Overclocked G5?

[disconap]

I hadn't heard of this being successful; then I came across this listing on ebay. My immediate thought was "scam", but the dude has a 100% positive rating with 600+ transactions. Thoughts? Has anyone heard of this working?

 

[imacintel]

How hot will it get when I overclock my G4?

 

[disconap]

How hot will it get when I overclock my G4?

Um, wrong thread?

 

[imacintel]

Kind of related...I guess I was just wondering if I could overclock my g4. Kind of similar.

 

[mad jew]

Getting the earlier G5 to 3.5GHz is insane. I can't imagine it'd be stable, but I'm still impressed the seller has managed to do it.

 

[~Shard~]

Getting the earlier G5 to 3.5GHz is insane. I can't imagine it'd be stable, but I'm still impressed the seller has managed to do it.

Yeah, however I'd really question the longevity of that system. It would be along the same lines of those idiots who put nitrous in their cars.

 

[Spanky Deluxe]

Holy cow, is this for real? A dual processor machine overclocked like that would be hands down the fastest OS X Photoshop machine on the planet! $800 isn't *that* much for an increase in speed of 75% and could work wonders for photoshop pros (providing they haven't already upgraded to newer G5s).

 

[Fearless Leader]

thats insane, kinda tempting for photoshop though. I wounder how he/she is doing it though.

 

[imacintel]

Either way, I call BS. This thing must get insanely hot and burn your place down. G5s got hot enough at regular speed I can't imagine how badly you could get burnt by touching an Oc'ed G5.

 

[Mernak]

Either way, I call BS. This thing must get insanely hot and burn your place down. G5s got hot enough at regular speed I can't imagine how badly you could get burnt by touching an Oc'ed G5.

Exactly, The G5s had a liquid cooling system and quite a few fans if I remember correctly. And that was just for normal operation, overclocking usually requires even more cooling than normal assuming you want it to be stable, and that is hard to do with the G5.

 

[aussiemac86]

Exactly, The G5s had a liquid cooling system and quite a few fans if I remember correctly. And that was just for normal operation, overclocking usually requires even more cooling than normal assuming you want it to be stable, and that is hard to do with the G5.

Only the Dual 2.5GHz had the liquid cooling, the lower and previous models didnt, he was only offering on 1.8/2GHz, so im assuming he doesnt want to deal wih the liquid cooling.

 

[4JNA]

...the benchmark number does not change very much from the standard Apple speed configuration. This is software up clock. ... Also the speed of the fans may increase which will equate with a higher decibel.

ummm, so i take my silent kick-ass dual 1.8 apart, send in both cpus, and get back a 'software' overclocked hair dryer. all for $500. fsb overclock. i'm sure that will do really nice things to pci, memory, and so on...

god bless ebay and the crap that can be marketed to the world from the comfort of the sofa... i should send him the g5 processor from my laptop and see how fast i can get it to run... might kill battery life though.

 

[theblotted]

in the eBay pic, where's the serial number? sure looks photoshopped... and the 3.5GHz number can easily be photoshopped also.

but the idea is great. i hope it's not a scam.

 

[simie]

Why is this guy not selling his mod on Ebay or offering to mod other Mac Users G5's for a fee.

He would make more money from this than selling his own over-clocked G5 on Ebay.

 

[theblotted]

Why is this guy not selling his mod on Ebay or offering to mod other Mac Users G5's for a fee.
.

i thought this IS what the seller is doing, no?

 

[RichP]

His list of constraints make it at least seem plausible. Although stock air cooling at that speed sounds suspicous as ever. Also, the overclock is a result of busspeed increase, not multiplier increase. I would think that the FSB control would be on the MB, not on chip?

So who is going to try it out?

 

[Sun Baked]

Here is an older post...

Here is an older post...

really, i thought that was one of the big selling points of the massive G5 towers, that they were quite upgradeable. the CPU wasnt much of concern to me when i bought it, but i did imagine it would be as well. is it just that with the switch to intel the demand for this is no longer there? if so, that kind of bums me out.

It can be done, heck they can probably be overclocked -- but it isn't a trivial job like the G4.

Hello,
I'm one of the engineers that did the 'Maple' design for IBM at Momentum(with a LOT of help). What you propose to do is achievable, but would have to be broken up into different steps.

As previously mentioned, there are a bunch of startup steps that are performed by a microcontroller, before the 970 ever fetches its first byte of code. Included in these steps is the setting of processor speed, bus speed (in both the CPU and the bridge) and bus ratio (the ratio of processor speed to bus speed... this is usually 2:1 for Macs)... This is all done over I2C to either GPIO expanders (I2C in, 8 discrete controllable signals out) or directly to the 970(s) and the bridge.

Using an I2C bus analyzer, one would be able to see what goes where and when. There really is a tremendous amount of traffic generated, but it is all understandable from the address it goes to. Figuring out which pins on the micro are for what bus would take more time... It would be easier to find the I2C pins on the processor module and the CPC925 bridge and attach wires for the probe...

I2C bus analyzers can be pretty cheap... look at BLACK-BOX... You don't ever need to Master the I2C bus, just 'snoop' by dumping everything it sees to an RS-232 port.

There may also be a certain amount of 'tuning' of various parameters required... This would be the 'we have a million years to spend on this, but want to understand everything' approach.

