Anand re-examines Conroe benchmarks

I think it's funny that not one single post here mentions the fact that when Anand went back and looked at some of the things people were complaining about, they actually found things that were wrong!


1) The BIOS was an problem that did skew the results, though it turned out to be a minor issue.

2) When Anand went back and looked at the FEAR benchmark, they found that the two boxes were not even running at the same resolution!

While our intention was to test both the AMD and Intel systems at the “Maximum” Computer settings and “High” Graphics settings, only the Conroe system was configured as such. We inadvertently left the AMD system at a higher resolution (1280 x 960) instead of the default resolution (1024 x 768) when you select the “High” Graphics defaults. The oversight was entirely our own doing as Intel was not running the benchmarks or configuring them, it simply happened while we were setting up both systems at the same time. We played with different resolution settings and while deciding that we would go with one, managed to configure the two boxes differently.
Click to expand...


According to the update, Intel still commands a lead, but the way many people handled and still handle these benchmarks really bothers me.

So many people rushed to defend the benchmarks, and they were so closed minded and naive, they insisted nothing was wrong with them. Now, it turns out that Intel's lead in one of the benchmarks was off by 50%! The first version told of a 40% gain, and the new versions reports only 20%; that's a 50% reduction!

One may be able to walk away from these benchmarks with a "general trend" sort of impression, but the bottom line is this: this was not a complete benching session done carefully and methodically by a third party. All this needs to be taken with a grain of salt.

Trusting the results between two systems Intel built themselves is like trusting the oil industry when they claim global warming doesn't exist.
 
I was expecting the DDR2 at 4cas to do alot for the scores. Does this mean latency isnt a big deal for DDR2 and the conroe chips?
 
visaris said:
I think it's funny that not one single post here mentions the fact that when Anand went back and looked at some of the things people were complaining about, they actually found things that were wrong!


1) The BIOS was an problem that did skew the results, though it turned out to be a minor issue.

2) When Anand went back and looked at the FEAR benchmark, they found that the two boxes were not even running at the same resolution!




According to the update, Intel still commands a lead, but the way many people handled and still handle these benchmarks really bothers me.

So many people rushed to defend the benchmarks, and they were so closed minded and naive, they insisted nothing was wrong with them. Now, it turns out that Intel's lead in one of the benchmarks was off by 50%! The first version told of a 40% gain, and the new versions reports only 20%; that's a 50% reduction!

One may be able to walk away from these benchmarks with a "general trend" sort of impression, but the bottom line is this: this was not a complete benching session done carefully and methodically by a third party. All this needs to be taken with a grain of salt.

Trusting the results between two systems Intel built themselves is like trusting the oil industry when they claim global warming doesn't exist.
Click to expand...

Don't be ridiculous, the BIOS made pretty much no difference, and the Conroe machine was running under specced as well.

All the benchmarks had the same results except the F.E.A.R which ANAND ran, not Intel. So this implication that Intel somehow cheated is ridiculous and is not bore out by the facts.

It needs to be taken with no grain of salt - it's pre-release benchmarking, treat it as such.
 
visaris said:
I think it's funny that not one single post here mentions the fact that when Anand went back and looked at some of the things people were complaining about, they actually found things that were wrong!


1) The BIOS was an problem that did skew the results, though it turned out to be a minor issue.

2) When Anand went back and looked at the FEAR benchmark, they found that the two boxes were not even running at the same resolution!




According to the update, Intel still commands a lead, but the way many people handled and still handle these benchmarks really bothers me.

So many people rushed to defend the benchmarks, and they were so closed minded and naive, they insisted nothing was wrong with them. Now, it turns out that Intel's lead in one of the benchmarks was off by 50%! The first version told of a 40% gain, and the new versions reports only 20%; that's a 50% reduction!

One may be able to walk away from these benchmarks with a "general trend" sort of impression, but the bottom line is this: this was not a complete benching session done carefully and methodically by a third party. All this needs to be taken with a grain of salt.

Trusting the results between two systems Intel built themselves is like trusting the oil industry when they claim global warming doesn't exist.
Click to expand...

There's was a link to the update about an hour after Anand posted it.
 
jebo_4jc said:
Keep in mind that the XE will provide about 12% more clockspeed than the 2.66ghz, so let's say the XE will beat the FX-62 by about 30-ish%.
Click to expand...


