PDA

View Full Version : Grantdale and Alderwood NDA lifted!! PCi - Express, DDR2, Hi Def Audio Galore!


MUYA
06-19-04, 04:20 AM
PCI Express platform reviews out

Chek the front page for more links..as I get them;

Dazz
06-19-04, 04:53 AM
I just noticed peformance wise there is no diffrence at all DDR2 Vs DDR & none Vs S775 vs 478, PCI 16x Vs AGP 8x remains to be seen. But the P4 will have a hard time against the Athlon 64.

MUYA
06-19-04, 06:22 AM
actually in some cases DDR2 is slower..check the Anandtech benches in the workstation test. Hmmm

Seems its early days yet with the drivers from Intel set to improve..maybe.

But I have seen other benches under NDA that suggest that PCI Express X16 will help somewhat more wrt to graphics cards rather than what has been seen in the benches so far.

Dazz
06-19-04, 06:24 AM
Yeah i saw it but the diffrence is minamal although DDR2 has a premium soooo. Well will have to wait for apples to apples setups. As for Didef audio they are infact powered by the Realtek ALC880 codec so it sows onboard auido an't what it used tobe. I am only sticking with my Auidgy 2 ZS simply due to EAX4.0.

MUYA
06-19-04, 06:30 AM
DDR2 isn't supposed to be beneficial till DDR 667,MHz...well only due to the fact that DDR1 cannot clock that high :D

MUYA
06-19-04, 06:32 AM
however, AthlonFX 53 does do incredibly well compared to the Prescotts here;

http://www.anandtech.com/chipsets/showdoc.html?i=2088&p=10

:D

Dazz
06-19-04, 08:03 AM
Then again the FX53 is rated for 3800+ so would go up against the P4 3.8E. But considering the huge gap in a fair few cases it's fair to say the 3500+ can easily go up against the P4 3.6E.

MUYA
06-19-04, 08:55 AM
clock for clock i was on about ;) but we all know the score about that ;)

FX 53 would be more than 3800+ methinks maybe 4000+

:)

|JuiceZ|
06-19-04, 09:47 AM
A64 is the shiz :cool:

Riptide
06-19-04, 10:05 AM
Yeah... Just read the anandtech article and <yawn>

GlowStick
06-19-04, 12:19 PM
I read the anand article and im happy i chose the FX-53 heh : )

Pandora's Box
06-19-04, 04:08 PM
mwahahaa! intel looses again!

god this feels so good. one of my friends is a die hard intel fanboy. he kept bugging me about how intel was owning intel in the past. now the roles have reversed. after i read the anandtech review i asked him what he thought and he said just wait till intel does a 64bit cpu that will totally own amd. :lol2:

Draconis
06-19-04, 05:04 PM
Well this doesnt sway me into getting amd im still sticking with intel. :)

jAkUp
06-19-04, 05:08 PM
Terrible... all this for nothing. meh... I'm persuaded to go AMD next time more and more everyday.

First the :lame: Prescott now the :lame: 775... Somebody make it stop :rolleyes:

Pandora's Box
06-19-04, 06:33 PM
its a lame company what do you expect? a good processor? lol

Arioch
06-20-04, 02:23 AM
AMD is really on a roll right now - I think Intel got a bit complacent in the last few years overall and now its starting to show. Hopefully they will get their act together but for now AMD is the way to go for gamers.

MUYA
06-20-04, 02:25 AM
the fact that Intel cannot answer the K8s is the reason why the k8s are so expensive...remeber the good ole days when AMD and Intel used to cut prices galore

netviper13
06-20-04, 03:57 AM
This makes me think back to last summer at the National Youth Leadership Forum on Technology I attended, where Craig Barrett (Intel's CEO) was asked about Intel's plans regarding 64-bit processors, and he replied "we at Intel do not see a market in the 64-bit desktop area and currently have no plans to release such a chip". Nice move.

Sazar
06-20-04, 04:34 AM
intel will answer when their on-die memory controllers go online...

@ the moment their marchitecture is a generation old...

LGA should help perhaps wrt temp control though given the relative heat output of some of these cpus...

really looking forward to their next stuff :)

Riptide
06-20-04, 10:44 AM
This makes me think back to last summer at the National Youth Leadership Forum on Technology I attended, where Craig Barrett (Intel's CEO) was asked about Intel's plans regarding 64-bit processors, and he replied "we at Intel do not see a market in the 64-bit desktop area and currently have no plans to release such a chip". Nice move.
This may not look like the greatest move, for sure. But what made the A64 is the 32bit performance isn't it? If it didn't have that going for it I bet it wouldn't sell nearly as well just because it's "64bit".

Intel is hitting a wall right now it hadn't planned on which is the heat issues and problems with continually needing to shrink the die to up their clocks. People in the industry seem to have been taken by surprise that they ran up against these issues as early as they have. If they hadn't run into these obstacles then they might not be looking so bad right now. Prescott's @ 4ghz would be pretty fast from what I hear...

But regardless, AMD has a chance to capitalize on their misfortune and miscalculation. And I hope they make the best of it. Intel can afford screwups more than they can. ;)

Viral
06-21-04, 07:25 AM
intel will answer when their on-die memory controllers go online...

@ the moment their marchitecture is a generation old...

LGA should help perhaps wrt temp control though given the relative heat output of some of these cpus...

really looking forward to their next stuff :)

That's what i would have thought, but then i read the tom's review.

http://www.tomshardware.com/motherboard/20040619/socket_775-49.html

Power consumption.

S775 P4's use up to 20watt more than S478 prescotts!.

On a side note, I never knew A64's used less power than AXP's at idle(even the 2400+ - by a lot too) and load(with comparable GHz models).. looks like SOI really payed off for AMD.

{Sniping}Waste
06-21-04, 01:21 PM
Inte's problem is what the die's are built on. They use Strained silicon, strained with Germaneum(spelling). The resson for germaneum is the transister voltage bias drop. Transister made on silicon needs .7 volts on the gate to bias the transister and at .09 micron .7V will fry the traces on the IC. Germaneum only needs .3V on the transister gate but leaks volts and amps alot higher then silicon. Another thing that hurts Intel is the Germaneum is not singel crystal but poly. That means that even if the proces is great for each die, the die will fail across the whole wafer.