Frying An Arm Chip

[Poem58]

Sorry I was editing and more posts came in I quickly googled overclocking the cortex 1-8 and the first result was this
55mhz 1-8 overclocked to 1.3Ghz and still no mention of it being fried.

that's encouraging.

http://blog.7touchgroup.com/tag/overclocking-the-arm-cortex-a8-a-550mhz/

 

[Poem58]

Isn't xScale an Intel design? Might explain a lot

yeah, an intel arm v5 chip
http://en.wikipedia.org/wiki/XScale

i think the arms designs have come a long way, but I am far from an expert I just read alot from time to time.

 

[craigix]

Sorry I was editing and more posts came in I quickly googled overclocking the cortex 1-8 and the first result was this
55mhz 1-8 overclocked to 1.3Ghz and still no mention of it being fried.

that's encouraging.

http://blog.7touchgroup.com/tag/overclocking-the-arm-cortex-a8-a-550mhz/

I know about the omap/a8, and I'm still pretty sure it can't be fried - but I am more interested in these other ARM chips which seemingly can be fried. I can't really find anything online about it aside from warnings/rumours.

Does anyone maybe have a forum link or something like that where someone has killed their chip and given some detailed information?

 

[mali]

Can you give exact details, we want to buy this chip/device, clock it and measure the temp.

What did you do to kill it exactly? This seems to be an area we need to document. ARM overclocking.

It was an Asus Mypal A600. I was active on a forum at that time where Pocket Hack Master was developed and provided betatesting info. I forgot which board it was, though :huh: I think it was the board where Dan East ported the first Pocket Quake. One day I set it to 448MHz or some such, I don't remeber accurately and it worked for a few seconds, then the screen froze and nothing. No softreset, no hardreset worked. I put that PocketPC on a shelf, because it wasn't possible to deactivate, as the battery was inside. I let it run out of juice for a week and tried to turn it on, but it didn't work.

Edit:
That other heat related failure I mentioned is the xScale in my x50v. It would crash after a while when playing PocketQuake at 702MHz(or 709MHz? I forgot) and the back of the device was very hot. I disassembled it removed the shielding of the processor area, cut a piece of copper and put it on the CPU with some thermal paste which touched the shielding too, to dissipate heat. I put it back together and it worked flawlessly afterwards at that mentioned speed.

 

[Poem58]

Sorry I was editing and more posts came in I quickly googled overclocking the cortex 1-8 and the first result was this
55mhz 1-8 overclocked to 1.3Ghz and still no mention of it being fried.

that's encouraging.

http://blog.7touchgr...ex-a8-a-550mhz/

I know about the omap/a8, and I'm still pretty sure it can't be fried - but I am more interested in these other ARM chips which seemingly can be fried. I can't really find anything online about it aside from warnings/rumours.

Does anyone maybe have a forum link or something like that where someone has killed their chip and given some detailed information?

i will look but keep in mind this is all form memory around the time My Axim was relatively new, and it's been nearly 7 years since. I really think we are no longer in the days of those old Intel Arm's.
here's another post I found when searching for fried Arm Cortex processors, still nothing...
http://phandroid.com...-up-to-1-1-ghz/

Edit: Article about PDA overclocking from 2003 being dangerous.
http://www.pcworld.com/article/109556/overclocking_poses_risks_to_pdas.html

 

[MDave]

I'm finding it real hard to find posts in old threads, but Squidge tested the temperature and it only got 1 degree Celsius (2 degrees Fahrenheit) hotter at 800mhz? Wish I could find that post again. Somewhere it was quoted in the 30+ page overclock warranty thread.

 

[Fishbong]

Just out of interest Craig, how many gp2xs have you had returned which have been fryed by overclocking? (I know you send them to GPH to test why they failed)

I'm not aware of any, I don't think the chip can get hot enough. I guess if you started to fuck about with the voltage by hand (ie doing wacky hardware mods) you could kill it.

That said it was a pretty common GP32 mod.

Somehow the MMSP2 inside the GP2X was protected from GP32-style voltage mods. I tried cranking up the voltage the same way it was done with the GP32 - exchanging the resistors that determined the output voltage of the step-up regulator. I managed to increase the output voltage from 3.3V to up to 4V (i didn't dare to go higher than that) but to no avail. Well, the display got brighter, but i couldn't overclock a single MHz higher than before. I'm not actually very knowledgeable around those things, so i had no choice but to give it up...