Couple of suggestions for approaches that would take far less time...
1. Get a 1.8/2.0/2.2GHz processor module and copy the I2C PROM. This has all the work already done for you. Burning a higher speed image into a lower speed module would effectively yeild one half of an overclock mechanism. BACKUP EVERYTHING FIRST!!! - As a side note, IBM is pretty good at figuring out how fast their processors will and will not run... what is unknown is if there is any de-rating added by Apple.
2. Get pictures of the resistor settings on 1.6/1.8/2.0/2.2GHz boards and compare the differences. These are likely set for a range of frequencies and must match what you are going to. Having pictures of different configurations would help decrypt what each resistor does.
3. Use an I2C analyzer to capture the H8 activity for both a 1.6GHz boot and a 1.8GHz boot and post the 'diff'. This would make obvious what needs to be done...
4. It's not obvious to me whether Apple hard-coded the motherboards for a certain speed, or did some sort of resistor/firmware funniness. Given the nature of the product, being in volume, you'd really want to only have to manage one design for multiple speeds... Even managing different H8 part numbers with masked ROMs would be non-trivial in volume. With this assumption, there would be a trial/error step involved in setting different resistors and viewing what the H8 outputs on I2C... ...until the motherboard and processor module agreed on what the new bus speed should be and all the software already written would 'just work'...

Trivial... not.
Risky... very.
BUT this does provide a deterministic means of getting there from here. If lots of folks got together and pooled their data for resistor settings and PROM images at different speeds, one could readily make some conclusions and eventually a 'patch' for overclocking chips... This would still require either soldering wires to processor modules, or a fixture to set the module into...

Anyways, just some thoughts.

Regards,

Bruce

and...

Thanks.

First - please le me apologize for the monster post... there is a lot of data to digest.

Secondly - Please let me clarify... Momentum (http://www.momenco.com), where I work, is not affiliated with IBM. We just want to bring the chips (& 64-bit computing) into the embedded and TELCO space.

A better picture of the Maple product is here:
http://www.momenco.com/products/xsc100.html

Also, we maintain the http://www.970eval.com site and its associated forum(s).

The purpose of those forums is to help people who are trying to understand the 'nuts-n-bolts' of 970 and how it works in a system. It is at an engineering level. While we can't talk about other companies specific implementation(s), we can talk about what has to happen at what time.

I am pretty motivated to help folks understand 970 and its associated quirks. I think that people must understand this chipset before they will really accept it - So without violating any NDA's, I'm happy to pontificate on hypothetical scenarios, like overclocking the 970.

'nuff said...

WRT YellowDog Linux... Kai Staats, the CEO of TerraSoftSolutions, who makes & maintains YellowDog Linux and I have spoken about how YellowDog starts up on 970 systems. There is an OpenFirmware tree structure in memory that is handed off to the Kernel, which describes everything that should have already been enumerated... like processor and bus speed.... so by the time an OS gets loaded... bus speeds are already set. This tends to point us back at the H8, towards figuring out what on the motherboard directs this micro to tell the Bridge and processor how fast to run the bus...

It all comes down to clocking...
If you go to http://www.970Eval.com and register into the forum, go to
PowerPC 970 Evaluation » Board Management Processor » 405EP BootLoader
and look at "System Clock speed, Problems working at 450 Mhz ?"
from Jan 14 2004
(Sorry, I can't re-post here without REALLY pissing someone off)
There, you will find a FULL description of PLL, multipliers/dividers & clocking requirements for 970 and the bridge. Examples are given for clarity.

So, from the data above, one should be able to see that there is a (programmable) clock source on the motherboard that is responsible for giving various clock speeds to the processors and bridge for different configurations. Seeing as the H8 is the 'Master' of configuration of this motherboard, as well as monitoring (for fan speed) - it is not a large leap to assume that the H8 probably configures this clock source at boot time, over I2C - since everything else is I2C, too. The clock speed generated for a 1.6GHz processor can not be the same speed as for a 1.8GHz processor, because the bus speeds have to be different for the math to work out... Therefore one must use an I2C probe to figure out what modifications to the baseboard (like resistor position) yeild the correct generated clock speeds to yeild the desired processor speed. It is also possible that there are resistors on the processor modules that direct this functionality.

One has to realize that this is only half of the hack... The other half lies in convincing the 1.6GHz processor module to run at 1.8GHz. Just like most modern processors, there is nothing inside the silicon of a 970 that limits its speed or configuration... IBM bins different speed grades based upon their characterization of the bare die. Convincing a 970 to run faster than what it has been 'binned' for is a simple matter. Keeping it running may be less trivial (Heat). Please remember that the thermal density of the silicon is four times that of a Pentium IV - Same Wattage in 1/4th the space... so thermal management becomes a real issue, thus the HUGE heat sinks... If you look at a processor module, you'll see that there is a programmable I2C part on them. I would assume that this is what holds the magic for specific speed configuration for a specific processor. This PROM is probably read out at boot time, telling the H8 what it's max speed and configuration is, and the H8 probably compares that against what it thinks the motherboard is configured for and activates the lowest common denominator. The second half of the hack would be to re-program this I2C PROM with the contents of a higher speed processor modules PROM. Hypothetically.

If you assume that the hypothesis stated above is correct, then you really have to hand it to Apple for creating a manufacturable, configureable machine. Its graceful in its configureability. They could stock up a number of generic motherboards and simply plop in pre-configured processor modules at build time, depending strictly upon demand - and know that the system will only run as fast as they want it to. Amazing.

Bruce

Compared to look up resistor settings, and add 1 or two or remove them on a G4.