Don't start again. Fugger will get the mother of all unlocked chips. Why can't you all just be happy for him? Stop the :banana::banana::banana::banana::banana:ing.
Printable View
Don't start again. Fugger will get the mother of all unlocked chips. Why can't you all just be happy for him? Stop the :banana::banana::banana::banana::banana:ing.
I think as long as it happens in a respective way, it is okay to talk about why intel cherrypicks and unlocks CPU's as a response to AMD maybe doing the same.
On a sidenote, this whole event credits the importance of XS as a community, which is a cool thing I think.
What happens on XS matters! :)
Damn, I didnt see this thread :D
News to me, Im kinda busy that week too.
Cooper, Lots of good ones deleted. slow down on the damage control.
Im on my own schedule, it was just known publicly that Intel was coming out this time and media spinning this, hopefully AMD can confirm or does anyone have a link to where they got that info?
Uhm yeah.... I don't see any reason why my posts would have been deleted from that thread...
Anyways... I hope Fugger will let some I7's bleed.. but I still feel Intel taking this way too far..
It kind of reminds when AMD presented it's S939 CPU's...
Here guys, read the first 4 pages of this thread: http://www.xtremesystems.org/forums/...d.php?t=208374
DrWHO earlier indicated he was sure AMD had disabled some features; interestingly, Informal confirmed his suspicion? by actually confirming it (imho)
Quoting Theo:
And DrWHO replies:Quote:
AMD decided that the time has come to regain the crown in overclocking, and the company means it. You can thank all of this to several key members of former ATI Technologies, now AMD GPG. I know it is not fair to name the few, since I can’t mention them all, but guys such as Godfrey Cheng, Ian McNaughton, Korhan Eben, Macci, and many unsung heroes combined with old-school AMD executives such as Patrick Moorhead, who keeps his desktop machines OC’ed to 3.2 GHz and arguing with IT Police… All in all, this effort brought a lot of fighting spirit to AMD and these guys are now biting at all sides. First with Radeon series, then with chipsets and now with CPUs.
Phenom II is going to be extremely overclockable. How overclockable? Much.
With good air-cooler, 4.0 GHz is a given on almost every Black Edition CPU that will hit the stores starting January 8, 2009. This is nothing special, since Intel can do the same with Core i7 series. But, with water-cooling or TEC-enhanced water-cooling (hint: CoolIT), you can easily reach 4.5 GHz and beyond.
But, the real show starts when we dip below the zero degrees Celsius/Centigrade. History of AMD and sub-zero cooling is quite interesting. AMD was the first manufacturer to showcase vapor chamber and liquid oxygen cooling (yes, oxygen, not nitrogen) with Athlon 550 MHz that reached 1.03 GHz and became the first CPU to pass the 1 GHz mark. As the time went by, AMD slowly started to move away from Swashbuckler attitude lead by Jerry Sanders and put a dull corporate Motorola face lead by Hector Jesus Ruiz. All of this is past now, since AMD put substantial effort to produce a CPU which operating temperature is massive 300 degrees Celsius!
Yes, you’ve read it correctly – from -200 to +100, you can be certain that your Phenom will not fail. On-die sensors are tweaked up and they will not lock the part at -100 or -20C, and you can use dry ice or LN2 to crank it up to the max.
The max reached so far by AMD team is… 6.0 GHz! Given that most LN2 clockers use dual-core CPUs to reach 5.5+ GHz speeds, seeing 6.0 GHz CPU running applications such as Crysis is a testament to monumental effort put by AMD’s CPU and Chipset team. Don’t think this was CPU team only effort – guys from Markham worked hard on Advanced Clock Calibration, or ACC. ACC is exactly “the secret sauce” AMD needed to unlock these levels of performance.
Hitting 4.0 GHz on air, 4.5 on water, 4.5+ on TEC+water, 5.6 GHz on dry ice and there was a post at 6.0 GHz. Given the fact that dry ice is “weaker” solution than LN2, it will be really interesting to see what will happen when hard-core overclockers get their hands on these 45nm ice-cold babies.