Anyway, why don't you just run some tests with overclocked devboards? Sure, the processors will all react a little differently, but as long as they don't get hotter than, say, 70°C, i don't see any way they could be destroyed.

 

[EvilDragon]

It was a first generation xScale, so old stuff compared to Cortex. You could either make a stress test by going to the highest possible stable voltage+clock combination and let it run till it breaks(or not) or just lock the voltage(or just don't expose it to the user, like it is now already) and let the user overclock without overvolting. This should be save enough to be covered under warranty, I suppose. And this would also be totally in 'community spirit'.

It was an Asus Mypal A600. I was active on a forum at that time where Pocket Hack Master was developed and provided betatesting info. I forgot which board it was, though :huh: I think it was the board where Dan East ported the first Pocket Quake. One day I set it to 448MHz or some such, I don't remeber accurately and it worked for a few seconds, then the screen froze and nothing. No softreset, no hardreset worked. I put that PocketPC on a shelf, because it wasn't possible to deactivate, as the battery was inside. I let it run out of juice for a week and tried to turn it on, but it didn't work.

Was that the same device? If could determine the highest possible stable voltage, it could definately change the voltage by software, and this could definately fry the chip.

I doubt overclocking without overvolting would fry the chip... it would crash before overheating.

 

[Poem58]

Simply put the old processors The Intel Xscale ones would be sold as say 300 or 400 Mhz models. the 300's failed at 400 and were binned at 300. the overclocking software would cause problems. If memory serves, they seemed prone to failure when overclocked to extremes because they did tend to get hot, and having one I can say it gets quite warm after running it at 400Mhz full bore for say 1/2 hour. So If it were to be extremely overclocked I could see it dying. I'm sure with no speed settings beyond the 400 and low power 200 Mhz mode on mine it was never designed to dissipate any more heat than it generates. So I could see how If I clocked it at 700+ or something I could kill it. But they don;t make 'em like they used to, and in this case that's a good thing. If the Dev's have tested emu's at 800 and barely any temp increase, I really don't see a problem aside from the estimations that you might shave a year of life off the 10 year guesstimated lifespan. Remember that was a 10 year non stop guess if my memory serves correct.

Anyway, that exhausts the limits of my knowledge and memory. hope that helped.

 

[Rathum]

Make sure you're clear that you're not talking about overvolting it. Overvolting will definitely kill the chip :lol:

Proof? You can't prove that a reasonable overvolt does squat. It generally will speed up electromigration and cause the chip to heat up faster. The exact effect varies between chips, but taking it too far™ can exceed the voltage specification for certain components on the die or even components off the die.

Well, if we're trying to get an OC warranty policy, I would want to see if there's anyways to possibly kill the ARM chip with no OV first and then test for a reasonable OV level if they allow one at all.

Honestly, the warranty should be voided on the CPU if you OV it. It's too easy for newbies to damage a chip via excessive voltage. And, yes, I do know that a reasonable OV doesn't cause any damage; I OC and OV my desktops.

EDIT: That's not to say OC should mean any non-OC related damage should be out of warranty. All or nothing warranties suck.

 

[mali]

It was a first generation xScale, so old stuff compared to Cortex. You could either make a stress test by going to the highest possible stable voltage+clock combination and let it run till it breaks(or not) or just lock the voltage(or just don't expose it to the user, like it is now already) and let the user overclock without overvolting. This should be save enough to be covered under warranty, I suppose. And this would also be totally in 'community spirit'.

It was an Asus Mypal A600. I was active on a forum at that time where Pocket Hack Master was developed and provided betatesting info. I forgot which board it was, though :huh: I think it was the board where Dan East ported the first Pocket Quake. One day I set it to 448MHz or some such, I don't remeber accurately and it worked for a few seconds, then the screen froze and nothing. No softreset, no hardreset worked. I put that PocketPC on a shelf, because it wasn't possible to deactivate, as the battery was inside. I let it run out of juice for a week and tried to turn it on, but it didn't work.

Was that the same device? If could determine the highest possible stable voltage, it could definately change the voltage by software, and this could definately fry the chip.

I doubt overclocking without overvolting would fry the chip... it would crash before overheating.

Yes, both of those quotes mean the same device. I don't know how much control Pocket Hack Master provided at that time, because the author was still developing it. It might have changed the voltage, I don't know.

