Phenom, debuting at 2.4GHz

[Hornet331]

Don't forget Phenom is already spoted at 3.33ghz

only spoted doesn't mean its gona be released... hell, there allready 6ghz QX9650 spotted, and it seems they run with no problem at 4ghz, still intel doesn't has released anything faster then 3ghz....

[justapost]

Slower than expected, AND higher TDP than expected.

But 9700 was expected in december and should now be released in time with 9500 and 9600, anyway i dont rust that news.

[terrace215]

ok but TDP is not ACP

Who cares about AMD's "ACP" marketing crap anyhow?

They wanted to talk about a lower "power" figure when comparing to Intel's upcoming low-TDP 45nm parts. So they invented a crummy "average" measurement that works out to be significantly lower than the TDP methodology they (and Intel) use. You're not going to fall for it, are you?

[terrace215]

But 9700 was expected in december and should now be released in time with 9500 and 9600, anyway i dont rust that news.

It's no longer a 2.6GHz part. It's become a 2.4GHz part. Hence, "slower". EDIT: oh I see. You mean there will be *some* 2.4GHz part in November. Ok, true. But then... NO 2.6GHz part in December. And the 2.4 is a power hog.

You don't trust that news. Well, Charlie is (if anything) pro-AMD, and the 9600 being 2.3GHz was backed up by a link on this very thread to a CPUZ shot of such a part.

AMD's general track record of disappointment with the K10 would also support this. I mean, you can't tell me you'd be SURPRISED at this point by another delay / speed cut / power increase from AMD, can you?

[Brother Esau]

Sheesh....Enough of the Drama already, give us the damn stuff and be done with it!

[Hornet331]

tell that amd.

[PhilDoc]

It wouldn't supprise me if this is true, but who knows, better to just wait. One thing, if this is true, then the chips we have been seeing reports on are at best a 2.4G cpu overclocking to 3G, most likely on a stock cooler. I'd take that and I'd probably give more than the 1.5V reported to get there.

I do agree with the Intel faithful, that I'd like to see AMD go back to bulk. Imho SOI is the reason so many AMD cpus have the cold bug.

[justapost]

It's no longer a 2.6GHz part. It's become a 2.4GHz part. Hence, "slower". EDIT: oh I see. You mean there will be *some* 2.4GHz part in November. Ok, true. But then... NO 2.6GHz part in December. And the 2.4 is a power hog.

That TDP at 2.4 is ugly. The inq does not cover release dates for 9800 or 9900.

You don't trust that news. Well, Charlie is (if anything) pro-AMD, and the 9600 being 2.3GHz was backed up by a link on this very thread to a CPUZ shot of such a part.

I'm abit confused about the FX numbering, thought it's comparable by clock to the 9xxx phenoms.
According to that VR-Zone post it's different but i think that changed a few weeks ago.
I can not believe that they are not able to handpick a small amount of fx cpu's being able to run above 2.6 GHz at stock, otherwise that numbering scheme is stupid.
But with the VR Zone's numbering FX-82 at 2.6 GHz the changed numbering for the AM2+ phenoms would make more sense.
EDIT: Looking at those screen shots there is a 2.0 GHz version labled GP-7000.

AMD's general track record of disappointment with the K10 would also support this. I mean, you can't tell me you'd be SURPRISED at this point by another delay / speed cut / power increase from AMD, can you?

I'm not surprised by another delay or speed cut. I'm just convused and curious about the numbering and the consequences.

[XS Janus]

http://my.ocworkbench.com/bbs/showth...697#post422697

vCore at 0.976v for 2.3GHz?? Sounds good...

[Pacha]

I guess it is cpu-z misreporting, otherwise 0.976V with such a high TDP... what did they do??

[Zytek_Fan]

I'm optimistic that Phenom will succeed, but I'll go blue if it doesn't for sure.

[Piotrsama]

Slow speeds at launch, lets hope they can ramp it fast.

[Ubermann]

Slow speeds at launch, lets hope they can ramp it fast.

Kinda SUX that they start showing a new gen CPU at 3.0ghz and then never release them ?!
Well its AMD..doing what their best at..and its not making cpus..

[OBR]

First 3 launched Phenoms will be a 2.3GHz, 2.4GHz, 2.5GHz (it cannot to compete with Intel Penryns) ...nothing more ... it is funeral march for AMD i think ...

[BrowncoatGR]

Who cares about AMD's "ACP" marketing crap anyhow?

They wanted to talk about a lower "power" figure when comparing to Intel's upcoming low-TDP 45nm parts. So they invented a crummy "average" measurement that works out to be significantly lower than the TDP methodology they (and Intel) use. You're not going to fall for it, are you?

Thats a lie. ACP is the same measurement way Intel uses. AMDs TDP is a theoretical max

[The_Beast]

INQ

[kl0012]

ACP is the same measurement way Intel uses.

