• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD FX Bulldozer CPU clocked to 8429 MHz - World Record

Last edited:
AMD also ran a smaller event in the UK, and I was privileged enough to be there along with Helios1234 from these forums on behalf of Vortez.net.

I've posted up some images and info about what we saw and did, no performance info yet though bar overclocking, everything is still under NDA.

Have a read here.
 
Gashman said:
what the hell? Bulldozer has been ripped so much on here over the past few months, how does this change anything? big deal it broke the world record for clock speed, but that has absolutely no relevance in the real world. let me think back in the day Prescott was breaking all the clock speed records, but the catch was it was terrible and got stomped by K8. don't understand why this news has suddenly changed the perception of the imminent Bulldozer...? :confused:
Click to expand...

Well it kind of is big news, recent AMD CPUs had both lower clockspeed and lower instructions per clock than Intel equivalents, this bench might not relate to real world performance but it does indicate that at least one of these two problems has been overcome. (Not that I think that a Bulldozer oc'ed to 5GHz will necessarily be faster than a Sandybridge @4.5, but at least it'll be closer than its predecessors.)

Also, remember that the previous record holder was a single-core Netburst Celly, the mere fact that a far more complex dual-core chip has beaten it is news.
 
Ehh high overclocks doesn't excite me much unless it has the performance to back it up. So hoping it will be about equal to Sandy Bridge at the same clock speed
 
alex24 said:
AMD also ran a smaller event in the UK, and I was privileged enough to be there along with Helios1234 from these forums on behalf of Vortez.net.

I've posted up some images and info about what we saw and did, no performance info yet though bar overclocking, everything is still under NDA.

Have a read here.
Click to expand...



2.004v :eek::eek::eek:
 
EniGmA1981 said:
Wonder what it could do with just one core enabled??

I think I would genuinely drool if I see a screenshot of one of these clocking up 10GHz lol
Click to expand...

You can't disable a core within a module, and it wouldn't have made any difference.

Hotwired said:
So by disabling 6/8 cores they manage to finally outclock the Celerons which were doing 8.2 - 8.3GHz

http://valid.canardpc.com/records.php

Crippled processors doing stunts to get impatient fans gibbering < proper benches at stock :mad:

Still, good on the OC team for the new record.
Click to expand...

What are you talking about, a Celeron IS a crippled processor to start with, and it is far MORE crippled than the Bulldozer that broke the record, hence it is fairly impressive. Most importantly, do Phenoms do 4.8Ghz on air/crap water AT ALL, therefore, bulldozer doing it..... very useful information.

What is a Celeron, it was a single core, most of the L2 cache removed, insanely cut down chip.

What does the bulldozer have, a module, 2 cores, far more circuitry, 2mb l2 cache, a lot more than the Celery has, and 8mb l3 cache, the celery had none. Would it have made a difference if AMD were releasing a chip called, Brocollon, which we'd call Brocolli, which had its l3 cache disabled, 6 cores fused off, and various other instruction sets not working(they normally remove cack like visualisation and a few other bits and bobs from Celerys) then did all the same tests, or "simulated" the same chip by just taking a full chip and doing it. Intel didn't break the world record with a "full" Prescott core did they, but a Prescott even more "crippled" than this Bulldozer was.

It is fairly impressive, and completely pointless as are essentially all world records.

That it should easily hit 5Ghz+ with good watercooling and will probably hit 5Ghz with a very very good air cooler(crap out the box watercooling isn't as good as top end air), is fairly positive, and thats with the full 8 cores.

I mean, people who bought them could have been limited at 4Ghz, it looks more like they'll be able to hit 5Ghz, so a 25% boost in in speeds seemingly possible over Phenom 2's, even if it had the SAME IPC, thats 25% direct increase in speed, and 4 extra cores, these are all fairly positive signs.
 
Last edited:
Helios1234 said:
Yup, Macci just whacked 2V on it and went straight to 6GHz to start off with then up to 8GHz before it crashed.
Click to expand...

i notice that on the world records list posted earlier most of those runs were done with similarly high voltage, but i suppose it is about pushing the boundries.
it will be interesting to see what the sandybridge E chips can do once they allow FSB adjustments as well as multiplier.
still good showing from AMD, all we need now is for them to actually release the damm things :)
 
come on, get the hell of it, how does the fact Bulldozer breaks a world record cooled by liquid Nitrogen and liquid Helium make any bit of difference, it doesn't! nobody uses that sort of cooling, guess what folks? a Ford Focus will do Mach 3 if you strap a Solid Rocket Booster to the top of it, but guess what, it make no difference because it has absolutely no real world relevance!

also I am finding the sudden change of tune hypocritical to the extreme, most of the people on here have been absolutely scything into AMD, writing them off at every single turn, coming up with the whole 'Instruction Per Clock is the only important factor, therefore Sandy Bridge will always be better than Bulldozer...' or things such like that. for goodness sake, the Official Bulldozer thread has suddenly changed its tune, people who have been slating the hell out of an unknown product for the past 6+ months are no suddenly interested, totally throwing out their 'its AMD, terrible IPC therefore rubbish...' argument out of the window...

we still know absolutely nothing of how Bulldozer performs, we know nothing about its Instructions Per Clock, nothing of its clock speeds other than some sketchy claims and this insane liquid Nitrogen/Helium run. so instead of sarcastically quoting what I said, how about go back and take a long think about a lot of the stuff that has been said in recent times, come back and stop seeming so hypocritical and changing the tune so quickly because of this little, somewhat pointless and irrelevant piece of information has came along. also how does it make any bleeding difference that Bulldozer clocks better than Phenom II, it was designed to clock higher from the off, so what does it really tell us, that AMD indeed did design it to clock higher...and?!

other than that sure, Bulldozer breaking the record is pretty impressive, suppose it 'hints' that AMD have managed to get their architecture taking to high frequencies like a boat to water. but crucially it tells us nothing we as potential customers what the performance is like compared to current processors. also don't bother with any grief because, well call it observations.
 
drunkenmaster said:
What are you talking about, a Celeron IS a crippled processor to start with, and it is far MORE crippled than the Bulldozer that broke the record, hence it is fairly impressive. Most importantly, do Phenoms do 4.8Ghz on air/crap water AT ALL, therefore, bulldozer doing it..... very useful information.

What is a Celeron, it was a single core, most of the L2 cache removed, insanely cut down chip.

What does the bulldozer have, a module, 2 cores, far more circuitry, 2mb l2 cache, a lot more than the Celery has, and 8mb l3 cache, the celery had none. Would it have made a difference if AMD were releasing a chip called, Brocollon, which we'd call Brocolli, which had its l3 cache disabled, 6 cores fused off, and various other instruction sets not working(they normally remove cack like visualisation and a few other bits and bobs from Celerys) then did all the same tests, or "simulated" the same chip by just taking a full chip and doing it. Intel didn't break the world record with a "full" Prescott core did they, but a Prescott even more "crippled" than this Bulldozer was.

It is fairly impressive, and completely pointless as are essentially all world records.

That it should easily hit 5Ghz+ with good watercooling and will probably hit 5Ghz with a very very good air cooler(crap out the box watercooling isn't as good as top end air), is fairly positive, and thats with the full 8 cores.

I mean, people who bought them could have been limited at 4Ghz, it looks more like they'll be able to hit 5Ghz, so a 25% boost in in speeds seemingly possible over Phenom 2's, even if it had the SAME IPC, thats 25% direct increase in speed, and 4 extra cores, these are all fairly positive signs.
Click to expand...

What are you on about.

You're ranting AND agreeing with the post I made, about which you ask what I'm on about.
 
I gotta admit, I was well up for moving to Sandybridge, but I will continue my love for AMD as this toy will be fun to play with. Curious to as why, 90% of people were dead against BD being any good at any level, be it a mad world record OC or simple usage, now it looks as though OCUK will have plenty of pre-orders
 
Gashman said:
guess what folks? a Ford Focus will do Mach 3 if you strap a Solid Rocket Booster to the top of it, but guess what, it make no difference because it has absolutely no real world relevance!
Click to expand...

No it wouldn't, it'd break apart when it reached Mach 1.
 
You must log in or register to reply here.
Back
Top Bottom