SGI: Hardware

600MHz O2 Is Up And Running!! - Page 4

LoWeN wrote: Hello,

Hummm "bad".
So what can we take as "base" for that kind of mod.
Any R7K sure?
R5200?
Because I never found any R7K here around only R5K/R10K/R12K and one time in my life I have seen 02+.

Kind regards,

Charles


All you need is writen on page 2.

http://www.nekochan.net/phpBB2/viewtopi ... c&start=15

You need a R5K motherboard with a RM5200 - 300 Mhz Cpu with a specific Cpu board revision.

RM5200 are hard to find, and even if you find it, its price will be very high.
RM5200 are hard to find, and even if you find it, its price will be very high.


What do you mean expensive? They're only $995 from SGI :wink:
chicago-joe wrote: LoWeN Wrote
would an O2 with an R5K@180Mhz be a "good base" for that mod.


The cpu chips on the 180 and 200MHz O2s are V5000 series chips made by NEC and are completely different, swapping a RM7000C chip for a V5000 chip is out of the question. This was my answer to a question Shtoink had earlier on the same subject:

I don't know much about Indys, but the 180 and 200MHz O2s use the NEC v5000 cpu chip. I'll bet the R5000 series Indys use the same series of cpu chips. NEC made a V5000A series chip (pin and command compatable) that maxed out at 266MHz and I have seen these around so they are available. I downloaded the data and use guides for both and the only problems I see are that the 150, 180, 200 cpus use a 3.3V core and IO voltage and the 250 and 266 cpus use a 2.5V core and 3.3V IO voltage and there is not a 2.5 mutiplier for the core speed. I don't know how hard it would be to modify the voltages on the cpu card, if it could be done at all. I don't know what would happen if you ran a 266 chip with 3.3V on the core, it would get HOT so you would have to cool it well and I would think the chip life would be shorter but it might still last quite a while and should be much better in performance. To get around the core speed problem you would have to change the control resistors on the PLL chip (ICS9159-10) to change the Xtal to output ratio to 93/20 (66MHz X cpu mult of 4) to run the chip at 266MHz, this should be easy enough to do. I might try this with an O2 200MHz cpu board, I would think if it works on an O2 board it could be made to work on an Indy cpu board.

[...]



I think Indys would be a much better candidate for the VR5k upgrade, since at least some of the CPU boards have voltage regulators on them (according to http://www.reputable.com/indytech.html ). There's one visible on http://www.schrotthal.de/sgi/indy/indy_r5000_cpu.jpg In an indy, there should be some glue logic between the 5V and 3.3 V parts, so with some luck, I/O and core voltages are separate already.

My O2 R5k modules don't look like it could be possible to convert them to 2.5V core without some kind of adapter socket.
canavan wrote: I think Indys would be a much better candidate for the VR5k upgrade, since at least some of the CPU boards have voltage regulators on them (according to http://www.reputable.com/indytech.html ). There's one visible on http://www.schrotthal.de/sgi/indy/indy_r5000_cpu.jpg In an indy, there should be some glue logic between the 5V and 3.3 V parts, so with some luck, I/O and core voltages are separate already.

My O2 R5k modules don't look like it could be possible to convert them to 2.5V core without some kind of adapter socket.


Actually the picture you show is the version of the Indy r5k that is without the DC-DC converter on it. I'm having a hell of a time trying to find a picture of a r5k module with one but I can't seem to find one right now. I don't understand why there are some with and some without because all Indy r5k CPU's use the same PCB.

The voltage regulator is located on the large white rectangles to the left of the CPU (in that picture).

I do own one of each (with and without) and I'll attempt to make some pictures as soon as I can take them and find a place to upload them.
I have (tried to) organize all the 600MHz O2 CPU upgrade into a .pdf file to make everything a little more coherent, I hope it is useful.

http://users.rcn.com/joepage/o2_600mhz.pdf

Nekochan Net user radrob is going to mirror this file on his website as well.

Joe
Awesome write up...I am seriously considering following this blue print. I have a 200Mhz R5000 O2 now. I suppose I will need to get the 300Mhz RM5000 upgrade from SGI ($995?), and then the new 600Mhz CPU.

