Deneb Samples are almost out

[Chad Boga]

I for one have never liked Overclocking that much, yes yes I know, WTH?!

But it's my personal preference, if AMD came out with a 3+Ghz product that performs as well of better than my current e6850 I'll happily upgrade.

If it only performs as well, what kind of "upgrade" would it be?

[Helmore]

If it only performs as well, what kind of "upgrade" would it be?

Some people just change system because the want something new and different from what they previously had, though they are probably a minority. Although I can imagine that most will want to change clothes quite often, in that case the opposite is usually the minority .

[inCore]

Yes but in comparison to clothes, it's harder to spot a new CPU when looking at a computer than it is to see my new pink collection of exclusive menswear.

[003]

That being said a 3.5ghz i7 will get romped by a 4ghz Deneb IF

No. There is no "if". Just no.

Deneb is just a die shrink of Agena, and at best will offer 5-10% performance improvement clock for clock. It's like the R600 to RV670... just a die shrink, with maybe some minor instruction set optimizations as well. How do we know this is the case?

AMD has said nothing about architectural improvements, because there are none. It's the same architecture, just a die shrink, with some very general talk of "25% performance boost". My arse.

It was the same with the transition from K8 to K10. AMD was silent on anything relating to architectural improvement, and instead talked about HT 3.0 and DDR2 and other things that don't really improve performance. And same with R600 to RV670.

But you notice that with RV670 to RV770, rumors of it's monster specs were flying out of control, and they proved to be true.

The only saving grace Deneb may have is ability to easily overclock to the range of 3.6GHz, but even now most Q6600s will go over that mark. And don't even try and compare Deneb to Nehalem ... it will be absolutely no contest. Nehalem is simply two or three leagues above Deneb in terms of performance, and it will rape (or is raping, I should say) in orifices that do not exist naturally.

[adamsleath]

Nehalem is simply two or three leagues above Deneb in terms of performance, and it will rape (or is raping, I should say) in orifices that do not exist naturally.

it would want to considering the price of the mobo and chip.

im keen on a deneb meself with one of those dfi 750sb 790 boards....maybe. at least i can afford this, although i am a little concerned about stability...

Some people just change system because the want something new and different from what they previously had, though they are probably a minority.

call it a fetish.

[Hornet331]

No. There is no "if". Just no.

Deneb is just a die shrink of Agena, and at best will offer 5-10% performance improvement clock for clock. It's like the R600 to RV670... just a die shrink, with maybe some minor instruction set optimizations as well. How do we know this is the case?

AMD has said nothing about architectural improvements, because there are none. It's the same architecture, just a die shrink, with some very general talk of "25% performance boost". My arse.

It was the same with the transition from K8 to K10. AMD was silent on anything relating to architectural improvement, and instead talked about HT 3.0 and DDR2 and other things that don't really improve performance. And same with R600 to RV670.

But you notice that with RV670 to RV770, rumors of it's monster specs were flying out of control, and they proved to be true.

The only saving grace Deneb may have is ability to easily overclock to the range of 3.6GHz, but even now most Q6600s will go over that mark. And don't even try and compare Deneb to Nehalem ... it will be absolutely no contest. Nehalem is simply two or three leagues above Deneb in terms of performance, and it will rape (or is raping, I should say) in orifices that do not exist naturally.

My personal ranking of cpus according to IPC (Desktop) -> Nehlaem>Yorkfield>Deneb>Kentsfield>Agena

But i think server market will be a interesting battle, shanghai vs gainstown will be the big showdown.

[Boissez]

but i really like to think about deneb as a RV770, which means a RV670 done right
i dont expect it to beat nehalem, but if it is a K10 done right, it may end op close in some scenarios

at least deneb will offer full control over the cpu, unlike nehalem where only the extreme part is usefull for overclocking and high mem speeds

I'd be more inclined to compare K10 to R600 and K10.5 to RV670 - iow Deneb will be nowhere near the top spot but at least it will be a sensible mainstream option - and maybe a return to profitability for AMD (although not likely since it's about to be spanked in serverland).

