Sign in with
Sign up | Sign in
Your question

Anandtech does Conroe vs FX-60 testing, admits some errors.

Last response: in CPUs
Share
March 10, 2006 2:45:49 AM

Interesting article by anandtech. Pointing out Bios issues, and other mistakes that they made during the FX vs Conroe test, and the new results they have after the test is over. FX still beats Conroe in their Divx, Itunes, and Windows media encoding. I am not really fluent in geekness so anyone who could further clarify this that would be nice.

Anandtech article
Anonymous
a b à CPUs
March 10, 2006 2:47:29 AM

Its been reported already in another thread 10h ago...
March 10, 2006 2:50:08 AM

Summary, tests were reviewed and modified with teh end result is that Conroe at 2.66GHz performs an average of 20% faster than a FX-60 overclocked to 2.8GHz (future FX-62 speed).
March 10, 2006 3:32:33 AM

If you look at those benchmarks, it appears that the FX60 is faster than Conroe. That's false because, if you take a closer look at those charts, they say at the top, lower is better.

The purpose of decoding tests is to see which processor has lower numbers and not higher as in all FPS tests.

Conroe clearly beat the FX60 in all tests in that article. I still believe that when the Conroe EE hits the streets clocked at 3.33Ghz, it will definately put out more numbers then the processor used in those tests. We do have to take that into consideration that Intel used one of their more midrange processors.
March 10, 2006 5:01:56 AM

I'm still catching up on the latest hardware stuff, but is this correct?

the Amd FX62 sucks already and it's not even out yet! :lol: 
March 10, 2006 5:10:40 AM

Probably, but there's no way of knowing for sure yet. Wait and see.
March 10, 2006 5:13:34 AM

Quote:
I'm still catching up on the latest hardware stuff, but is this correct?

the Amd FX62 sucks already and it's not even out yet! :lol: 
It's behind on a benchmark to a CPU not in production yet by a system put togeather by Intel.

That's not to say that the Conroe benchmarks aren't right, but it's a final production sample, I do anything more than hope it keeps up that level of performance.

Plus, we don't know what AMD has up it's sleeve beyond AM2. They certaintly won't go down quietly, and my guess is that they'll still be a better value per performance than the Intel chips, which tend to come at a premium.

Of course, that's never ment anything to you Melty, as long as you could sling mud at AMD.
a b à CPUs
March 10, 2006 5:39:54 AM

Quote:
I'm still catching up on the latest hardware stuff, but is this correct?

the Amd FX62 sucks already and it's not even out yet! :lol: 


well it sucks compared even to a lower clocked standard intel desktop cpu, but still - its one mighty processor (dont go as far as saying it sucks) - to the current standard its fast.

But what the heck im gettin a conroe!!!!!!
March 10, 2006 5:49:49 AM

To all fanboys responding to these threads.
They are a permanent record. What you say here can and will be reviewed when the conroe chips are released.
Go ahead, make a fool of yourself. we can all read them in September, and laugh at you.
March 10, 2006 5:58:42 AM

At least people can no longer say AMD is the best for gaming. Those tests for with an fx-60. Why not test conroe against the x2 cpu's.
March 10, 2006 6:10:34 AM

Quote:
Interesting article by anandtech. Pointing out Bios issues, and other mistakes that they made during the FX vs Conroe test, and the new results they have after the test is over. FX still beats Conroe in their Divx, Itunes, and Windows media encoding. I am not really fluent in geekness so anyone who could further clarify this that would be nice.

Anandtech article


BAHAHAHAHA "FX still beats Conroe in their Divx, Itunes, and Windows media encoding."

Dude, You do know that the LESS time it takes to finish those tests THE FASTER it is ?

I'd rather it take 5 seconds to encode a 1 hour video, than 10 hours.
Although according to you the LONGER BAR GRAPH = ALWAYS FASTER.

ROFL.

