AMD's 3GHz K10 to break 30,000 3DMark06 (Inq)

[hipro5]

Thanks @ CraptacularOne
Learning everyday....

[T2k]

Sry for posting this more than once, but let's get this out in the open.

Enough crap. Let's show the official spokespeople.


AMD's Quads to outperform Intel's by over 40%

http://www.youtube.com/watch?v=G_n3wvsfq4Y



3GHZ demo is TRIFIRE:

http://www.youtube.com/watch?v=R7EZmYth6TM


So, unless these people are in fact lying(hence the Fudzilla comment yesterday), maybe it is far more possible than most think, and then again, maybe not.


And i gotta get off the internets.

If something then it IS CRAP. Typical PR BS with no substance whatsoever.

[Aussie FX]

lol...

re·tard /rɪˈtɑrd, for 1–3, 5; ˈri-tɑrd for 4/[ri-tahrd, for 1–3, 5; ree-tahrd for 4]

1. to make slow; delay the development or progress of (an action, process, etc.); hinder or impede.
–verb (used without object)
2. to be delayed.
–noun
3. a slowing down, diminution, or hindrance, as in a machine.
4. Slang: Disparaging.
a. a mentally retarded person.
b. a person who is stupid, obtuse, or ineffective in some way: a hopeless social retard.
5. Automotive, Machinery. an adjustment made in the setting of the distributor of an internal-combustion engine so that the spark for ignition in each cylinder is generated later in the cycle.

You forgot number 6. So I will fill it in for you.

6. CraptacularOne.

People like you are the reason I don't visit this forum much these days.

Bye.

[Movieman]

You forgot number 6. So I will fill it in for you.

6. CraptacularOne.

People like you are the reason I don't visit this forum much these days.

Bye.

And I will give you an infraction for not listening when someone nicely asks that the personal attacks stop..Keep it up and we can make it a vacation.
Thanks for reading!

[xlink]

The architecture is different, L3 cache etc. and everyone knows that when you overclock a processor it does not necessarily scale well... Explaining the disparity between a 5ghz c2d and a 3ghz Barcelona

unless the cache in unlinked from the processors, an increase in switching speed should result in a rather linear increase in overall CPU performance. Whether or not the memory subsystem and the system bus scales well enough is a different matter but hey highend DDR3 + a well OCed HTT bus probably scales well enough as compared to mid-highend DDR2 and a traditional 2GT/s HTT bus

[Movieman]

And you never stop being a mindless troll. If you have any idea of how a system allocates memory addresses you would understand what I was talking about.

You are easily the stupidest person I've ever seen in all of XS forums. Now how about you go out side and play hide and go f@#k yourself"?

And you just bought yourself a 3 day vacation for your nasty mouth.
Enjoy!

[Leeghoofd]

Admin powaaaaaaaaaaaaaaa !!!

[[XC] leviathan18]

oh MM with admin hat nice one.... where did you send him on vacation?

[Sparky]

The moderator speaketh!

Now, with the mod here, things will be nicer, no?

[Movieman]

oh MM with admin hat nice one.... where did you send him on vacation?

Just a short 3 day one..If it doesn't stop I'll ask NJ to come in and make them intergalactic vacations..

[rekleif]

And you just bought yourself a 3 day vacation for your nasty mouth.
Enjoy!

Thank you!

I think Theo is telling the truth, but he probably never knew the correct specs of the machine he was benching on. The only two possible answers imo is that it was run in trifire or in a less than native resolution. 30k on air is probably impossible if we take what the x.i.p's say at face value. However i would love to be able to recomend and build AMD boxes for highend customers again.

[Donnie27]

And I will give you an infraction for not listening when someone nicely asks that the personal attacks stop..Keep it up and we can make it a vacation.
Thanks for reading!

Eactly why I stayed the heck out of this one.

[doompc]

Link?
If this were true...

I went to the Barcelona presentation and asked directly to the presenter about this. He confirmed the integer divider limitations were solved.

I'm prety sure we'll have 1MHz step memory clock adjustment all the way up to about 750MHz (DDR2-1400).

Also, the IMC received a deep upgrade (larger write bufer, dual channel is now two independent 64bit channels, DDR2-1066 official support, the data path to the northbridge was extended to 128 bits, and so on).

30k still seems unreal, but I think it will be at least 10% faster per clock than Penryn in general application, and way more in highly multithreaded workloads.