Anyways I hope that AMD will manage to match Penryn clock for clock. Then AMD will at least have a reasonable platform for the desktop and mobile markets.

[adamsleath]

i wanna see deneb vs yorkfield.

[Caveman787]

003 stop flaming this is a DENEB thread.

It's already been stated several times that ipc it's 15-20% faster not counting the clock speed increase.

i7 is awesome no doubt but don't reign on Deneb just because i7 shows good promise now.

Deneb shows promise as well and until we get some good numbers don't go counting it out yet.

[003]

003 stop flaming this is a DENEB thread.

It's already been stated several times that ipc it's 15-20% faster not counting the clock speed increase.

i7 is awesome no doubt but don't reign on Deneb just because i7 shows good promise now.

Deneb shows promise as well and until we get some good numbers don't go counting it out yet.

Not flaming, telling it like it is. I really don't care about intel or AMD, and I am not a fanboy of anything. Will deneb be a viable option for people in different situations? sure.

Will it even match kentsfield clock for clock? Nope.

Go on and tell me I am flaming, I honestly don't care. You will all see when deneb is out in the wild, so it makes no difference if you try and disagree with me at this point.

I'm sorry but I can't just sit around with my thumb in my ass while there is an entire thread of people who seem to think that deneb is going to match yorkfield or even kentsfield clock for clock. The difference between yorkfield and kentsfield is only 10% at best anyway, so people saying deneb will fall right between them are even more out of their mind.

[adamsleath]

i still wanna see deneb vs yorkfield.



Black Edition

[malice85]

Not flaming, telling it like it is.

How do you know? I like my facts proven, so if you don't mind to do so I'd be very thankful

[Cooper]

Not flaming, telling it like it is. I really don't care about intel or AMD, and I am not a fanboy of anything. Will deneb be a viable option for people in different situations? sure.

Will it even match kentsfield clock for clock? Nope.

Go on and tell me I am flaming, I honestly don't care. You will all see when deneb is out in the wild, so it makes no difference if you try and disagree with me at this point.

I'm sorry but I can't just sit around with my thumb in my ass while there is an entire thread of people who seem to think that deneb is going to match yorkfield or even kentsfield clock for clock. The difference between yorkfield and kentsfield is only 10% at best anyway, so people saying deneb will fall right between them are even more out of their mind.

Assuming what you telling is true you already have working Deneb platform and done all sorts of benches Vs Yorkie and Kent. Perhaps you might as well share the results instead of showing us crystal ball ?
Just by dumb arithmetics Phenom+15-25% improvements will go toe to toe with Yorkie in most cases.

[003]

How do you know? I like my facts proven, so if you don't mind to do so I'd be very thankful

Just wait until it's released and benched on Guru3D and all the sites of the like. If you don't believe me now, you will find out soon enough.

[informal]

Not flaming, telling it like it is.

Will it even match kentsfield clock for clock? Nope.

You claim this based on what? You do know that Agena is 5-8% slower per clock then Kenstfield?How can you claim AMD can't manage ever so small IPC gain(3x larger L3 alone can probably account for this)?

Simple question: Have you seen Deneb tests? Can you prove you saw them?Can you prove Deneb can't "match" kentsfield? If not you are flame baiting people here and trolling along the way.

edit:

Just wait until it's released and benched on Guru3D and all the sites of the like. If you don't believe me now, you will find out soon enough.

Oh and since when has Guru3d become an "expert" site for CPU evaluation and testing? If one anandtech which has been around for so many years still can't be called that,how can Guru3d?(known mostly because of their GPU reviews and nothing else)

[stangracin2]

Just wait until it's released and benched on Guru3D and all the sites of the like. If you don't believe me now, you will find out soon enough.

quit being a troll


imo deneb will just beat the yorkies
and for some reason i get the feeling well be seeing a phenom @4.1-4.2

[Helmore]

Just wait until it's released and benched on Guru3D and all the sites of the like. If you don't believe me now, you will find out soon enough.

