AMD states K8L aka Barcelona faster than all Intel cores

[MU_Engineer]

The Conroe runs at a 1.30V Vcore? That seems a little high, especially for 65nm. My laptop's 130nm Pentium 4-M runs at that same voltage, and my desktop's Athlon X2 runs on 1.35V. Supposedly K8 Brisbanes (65 nm) need 1.20V. Do you know why Conroe's Vcore is that high?

 

[hball]

AMD saying their new cores beats INTEL's new cores in performance

is ALOTTTTTTTTT better than going right to POWER CONSUMPTION numbers

leads me to believe they are either right...or believe they are right that BARCELONA is faster thant the core 2 architechture

You can say that again. Can't wait to see how it runs in tests. Could mean more price drops on existing processors for all concerned.

hball

 

[ElMoIsEviL]

It's nice to see that AMD seems ready to take on Intel again allready when the Quadcores (AKA Barcelona and Kentsfield) will meet.

I personally especially like the idea of the processor not using more power than needed, which might be the weak point in Intels armor, especially when the whole low power platform is developed(One of the reasons ATI will be AMD).

The only sad fact is we'll have to wait a while before Barcelona comes :? But that might just mean that Intel will have something to respond with(=Competition, which is good for us all)

CLICK ME!
CLICK ME 2!

Not really a weak point of Intel's. One must remember that Barcelona is a 2Q07/3Q07 product (and not K8L, it's based on the H revision of the A64 Greyhound will be AMD's first K8L based processor and it's release date is in 2008 (1H08 )). When you look at the big picture, you see just how much Bullsh!t is being fed to us by the likes of fanboi's everywhere. Wether they go by the name of Sharikou PHd etc they're all telling us and have us believing that K8L will be here by 2Q07, thats not true. K8L is a 1H08 product. Until then Intel will rule the roost. Barcelona will have a lower TDP and offer more performance per watt but not more performance per clock or more performance in general over Core 2 Quadro. Intel will rule the roost until K8L, only benchmarks that are incredibly memory bandwidth dependant will run better on 4x4 setups with AMD Dual Cores. Incredibly threaded apps will run better on 4x4 setups with dual AMD Quad Core processors. But all the apps we use as gamer's and enthusiast, for the most part will run better on Core 2 based processors.

By the time AMD's Barcelone hits Intel's roadmap shows a few months later (in 4Q07) the appearances of quite a few new technologies. One of which when implemented (45nm) will give Intel back the performance per watt lead.

Intel Xeon MP (Tigerton) CPU is expected to be released in 2007 as part of the Caneland platform. Tigerton is the replacement of the now abandoned Whitefield CPU and is expected to be a 4 core CPU possibly based on the Merom core and using Intel's CSI bus - akin to AMD's HyperTransport. Like Whitefield, Tigerton is being designed in Intel's design centre in India.

Intel 45nm process is expected to come online in 2007.

As for AMD, here's a description of Barcelona, Brisbane, Budapest and then K8L Greyhound.

2Q07

AMD Athlon X4? (Barcelona) is expected to be released in Q2. Barcelona is AMD's first quad core CPU based on a the 65nm process. Barcelona additionally introduces the Revision H core for quad processors.

AMD Athlon X2 (Brisbane) is expected to be released in H1. Brisbane is the 65nm successor to Windsor and introduces the K8 Revision G core. Brisbane features either 2x1MB L2 cache or 2x512KB L2 cache and the Brisbane core will effectively replace all Athlon X2 and Opteron Dual Core Revision E and Revision F processors with a single architecture.

4Q07

AMD Athlon X4? (Budapest) is expected to be released in Q4. Barcelona is the Hypertransport 3.0 version of Budapest, featuring the same Revision H core. Like Barcelona, Budapest features a Quad Core design built on a 65nm process. Barcelona will require updated chipsets due to the use of the HT3 protocol.