Thanks Joe for:
1) having enough of an adventurous spirit to take these risks with your hardware
2) freely sharing what you learned
chicago-joe wrote: I don't know much about Indys, but the 180 and 200MHz O2s use the NEC v5000 cpu chip. I'll bet the R5000 series Indys use the same series of cpu chips. NEC made a V5000A series chip (pin and command compatable) that maxed out at 266MHz and I have seen these around so they are available. I downloaded the data and use guides for both and the only problems I see are that the 150, 180, 200 cpus use a 3.3V core and IO voltage and the 250 and 266 cpus use a 2.5V core and 3.3V IO voltage and there is not a 2.5 mutiplier for the core speed. I don't know how hard it would be to modify the voltages on the cpu card, if it could be done at all. I don't know what would happen if you ran a 266 chip with 3.3V on the core, it would get HOT so you would have to cool it well and I would think the chip life would be shorter but it might still last quite a while and should be much better in performance. To get around the core speed problem you would have to change the control resistors on the PLL chip (ICS9159-10) to change the Xtal to output ratio to 93/20 (66MHz X cpu mult of 4) to run the chip at 266MHz, this should be easy enough to do. I might try this with an O2 200MHz cpu board, I would think if it works on an O2 board it could be made to work on an Indy cpu board.Joe


So... I am assuming the actual socketed chip is the same among the Indy and O2 right?

The thought of owning the worlds fastest Indy just makes me drool. Well actually lets just make it an even 300. :lol:
- Stonent -
The preceding post has been brought to you in Infinite Reality!
Image
Yes, read:
http://www.ff-net.demon.nl/papa/overclo ... eite3.html

The Indy R5K@150MHz and the O2 R5K@180MHz are identical in appearance. Gemm swapped them and both machines were happy with eachothers CPU.
BTW, the 180 MHz Indy cpu has a voltage regulator feeding the CPU at 3.6 volt, which the 150 MHz is lacking.

Triox has pioneered Indy overclocking, together with my feeble attemps at overclocking 5 Volt R4K's. We are now focusing on modifying the EEPROM to overclock the R5K@150MHz to 200 MHz. If the O2 can take that, the Indy should work as well.
chicago-joe wrote: I don't know much about Indys, but the 180 and 200MHz O2s use the NEC v5000 cpu chip. I'll bet the R5000 series Indys use the same series of cpu chips. NEC made a V5000A series chip (pin and command compatable) that maxed out at 266MHz and I have seen these around so they are available. I downloaded the data and use guides for both and the only problems I see are that the 150, 180, 200 cpus use a 3.3V core and IO voltage and the 250 and 266 cpus use a 2.5V core and 3.3V IO
[...]


Voltages will be the smallest problem, as opposed to packaging - O2 and Indy use PGA CPUs, the VR5000A is available as BGA only, with a different package size (29mm edge length vs 47mm) and different pin count. Someone would have to design something like this

http://www.isipkg.com/images/adp_bga_pga.jpg

and fitting a voltage regulator on that pcb should be easy, compared to the design of the thing itsself.
Here's a O2 RM5200 starter system. http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... gory=11223

But at $600.00 OUCH!!
- Stonent -

The preceding post has been brought to you in Infinite Reality!

Image
Ok after reading the O2 overclocking guide, I see that the Indy R5k module has J1 - J5 on the side. Could these have a similar function to the resistor matrix on the O2?
http://132.180.68.114/~gemm/images/o2/i ... _modul.jpg
- Stonent -

The preceding post has been brought to you in Infinite Reality!

Image
Stonent wrote: Ok after reading the O2 overclocking guide, I see that the Indy R5k module has J1 - J5 on the side. Could these have a similar function to the resistor matrix on the O2?
http://132.180.68.114/~gemm/images/o2/i ... _modul.jpg


Supposidley you can only change the multiplier on a Indy CPU by hacking the EPROM on the CPU board, however the r5000 CPU's may be different...
Oh well, if I feel brave, I'll try a faster crystal.
- Stonent -

The preceding post has been brought to you in Infinite Reality!

Image
Stonent Wrote
So... I am assuming the actual socketed chip is the same among the Indy and O2 right?


canavan wrote
Voltages will be the smallest problem, as opposed to packaging - O2 and Indy use PGA CPUs, the VR5000A is available as BGA only, with a different package size (29mm edge length vs 47mm) and different pin count.