AMD is back. And yes, Intel’s Core i7 thunder has just been stolen completely. Core i7 is a great CPU, but quite expensive platform, with three channels of DDR3. AMD Phenom II comes out with 16 GB/s of memory bandwidth using only two channels, and clocking like there’s no tomorrow.
The best part of them all is the price: AMD Phenom II 940 Black Edition will set you back for 40% of the amount you have to shell out for Core i7 Extreme 965 and yet, it comes with a radically cheaper platform of equal or even better overclocking capabilities.
So this is where the real debate lies, Theo claims every PII chip should do -200 to 100+, and Francois seems to claim that the tweaks were only applied to the bench-marked chips. Only time will tell.Quote:
AMD has tweaked the on-die sensor to not lock the part
On another note, Theo's comment about AMD's effort to claim the overclocking record should not be overlooked; imho it legitimizes every move Intel makes to counter than notion. So far, Intel is ahead, but like I said in another thread, they have learnt their lesson in the past and are not going to make the same mistake twice. I don't blame them if they're going after AMD, after all what was the purpose of AMD demoing the chip to the media, alonside a i965 XE?
It seems what AMD was hoping to avoid, a clockspeed war with Intel, is not going to happen. Personally, I love it; and I'm praying those PII tweaks come stock, who knows, Intel may follow suit and before you know it, we'll be getting BE (Blue Edition) Nehalems with unlocked multis (dreaming).
Yeah, my opinion on this is probably clear enough.
Still, I'm curious about the poopstorm that's going to down on Jan 9th (?).
January 9th.
LN2 galore. CES showfloor is snowing. AMD and Intel PR guys have snowfight and momentarily forget about competing in real life.
IF Intel is going with this, i want to see the quote up there accomplished first by Intel and than the 7ghz fiasco. Let the fun begin... :DQuote:
...seeing 6.0 GHz CPU running applications such as Crysis is a testament to monumental effort put by AMD’s CPU and Chipset team.
Yeah, damn it! It turns out he's an actual human with human feelings who likes to work for his company, how unfortunate.
I've only gained respect for him, when I first read his posts I thought he's a troll pretending to be someone from Intel. :D
Still I believe it's all FUD.
Or, and since it took so long for a reply, it's more likely that because of the backlash of this PR stunt, the event was pulled. Looks like not even 'the borg' could pull this one off. :D
I think there is no point in INTEL making a move, we all know intel overclocks good already.
And where did you get that idea; it's quite clear Theo was all over the place with that 6ghz figure; ie. OC or stable OC. Not to say this can't be done, but imho 6Ghz is a mighty oc on Nehalem, and if Fugger manages to get 7Ghz, 6ghz Crysis will be a walk in the park.
I'm talking about that 3GHz on stock. ;)
This demo http://www.youtube.com/watch?v=R7EZmYth6TM
You have an excuse because you've been busy fighting ghosts. :) But man go here: http://www.anandtech.com/cpuchipsets...spx?i=3360&p=4 or just use google. The web is riddled with reports of 3+ GHz Phenoms.
hummmm ....
read better the post from theo:
"On-die sensors are tweaked up and they will not lock the part at -100 or -20C, and you can use dry ice or LN2 to crank it up to the max."
:shrug: can you read it?
by the way, with Charles, we decided to stay without mods of the CPUs.
That AnandTech was using the SB750 which did not exist when the 3GHz demo was shown. :rolleyes:
I'm talking about cherry picked ones (like the one's FUGGER's getting). The 3GHz demo was using a standard heastsink. AMD didn't say it was overclocked either. :doh:
EDIT: Back then you should know the first Phenoms were the 9x00 series.
Actually Theo Said:
"Yes, you’ve read it correctly – from -200 to +100, you can be certain that your Phenom will not fail. On-die sensors are tweaked up and they will not lock the part at -100 or -20C, and you can use dry ice or LN2 to crank it up to the max."
Not sure why you decided to leave out that part of the quote, but that would implie (at least to me) that it was a tweak made to the core. Not diasabled sensors...