News - AMD FX CPU sets overclocking record

[HEXUS]

Bulldozer at 8.4GHz inducted into Guinness Book of World Records.

Read more.

[CAT-THE-FIFTH]

Bulldozer looks like it might be very interesting for overclockers.

[CAT-THE-FIFTH]

Here are some more details about the overclocking event:

http://www.overclockers.com/amd-fx-b...-world-record/

It seems many CPUs hit 8GHZ.

[jim]

5GHz on air?

This is getting very exciting all of a sudden! After the C2D release I never thought we'd see the advent of the mythical netburst clocks, and now it's really happening.

[00oceanic]

Getting a little skeptical with all this hype and no benchmarks... Maybe the real world performance is terrible?

[Roobubba]

Great way to drum up interest! Anything that makes the marketplace a bit more competitive is a good thing, in my opinion.

Would be nice to see some benchies, as mentioned, but can't take away from the marketing success that this is! Hopefully it breathes a bit of life into Intel's efforts again, they've had such good chips for so long that things seem to have slowed down a little from the consumer perspective. Fingers crossed AMD's catch-up game is beginning to take effect, and based on the early reports, these certainly do look like good old-fashioned overclocking parts

[krazy_olie]

sort of a meaningless statistic really

[kingpotnoodle]

I've never really been interested in out and out speed like this... it's like those "land speed records" set by modified fighter planes with wheels - kinda pointless for anything and IMHO cheating as it's hardly real world conditions (how many cars do you know with an afterburner?).

I'm intrigued by the 5GHz+ on air far more, but what I would really like to know is how fast it can go within a reasonable TDP, so that I could cool it quietly... for me that is more impressive, being fast AND efficient AND cost effective.

I'm not interested in a chip that goes at 5GHz, but requires 350W and a cooler that screams like a jet plane, it's as useless to me as a car that goes 200mph+++, I'm never going to go that fast except to test it out [on the track].

[keithwalton]

I wonder how many of the 8 cores were active at that speed.

My money is on just the one core, this would give it a much lower heat output per area that we've seen for a while.

I remember one of the early athlon die shrinks didn't yield much of a clock speed increase despite the drop in thermal output the chip was significantly smaller so the thermal density went up and heatsinks of the time couldn't reliably cope. Intel was already fitting heat spreaders and increasing cache sizes to increase silicon size to improve the cooling

[Roobubba]

I've never really been interested in out and out speed like this... it's like those "land speed records" set by modified fighter planes with wheels - kinda pointless for anything and IMHO cheating as it's hardly real world conditions (how many cars do you know with an afterburner?).

I take your point to some extent, but without people pushing boundaries, development wouldn't be nearly as fast as it has been.

How many inventions from the space programmes (especially NASA) have made it (eventually) into our every day lives? There's not really much point in putting a man on the moon, but the benefit to humankind of doing so is actually quite tangible - water filters, memory foam, cordless tools, improved shoe soles and insoles, in-ear thermometers to name but a few.

So while the interest for consumers in some guys artificially pushing a new boundary back is at first rather remote, often the implications of those endeavours can be quite surprisingly wide-ranging!

[Terbinator]

I wonder how many of the 8 cores were active at that speed.

My money is on just the one core, this would give it a much lower heat output per area that we've seen for a while.

I remember one of the early athlon die shrinks didn't yield much of a clock speed increase despite the drop in thermal output the chip was significantly smaller so the thermal density went up and heatsinks of the time couldn't reliably cope. Intel was already fitting heat spreaders and increasing cache sizes to increase silicon size to improve the cooling

2 cores going off the CPU-Z screenshot.

[CAT-THE-FIFTH]

Interview with one of the chaps who was involved in the overclocking attempt:

http://translate.google.com/translat...srekordet.html

"NordicHardware: "It says you reached Frequencies" well above 5 GHz, "with only air and sub-$ 100 water cooling solutions. Mind Going Into more detail about your ventures on air and water?"
Sami Mäkinen: "In terms of CPUZ MHz I've seen VIRTUALLY every CPU reach over 5GHz When Using a good air / water cooler. I believe the highest result I saw with aircooling was around 5.5GHz CPUZ. This CPU hit around 7.8GHz on LN2 '.
"

[kingpotnoodle]

I take your point to some extent, but without people pushing boundaries, development wouldn't be nearly as fast as it has been.

How many inventions from the space programmes (especially NASA) have made it (eventually) into our every day lives? There's not really much point in putting a man on the moon, but the benefit to humankind of doing so is actually quite tangible - water filters, memory foam, cordless tools, improved shoe soles and insoles, in-ear thermometers to name but a few.

So while the interest for consumers in some guys artificially pushing a new boundary back is at first rather remote, often the implications of those endeavours can be quite surprisingly wide-ranging!

But isn't fast and useful more benefit? Wouldn't pushing these boundaries in ways which also make a practical object be better?

Making a faster car that's also able to turn corners and doesn't need an airforce pilot to drive it? Or making a chip that can go faster without gimping it to two cores and need a power station in the garden? (of note, a 2 core BD @ 8.5GHz is probably not the FASTEST chip in the world and the one that does the jobs quickest, just the one with the highest clock speed). Or how about going to the moon to do something better than admire the view, take a photo and come home... perhaps building a moon base to look for the Mysterons and help Captain Scarlett out a bit more would have been a more productive use of all that money...

[HalloweenJack]

reading some replies here it seems the old days of challenging everything and being the best has been passed over for metro fuzziness in your £200 t shirts and your perfect skin....


AWESOME they broke the world record! now push it even higher!

[llololloy]

I'm firmly on the side of the overclockers on this?
Why do it? To see if you can.
Why run the 100 metres race? To see if you're better than the rest? competition is a good thing.
kingpotnoodle, you say why not do something useful? It becomes useful when the knowledge gained from this trickles down to other products.
The stuff that a top of the line merc/bmw/jag/bently/rolls had a decade ago filters through to current astra/fiesta, so what they learn from pushing this cpu will help design future cpu/chipset and coolers.

[Fraz]

Or how about going to the moon to do something better than admire the view, take a photo and come home...

I'm shocked you think that's all they achieved with the moon landings, although I suppose you're being somewhat flippant. They were an incredible force for technological progression, and also had a very positive effect on the American economy. Even with modern-day hindsight, I doubt I'd change much of what they did. What's more, it's an achievement we currently cannot replicate, despite being 40+ years more advanced.

Regarding what you're saying about pushing "real-world" things, such as power efficiency, etc. Well, that kind of middle-of-the-road optimisation is what the chip manufacturers are already doing. That's their job, to make a product that's as powerful as it can be whilst still being efficient as possible. It's not all that captivating though, which is why it's nice to see someone take such a chip and then find its ultimate limits.