Tuesday, September 13th 2011

AMD FX Sets Guinness Record for Clock Speed

Weeks ahead of its market launch, AMD pulled off a nice PR feat by setting making its trusty squad of overclockers, Sami Mäkinen, Brian Mclachlan, Pete Hardman, and Aaron Schradin set a new clock speed world record (as in Guinness World Record). With just one of its four modules enabled, the eight-core FX-8150 engineering sample was overclocked to a stunning 8429.38 MHz. The chip was able to tolerate a brutal core voltage of 2.016V. Even for a one-in-a-million cherry-picked chip, those are staggering numbers.

8429.38 MHz was achieved using a base clock of 271.92 MHz, with 31.0X multiplier. The memory used was a Corsair Dominator GT single module, which apparently tolerated 3:10 DRAM ratio and timings of 2-16-2-22. That's right, 2-16-2-22. ASUS Crosshair V Formula seated the platform. Cooling was care of a custom liquid-nitrogen evaporator setup. The team used liquid nitrogen as its cooling medium, and switched to liquid helium halfway, which has a lower boiling point. The team cherry-picked chips from the best lots on-site.
A video of the feat follows.


This feat was more of a hit-and-run, in which the system could run at the desired frequency stable enough to make a CPU-Z validation, no proper stability testing was done. AMD claims that frequencies over 5.00 GHz were possible using sub-$100 cooling solutions (now that can be anything between a high-end heatsink and a cheap closed-loop liquid cooler). AMD did a similar overclocking feat ahead of its Phenom II processor launch.
Source: Overclockers.com
Add your own comment

225 Comments on AMD FX Sets Guinness Record for Clock Speed

#26
[H]@RD5TUFF
Completely BonkersEr, how is revving your 50cc motorbike/vespa to breaking point anything of interest. Seriously, if they had all 8 cores blowing and stable with world record benchmark numbers, then that would be worth noting. This isnt. Q. Can it run minesweeper? A. No.

Message to these boys: Listen. If you want to be a geeky nerd, fine. But do something to impress us. Don't have a camera focused on yourself like you are doing something that will put us in awe. When it didn't. Fail. Even bigger fail to AMD that thinks this is worthy. Corporate Fail. (And that's even worse!)
Gotta agree, this is a pretty fail attention grab.
Posted on Reply
#27
1Kurgan1
The Knife in your Back
I don't understand the flaming here. Anyone with half a brain knows this isn't a performance test, this isn't showing anything about what we will be receiving. They are just breaking a number that has stood for I believe 5 years or so, which is a long time for records to stand in the ever evolving PC world. You will see benches soon enough, if anyone was to go out and base their processor purchase (for daily use) on benching done using anything above water cooling, thats your own fault.
Posted on Reply
#28
DanishDevil
IMO, if this news doesn't make your nuts tingle, you don't belong on this forum ;)
Posted on Reply
#29
EarthDog
tricksonI agree . Lots of fluff right now nothing indicating any performance at all .
Fluff

Fluff

Fluff

Fluff

Fluff

Fluff

Fluff

Fluff


Ok.. we get it. :p
IMO, if this news doesn't make your nuts tingle, you don't belong on this forum
OMG That is sig worthy!
Posted on Reply
#30
trickson
OH, I have such a headache
What would WOW me is if they did this on all 8 cores ! Kind of seems a bit lacking when you have an 8 core CPU and have to disable all the cores but 2 to achieve some thing meaningless like this . But I guess if AMD wants to show off what they can do like this , They can . I want to be WOWED and this did nothing but piss me off . Show some real substance show some real performance not this BS fluff job ! AMD 8 core CPU can hit 8+ GHz if you have LN2 and all but 2 cores disabled ! WOW !
Posted on Reply
#31
linoliveira
gotta agree... all the flaming for what? this is an enthusiast run there... isn't where we all interested in here? i did get impressed with the high freqs, now just wait for RL benches and stop flaming like an Intel fanboy who got crashed by a Bulldozer.
Posted on Reply
#32
cadaveca
My name is Dave
DanishDevilIMO, if this news makes your nuts tingle, you don't belong on this forum
Survey says:


Corrected! :p
Posted on Reply
#33
trickson
OH, I have such a headache
DanishDevilIMO, if this news doesn't make your nuts tingle, you don't belong on this forum ;)
LOL . Honestly just how many of these will get in the hands of US ? Nothing news worthy here . I mean yeah good job getting that CPU up there and all , I do not see how this is relevant to any end user .
Posted on Reply
#34
EarthDog
tricksonLOL . Honestly just how many of these will get in the hands of US ? Nothing news worthy here . I mean yeah good job getting that CPU up there and all , I do not see how this is relevant to any end user .
This is clearly not for you...and you have made that abudently, flufflily, clear. :roll:

Some people, albeit a tiny minority are excited about this though.
Posted on Reply
#35
trickson
OH, I have such a headache
EarthDogThis is clearly not for you...and you have made that abudently, flufflily, clear. :roll:

Some people, albeit a tiny minority are excited about this though.
I just do not see what is so exciting is all . I am really wanting this chip so I am looking at it with eyes wide open . And if you know me then yes I like a great over clocking Chip as in my sig my CPU now has been going strong at 4.2GHz ! I hope these will do that and more yes 24/7 fully stable .
Posted on Reply
#36
repman244
I really can't believe some of the posts here....

First of all they need a design which they can improve (they don't have the money to make new designs all the time like Intel does) and a design where you can gain performance by increasing frequency and by doing small improvements (which are cheap!).
Second: a CPU to compete with mid-range CPU's (2500k for an example) because that's where the money is and not high-end $1k chips.

While doing all of that why not break a WR? I don't see a problem with this, are you all jealous of those clocks?
I think it's a nice way to promote the chip and don't see anything wrong with it.

All of those whining about that they only used 2 core...well did you even take a look at the current top 20 list? I guess not, so here you go: hwbot.org/benchmark/cpu_frequency/
Weird isn't it, all those celerons....
When you are after a frequency world record you do not care about the number of core (it is pointless...) all you care is the maximum you can get out of the CPU, that's how freq. WR's are set just deal with it.

Here comes the more important part:

This WR may be pointless to some but then again it shows just how much clock potential the CPU has for future (and this is only the beginning!).
Just like the first Phenom II's that barely made it to 6.4GHz on LN2 but at the end they got all the way up to 7.4GHz (and you could see that frequency increase on stock frequency at the same TDP).
If you think all of this is pointless, well you are on the wrong forum...

I could be wrong on this one but as far as I know JF-AMD said that IPC increases compared to the current Phenom II's, and if you combine that IPC increase with a much higher clock potential, isn't that a win?
Posted on Reply
#37
trickson
OH, I have such a headache
OK I guess you are right . I am looking at this all wrong . AMD hit a WR overclock now that is great YES it is . The wow factor is there for sure .
Posted on Reply
#38
Steevo
If you look at raw integer performance on the server chips compared to the same core architecture in server chips Intel is making we see they are still behind in CORE IPC, but in threaded IPC they are 30% ahead at the same clock rate from my math.

So multithreaded applications will benefit from these new chips, single threaded applications will require a higher frequency than the competitive Intel chip to attain the same performance. Single threaded apps currently are going the way of the dodo, and those still in use can be ran effectively by older processor, meaning current gen processors have more than enough to run them fine.

So at the end of the day, what does this mean?

I will take a overclockable 5+Ghz 8 core chip that offers superior performance than a 4 core hyper-threading chip, and does it at a lower price any day. Especially when the whole platform costs less than the equivalent platform from Intel.
Posted on Reply
#39
repman244
SteevoSo multithreaded applications will benefit from these new chips, single threaded applications will require a higher frequency than the competitive Intel chip to attain the same performance
Yes, IMO that's how it's going to be, the aggressive high frequency turbo is there for the ST workloads.

I think everyone should be happy with performance that is close to 2500k-2600k at same/lower price. I mean look how much behind SB Phenom II is...they are really doing a massive leap in performance to come close to it (everyone is comparing the FX chips to SB but no one compares them to Phenom II).
Posted on Reply
#40
trickson
OH, I have such a headache
repman244Yes, IMO that's how it's going to be, the aggressive high frequency turbo is there for the ST workloads.

I think everyone should be happy with performance that is close to 2500k-2600k at same/lower price. I mean look how much behind SB Phenom II is...they are really doing a massive leap in performance to come close to it (everyone is comparing the FX chips to SB but no one compares them to Phenom II).
Now I am getting all excited . My new build will be so cool BD !
Posted on Reply
#41
repman244
tricksonLOL . Honestly just how many of these will get in the hands of US ? Nothing news worthy here . I mean yeah good job getting that CPU up there and all , I do not see how this is relevant to any end user .
Well you might be interested in this then, directly from chew*:
I would not go as far as to say cherry picked. I blind pulled with no windows testing 24 chips based on there VID, the 4th chip we tested (note not a pretested ever chip ) did 8.4

We still had twenty chips left to try out but really didn't care after the 4th one.......
www.xtremesystems.org/forums/showthread.php?265710-AMD-Zambezi-news-info-fans-!&p=4948757&viewfull=1#post4948757
Posted on Reply
#42
EarthDog
Glad you/he posted that. I knew that tidbit but wasnt sure if I could share. :)
Posted on Reply
#43
exodusprime1337
SihastruIt's not bashing, it's observing a few inconsistencies. Is it not true that most of the chip is disabled? Is it not true that it says there on the validation page "2 cores, 2 threads"?

