Intel's Conroe Now In Hands Of A Few

[lopri]

Well with a 165 and a 146, there are not many choices to reach 3GHz, you know..
165 -> 9 x 333 (Divider 2:3, 2-3-2-5-1T)
146 -> 10 x 300 (Divider 5:6, 2.5-3-2-6-1T)

 

[Extelleron]

grrr........ are the XS forums down again?

 

[Keysplayr]

Originally posted by: Hard Ball

Originally posted by: keysplayr2003

Originally posted by: lopri
If you check the screenshots I posted earlier, there is a test where both the A64 3Gs w/6200TC beat Merom 2G w/X1400. Game 1 High Detail. (3DMark01 SE) Both dual and single core A64s are leading in that specific test. Does anone know what it tests?

Well, both AMD's could be benefitting from the integrated memory controller and 2GB of RAM. Merom has 167 FSB and 512MB RAM. Don't really know if this would make a difference in 3DMark2001SE.

Merom has 167 FSB?? What are you on?

quad pump it dufus. joke. Seriously, quad pump it to get yer 667.

 

[designit]

Originally posted by: Extelleron
grrr........ are the XS forums down again?

Do you think conroe has voodoo power and jinked the website's admin/mods?
Hell I aint gonna go near that chip if it's 5ghz for $50 and does 32M Spi in 10 seconds.
I will just stay w/ good ole AMD and hang garlic on my PC case.

 

[Viditor]

Mmmmmmm....Garrrrrrliiiiiiic...

 

[BrownTown]

clearly a 32M @ 10s run would be worth having voodoo hexes on you.

 

[jlbenedict]

XS seems to be back online.

 

[BrownTown]

still got a chip but no motherboard, as yet no timeline for motherboard arrival, so I wouldn't be getting my hopes up for anything today except for a better timeline of when all the parts will be in one place.

 

[Hard Ball]

Originally posted by: keysplayr2003

Originally posted by: Hard Ball

Originally posted by: keysplayr2003

Originally posted by: lopri
If you check the screenshots I posted earlier, there is a test where both the A64 3Gs w/6200TC beat Merom 2G w/X1400. Game 1 High Detail. (3DMark01 SE) Both dual and single core A64s are leading in that specific test. Does anone know what it tests?

Well, both AMD's could be benefitting from the integrated memory controller and 2GB of RAM. Merom has 167 FSB and 512MB RAM. Don't really know if this would make a difference in 3DMark2001SE.

Merom has 167 FSB?? What are you on?

quad pump it dufus. joke. Seriously, quad pump it to get yer 667.

Oh, I see, you mean the system bus, not the FSB signal freq.

 

[sindows]

http://www.xtremesystems.org/forums/showthread.php?t=95021

 

[TuxDave]

Wow... and all this time I was waiting for a big 'April Fools' reply to come out of all of this. Impressive scores btw

 

[jlbenedict]

Originally posted by: sindows
http://www.xtremesystems.org/forums/showthread.php?t=95021

You beat me to it.
First sign of any benchmarks....

 

[996GT2]

WOAH...21.5 Second SuperPi 1M is AMAZING...that's only like 5 seconds off the WR

 

[GuitarDaddy]

I see he stopped the 32m SPI run because of bios problems? I'm guessing he didn't like the numbers he was seeing.

I wish they would run PCmark or some encoding benches, SuperPI is a pretty crappy test of CPU

 

[Thor86]

Originally posted by: GuitarDaddy
I wish they would run PCmark or some encoding benches, SuperPI is a pretty crappy test of CPU

Seriously.

 

[dmens]

How is superpi a crappy test? It lines up with games pretty well.

 

[Furen]

Originally posted by: dmens
How is superpi a crappy test? It lines up with games pretty well.

No it freaking doesn't. Dothan crushes anything AMD in SuperPI 1m but this hardly happens in games.

Originally posted by: Hard Ball
Oh, I see, you mean the system bus, not the FSB signal freq.

The FSB/system bus operates at 166MHz. The whole "Quad-pumped" thing just means that it operates at quad data rate, not that it magically runs at 4x the clock speed. So even if it is "effectively" 667MHz (compared to single data rate), it has a 166MHz granularity and, thus, a 166MHz signal frequency.

 

[Viditor]

Originally posted by: dmens
How is superpi a crappy test? It lines up with games pretty well.

Have to agree with Furen...SuperPi is a very poor indicator of gaming performance. In fact, I can't think of anything it's a good indicator for.

 

[aLeoN]

Originally posted by: Viditor

Originally posted by: dmens
How is superpi a crappy test? It lines up with games pretty well.

Have to agree with Furen...SuperPi is a very poor indicator of gaming performance. In fact, I can't think of anything it's a good indicator for.

How about counting pi to a million digits really really really quick?

 

[imported_Questar]

Have to agree with Furen...SuperPi is a very poor indicator of gaming performance. In fact, I can't think of anything it's a good indicator for.

In fact, you are so AMD biased that you would not say anything is a good idicator for performace that didn't show an AMD cpu in absolutly the best light.

 

[dmens]

Originally posted by: Furen
No it freaking doesn't. Dothan crushes anything AMD in SuperPI 1m but this hardly happens in games.

Doesn't an overclocked dothan hold its own against a similarly clocked AMD64 in games, even with a few benchmarks? Maybe it's not a perfect correlation, but I'm sure there is some.

 

[Furen]

Originally posted by: dmens

Originally posted by: Furen
No it freaking doesn't. Dothan crushes anything AMD in SuperPI 1m but this hardly happens in games.

Doesn't an overclocked dothan hold its own against a similarly clocked AMD64 in games, even with a few benchmarks? Maybe it's not a perfect correlation, but I'm sure there is some.

Oh yes, a Dothan does about the same as a similarly-clocked K8 in games, that's not what I said. I said that Dothan ABSOLUTELY CRUSHES K8s in SuperPi 1M. Now, before anyone takes me out of context, I think Conroe will be a marvelous performer, I just dont think SuperPi (or PCMark and 3DMark, for that matter) really give us a good approximation of the real-world performance we can expect from these chips.

 

[carlosd]

NO, 3DMark, PCmark and Superpi don't give aproximations of real world apps specially when comparing different architechtures. That's it don't waste time with synth apps , look at the real world son. As you said dothan crushes K8 in superpi, but in a real app like games, they are about the same....so tests like superpi are unaccurate when comparing different architechtures for real world performance.

 

[Absolute0]

the "real world" performance is just running windows, where i can't tell the difference between my 170 @ 3.15 Ghz and a Pentium 4 3.2C.

We have benchmarks for a reason, so we can quantify performance, now if you say that SP, 3dmak, and PC mark are "time wasting synthetic apps" then it seems you don't really care about performance. As stupid as 3dmark can be, it is designed to be indicative of performance, and even though it isn't useful to us to calculate Pi, it is an actual application that runs, and runs damn fast on Dothan/Yonah/Conroe. You can't just dismiss it because you don't like the results and be like "Only gaming results matter." Gaming is good, but he has 512mb of ram and we won't see any good gaming results yet.

AFAIK we are all AT readers and everybody has already seen the gaming benchmarks, 2.66 Ghz Conroe running memory at 5-5-5-x consistently beating the FX60 at 2.8 Ghz by 25% or so.

 

[Furen]

Originally posted by: Absolute0
the "real world" performance is just running windows, where i can't tell the difference between my 170 @ 3.15 Ghz and a Pentium 4 3.2C.

We have benchmarks for a reason, so we can quantify performance, now if you say that SP, 3dmak, and PC mark are "time wasting synthetic apps" then it seems you don't really care about performance. As stupid as 3dmark can be, it is designed to be indicative of performance, and even though it isn't useful to us to calculate Pi, it is an actual application that runs, and runs damn fast on Dothan/Yonah/Conroe. You can't just dismiss it because you don't like the results and be like "Only gaming results matter." Gaming is good, but he has 512mb of ram and we won't see any good gaming results yet.

AFAIK we are all AT readers and everybody has already seen the gaming benchmarks, 2.66 Ghz Conroe running memory at 5-5-5-x consistently beating the FX60 at 2.8 Ghz by 25% or so.

We saw an Intel provided FX-60 get crushed by an Intel provided Conroe. Not being able to actually set up both systems makes the results a bit suspect. 3DMark is a video card benchmark. It's CPU component is mediocre, to say the least. Remember that P4s consitently beat AMD CPUs in the 3DMark CPU benchmark, which does not happen in most real world applications. By real-world test, by the way, I mean rendering, encoding, encryption, compression, office applications and, yes, games. I'm hoping someone out there will confirm Intel's IDF performance wins but instead of that we get SuperPi 1M and SANDRA.