Y
YKhan
I don't really buy his arguments, but I'm posting it anyways.
AMD's K10 Is delayed or dead
http://www.theinquirer.net/?article=27421
Yousuf Khan
AMD's K10 Is delayed or dead
http://www.theinquirer.net/?article=27421
Yousuf Khan
T
Tony Hill
I'm not sure whether it's true or not, but AMD sure hasn't said much
about any future generation cores. All they are talking about
publicly are tweaks to the existing K8 core.
I'm not sure that AMD is necessarily in desperate need of a new core
just yet, but if they want to keep the advantage they have now they
are definitely going to have to continue pushing forward. With the
way things are going it seems that Intel could easily catch up with
AMD by sometime in 2007 if all AMD is planning on doing is tweaking
their existing core.
Y
YKhan
Tony said:
My feelings too, but I'm not sure a new core is all that important to
them. It seems all of their performance improvements were as a result
of stuff that hangs off of the core, but isn't really a part of the
core, like HTT and memory controller; making further improvements on
those seems to be the path of greater return. They can probably make
even lower power AMD64's by adding circuitry into the existing core
like Intel did with the Pentium 3 to come up with the Pentium M, which
would be the intelligent power management stuff.
Yousuf Khan
D
David Kanter
YKhan said:
The K8 had several improvements that were not related to their memory
controller or HT that were significant. Certainly, the most noticeable
changes were those two, but they also did some interesting stuff with
distributing and slightly enlarging instruction queues.
I think that depends if you are talking about MP performance or single
socket performance. I don't see a big advantage in the case of single
socket systems. In fact, I think AMD will be at a slight disadvantage
when they are still using shared interface systems and Intel is using
shared cache (again, for the 1 socket scenario, unsure about MP).
I think you are trivializing what Intel did with the Pentium M. The
Pentium M is a radically different chip from the P6 or Pentium 3.
Sure, there was some different circuit design stuff done, but there was
also quite a bit more. Intel has been steadily improving their
decoders, so that some operations which used to be complex are now
simple, etc. etc.
If you want to get a grasp on the differences between the P!!! and the
PM, check out:
ftp://download.intel.com/technology/itj/2003/volume07issue02/art03_pentiumm/vol7iss2_art03.pdf
I'm assuming you are familiar enough with the P!!! uarch to make the
implicit comparisons...
David
R
Rob Stow
Tony said:
Paxville's abject failure and Intel's recent cancellations have
taken a lot of pressure off of AMD. Since AMD now knows that
they have a little more time to tweak and debug, perhaps that is
all they are doing: using the available time for exactly that.
Y
Yousuf Khan
Rob said:
Or as Sander Sassen, at Hardwareanalysis, puts it: "Cedar Mill, Intel's
65-nm Pentium 4, will finally put the Pentium 4 on par with AMD's Athlon
64 in terms of performance, power-consumption and heat-production...."
"Intel's new 65-nm processors"
http://www.hardwareanalysis.com/content/article/1820/
"AMD’s new sockets and DDR2 support"
http://www.hardwareanalysis.com/content/article/1820.2/
Me thinks there's a just a wee bit of an Intel bias in that boy's articles.
Yousuf Khan
R
Rob Stow
Yousuf said:Or as Sander Sassen, at Hardwareanalysis, puts it: "Cedar Mill, Intel's
65-nm Pentium 4, will finally put the Pentium 4 on par with AMD's Athlon
64 in terms of performance, power-consumption and heat-production...."
"Intel's new 65-nm processors"
http://www.hardwareanalysis.com/content/article/1820/
"AMD’s new sockets and DDR2 support"
http://www.hardwareanalysis.com/content/article/1820.2/
Me thinks there's a just a wee bit of an Intel bias in that boy's articles.
All he has to work with are Intel and AMD press releases, so what
I was seeing in those articles is exactly what those two
companies have been saying - nothing more and nothing less.
G
George Macdonald
Don't you think though that it is also typical of a mindset which has not
grasped just how far ahead AMD is right now? We see it all the time here.
R
Rob Stow
George said:
You are missing the point: Sassen said *nothing* to indicate
*his* mindset. One of his two articles regurgitated AMD's press
releases and the other regurgitated Intel's. Period. Full stop.
In those two articles the content/opinions from Sassen=NIL.
N
nobody
Intel press releases were always more marketing-geared and, let's put
it this way, overly optimistic. According to some past Intel press
releases and road maps, NetBust was supposed to surpass 4GHz by this
time. "Past performance is not a guarantee of future results"
NNN
Y
YKhan
Rob said:
So you're saying that on AMD's press release, AMD said that Intel's
Cedar Mill will catch upto them finally? Because that above statement
came from the AMD side of the article. If it's just a press release
from AMD, then it's a bad press release.
Also Intel typically does not say these sort of things either. They
typically try to downplay any competitor's products or company in their
own press releases.
Yousuf Khan
R
Rob Stow
YKhan said:
No, I didn't intend to say that. Instead of saying that one
article is an AMD release and the other an Intel, I should have
said that neither article contains anything that isn't in the
press releases from those two companies.
To put it another way, you will fail if you try to find
something in either article that Sassen got from anyplace other
than Intel or AMD. Neither article has anything about either AMD
or Intel's future chips that is not obviously copied or
paraphrased from the AMD and Intel flacks.
G
George Macdonald
Huh? Intel said this: "Cedar Mill, Intel's 65-nm Pentium 4, will finally
put the Pentium 4 on par with AMD's Athlon 64 in terms of performance,
power-consumption and heat-production..." - just "on par"?
As for AMD, the last PR I saw did not have "will not make the transition to
65-nm in 2006"; in fact, IIRC, it stated 65nm some time in 2H2006. The
article certainly suggests to me that the author is indicating that AMD
needs 65nm to stay in the game, which is not at all obvious.
T
Tony Hill
Or as Sander Sassen, at Hardwareanalysis, puts it: "Cedar Mill, Intel's
65-nm Pentium 4, will finally put the Pentium 4 on par with AMD's Athlon
64 in terms of performance, power-consumption and heat-production...."
From the early samples I've seen, Intel still has a ways to go. Their
65nm P4 chips, while consuming less power than their 90nm
counterparts, will still consume more power than AMD chips. They also
won't change performance by any meaningful amount (the single-core 600
series are identical while the dual-core 900 series double the L2
cache but otherwise are the same). It remains to be seen if 65nm
production allows Intel to crank clock speeds up at all, but if they
can't, AMD should still be comfortably ahead.
That being said, Intel isn't standing still. Their plans for a whole
new architecture, evolved from the Pentium-M line, seem fairly
impressive and are expected before the end of 2006. This is why I
think Intel will have caught up to AMD in 2007 if things keep going as
is.
Y
Yousuf Khan
Rob said:
Well, that can be said about just about any announcement article out
there. However, this article came not from his news announcements pile,
but from his editorial section. And in editorials, it's all of the
little flourishes and spins that they add that indicate their
preferences. Among the editorial conclusions that he makes, besides that
Cedar Mill will finally catch upto AMD, he also makes the conclusion
that only Pentium M is worth having in a laptop and that AMD has nothing
worthy to compete against that, despite the fact that AMD has already
said that they experienced 72% growth rates in the amount of laptop
chips they sold.
G
George Macdonald
Possibly the Cedar Mill comment is based on:
http://www.tomshardware.com/cpu/20051007/a_sneak_peak_at_intels_65nm_pentium_4-02.html
- a grrrreat umm, source?