Wooden screws can be used in many instances, nVidia does not hold the IP on those. A "fanboi" would be someone blinded by their infatuation, 2 out of 8 is just 25%, and that's a big difference.

Wasn't AMD that complained recently about "dark silicon"? ( <== now that's bashing)
It's completely valid and it's how it's done on both teams amd and intel. I personally know chew who is in that video and i've been to his house and benched with him. The validation is only to see how hi the chip goes under any condidition, each core is raised independently of each other to find the one that goes the highest, then when they find that, they push that one the hardest. intell does the same thing.
Posted on Reply
#44
TheMailMan78
Big Member
1Kurgan1I don't understand the flaming here. Anyone with half a brain knows this isn't a performance test, this isn't showing anything about what we will be receiving. They are just breaking a number that has stood for I believe 5 years or so, which is a long time for records to stand in the ever evolving PC world. You will see benches soon enough, if anyone was to go out and base their processor purchase (for daily use) on benching done using anything above water cooling, thats your own fault.
DanishDevilIMO, if this news doesn't make your nuts tingle, you don't belong on this forum ;)
If this was an Intel chip the tune on here would be much different. But hey, thats life. I mean in the end its just a record which is in no way indicative of consumer performance. Its like drag racing is no way indicative of an every day driving experience.

However I do find it funny how defensive people get of one brand showing off something that couldn't count for less against another brand. I had no idea this forum was so eat up with fanboys. To much new blood.

This post made me lol
www.techpowerup.com/forums/showpost.php?p=2393800&postcount=2

Thats EXACTLY how it should be taken.
Posted on Reply
#45
TheoneandonlyMrK
imho its just nice to get some rock solid facts goin on, cant be long now untill we all Know what they will do in benches:D
Posted on Reply
#46
NC37
WarraWarraI wonder if this FX also suffers from Intel cold / flu when Intel i7 reaches 18C or lower and starts to work worse than above 18C or something like that.
AMD CPUs won't. One of the reasons they are behind in tech is because they make their CPUs able to handle temperature extremes, extreme cold mostly. I remember reading it takes extra work for them to keep designing for this. Intel doesn't do this. But this is more of a thing scientists would care about.
Posted on Reply
#47
TheMailMan78
Big Member
NC37AMD CPUs won't. One of the reasons they are behind in tech is because they make their CPUs able to handle temperature extremes, extreme cold mostly. I remember reading it takes extra work for them to keep designing for this. Intel doesn't do this. But this is more of a thing scientists would care about.
You don't think it has anything to do with the fact Intels R&D department has a budget the size of an aircraft carrier?
Posted on Reply
#49
lashton
really?

Peop,e are completely missing the point it shows the FX could (possibly) iverclock better than the 2600K but intel fanboys just shuddered in thier pants at this news
Posted on Reply
#50
cadaveca
My name is Dave
lashtonPeop,e are completely missing the point it shows the FX could (possibly) iverclock better than the 2600K but intel fanboys just shuddered in thier pants at this news
I think you mistake some people as intel fanboys, when really, this info really tells them nothing, and that's why they are not impressed. Most users care about what BD can do for THEM, 24/7, not what it can do for a handful of people, for a few minutes.

If it overclocks to 60 GHz, but is slower than the 2600k, does it matter that it can hit 60 GHz?


NOPE.

This event should have been saved for the launch, IMHO, and that's MY personal issue with it. Again, AMD marketing missed out on a golden oppotunity. Done right, they could have sold a CPU to every single person at the event, but htey cannot, beucase the CPU is still unreleased.

Great, a CPU that can set records...that isn't even for sale yet. I'd open up my wallet...but there's no point. And if htere's no point in me opening my wallet, there's no point in this news, IMHO.

Don't get me wrong though, it's good to see high clocks.

Of course, you also need to keep in mind that Guinness also has a record for the longest nipple hair. Yeah, that's where I want to be, next to the dude with the world's longest nipple hair!
Posted on Reply
Add your own comment
Apr 30th, 2024 02:26 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts