lol AHHAHAHAH IM SCREEWWEEEDD!!

[zmatt]

If that was true than I can say that Core2 was not made for 65nm either, as evidenced by the performance bump in Penryn. And Nehalem isn't made for 45nm because they will have a 32nm die shrink of that next year. You can't make such arguments. That is pure BS.

 

[ricanflow]

See but it is faster at the same clock speeds, plain and simple:

 

[Apokalipse]

^Not true. Penryn is a 5-10% faster than Conroe. I don't know if you could say the same for Deneb.<!-- google_ad_section_end -->

Shanghai has a higher increase than that over Barcelona.

That hardly means Phenon wasn't made for 65nm. Nearly every die shirk performs better than it's predecessor.

But how much better?
Usually not as much of an increase as Deneb has over Agena, especially in overclocking.

Also, you know as well as I do that Agena didn't live up to what a lot of people were expecting.
AMD did want 2.8-3GHZ out of Agena. It didn't get that far at stock. And it isn't because of the architecture.

 

[ricanflow]

And i thought overclocking was such a big deal to you guys??

But because its an amd chip it doesnt matter right?

Even if it performed the same clock for clock but oc'd higher, it would still be a better buy just because it can go higher. But not only can it go higher, but it performs faster clock for clock and uses less power.

There are no negatives here, just more of the usual nitpicking.

 

[Puddle Jumper]

But how much better?
Usually not as much of an increase as Deneb has over Agena, especially in overclocking.

Also, you know as well as I do that Agena didn't live up to what a lot of people were expecting.

I'm not saying that deneb isn't better than agena. I just think that deneb being faster than agena by no means proves phenom was not made for 65nm.

And i thought overclocking was such a big deal to you guys??

But because its an amd chip it doesnt matter right?

Even if it performed the same clock for clock but oc'd higher, it would still be a better buy just because it can go higher. But not only can it go higher, but it performs faster clock for clock and uses less power.

There are no negatives here, just more of the usual nitpicking.

I'm all for overclocking but I haven't seen any evidence that deneb can overclock higher than nehalem. It has nothing to do with it being a amd chip, it has everything to do with nehalem clocking higher and being faster clock for clock.

We haven't see any detailed benchmarks comparing deneb and yorkfiled so it is to early to tell which one of them is faster.

 

[Apokalipse]

I'm not saying that deneb isn't better than agena. I just think that deneb being faster than agena by no means proves phenom was not made for 65nm.

Well it obviously was made in 65nm. Perhaps not particularly well suited to it is a much better explanation.

And that's based on experimental evidence.

Even Intel did not think a quad 65nm die would do particularly well. That's why they didn't bother.

*edit*
In fact, I think logistically it would have been very bad for Intel to have tried to make Nehalem in 65nm. Their TDP is high enough as it is with the introduction of their IMC, their new complex architecture, and monolithic die.

 

[ricanflow]

I'm all for overclocking but I haven't seen any evidence that deneb can overclock higher than nehalem. We haven't see any detailed benchmarks comparing deneb and yorkfiled so it is to early to tell which one of them is faster.

Again, not talking about nehalem.

We are trying to say deneb is better than agena...jesus. How hard it is to realize that? Theres hard evidence in front yet you guys still wanna stretch this out over and over.

It is known that agenas dont oc well...and the ones that do use lots of power
and put out a lot of heat. 45nm solves this.

 

[Puddle Jumper]

Well it obviously was made in 65nm. Perhaps not particularly well suited to it is a much better explanation.

And that's based on experimental evidence.

Even Intel did not think a quad 65nm die would do particularly well. That's why they didn't bother.

Intel didn't release a 65nm native quad because it didn't make since for the to do so. Because of the problems with netburst they were forced to switch to the only other architecture they had, Pentium M which resulted in core 2 being brought to market as quickly as possible to reduce there losses to the Athlon x2's. because of the way their development cycle works a new architecture wouldn't be released until '08 (nehalem) so they made a smart move and released a quad using the architecture they had rater than waiting for a native quad.

Again, not talking about nehalem.

We are trying to say deneb is better than agena...jesus. How hard it is to realize that? Theres hard evidence in front yet you guys still wanna stretch this out over and over.

It is known that agenas dont oc well...and the ones that do use lots of power
and put out a lot of heat. 45nm solves this.

When did I say deneb wasn't faster or a better overclocker than agena?

I have been debating whether the statement that phenom was not made for 65nm is accurate not whether deneb is faster than ageana.

 

[ricanflow]

This whole debate is about you guys not accepting the fact that deneb is great improvement over agena. 45nm, higher ipc and less power usage and higher overclocking ability.

Why argue it? Its not like we are comparing them to intel chips and saying theyre faster.

 

[Apokalipse]

Intel didn't release a 65nm native quad because it didn't make since for the to do so.

It didn't make sense because Nehalem was already under development. Though a monolithic die without an IMC would have been a lot easier.
Again, logistically it would have been very bad for Nehalem to be produced on 65nm. And in a lot of ways, Nehalem is like Phenom.

Because of the problems with netburst they were forced to switch to the only other architecture they had, Pentium M which resulted in core 2 being brought to market as quickly as possible

Core 2 wasn't developed quickly. It just appeared on the market suddenly, and surprised people.
It was a fairly smooth transition from the Core Duo's (the first generation "core" chips), which were a fairly smooth transition from Pentium M's.

They wouldn't have wanted people to have a very advanced warning that a new CPU was coming out. that would have prematurely killed Pentium 4 sales a lot before the Core 2 would have even come out.