You are stating things as if they are facts, while there is no proof to it as of yet. Maybe you should on make predictions and when you do you should make it clear that you are doing so. Or you could provide hard proof or something, but for now it only seems to most other that you are flaming.

BTW, my prediction is a 10% IPC average improvement over Agena for Deneb, just a prediction though. That should put it right between a Kentfield and a Yorkfield, which would make it an interesting option for now. I've no idea how things will turn out for the next year though, I have this feeling that Deneb will pretty much loose its value once Intel launches their socket LGA1156 Nehalem derivatives.

[003]

Oh and since when has Guru3d become an "expert" site for CPU evaluation and testing? If one anandtech which has been around for so many years still can't be called that,how can Guru3d?(known mostly because of their GPU reviews and nothing else)

Well you got me there I must say. Yeah, all the benchmark sites are wrong. How could I have been blinded for so long?? In fact, Agena is almost neck and neck with kentsfield as it is, looks like deneb will be a formidable competitor to i7 after all. Intel better be worried, AMD is making their comeback.

[Cooper]

Guys let's just agree to disagree. Everyone has a right for his own pink world.

[AbelJemka]

Actually yes

http://www.overclock.net/intel-cpus/...-out-wild.html

Same picture than 2 days ago...
So one guy manages to get one...
If AMD do launch like this it's a "paper lauch", no? (kidding!)

[Hornet331]

BTW, my prediction is a 10% IPC average improvement over Agena for Deneb, just a prediction though. That should put it right between a Kentfield and a Yorkfield, which would make it an interesting option for now. I've no idea how things will turn out for the next year though, I have this feeling that Deneb will pretty much loose its value once Intel launches their socket LGA1156 Nehalem derivatives.

yup same here, i predict up to 10% IPC increasment for Deneb over Agena which will put it right inbetween Yorkfield and Kentsfield.

20-25% is just insane, maybe some rare cases (like spec_rate score, which to according to amd we all know is a "wide variety of workloads" *scnr* ) but most of the time i doubt we see more then 10%.

Remember that Deneb is still the 3 issue architekture that barcelona was and 3rd level cache cant fix everything, espcial when you already got a IMC.

[Shintai]

Same picture than 2 days ago...
So one guy manages to get one...
If AMD do launch like this it's a "paper lauch", no? (kidding!)

You asked if it was available. It was.

Would be nice if AMD could release the chips soon. But being quiet usually means there is no bang. Plus it might not be before 2009. I wonder if they stretch it like 65nm, just to be able to put a 45nm checkmark in 2008.

Also for people to expect 15-25% of deneb? Jeebus...Be happy if its 5%. 15-25% is more than K8->K10.
45nm K10s primary goal is to save them from the current suicide train in productioncost. The L3 less dualcore is also a demonstration of it.

[stangracin2]

shintai you do realize noone really pays attention to your ramblings because you sit there witing for an amd thread to come up so you can troll

[Hornet331]

shintai you do realize noone really pays attention to your ramblings because you sit there witing for an amd thread to come up so you can troll

actually you just did.

[Epsilon84]

yup same here, i predict up to 10% IPC increasment for Deneb over Agena which will put it right inbetween Yorkfield and Kentsfield.

20-25% is just insane, maybe some rare cases (like spec_rate score, which to according to amd we all know is a "wide variety of workloads" *scnr* ) but most of the time i doubt we see more then 10%.

Remember that Deneb is still the 3 issue architekture that barcelona was and 3rd level cache cant fix everything, espcial when you already got a IMC.

I'm sure there will be wild debate as to where exactly Deneb falls on the IPC spectrum. I remember you doing the analysis on Agena a while back, but in the end there is no 'law' on what constitutes IPC and it really depends on the benchmark suite(s) used for comparison.

For example, in the recent 100 CPU roundup, Yorkfield is some ~30% faster than Phenom per clock: http://www.matbe.com/articles/lire/1...urs/page30.php