The 200MHz O2 CPU board uses the V5000 CPU chip in the BGA package, I have not seen a 180MHz CPU board but I would think it also uses the BPG form of the chip. If the Indy uses the PGA chip, it looks like you could at least upgrade to a 200MHz CPU and maybe overclock that by a few MHz. http://www.necelam.com/microprocessors/uPD30500-200.cfm It should be possible to upgrade a 200MHz O2 to the 266MHz chip since they are the same form factor, as long as the extra voltage on the core doesn't damage the V5000A chip.

Joe
I found this:

SR71040A, a MIPS64 600Mhz, superscalar embedded microprocessor represents the highest value in terms of performance to the dollar available in the MIPS64 architecture today. The SR71040A complements SandCraft’s SR71010A, the industry’s highest performance production released MIPS64 CPU, by adding a lower cost option to the SR71000 family of processors. The SR71040A utilizes the same CPU pipeline as the SR71010A, but with smaller caches and in a lower pin-count package.

The SR71040A provides an upgrade path for users of the R4000 and R5000-class processors in low to mid range embedded systems, and also enables SandCraft to address more value oriented applications, including enterprise LANs, Storage Area Network systems, security processing systems, remote aggregation systems such as DSLAMs and wireless head ends, and office automation products (laser printers and multifunction peripherals).


These processors are $20 USD in lots of 10,000.

I doubt that they are in a compatable package though...


Here's a link for more info: http://www.morphoses.com/portfolio/sand ... 040.ai.pdf

256 pin TBGA package

The VR5000 is a 223 pin Ceramic Pin Grid Array. Maybe an adaptor? :lol:

http://www.info.fundp.ac.be/~bqu/downloads/vr5000ds.pdf
- Stonent -

The preceding post has been brought to you in Infinite Reality!

Image
edefault wrote:
Just curious:

is anybody planning to exchange a R5200 CPU with a R7900 to see if a patched PROM image will work?

And if yes, when eventually?



Walther,

Count me in for first batch of RM7900, hopefuly...
I have RM7000 board but was hasitating to go for 600MHz since is not exactly cheap and I have already R12k 400 o2, so it's not big gain for me.
RM7900 move makes more sense to me.

waiting to see some 7900 benchmarks :roll:

z
jan-jaap wrote:
Quote:
Of course, if the RM7900 is not functionally identical to the RM7000, you're looking at a different problem. You'd have to write TLB and cache setup code and inject it into an existing PROM. Tricky business.

FWIW - The L2 and L3 cache can be turned off in the boot-time PROM program, I would say get the RM7900 chip running and address the cache and other issues later. It might be easier just to replace the PRId number of the RM7000 with the one from the RM7900 in the IP32PROM, write a new boot-time PROM program, burn a PROM chip, flash the IP32PROM to the O2 motherboard and see if the system will boot. If the O2 were booted from a terminal, it might be easier to tell what's going on as the system starts.


edefault wrote:
Quote:
Perhaps we´d to decide to go straight through and order R7900 CPUs instead?
(OK, just dreamin´)

Yes, just dreaming. :wink:
The RM7900 CPU chip swap is a "long term" project at the very best.

Joe
I'm ignorant of the concept of checksums (exact byte count??), but I suspect it's safe to assume that the PROM image has one and that before the flash program will load it and flash do the upgrade it checks.

The only consequence I can see of hijacking the RM7000 settings is if any of the TLB exceptions or associated programing adds more data than it replaces then it would possibly affect execution if instructions point to a specific piece of code at a specific address. If the two are completely command compatible then there may be no hassle.

I've got the data sheet for the PROM chip so I'm going to sit down with a DMM later and see if I can find external contacts for programming the PROM. Common sense tells me there should be some, just gotta find them.

Has anyone with both a DUMP of the PROM as well as the actual PROM image that came with IRIX compared them to see if there are any differences?

Glenn
Awesome!!

*drooling*

So what's the next step in this project?
Broadcom still make workstation-class multicore mips64 CPUs, but the interconnects are far more modern and so I doubt they could be shoehorned into SGI hardware. Irix could be ported to them though, if the source was ever to be released. Rumor has it it may have even been done internally at some point.

For now though, I think the MIPS Inc 1ghz r16000 is the fastest we'll see in sgi hardware.

_________________
:0300: <> :0300: :Indy: :1600SW: :1600SW: