!!!The Ultimate K8L Thread 2007 & Beyond!!!

[Serge84]

AMD is a serious company. If they are not hyping and are telling the truth this will kill the compatition for another 3 years specially in the 4 socket and 8 socket, nobody can touch them here. the other company will lose even more market to them.

If not then AMD will just be like that other company was to AMD like it was for the past 3 years. If so again we have the P4 days all over again with the P4 struggling to a higher clock speed and AMD with a even lower one out performing it.

[nn_step]

Often companies don't do this because they want to avoid the "Osborne effect" This is when the hype for future products causes demand for current products to drop off. When this happens, you have to slash the prices on your current products just to sell them.

However, if your new product doesn't really compete with your current products but does compete with your competitor's product, you want to talk it up a lot.

I think AMD is in both situations now with rev. H. On one hand, they want to hype up Barcelona because they currently have nothing to compete in the quad-core space, and they want to make potential quad-core customers delay purchasing decisions until rev. H is released.

On the other hand, if they hype it up too much they risk hurting their current dual core sales. Eventually their dual core prices will be so low anyway that the Osborne effect can be expected to be small.

With every X2 price drop, they get closer to the point at which it makes sense to start hyping rev H. I think they're close to that point already. I find it interesting that the most recent dual-core price cuts are rurmored to take place on February 12th. On the same date, AMD will be giving a talk at ISSCC about Barcelona. Maybe it's just wishful thinking on my part, but that would seem like a good time to release the first benchmarks.

I do find it suspicious that recently they seem to be trying to hype rev. H with no real data to back it up. The longer they keep this up, the worse it looks for rev. H.

Exactly, right now AMD will not do ANYTHING that could possibly hurt their market share. Besides not even with Conroe can Intel possibly touch AMD in the 4+P area (until tigerton finally works).

[ramenchef]

I dunno if Barcelona will have a 40% performance advantage, but at least it will put amd back into competition with Intel. Hopefully this will put them back on the path of performance dominance they once had.

[nn_step]

I dunno if Barcelona will have a 40% performance advantage, but at least it will put amd back into competition with Intel. Hopefully this will put them back on the path of performance dominance they once had.

http://news.com.com/AMD+Go+to+Barcel...l?tag=nefd.top

The quad-core chip also will outperform AMD's current dual-core Opterons on "floating point" mathematical calculations by a factor of 3.6 at the same clock

umm yeah.

And a little bit of facts for Serge:
Power6 has exactly 16 stages and it runs at 5Ghz (something intel couldn't do with 31 stage ) [Random associated fact, K8 has 12 and conroe has 14]

[Shintai]

Can we drop the 40% performance BS? Its SpecFP_Rate!

Its more or less like a memory bandwidth test.
For those that still don´t get it (Like some diehard fanbois above)...visit this:
http://www.spec.org/cpu2000/results/rfp2000.html

As example from same vendor, same modelrange:
2220SE 4 cores, 2 chips, 2 cores/chip 83.3 90.4 (2.8Ghz)
5160 4 cores, 2 chips, 2 cores/chip 82.0 83.0 (3.0Ghz)

That alone is the first 9% if you are into marketing on old K8s.

And in case the fanbois didn´t get it yet:
854 4 cores, 4 chips, 1 core/chip 90.5 98.3 (2.8Ghz)

OMG...4 singlecores are way better!!!! 18.5%! (And thats with DDR1! With DDR2 it would have been even higher.)

There is a reason AMD selected the benchmarks as they did. If you eat marketing PR spins like candy. Then don´t post in a place with speculation and rumour. unless you wanna join the console hype bandwagon.

So lets drop the BS numbers, its like saying Core 2 is 300% faster than K8 after running the Sandra SSE benchmark. Or the other way around with the memory benchmark.

[s7e9h3n]

I have a contact that says A2/A3 ES chips are in the wild, but he refuses to give me numbers yet (NDA). The architectural changes we know thus far put K8L/K10 in the same league (probably a step better) than C2D, but it is probably wishful thinking that K8L is going to stomp C2D in a manner similar to what K8 did to P4.

Hmmm....not too shabby of a guess Now I'm not going to get into specifics, but EARLY (note the emphasis on EARLY) ES AM2 quads are showing approximately a 10% performance advantage over Intel's offerings at the same clockspeed. Keep in mind that there may be/ will be changes before the design is finalized.....