 

[fischju2000]

There was conflicting information in the massive thread. What power profile does the Pandora operate on at 600mhz? A while ago there were plenty of posts about the power profiles. I believe they are all within spec but the higher you raise it, the shorter the life of the CPU. This might effect other components because of the SoC design though. And IIRC you can't raise the voltage higher than the maximum power profile in software.

 

[Poem58]

i think all of the links are dead, but still trying to show it was the old Intel's or early Pocket PC ones...

Can I overclock my PDA?
WARNING...If you don't know what Overclocking is, then it is probably not for you. Though in the end, this should make your PDA go much faster, it can ruin it. Since you are voiding your warranty by trying this, you will not be able to get your device fixed without paying, should you have problems in the future. Casio E-115 -
http://www.dalecoffi...15overclock.htm
http://www.truthring...clock/intro.htm
Casio E-125 -
http://www.hi-pda.co...erclocke125.htm
http://www.geocities...5OVERCLOCK.html
http://www.rojakpot....E-125_OC_01.htm
Casio EM-500 -
http://www.geocities...00_WARNING.html
Compaq iPAQ -
http://www.jimmysoft...ock/IPAQCLK.EXE​

I still haven't found an instance of a Cortex being fried, just the old PDA ones.I see the warnings but unfortunately most of the posts and links for all the old stuff are long since dead, as I believe are the prospects of frying a Cortex without doing something really stupid.

 

[petelewin]

We could also ask to TI ...the producer of the Chip maybe have some more experience on it ...no ?

This is a good idea, but remember some chip makers are cautious about overclocking, it allows you to save money by just OC'ing their lower binned chips.

Doesn't it all depend on TI, as supplier of the OMAP parts? If they will guarantee the OMAP parts to OpenPandora when they've been overclocked, OpenPandora should be happy to pass that guarantee on. If TI will not, then OpenPandora should not - regardless of any suspicions we may have about the silicon suppliers motives.

Technically I'm sure overclocking (without overvoltage) will cause no damage to an OMAP, and I will be happy to overclock mine at my own risk. However I don't see OpenPandora sensibly can if TI don't.

And of course that would allow the possibility of an 'overclock safe' Pandora using higher spec'ed chips, at a higher price ;-)

 

[mali]

@Poem
I had an E115 too and did the pencil trick, it's a MIPS though and no ARM. I crammed a grinded 4x86 cooler in the CF card slot to passively cool the PCB underneath the processor :lol:

 

[Poem58]

@Poem
I had an E115 too and did the pencil trick, it's a MIPS though and no ARM. I crammed a grinded 4x86 cooler in the CF card slot to passively cool the PCB underneath the processor :lol:

Oops, see that's why I'm not an expert! I forgot about the MIPS ones, I had an E-100 myself. The XScales were much faster, but don;t you remember them getting a bit warm after suing them for a bit? Mine still gets warm enough to make me wonder at times, not HOT, just really really warm. Noticeable warm.

What was the pencil trick!? I never did anything but use it normal til it died.

 

[mali]

^Yep. They are just Intel crap
Marvell bought them later. The newer Marvell SoCs look promising, though.

edit:
The pencil trick was to make a contact on the board to get it from 133MHz to 168MHz(?) or 200MHz(?).

 

[calc84maniac]

The XScales were much faster, but don;t you remember them getting a bit warm after suing them for a bit?

That's why you shouldn't do frivolous lawsuits

 

[craigix]

i think all of the links are dead, but still trying to show it was the old Intel's or early Pocket PC ones...

Can I overclock my PDA?
WARNING...If you don't know what Overclocking is, then it is probably not for you. Though in the end, this should make your PDA go much faster, it can ruin it. Since you are voiding your warranty by trying this, you will not be able to get your device fixed without paying, should you have problems in the future. Casio E-115 -

​

Like you said the links don't work, and the warning is just that - a warning. We saw this in the early GP32 days too.

But was it actually possible? We need to see some real evidence. Surely with all those devices out there someone would have documented something? There is tons of stuff about frying X86 CPUs.​

 

[Poem58]

The XScales were much faster, but don;t you remember them getting a bit warm after suing them for a bit?

That's why you shouldn't do frivolous lawsuits

HA HA pick on my crappy keyboard, I can only make super gluing may laptop keys work so well you know. Combine them with bad hunt and peck typing and just be glad I can even be readable. once I get my Pandora, hopefully i won't be typing on this anymore.