888

uhh...
http://bwrc.eecs.berkeley.edu/CIC/annou ... 0-300.html

The downsizing of computing has made performance issues even more vital for workstations and highly functional personal computers. VR12000-300 microprocessor provides up to 20% higher performance than the previous model (the VR10000-250), while its binary and pin compatibility assure a smooth transition with no changes, other than power supply voltage.

TeeTylerToe Wrote:


The 225 and 250MHz R10K and 275 and 300MHz R12K are 100% pin and command compatable and can be interchanged by changing the SysClk multiplier on the CPU board.

The 360 and 400MHz R12KA chips have a different pinout (soldered to the PC board) and a lower core voltage requirement.

I have not tried overclocking a 250MHz chip to 300MHz, if it works, expect the CPU chip to run very hot.

Joe

thanks

In my O2 it works for about thirty seconds before everything freezes.

Oh well.

thanks hamei, looks like instead of getting the bare part, it'll cost less to pick up an octane 300MHz proc (possibly with mobo) or some other sgi equipment with that sort of thing...

Is it possible obtain 450 MHz? I would be intersted in overclocking an R12K/400 module in O2.

I figured something must have been different with the 400 MHz model. It seemed like my R12K/400 O2 ran cooler than my R12K/300 did.

Is it the soldered-on ball grid array type chip that the Octanes have ? If so, the chip itself seems to be okay at 450mhz ....