[railer]

Just a short 3 day one..If it doesn't stop I'll ask NJ to come in and make them intergalactic vacations..

however you gave him a vacation for the post prior to you warning.............

[Donnie27]

however you gave him a vacation for the post prior to you warning.............

No, he warned the other guy

[Donnie27]

I went to the Barcelona presentation and asked directly to the presenter about this. He confirmed the integer divider limitations were solved.

Means absolutely nothing. Most of us don't believe anything Intel or AMD says without proof.

I'm prety sure we'll have 1MHz step memory clock adjustment all the way up to about 750MHz (DDR2-1400).

Also, the IMC received a deep upgrade (larger write bufer, dual channel is now two independent 64bit channels, DDR2-1066 official support, the data path to the northbridge was extended to 128 bits, and so on).

30k still seems unreal, but I think it will be at least 10% faster per clock than Penryn in general application, and way more in highly multithreaded workloads.

3DMarks has always proven its uselessness for anything other than like platform stability testing and etc..... Since it doesn't correlate to any real-world game/s, trying to use it as a standard is misleading to say the least. Please note, I said the same thing when P4 was winning*.

How much faster is K10? I don't think there will be a single Performance number advantage. Some apps it might be 25% faster, other apps might show it as being 20% slower.

[informal]

Eactly why I stayed the heck out of this one.

You are back in this topic now it seems.


Anyhow,2 weeks to go until we find out the truth.
Based on TechArp article,K10 will have 20-30% advantage against intel's counterparts(clock /clock) with excesses of 170% advantage in rare cases.
Sounds better than C2D did against K8.

[xenolith]

Hmmm...

K10 x RD790 x X2900XT CF x engineered and optimizated from the ground up within one Co. that has a proven track record for innovation = 30K 3DMark06? It's not impossible!

Hey... get away from my dream... it's all mine... *walks away laughing sinisterly*

[n91htmare]

Chances are he probably did see the 30K score, the problem is he's not sure of the test details or the specs.

All the demos we have seen of the machine include a Tri-fire system.

How much extra pump can we expect from a tri-fire Vs. a 2 card crossfire?

Also someone requested 3D06 marks on different resolutions on same system.

I would like to see such a comparison to see how much of difference test resolution makes.

[strange|ife]

dude its the inq. And this kids laptop was stolen? by whom the NDA thugs?

at any rate K10 is gonna be sweet, and so will penyrn/wolfdale

but yeah that had to of been mark05


I would like to see AMD gain ground next year

[TigeriS]

I'm not a cpu designer but I do however know a lot about the process of cpu design and manufacturing. It's my hobby. I love technology and innovation. I dislike "patches" and workarounds (all to an extend of course).

Anyway I'd like to share my 2Mhz.

IMO most posters forget one important fact. That is that noone knows how efficient a true quad core cpu is compared to a 2x dual core mcm solution.
Sure 30k is amazing and I also believe it was a trifire setup. But still it is not "impossible".

One must know the disadvantages AMD's K8 has over intel's Conroe. And then compare that to the K10. Most of Conroe's advantages are taken care of by the K10. One of K8's biggest problem is the "interchip bandwidth". The memory throughput is bottlenecked. That's why you don't see much efficient use of DDR2's bandwidth on K8's.

Almost every bottleneck has been taken care of. And remember we are talking about A Quad CPU Core, not a Quad Core CPU It's 1 Core with 4 CPU Units sharing 1 L3 Cache of 2MB. So who can tell what performance advantage you can get? It's probably not much in some benchmarks but it could make a difference in ohters. Also the memory controler seems to be clocked either higher or slower than the core it self.
That will also mean that the system memory could be able to run at default clock no matter the cpu clock. There are many more enhancements that could make the K10 perform well beyond expectations.

But to get the most of a K10 software must be recompiled. To make fully use of the FPU you don't have any choice anyway.

[ak_47_boy]

They must have had the gfx cards on liquid helium.

[CraptacularOne]

Well, 3 cards in Trifire isn't as impressinve at it sounded from the beginning is it now? I have no doubt that Barcelona is going to be good, but it isn't enought to save the R600

[Trunks]

They must have had the gfx cards on liquid helium.

lol liquid helium for a core that runs at 830MHz
I dont think so

[cky2k6]

i am almost positive that this is tri crossfire. the demo system going around, the "other announcement", and the fact that face it, its mathematically impossible for k10 to be THAT good. it still seems that its beast of a cpu, just not enough to beat a 5ghz kentsfield into a bloody pulp...