April fools is still 22 days away dude, this is a little early.
a b à CPUs
March 10, 2006 6:52:25 AM

Quote:
To all fanboys responding to these threads.
They are a permanent record. What you say here can and will be reviewed when the conroe chips are released.
Go ahead, make a fool of yourself. we can all read them in September, and laugh at you.


and that goes your you too :D 

*waiting patiently hoping he was right all along*
March 10, 2006 9:23:05 AM

Less time is better. Conroe beats the FX in all tests.
I very much hope Conroe will be a gaming chip.
BTW Does anyone here know what "The Oblivion" will stress mostly, CPU or GPU? THx.
March 10, 2006 9:32:54 AM

nothing extra ordinary about conroe...coz conroe based on pentium m/pentium iii architecture...basically they have same IPC like athlon but the different is they have more cache than athlon x2.

athlon x2 1mb can beat p4 that have more cache becoz p4 have a stupid architecture...however conroe not using that stupid netburst ...and of course conroe win coz they using 4mb x 2 cache vs 1mb x 2 cache of athlon...

so what amd need to do...is just load more cache on athlon .... done...
a b à CPUs
March 10, 2006 10:26:37 AM

Quote:
nothing extra ordinary about conroe...coz conroe based on pentium m/pentium iii architecture...basically they have same IPC like athlon but the different is they have more cache than athlon x2.

athlon x2 1mb can beat p4 that have more cache becoz p4 have a stupid architecture...however conroe not using that stupid netburst ...and of course conroe win coz they using 4mb x 2 cache vs 1mb x 2 cache of athlon...

so what amd need to do...is just load more cache on athlon .... done...


P4's architecture aint stupid it took on the K7 with ease but the K8 was designed to beat it and it did the same as conroe beats the K8, and was about time it was replaced.

Conroes cache is 1x4mb shared not 2x4mb and cache doesnt matter as much depending on architecture, ram speed/location and FSB, twice the cache might translate into 10% performance at best depending on original size.

Conroe is P6 based (pentium pro -> P2 -> P3 -> pm).
a b à CPUs
March 10, 2006 10:29:31 AM

"If you look at those benchmarks, it appears that the FX60 is faster than Conroe. That's false because, if you take a closer look at those charts, they say at the top, lower is better. "

LOL!

Agreed, not always is a higher number... better! :-)
March 10, 2006 11:45:02 AM

Quote:
nothing extra ordinary about conroe...coz conroe based on pentium m/pentium iii architecture...basically they have same IPC like athlon but the different is they have more cache than athlon x2.

athlon x2 1mb can beat p4 that have more cache becoz p4 have a stupid architecture...however conroe not using that stupid netburst ...and of course conroe win coz they using 4mb x 2 cache vs 1mb x 2 cache of athlon...

so what amd need to do...is just load more cache on athlon .... done...


That is as stupid as saying the Radeon 9600 Pro will beat the Radeon 9800 Pro because it is clocked faster, even though they are based on exactly the same GPU architecture.

Remember: The 9800 Pro had 8 pipelines, the 9600 Pro had 4.


The Conroe, in terms you'll understand, has 4 'pipelines' per core, the Pentium III / M / Core Duo only had 2-3 'pipelines' per core. Thus Conroe can process twice (or 4/3rds more depending what you compare it to) the information per clock cycle. Much like the Athlon 64 could process apx +50% more per clock cycle than the Pentium 4.

Hopefully when put in a light similar to how video card GPUs are compared a gamer might actually understand this now, and why it is such a massive leap.

When nVidia goes from 24 pipeline to 32 pipeline video cards do these same idiots say performance won't rise unless clock speed rises ?, Yet they understand AMD Athlon 64 at lower clock beats the Intel Pentium 4.

I am sure with in 12 months (or less) they'll finally get it.
Anonymous
a b à CPUs
March 10, 2006 11:52:02 AM

Where did you guys(RichPLs and you) got the info about a 3.33 EE? I only saw a possible 3ghz EE part at lauch and thats not confirmed. 3.33 is certainly possible just wondering where you guy got that or if its just speculation.

Oh and lets post that here, some ppl say the 2.66 is something like 300$ its more around 500$.

The 2.4 with 4meg will probably be a good choice at 300$ price range!