1H08

AMD Athlon 64 (Greyhound) is expected to be released in H1 2008 on a 65nm process. Greyhound is expected to be the first K8L based CPU, featuring 4 cores. Each core will feature 64KB - 32KB Instruction, 32KB Data - of L1 (down from 128KB in the K8 architecture), 512KB of L2 cache per core and - in it's Opteron form, 2MB of shared L3 cache. K8L will also feature AMD's DICE (Dynamic Independent Core Engagement) power saving technology which enables each core to alter it's own p-state (Power state) right down to putting a core in a full Halt condition and will introduce HyperTransport 3. Hypertransport 3 will introduce a number of improvements. Firstly, the HT speed will be increased to 2.6Ghz, which will allow for 5.2GT/s, compared with a maximum of 1.4Ghz in HT2 (1Ghz in the K8 architecture). Secondly HT3 will introduce 'Un Ganging', which will allow either one 16-bit link or two 8-bit links to be created on the fly. This will be particularly useful with multi-socket Opteron servers as it can allow for single memory hop access to memory which would previously have taken two hops. Additionally the K8L core will have an enhanced instruction set, Indirect branch prediction, 32-byte prefetch (compared with 16 in the K8 architecture), 48-bit addressing with 1GB pages, better cache coherency, I/O virtualisation, Memory mirroring, data poisoning and HT retry protocol support, and 2x128-bit SSE units (compared with 2x64-bit units in K8) featuring support for single cycle 128-bit instructions. Greyhound will interface to DDR2 memory, with the K8L core featuring support for FBD and, in a future memory controller revision, DDR3 and FBD2 support.

I'd love to see what rose colored glasses some fanboi's on here are looking through. Seriously.

 

[shabodah]

Here's an idea:

Quit worrying about CPU's and pressure Intel and AMD into getting hard drive performance up!

 

[ElMoIsEviL]

Here's an idea:

Quit worrying about CPU's and pressure Intel and AMD into getting hard drive performance up!

I don't think we need much more performance right now from the Hard Drive really.

It does help in Synthetic benchmarks and it would help under Virtualisation or heavy server usage but for the average Home User who either plays games or does work on there PC, a quicker HD doesn't make THAT big of a difference.

 

[shabodah]

For the AVERAGE user, these new CPU's aren't creating that much of a difference.

 

[jagmuss]

BS,

Although the link you provided does show 2 cores running 85% faster than one core, it is a large leap in logic to say that a single core Core 2 would be 54% slower than Core2Duo.

The bench marks provided were using CoV and were intended to demonstrate the effect of the Physics card and 2 cores while running complex PhysX alogrithms. I'm definitely not an expert (just a mid-level noob), but to extrapolate such a software specific application into a general statement that CoreSolo will generally suck seems somewhat...naive and misleading.

 

[1Tanker]

BS,

Although the link you provided does show 2 cores running 85% faster than one core, it is a large leap in logic to say that a single core Core 2 would be 54% slower than Core2Duo.

The bench marks provided were using CoV and were intended to demonstrate the effect of the Physics card and 2 cores while running complex PhysX alogrithms. I'm definitely not an expert (just a mid-level noob), but to extrapolate such a software specific application into a general statement that CoreSolo will generally suck seems somewhat...naive and misleading.

As are all of BM's posts.

 

[GlacierFreeze]

In response to Elmoisevil: Your sites say '08 and this one says '07.

K8L Q2 '07

Too many discrepencies in dates between sites. I guess we'll have to wait and see when they're in stores.

 

[HoldenMcGroin]

The only sad fact is we'll have to wait a while before Barcelona comes :? But that might just mean that Intel will have something to respond with(=Competition, which is good for us all)

Except AMD.

 

[turpit]

You make an excellent point. The C2D's individual core performance is what is powering the CPUs success. Were Intel to market a single core varient, it would outclass, by no small margin, any of AMDs A64 single core CPUs.

The last tests Anand ran showed an 85% increase going from 1 core to 2.

http://www.anandtech.com/video/showdoc.aspx?i=2828&p=3

That means if two cores get 45 fps one core will get less than 22fps.

Did you actually read the post you are replying to?

Your reply is pointless and has 0 (zero) bearing on the post subject

Yeah right dufus. He said Core2 single core would be faster than Athlon single core. Unless X2 is 85% faster than single, Conroe-L as it's called will have no chance except against Sempron.

Once again, did you actually read the post? Clearly not. Try again, and actaully read it this time.

Your second reply is as pointless as your first and has 0 (zero) bearing on the post subject

 

[carlhungis]

I am curious why this thread is still not locked?

I am also shocked to see that Jack finally got an avatar. Only took what.... 4500 posts?