Deneb just around the corner?

[nanohead]

deneb is appearing to be coming out around the time of shanghai. really tho deneb looks like a big success. agena maxed at 2.6ghz stock and deneb is coming out with 2.8 and 3.0 plus the 15-20% bonus. seems to be overclocking well and if people overclock it to 4ghz it will be a big difference. 15-20% more performance plus another 800 mhz overclock will be a big performance change. (assuming 3.2ghz is average agena overclock and 4ghz is average deneb overclock)

Heck, I'll just be happy if they

1) Get the thing out the door this year
2) It actually works without any embarrassing bugs
3) Is delivered with respectable clocks
4) Doesn't require a small nukular pile to run it
5) Overclocks respectably, and doesn't require LN02 to cool it
6) Actually works with the existing AM2+ boards and DDR2 8500 with little or no trouble

If they deliver this, most AMD'ers will be happy, and maybe some of the intel fambois who are broke will come back to where it all started. And maybe the shedload of AMD stock I've been buying will be worth more than the pittance it is now.

No more motherboards for me until DDR3 comes WAY down in price, and until then, its just CPUs and SSDs. So Deneb better friggin work

[Aussie FX]

I think Denb is going to be be much better than most people realise. AMD have said themselves that it will bring 20% ipc improvements over Barcelona and we have all seen the 4+ghz overclocks on C1 silicon.
What makes me so positive is the way intel are going about things with this massive paper launch of nehalem, they are throwing everything at it to the detriment of the current penryns which are a great cpu in their own right.
I just get the feeling intel are running scared atm. I have never seen them react this way in my 20+yrs of being around pc's.

Here's hoping.

[YukonTrooper]

I think Denb is going to be be much better than most people realise. AMD have said themselves that it will bring 20% ipc improvements over Barcelona and we have all seen the 4+ghz overclocks on C1 silicon.
What makes me so positive is the way intel are going about things with this massive paper launch of nehalem, they are throwing everything at it to the detriment of the current penryns which are a great cpu in their own right.
I just get the feeling intel are running scared atm. I have never seen them react this way in my 20+yrs of being around pc's.

Here's hoping.

Right. Intel is running scared at the moment. Even while Nehalem is proving itself on dozens of review sites, and proving itself with a handful of overclockers right here on XS. You don't have to go too far to see what kind of numbers it's putting up. It's quite the chip.

And don't forget what happened the last time AMD threw around performance numbers, because they only managed to talk the talk.

Look, I'm hoping for the best for AMD, and I'm really hoping their new chips do well in games, because I'll switch in an instant, but let's be realistic and realize they have lots of ground to make up against the performance that Intel is producing. I just think "Intel running scared" sounds a little silly at this point, especially considering all the evidence argues against it. But, of course, your 20+ years around PC's must be taken into consideration, because that is a very notable qualification.

[keithlm]

Actually I think we are seeing many Intel fans getting nervous.

We have the B2 that came out with a "problem" that AMD admitted. There was a bios update that "stopped" that problem but was not really a "solution".

Then the B3 came out with quick fix to that problem in "hardware." But was it really a FIX? We must ask: Did they FIX that problem with B3 or just throw in something that made it not happen? And did that problem affect actual performance?

If they actually had a problem that was only "masked" by the "fix" then it is VERY possible that the Deneb, if it actually FIXES that problem could be much more than a mere "shrink".

Okay. They add cache. They change a few other things. They actually fix the problem they had. This could be MUCH more than a mere "shrink" from 65nm to 45nm. It could actually be MUCH more than that.

On the other hand we might just see Deneb suddenly make Yorkfield look bad. If that is the only thing that happens... that would be VERY good indeed.

But then based on the fact that it is also an IMC... if it does well against Yorkfield it could also do VERY WELL against the i7. Even if it ONLY does well against Yorkfield that is GREAT. But what I am pointing out is that it could also do very well on all benchmarks that make the i7 look "great". That would be predictable and only someone that is biased would not see that. But then even if it did... it still wouldn't stop the "OMG i7 is going to be the better" posts on this or other forums. Even if it only matched i7 at the same speed... you'd still have the "OMG it's not Intel" crowd. Although it would stop the "I only really care about the price per performance" crowd from making stupid posts.

Will it happen? We do not actually know yet. If might merely be better than what is available now. On the other hand... the Intel "natives" are getting restless. (Which I must admit is very enjoyable to see.)

OTOH: I am tired of the "Q6600 is better than the best AMD has to offer" crowd posting their tripe. That hasn't been true for months but they still post it in a sad attempt to make their opinion "more true" than reality. It would be nice to see the "best" Yorkfields fall short. And guess what? It could happen. If they didn't have a "secret" we probably would have seen leeks. We haven't. That means SOMETHING is going to happen soon. Or not. The only thing we can really say NOW is that we don't know yet.

[Clairvoyant129]

Actually I think we are seeing many Intel fans getting nervous.