*I'm already going out on a limb by disclosing this info so please excuse me if I choose not to comment any further...(at least for now)

[dinos22]

Hmmm....not too shabby of a guess Now I'm not going to get into specifics, but EARLY (note the emphasis on EARLY) ES AM2 quads are showing approximately a 10% performance advantage over Intel's offerings at the same clockspeed. Keep in mind that there may be/ will be changes before the design is finalized.....

*I'm already going out on a limb by disclosing this info so please excuse me if I choose not to comment any further...(at least for now)

wow looks forward to hearing more....let me know if you want me to edit out your quote in case you're asked to pull the info

[Shintai]

Hmmm....not too shabby of a guess Now I'm not going to get into specifics, but EARLY (note the emphasis on EARLY) ES AM2 quads are showing approximately a 10% performance advantage over Intel's offerings at the same clockspeed. Keep in mind that there may be/ will be changes before the design is finalized.....

*I'm already going out on a limb by disclosing this info so please excuse me if I choose not to comment any further...(at least for now)

Nice to see alot more realistic numbers

10% would be good. Hopefully abit more to keep the competition up due to their low clockspeeds.

[MAS]

EARLY (note the emphasis on EARLY) ES AM2 quads are showing approximately a 10% performance advantage over Intel's offerings at the same clockspeed.

it looks like truth
my own suggestion was K10 is 7 percent in 32bit and 12 percent in 64bit mode faster than C2D

[LowRun]

Thanks for the info S7e9h3n

[cumec]

@S7e9h3n

I hope it will become like you say!

[SOLDNER-MOFO64]

7% & 10%?

Yeah, whilst it boots up.

I think everyne knows AMD's track record. They didn't lie about they're performance increase with AXP over PENTIUM so I reckon it's a safe bet they've got it right this time also.

They ain't made this many revisions for a 10% BOOST. I reckon we're talking 15%-20% minimum.....and a lot higher on the HIGH-END OC'd baby's.

[dinos22]

7% & 10%?

Yeah, whilst it boots up.

I think everyne knows AMD's track record. They didn't lie about they're performance increase with AXP over PENTIUM so I reckon it's a safe bet they've got it right this time also.

They ain't made this many revisions for a 10% BOOST. I reckon we're talking 15%-20% minimum.....and a lot higher on the HIGH-END OC'd baby's.

now why post that when an AMD guy a few posts above you just told you what the numbers are

[Motiv]

now why post that when an AMD guy a few posts above you just told you what the numbers are

I think what SOLDNER-MOFO64 is saying is that s7e9h3n and his figures, will be improved upon by release (going off cpu optimisations and chipset improvements).

Well I think he's saying that.


@s7e9h3n - Thanks for the info. Much appreciated.

[MAS]

yeah 10% faster all the way huh?

10% faster in average like C2D faster K8 by 20% in average as well

[dinos22]

s7 do you know what sort of ihs the new cpus will have

i understand if you can't reply to this

[LowRun]

Any hints as to clock speeds (particularly how these might OC on decent AIR)?

He's risking NDA beatdown just by telling us what he has.

I think you just answered your own question.

[GMX]

lol

-K8L is not a K8 design.

-3-3.2 is a normal overclock for higher end K8 chips, not super chips. Super chips do 3.3 to 3.5 to pull rough figures out of my head. Get with the times.

- what the hell has anyones response to your trolling had to do with disagreeing with an over hyped pipe dream???

I agree there is a lot of eye rolling hype and whatnot in here, but it's to be expected. Unfortunatly for you though, the hype is a lot less annoying than your trolling.

It's quite obvious the soul purpose of your replies is to bag AMD, (hence you being labelled a troll) which i just DON'T understand, when, as an intel fanboi, you should be content and happy that intel are leading the cpu race atm. Seriously, it's bizare.

2.7ghz-3ghz on AIR is more common. Want to exaggerate more? This, my friends is what you call a fan-boy. Something this forum had very little of a while ago.

I think it may be you who's dreaming bud, I've only purchased 3x A64's that couldn't go over 3Ghz. My 3500+ WINNY does 3.1ghz, my 3700+ does 3.25ghz and my X2-4400 does 3ghz also...and those are just the ones I still own, nevermind the others.

As for superchip A64's, well....those goto 3.4ghz & 3.8ghz as I'm sure you know since they're ALL on this very forum

When it comes to nightmares I'm afraid INTEL are the only ones in REM-STATE. You remember 5yrs ago INTEL had a cpu that outperformed every AMD at the time. All the talk was INTEL,INTEL,INTEL. Then, from nowhere AMD released AXP and INTEL had trouble. They've since been inferior to AMD and that lasted until C2D's launch.

Now AMD are gonna let INTEL have some time in the sun, god knows they need it. That's almost a year INTEL have held the performance crown, only 3 more to go to equal AMD's previous title reign. Ofcourse that won't happen, since AMD are gonna launch K8L and INTEL are going back in the shade. Except this time it'll be more like the dark side of the moon

Clockspeeds mean nothing, 2Ghz A64's rape 4Ghz PENTIUMS, so whats the point in chasing them?

Regardless of TDP's, CLOCKSPEEDS, projected clockspeeds, DIE sizes, architecture ancestry......K8L are gonna make C2D look like PENTIUM. It'll be AXP v PENTIUM all over again

AMD FTW

Ummm... The Palominos were shocking clockers and P4A series were the most common amongst the community and also 2.4B. Intel gained the performance crown with P4C series. At that time it was really back and forth but the P4C just settled it. Also you are another example of a blind over-exaggerating "fan boy" Jeez, its a disease on this forum recently, turning it into dumb/blind/propaganda/insult central. Why don't u go join [H] or something.


My Opinion:
I think this little battle coming up will resemble something like AXP (Barton, Tbred B) vs P4C.
Can't wait to see initial results/previews.

[MAS]

I think this little battle coming up will resemble something like AXP (Barton, Tbred B) vs P4C.

imo like k8 vs. P4

[nn_step]

Not if K8L is only 10% faster IPC. K8 vs. P4 was a bigger gap (more like 20%).

so basically you think the Conroe vs K8L will be like P3 vs K7

[informal]

Not if K8L is only 10% faster IPC. K8 vs. P4 was a bigger gap (more like 20%).

If a 2Ghz K8 was roughly equal to 3Ghz Prescott,than its a bit more than those 20% of yours...

And BTW Brent,we don't know how much the K10 will be faster than Conroe.We have speculations tho.No hard numbers.So by stating the 10% general figure(what ever it means..),you risk being embarrassed by the real numbers.

[GMX]

If a 2Ghz K8 was roughly equal to 3Ghz Prescott,than its a bit more than those 20% of yours...

And BTW Brent,we don't know how much the K10 will be faster than Conroe.We have speculations tho.No hard numbers.So by stating the 10% general figure(what ever it means..),you risk being embarrassed by the real numbers.

No Steve, a highly informative AMD connected guy risks being embarrassed. And considering his strong links from the past, i doubt he will be.

All i'm looking forward to now is how far this thing will overclock on average. Then compare to a C2D/Q at an average overclock. That will make my decision.

[savantu]

10% faster on average at the same clockspeed ? Better be since it will be 25% slower wrt to clockspeed.

[Carfax]

A 10% average boost in performance over Conroe, doesn't seem like much, but when you think about it, it's very impressive because Conroe already has very high IPC.

It will be tough for AMD to beat Intel though. Even though there IPC may be higher, Intel's manufacturing process is second to none, and Core 2 can scale very well.

For quad core, AMD will probably exceed Intel though, because the monolithic design and higher bandwidth counts for more than in dual core.

[SOLDNER-MOFO64]

Ummm... The Palominos were shocking clockers and P4A series were the most common amongst the community and also 2.4B. Intel gained the performance crown with P4C series. At that time it was really back and forth but the P4C just settled it. Also you are another example of a blind over-exaggerating "fan boy" Jeez, its a disease on this forum recently, turning it into dumb/blind/propaganda/insult central. Why don't u go join [H] or something.

Now now, don't get out your pram over it m8......I'm an AMD FAN...so what?

I ain't being rude to anyone, and I don't blindly slate INTEL products. I own 2x INTEL pc's and 3x AMD aswell as numerous cpu's from both manufacturers. I merely prefer AMD is all.

No one in their right mind would insist C2D are garbage, or INTEL are useless. I've implied no such thing, only that I believe AMD's new architecture will run circles around the competition. Nor am I doubting the AMD GUY who posted figures, I am merely stating I think there will be even more of a performance gap come launch.

As for the remarks about joining other forums, you really need to get a grip of yourself and stop acting like a little INTEL FANBOY yourself. Jumping in and flaming ppl simply for expressing their opinion

No Steve, a highly informative AMD connected guy risks being embarrassed. And considering his strong links from the past, i doubt he will be.

I doubt he will be either. Why on earth would he be embarrased after claiming X% of performance gain if it actually turns out to be higher when the benchmarks are released? I just don't get that.