AMD K10 benches starting to leak: K10 20% slower than Clovertown c-f-c

[Wombat2]

http://www.xtremesystems.org/forums/showthread.php?t=157136

 

[jwlangs]

How come CPU-Z says the Phenom only has 1 core?

 

[Wombat2]

How come CPU-Z says the Phenom only has 1 core?

Its not a phenom ... its two quad core K10s i.e. a dual socket server rig.

 

[teh_boxzor]

how do you link and quote and do all that fancy forum crap.

 

[jwlangs]

So, CPU-Z sees 8 single cores in the dual opteron setup, and 2 quad cores in the dual xeon setup. Is this just some sort of nuance within CPU-Z?

 

[speedbird]

The second page of that link shows the barcelona completing a SuperPI test in 39 Seconds. My 4400+ X2 does that superPI test within a similar time. I'm not completely sure these Barcelona benchmarks are trustworthy, but we'll find out when it's actually released. Such benchmarks are not to be taken seriously anyway because Game benchmarks and real world benchmarks will give a far more accurate indication of performance.

 

[Wombat2]

So, CPU-Z sees 8 single cores in the dual opteron setup, and 2 quad cores in the dual xeon setup. Is this just some sort of nuance within CPU-Z?

Its the type of nuance that occurs with every new chip the cpuz author has not updated cpuz for.

 

[the_vorlon]

Ouch

If this is true, at least on these benchmarks, AMD is in deep sh*t

8 Cores of Deepfritz gets 8000 ish on Barcelona, gets 13000ish on a clovertown.

bad.

 

[Falken699]

Woah, bad news!!! No wonder Henri Richard bailed AMD... EVERYONES sinking ship rat analogy was correct! DAMN THOSE RATS!!! Somehow I think the architecture will only really start to shine over 2.6GHZ, so I hope they are big power savers in the AM3 boards, or are priced really low, we'll see how their stock performs when this news is mainstream...

 

[BaronMatrix]

http://www.xtremesystems.org/forums/showthread.php?t=157136

Not to be a stickler but the AMD system s running Generic GDI while the Intel system is running a 7950GX. Also, if you notice, the 2.4GHz Intel chip scores THE SAME as the 3.0GHz Intel, around 17000.

Hmmmm.

 

[Wombat2]

Not to be a stickler but the AMD system s running Generic GDI while the Intel system is running a 7950GX.

Cinebench has nothing to do with GFX cards :lol:

Prediction: CPUs are going to get a lot more expensive in the next 18 months.

 

[BaronMatrix]

Cinebench has nothing to do with GFX cards :lol:

Prediction: CPUs are going to get a lot more expensive in the next 18 months.

The point was that when comparing systems, everything should be the same. Also, all the Intel chips are clocked higher. The real SPEC numbers will come out on the 10th along with any other official benches they did.

I can wait.

 

[chunkymonster]

The point was that when comparing systems, everything should be the same. Also, all the Intel chips are clocked higher. The real SPEC numbers will come out on the 10th along with any other official benches they did.

I can wait.

That was my initial thought as well, mostly tho; what mobo's and chipset were these "tests" run on. I'm always suspect of benches like this especially when CPU-z hasn't been updated to fully recognize the proc. What else in the test is not fully recognized?!?!

I can wait too. Especially for more "reliable" site's and benches.

 

[3Ball]

Hopefully this isn't true, but we will know soon enough. I may even get a chance to test one out myself around launch because even if these things are terrible im sure my roommate will be getting one! woot! lol

Best,

3Ball

 

[BaronMatrix]

For those who say Phenom is nowhere to be found and is slower here is a quote from Rahul Sood of VoodooPC.



So, AMD decided to unveil Phenom running at 3.0GHz without showing actual benchmarks. What it showed was a game running smoothly with all details enabled, which makes perfect sense to me. And for the record, if you were to benchmark Phenom at 3GHz you would see that it kicks the living crap out of any current AMD or Intel processor%u2014it is a stone cold killer (at 3GHz, now imagine how it would perform if they could squeek some more juice out of it?).


Now here it is from a reputable OEM who has no reason to lie and definitely has access to the chips.

http://www.rahulsood.com/2007/08/benchmarks-are-wiggedy-wiggedy-whack.html


He also reiterates what I said last year BEFORE AMD, benchmarks aren't the be all and end all of a computing experience.

ALL HAIL THE DUOPOLY!!!

 

[lghtmup98]

I would see that as being more of a gpu demonstration then a cpu demo.....

 

[Mephistopheles]

it kicks the living crap out of any current AMD or Intel processor - it is a stone cold killer

Still don't buy it. I don't care, even if the pope were to say "Phenom is great! Have faith!".

And about the 3Ghz "kicking the living crap" out of any current AMD or Intel processor, who the hell cares? Current processors will always get kicked around by future processors.

I'm guessing that AMD will be able to launch some parts at higher clocks than it is currently showing in its roadmaps

Yeah sure, you're guessing. It's all like a poker game, sure. They're bluffing. They're not betting big money, but they actually have a royal flush in their hands. They're going to surprise everyone by actually winning! Those treacherous bastards!

It's interesting to note that AMD isn't showing benchmarks on a part that delivers the goods - perhaps it too is seeing that performance benchmarks are only a small piece of the overall experience puzzle.

OK, the whole "Experience" thing stands to reason, but the big question here is: if, like, 90% of IT people base their decisions on performance benchmarks and 10% in the "Experience", why wouldn't you show the vast majority of potential customers what they want to see? I don't care if this guy owns Vodoo or even is Bill Gates himself. He missed a very important point: show the customers what they want to see, not what you think they should want to see.

Simple enough, eh? It's not rocket science.

Also, gamer experience as in image quality is much more video-card-dependent. Heck, even system responsiveness depends a lot on video cards nowadays, with Vista's interface. My conclusion from this article is much more in the lines of "get a better video card, it'll give you a better experience" than "Phenom rocks".

 

[Rabidpeanut]

Why speculate, we will either see amd get kicked in the nads again or we won't everything out there now i likely a lie and thus it is hard to tell which bench is actually real. so just wait until it comes out.

 

[Reynod]

Like the baron said ... I can wait.

In the meantime Intel will go on a massive campaign of FUD.

AMD will stay silent ... laywers re-reading the NDA's with the partners ... trigger fingers twitching silently.

Anand will have rewritten his top shelf analysis of the state of play about 30 times by then.

I'll read what he has to say first ...

 

[function9]

Like the baron said ... I can wait.

If by waiting you mean waiting a couple hours so you can find a post that backs up your point of view, no matter how obscure it is, then ok.

In the meantime Intel will go on a massive campaign of FUD.

Yes, because it has been them all along that's been releasing slides about their next chip being the second coming and pointing out every obvious shortcoming of competitor's chips. All the while not releasing a single bit of hard data. Yes Intel are such FUD-mongers.

 

[function9]

For those who say Phenom is nowhere to be found and is slower here is a quote from Rahul Sood of VoodooPC.



So, AMD decided to unveil Phenom running at 3.0GHz without showing actual benchmarks. What it showed was a game running smoothly with all details enabled, which makes perfect sense to me. And for the record, if you were to benchmark Phenom at 3GHz you would see that it kicks the living crap out of any current AMD or Intel processor%u2014it is a stone cold killer (at 3GHz, now imagine how it would perform if they could squeek some more juice out of it?).


Now here it is from a reputable OEM who has no reason to lie and definitely has access to the chips.

http://www.rahulsood.com/2007/08/benchmarks-are-wiggedy-wiggedy-whack.html


He also reiterates what I said last year BEFORE AMD, benchmarks aren't the be all and end all of a computing experience.

ALL HAIL THE DUOPOLY!!!

Yes because salesmen are some of the most trustworthy people on the face of the earth.

 

[turpit]

Not to be a stickler but the AMD system s running Generic GDI while the Intel system is running a 7950GX. Also, if you notice, the 2.4GHz Intel chip scores THE SAME as the 3.0GHz Intel, around 17000.

Hmmmm.

Not to be a stickler, but AMD has been basing benchmarks off simulated CPUs at estimated clockspeeds against Intel processors using integrated graphics

And AMD has been caught and called on it by so many sites now that not only have they removed the graphs from their site, but also any mention of comparitive performance.

http://multicore.amd.com/us-en/AMD-Multi-Core/Products/Barcelona/Performance.aspx

Little more than one month ago, George Ou of ZDnet posted this story:

http://blogs.zdnet.com/Ou/?p=567

AMD posts blatantly deceptive benchmarks on Barcelona

George Ou of ZDnet wrote:

* Not real product. Fastest Barcelona being released in September is 2.0 GHz

As you can see from above, AMD’s claim that they have a 20% clock-for-clock advantage with Barcelona is simply wrong. Based on the latest certified SPEC.org results, AMD has a little more than a 1% clock-for-clock performance advantage in a dual-socket 8-core Server configuration but they have 50% clock speed deficit when the Barcelona finally launches in September. That means Barcelona will not be the Intel quad-core killer that AMD has been promising for most of this year and it won’t even be close.

The deception doesn’t end with the quad-cores; AMD is also claiming to have an advantage on dual-core processors when in fact they have a major performance deficit. AMD claims to have a 2.5% advantage when Intel actually has a 14.7% advantage when you’re looking at the certified SPEC.org scores.

 

[Reynod]

Hey Baron ... I'm with you ... hoping AMD deliver something competitive.

I just hope I can get a quad core cheap and some decent performance improvement over my dual cores.

I'd also like the possibility of doing that by just swapping a chip ... not a mobo transplant. Hence ... fingers crossed.

As that such an inrealistic ask?

And if AMD does a bit of creative markinging to boost their shakey financial position it isn't as if Intel hasn't done that before.

We remember the last of the netburst rubbish they put out to compete with the A64's.

We remember the massive FUD campaign there.

We hope the marketing practices Intel have been "allegedly" engaging in over past years earns them their just penalties in court.

AMD could do with a few billion to wipe their debt.

All hail the duopoly as previously aid.

I think the server world is going to benefit the most ... performance per watt ... from the new AMD quads.

I think the high end gaming wold is probably going to continue to see Core2 keeping most benchies ...

Happy with that.

Looks like they have optimised the K10 with a key priority for multithread applications, inter core communication, and tried to improve IPC without a drastic redesign. Plus also focussed on a huge emphasis on power savings ... for servers.

Lets see what pops up then.

 

[Harrisson]

Yes because salesmen are some of the most trustworthy people on the face of the earth.

Hmm, doesnt he sell BOTH cpu's? If so, its the same for him which sales best and your sarcasm is out of place here, HP most likely have few babies up and running while preparing their new lines of PCs for the autumn/winter seasons.

 

[function9]

So hyping a product you sell isn't good business? Ok then. It had nothing to do with Intel or AMD. But with the fact that the guy is a salesman first and foremost. I fail to see how my statement about salesmen is out of place. I would love to know though, what one has to do with the other. Please inform me.