DailyTech -Conroe pricing
March 10, 2006 12:04:54 PM

Quote:
I'm still catching up on the latest hardware stuff, but is this correct?

the Amd FX62 sucks already and it's not even out yet! :lol: 


It was an FX-60 clocked at FX-62 speeds. The FX-62 will be on the new AM2 line. I don't believe there has been any testing on an actually FX-62.
March 10, 2006 12:15:30 PM

The FX-62 will have a hard time even MATCHING the performance of the FX-"62" (the OCed FX-60).
March 10, 2006 1:03:29 PM

Why do you guys insist on comparing things that arent even available yet?? AMD has till June to release it's new thing, like they arent tweaking it right?
As if AMD "THE" performance king for 2 years now, doesnt have an ace up it's sleeve. Even if they dont, its about time Intel took the lead for once, it means nothing. Just like ATI and Nvidia, it will always go back and forth. With the size of Intel and the amount of engineers they have, you think they would always be king. NOT. Get real people. Little AMD has trounced big Intel and embarrassed the hell out of them for the last 2 years.
March 10, 2006 1:14:46 PM

Quote:
The FX-62 will have a hard time even MATCHING the performance of the FX-"62" (the OCed FX-60).
another mindless fanboy huh cant admit that amd pwned prescott
March 10, 2006 1:28:55 PM

Quote:
The FX-62 will have a hard time even MATCHING the performance of the FX-"62" (the OCed FX-60).


I doubt that, but only time will tell right?

Now let's all go play some golf or go bowling until they release the new AM2 chit. 8)
March 10, 2006 2:10:17 PM

Yay, Bowling!
March 10, 2006 2:21:49 PM

Since gaming it my karma, I pick the processor that will give the best frames, as shown in independant testing benchmarks, not ones from Intel. That being said, it is premature to think that just because Conroe appears to beat the FX series, that Intel is the new king. It might be true for a few months, but the only true comparison will come when we have a next-gen AMD processor at the same price point as Conroe to compare with.
March 10, 2006 4:09:53 PM

Quote:
The FX-62 will have a hard time even MATCHING the performance of the FX-"62" (the OCed FX-60).
another mindless fanboy huh cant admit that amd pwned prescott

there is no doubt in my mind that Prescott was a huge POS (piece of sh!t), not to say that I don't like Intel just stating the fact that the last good CPU up until the Conroe was the Northwood Core. Clock for Clock when Prescott was first released was actually slower then Northwood, add all the heat and power issues and there you have it. Now Intel after some hard work has found what appears to be a good CPU...good for them. I am sure AMD is not going to take this lying down. They never have..
March 10, 2006 8:46:22 PM

i agree northwood was great used to have one :D 
March 10, 2006 10:08:42 PM

The 3.33GHz Conroe EE can't really be confirmed until it's released, but the earliest reference to it was this article:

http://www.theinquirer.net/?article=29031

Quote:
Since Intel is sending out Conroe XEs at 3.33GHz to friends and people buried in NDAs until you can't find any trace of their existence, around the end of this month, we think things are in great shape.

This article was released on January 17th, so if they were given to developers on schedule they would have had them for more than a month now.

Personally, I don't think that Intel will release the Extreme Edition at 3.33GHz since that's way to high and uses up much of their clocking room too early in the architectures product life. With the highest mainstream product at 2.67GHz, it would make sense for the EE to launch at 3GHz with a 1333MHz FSB. Intel will be refreshing the line in late Q4 or early 2007 with the 2.93GHz mainstream model and so it would make sense for the 3.33GHz EE to be released then. I'm not sure about the clocking limits of Conroe, but I think that 3.33GHz is probably pretty much it for sufficient volumes so the next major bump probably won't come until H2 2007 when 45nm is available. The cache will be moving up to 6MB at that time.

The quad core Kentsfield will also be available H1 2007, but there's no word on it's clock speeds yet. It'll probably be in the 3-3.33GHz range and replace the 3.33GHz Conroe as the Extreme Edition.
a b à CPUs
March 10, 2006 10:33:19 PM

i read 3ghz by the end of the year which makes more sense
March 11, 2006 12:20:53 AM

Only Intel knows for sure, and it appears pretty solid the 3Ghz extreme version will be released, and I suspect if stablity and headroom is there in time for the launch, a surprize 3.33GHz version will round out the package and show how far and well designed their new line of cpu's are.
!