Thats a lie. Intel's TDP isn't an "average CPU power".

[red]

Intel TDP is the max power that a CPU can run over a good amount of time is how I understand it, not "average power consumed". Yes this means that a 120W Intel TDP can consume more than 120W but throttling will fix itself if it has to. Hence why 120W Yorkies can be cooler than 95W Kentsfield. This is like how G0 Q6600 have thermal spec of 71C rather than B3 spec of 62.2C, meaning G0 can tolerate higher temps but is still cooler.

AMD TDP is max power that a chip can output before it is out of spec.

ACP is average power consumed. Specifically, it is the geometric mean (all power measurements added, then square rooted) of measured power of TPC-C, SPECcpu2006, SPECjbb2005, and STREAM.

Intel TDP and AMD TDP were fine as cooling specs for manufacturers. AMD ACP is a marketing thing.

[thenrz]

Well, you've got half the people around here saying that Phenom is going to be a flop, and the other half saying Phenom is only going to be competition in the lower market segment, at a low price.

Intel's Conroe/C2D/C2Q/Penryn is a powerful series of chips. It's been proven time and time again in every benchmark. What makes you think that AMD is not giving everything they have to remain competitive? They are a smaller company who are keeping the wraps on this one to a crazy exponential term. We've heard of a crazy 30000 something 3dmark(?) score with guys like Tony telling us its true, while other threads indicate that the X4s aren't much better than aging K8.

Right now, we've got just a huge 'blob' on info floating around. Intel fanboys use it as leverage to make themselves feel as if they have been right all along, AMD fanboys constantly pump new hope about some killer-ass product that will change the face of computing altogether. Who is right?

Nobody who is talking knows. In six months, everything will be apparent. For now, the intense lack of plausible information is cause for Intel vs. AMD catfights on every front. For the sheer sake of competition, AMD needs to pull the fourth ace on the river to make quads. Haha, the pun!

Honestly, threads like this just add to the confusion. Nobody knows who is right, no one has any clue what most of the facts are due to eleven million different threads spewing different information, but somehow the same people are in on each one. If you can find an AMD press release, I'd love to see it.

[Shintai]

Intel TDP is the max power that a CPU can run over a good amount of time is how I understand it, not "average power consumed". Yes this means that a 120W Intel TDP can consume more than 120W but throttling will fix itself if it has to. Hence why 120W Yorkies can be cooler than 95W Kentsfield. This is like how G0 Q6600 have thermal spec of 71C rather than B3 spec of 62.2C, meaning G0 can tolerate higher temps but is still cooler.

AMD TDP is max power that a chip can output before it is out of spec.

ACP is average power consumed. Specifically, it is the geometric mean (all power measurements added, then square rooted) of measured power of TPC-C, SPECcpu2006, SPECjbb2005, and STREAM.

Intel TDP and AMD TDP were fine as cooling specs for manufacturers. AMD ACP is a marketing thing.

Wrong, TDP is TDP. Nomatter if you are AMD or Intel.

Yorkfield just got an heavy overrated TDP due to no competition. But who would run the Yorkfield EE at stock speeds for 65-75W? Also introducing a lower than 130W TDP for the top line could penalize you in the future. Its easier to give than take. Aka. easier to lower than increase the TDP in a PR view.

[Zytek_Fan]

Wrong, TDP is TDP. Nomatter if you are AMD or Intel.

Yorkfield just got an heavy overrated TDP due to no competition. But who would run the Yorkfield EE at stock speeds for 65-75W?

Agreed.

I at your sig.

[oldblue]

ACP is average power consumed. Specifically, it is the geometric mean (all power measurements added, then square rooted) of measured power of TPC-C, SPECcpu2006, SPECjbb2005, and STREAM.

Close -- the geometric mean is the nth root of the product of n values. There is no addition.

[BrowncoatGR]

ACP is average power consumed. Specifically, it is the geometric mean (all power measurements added, then square rooted) of measured power of TPC-C, SPECcpu2006, SPECjbb2005, and STREAM.

You got a link for this because ive been lookin at various places and there's only a general description as the power consumption for a typical server load over a certain period of time.

[oldblue]

Wrong, TDP is TDP. Nomatter if you are AMD or Intel.

From a sense that it's an engineering target for both companies, that's right. But the targets are set in different ways. There's no standard for how to define TDP for your processors, and each company uses its own method.

Yorkfield just got an heavy overrated TDP due to no competition.

That makes no sense. If they sold Yorkfield with a lower TDP, it would be more valuable in the eyes of consumers (and system builders) and they could charge more for it. Competition or not, Intel is not going to leave money on the table.

[oldblue]

You got a link for this because ive been lookin at various places and there's only a general description as the power consumption for a typical server load over a certain period of time.

http://www.amd.com/us-en/assets/cont...A_ACP_WPv7.pdf