OTOH: I am tired of the "Q6600 is better than the best AMD has to offer" crowd posting their tripe. That hasn't been true for months but they still post it in a sad attempt to make their opinion "more true" than reality. It would be nice to see the "best" Yorkfields fall short. And guess what? It could happen. If they didn't have a "secret" we probably would have seen leeks. We haven't. That means SOMETHING is going to happen soon. Or not. The only thing we can really say NOW is that we don't know yet.

I don't think Intel fans are nervous, with the slowest 920 Nehalem outperforming QX9650 occasionally in several benches. I'm one of the few people who would like AMD to do better but very skeptical after the Barcelona debacle. 40%+ performance improvement over Clovertown and all we see is actually the opposite of that.

And are you implying that the Phenom 9950 outperforms Q6600? I highly doubt that.

http://www.matbe.com/articles/lire/1...urs/page30.php

In terms of relative performance, Q6600 does outperform the Phenom 9950. If you're talking about $/performance ratio, that's a different story.

[Baam]

I don't think Intel fans are nervous, with the slowest 920 Nehalem outperforming QX9650 occasionally in several benches. I'm one of the few people who would like AMD to do better but very skeptical after the Barcelona debacle. 40%+ performance improvement over Clovertown and all we see is actually the opposite of that.

And are you implying that the Phenom 9950 outperforms Q6600? I highly doubt that.

http://www.matbe.com/articles/lire/1...urs/page30.php

In terms of relative performance, Q6600 does outperform the Phenom 9950. If you're talking about $/performance ratio, that's a different story.

The 9950 does outperform a Q6600

http://www.guru3d.com/article/amd-ph...cessor-tested/

[Glow9]

The 9950 does outperform a Q6600

http://www.guru3d.com/article/amd-ph...cessor-tested/

In this review it's a back and forth battle, this was kinda neat

http://www.xtremesystems.org/forums/...d.php?t=206025

[Aussie FX]

But, of course, your 20+ years around PC's must be taken into consideration, because that is a very notable qualification.

Sarcasm is the lowest form of wit.....
What have I done to you that makes you want to launch into me for making an observation?

Please give me an explanation as to why intel have paper launched nehalem. Seeing as though you appear to know everything.

[YukonTrooper]

Actually I think we are seeing many Intel fans getting nervous.

Even if true, which I highly doubt, it has absolutely nothing to do with Intel itself running scared, which is the context of the discussion, not the fans running scared.

I dunno. Maybe we should ask Dave (Movieman) if he's running scared of AMD with his brand new i7 hardware?

OTOH: I am tired of the "Q6600 is better than the best AMD has to offer" crowd posting their tripe. That hasn't been true for months but they still post it in a sad attempt to make their opinion "more true" than reality. It would be nice to see the "best" Yorkfields fall short. And guess what? It could happen. If they didn't have a "secret" we probably would have seen leeks. We haven't. That means SOMETHING is going to happen soon. Or not. The only thing we can really say NOW is that we don't know yet.

It's better when both chips are at their max average overclock, so I think there are two sides to that argument: At stock vs. overclocked. I don't think it has to be anymore complicated than that, at least in my opinion.

Sarcasm is the lowest form of wit.....
What have I done to you that makes you want to launch into me for making an observation?

I thought it was appropriate for the situation. Saying that you think Intel is running scared of an unreleased chip, all while their own newest chips are dominating benchmarks, is quite the assumption, is it not?

[Aussie FX]

I thought it was appropriate for the situation.:

Well you are wrong and way out of line.
It is ok to post and disagree with what I said.
It is not ok to make me out to be an idiot. It shows a distinct lack of respect for a fellow human and it also shows the level of your intelligence.
I made an observation based on 20 years of watching the industry and I have never seen intel react like this before. Why that warrants a personal attack is beyond me.
The fact is AMD has a cpu that has been clocked to over 4ghz on air several times now and I think that would be enough to make intel sit up and take notice.

[soundood]

The fact is AMD has a cpu that has been clocked to over 4ghz on air several times now and I think that would be enough to make intel sit up and take notice.

100% agree with this statement, all the while they are making a transistion to IMC? you better believe they are quaking, one false move here and the middle of the road intel guys will go AMD.

i have found that Intel guys dont give a monkeys, they go where the speed and stabilty is, if they think that for one second AMD is better, they will move.

More loyalty is held on the AMD side of things i believe, the Intel guys just want to 'drive' the fastest car, whether that is Ferrari or Porsche is irrelavent for them.

[Chad Boga]

I think Denb is going to be be much better than most people realise.

I think it will be much worse, but I hope you are right.

AMD have said themselves that it will bring 20% ipc improvements over Barcelona

They also said that Barcelona would bring 40% over Core 2 and the only people who don't realise how well that turned out are the clowns on AMDZone.

What makes me so positive is the way intel are going about things with this massive paper launch of nehalem, they are throwing everything at it to the detriment of the current penryns which are a great cpu in their own right.

Is the launch of Nehalem any more or less papery than previous Intel new architecture releases?

I just get the feeling intel are running scared atm. I have never seen them react this way in my 20+yrs of being around pc's.

Here's hoping.

Hope springs eternal.