Another disaster for AMD: Phenom X4 slower than 65 nm Intel quads

[Wombat2]

http://news.expreview.com/2007-10-29/1193590532d6599.html

 

[chechnyan]

D A A M I T , i am afraid to believe this, if this is true then Intel make us pay 1xxx buck without mercy

 

[zenmaster]

And it may be worse than the chart shows.
The QX6850 and QX9650 scored identical marks on the tests despite the fact the QX9650 is the faster CPU due to larger/faster cache.

This indicates these CPUs hit a bottleneck and performance was limited by a factor besides the CPU.

Due to a slightly different design, the AMD may have hit the same bottleneck but not as certain since it was clearly slower across the board.

 

[Reynod]

What stepping was that ?? Oh ... an ES.

Right.

So we expect some improvement on the retail chip then ??

 

[Thanatos421]

Also note that the 6850 is dual-core.... very scary although I don't think I'll believe these numbers just yet. A dual core vs quad core on a game that supposedly takes all cores into account shouldn't even be close. I'll reserve judgement for now. Something is askew...

 

[JimyKing]

I still believe Phenom X4 could provide better C/P value.

 

[buzzlightbeer]

I dont think these benchmarks are true . I am an AMD supporter but seeing the dual-core intels get around the same as the quads something must be wrong and plus i think Phenom can do alot better .

 

[zenmaster]

I dont think these benchmarks are true . I am an AMD supporter but seeing the dual-core intels get around the same as the quads something must be wrong and plus i think Phenom can do alot better .

I will give you that something seems amiss.
I doubt the source tried to fake the numbers, but the range of testing was insufficient, and something seemed to be limiting the CPUs due to nearly identical numbers from the Intel CPUs which those CPUs show a broader range by other benchmarkings. I would not be surprised if the AMD chip's performance was not properly measured either.

I would really love to see AMD beat the Penryn Clock for Clock and have lower thermals which I think it may based on some of the Barcey tests I've seen.

 

[dragonsprayer]

zen your amd-fantasism will not save amd from the second place standing and future as a wallmart pc supplier of choice!

could be drivers - the drivers have been a sticking point in amd's side

 

[onestar]

Something is indeed amiss. I am in agreement with zenmaster.

 

[epsilon84]

Damn, so let's get this straight - an OVERCLOCKED Phenom X4 @ 3GHz fails to beat a stock 3GHz C2D, let alone a 3GHz C2Q.

Considering Penryn chips are routinely hitting 4GHz overclocks... while AMD is barely hitting 3GHz, AND with a clock for clock deficit - need I say more?

 

[chechnyan]

hey guys what happened to that AMD's 3GHz to break 30,000 3DMark06
http://www.theinquirer.net/en/inquirer/news/2007/08/28/amds-3ghz-k10-to-break-30000-3dmark06
i am not sure about that that either but still promising for something

 

[BaronMatrix]

There are two things noticeable right away. First, either they didn't try or AMD still hasn't fixed the RAM divider. 50MHz to 800DDR2 isn't earth-shattering but it would make up the 2-3% deficit. Second, HT3 is NOT 1600MHz, it's at least 2.4GHz. And the Agena is said to run at 2.6GHz. That's probably another 5-7%.

Admittedly Penryn could get to 4GHz under 200W so we'll hope that the seeming ease they got Agena to 3GHz will mean the retail Agena - the last quad due - will get the stepping that has been said to hit 3.2 for quad.

That was a quick and dirty test though so we don't even know if they have a retail mobo sample. Either way, it's pretty close. CrySis is TOTALLY GPU dependent though so that's a bad test to use.

According to benches even a GTX can barely get above 50fps at even 1280x1024 WITH NO AA. Maybe they should use a DX9 game which will be CPU dependent at 1280.

 

[zenmaster]

zen your amd-fantasism will not save amd from the second place standing and future as a wallmart pc supplier of choice!

could be drivers - the drivers have been a sticking point in amd's side

Are you completely unable to grasp benchmarks and analyze them like everyone else in this thread?
The Penryn is losing to the old QX6850 on two runs, basically ties two others, and really only wins one.
Also The Quads are not showing a significant difference to the Duel.
All signs the benchmark has hit a non-CPU bottleneck.
This makes the results unreliable for determining the full potential of the chips.

Add to that the Phenom is in a legacy MB.

I'm really shocked you are the only one who fails to grasp that the Intel Chips may be significantly better than what is shown. Or that possible ALL of the chips are much better than what is shown.

What everyone else in this thread has realized is that the benches do not reveal any real potential.

 

[bfellow]

How about taking into effect Crysis probably isn't optimized

 

[Falken699]

Well, on the flipside, at least Phenom will be less costly.

AND games are still GPU pigs.

AND you get a proper camlock HSF, no need to reinstall pushpin cheepy BS. Set it and forget it.

Thank god for competition.

 

[jwolf24601]

What's the deal with the 200mhz x15?

Shouldn't they run at higher FSB? Or is that one of those hyperthreading things?

 

[yomamafor1]

There are two things noticeable right away.

Some good points... and some bad points...

First, either they didn't try or AMD still hasn't fixed the RAM divider. 50MHz to 800DDR2 isn't earth-shattering but it would make up the 2-3% deficit.

Good point, but I'm afraid that RAM divider will continue to haunt AMD in the future. Since the die area is limited, AMD simply didn't have enough space to implement complex RAM logic in the memory controller. In addition to that, AMD would need to go back to the drawing board to come up with a new IMC. In Intel's approach though, since the memory controller is on the chipset, Intel can implement complex logic to accommodate more RAM speed.

I agree that RAM speed will indeed hinder Phenom's performance, but I'm not sure if AMD will do something to rectify that.

Second, HT3 is NOT 1600MHz, it's at least 2.4GHz. And the Agena is said to run at 2.6GHz. That's probably another 5-7%.

Bad point... bad bad point. Contrary to public, and especially your belief, HT3 will not cause a significant performance difference in desktop applications. Phenom is not data hungry, and having a wider (not faster) bus will not help at all; maybe on server, but definitely not on desktop.

Admittedly Penryn could get to 4GHz under 200W so we'll hope that the seeming ease they got Agena to 3GHz will mean the retail Agena - the last quad due - will get the stepping that has been said to hit 3.2 for quad.

Another bad point. So far we've not seen a 3.0Ghz Phenom demonstrated by a 3rd party (ie. Not AMD), and the faster Phenom tested to date is 2.5Ghz. According to VR-Zone, Phenom 2.5Ghz will likely to have a TDP of 125W, and a lot more by 2.7Ghz. Therefore, I'm not sure if 3.0Ghz Phenom will have lower TDP than Penryn at 4.0Ghz.

Thing could change though, as AMD will bin significantly lower leakage part for Phenom 3.0Ghz.

That was a quick and dirty test though so we don't even know if they have a retail mobo sample. Either way, it's pretty close. CrySis is TOTALLY GPU dependent though so that's a bad test to use.

I'm not entirely sure if Crysis is totally GPU bound. Its developer highly recommend the use of quad core over any other hardware component.

http://www.shacknews.com/featuredarticle.x?id=639

According to benches even a GTX can barely get above 50fps at even 1280x1024 WITH NO AA. Maybe they should use a DX9 game which will be CPU dependent at 1280.

Again, even though the demo seemed to have minimal effect on multicores, it appears that game is still slightly CPU IPC bound.

Although I agree it is still too early to hand down a verdict in terms of Phenom's performance, bear in mind that Phenom is slated to launch next month. Therefore, we probably won't see significant performance difference between the ES and the finalized version.

 

[Thanatos421]

There may not be significant performance difference between the ES and the production stepping, but can you at least see that there is something very wrong with this benchmark? Even Intel's own quad is neck and neck with it's little dual-core brother. Makes no sense.

 

[yomamafor1]

There may not be significant performance difference between the ES and the production stepping, but can you at least see that there is something very wrong with this benchmark? Even Intel's own quad is neck and neck with it's little dual-core brother. Makes no sense.

As I said, the game seems to be not multi-core optimized... but that doesn't mean its not CPU bounded. It could be clockspeed bounded, or IPC bounded. I'm leaning towards the latter, as there is still performance difference between Phenom and Core 2, despite a slight one.

What I find weird is that Crysis is not multicore optimized. According to its developers, having a quad core will result in a measurable, if not significant boost in performance. I guess we'll know after the real game comes out.

 

[Thanatos421]

Indeed. Maybe I'm just trying to justify poor performance from the Phenom. I love AMD, and I'm really looking for a reason to buy one over a Penryn quad. If this is any indication, then I guess I'll just be "blue".

 

[yomamafor1]

Indeed. Maybe I'm just trying to justify poor performance from the Phenom. I love AMD, and I'm really looking for a reason to buy one over a Penryn quad. If this is any indication, then I guess I'll just be "blue".

Well, TBH, the final Phenom speed is still up for grabs. Even though the chances don't look good (at all), there's still a possibility that AMD may scale Phenom without too much heat penalty. Given Phenom's near perfect scaling in clockspeed, and if Phenom and Core 2 has similar IPC (not likely though), Phenom should perform head to head, if not better than Core 2 at the same clockspeed.

We'll see though.

 

[enewmen]

Not upset. The Phenom is at least close. Good enough to keep Intel from dragging it's heels.
The final Crysis isn't released yet anyway.

 

[Heyyou27]

Also note that the 6850 is dual-core.... very scary although I don't think I'll believe these numbers just yet. A dual core vs quad core on a game that supposedly takes all cores into account shouldn't even be close. I'll reserve judgement for now. Something is askew...

Neither the beta or demo has taken advantage of more than two cores.

 

[TechnologyCoordinator]

I'd like to see a benchmark that tests the CPU more than the graphics card.