However the Conroe XE will have more advantages over the "12%" gain in clock speed than the midrange Conroe being displayed.

Still tough as an AMD fan you should know that, with these sort of architectures clockspeed does not equal a linear performance gain.
 
More power to Intel!
I say it's time there was some competition out there again!
 
Any of you guys rememder what site did the fear preview at super low res. so that the cpu was doing all the work. In that preview Conroe still beat AMD by 40%. That was a ques. I was wondering about because I thought that 1024X 768 was to high to test cpu .

800 x 600 should have been used . I just thought that AMD was given every break in the book. Intel gave just enough to turn the forums upside down and inside out

Here it is SO I was off Its more than 40%

http://www.pcper.com/article.php?aid=219&type=expert

http://www.bit-tech.net/hardware/2006/03/10/intel_core_microarchitecture/1.html

http://www.extremetech.com/article2/0,1697,1935745,00.asp
 
LstOfTheBrunnenG said:
But noes!!! AMD might release a 3.6 GHz FX-70 the day after Conroe launches!!! Just because they're not bragging about their performance, does that mean it isn't there???? They've probably got an army of 4 GHz quad-core Athlons waiting in the wings!!!!!!!
Click to expand...


*sighs* lol. If AMD had the money, they wouldn't be in this situation (where a non-XE lower clocked processor beats out AMD's)
 
UltimaParadox said:
However the Conroe XE will have more advantages over the "12%" gain in clock speed than the midrange Conroe being displayed.

Still tough as an AMD fan you should know that, with these sort of architectures clockspeed does not equal a linear performance gain.
Click to expand...

Semi-false. Clockspeed increases _within a processor family_ can come pretty close to linear scaling.

It's when you try to compare between e.g. a P4 and an X2 that the clock scaling is nowhere near linear.
 
StealthyFish said:
*sighs* lol. If AMD had the money, they wouldn't be in this situation (where a non-XE lower clocked processor beats out AMD's)
Click to expand...
Evceryone is forgetting the this is not an apples to apples comparison. Intel tested future technology against todays. The conroe is still 6 months from release, we should wait for the production verision to hit the streets before we make a conclusion.

I wonder why Intel chose not to compare it to there current EE chip? I think we all know the answer to this. There is no doubt the Conroe is going to be kick ass when it's released and should beat today technology. If it cant then that would spell trouble for Intel. So what is all the fuss over?It took Intel long enought o pick itself up, dust off, and get back into the game. Congrats Intel! I hope to hell any chip AMD plans to release in 6 month or a year beats the chips they have oput today.

The main complaint about the initial test results were that the Conroe spanked the FX. I can assure you the Intel rig was tweeked to the hilt but the AMD rig had a dated bios and dated ATI chip set. There is no doubt that the conroe would still come out on top but by how much. Screw bragging rights! If I am going to put off an upgrade based on this info, I want it to be as accurate as possible.
 
StealthyFish said:
*sighs* lol. If AMD had the money, they wouldn't be in this situation (where a non-XE lower clocked processor beats out AMD's)
Click to expand...

What does money have to do with it???? AMD has been spanking Intel for the past couple of years and they have loads of cash. :rolleyes: Future Technology should always beat current Technology. No doubt the Conroe is a great chip but you have no Idea what the release versions will be like. For all we know this could have been the XE version. Oh wait, Intel wouldn’t deceive us right? How about the clock is king crap? Or the 35 different chipsets they released in the last 2 years? All crap!! I am glad I am not an engineer of the P4 team. I wonder if they still have a job?
 
I finished reading over several of the threads in here on the conroe and several of the writeups from anandtech... Take this as fan boy-ism if you'd like, but the only thing I see happening in here is a group of intel fans high fiving eachother for leaked benchmarks of a next gen chip outperforming a current gen chip...

Yes, the conroe does CURRENTLY beat the FX-60 by 20%... yet you are comparing a 90nm chip of the current gen vs the 65nm chip of next gen...

I've done some reading up on the AM2 as well, and you guys are so blinded by these conroe samples that you forgot the competition has a timeline of its own. Yes, everyone knows the AM2's are coming in June, and the conroes are coming in Q3 (my guess is late August or early September)...

The AM2's will gain about increase their performance by about 10-15% with the intro of the DDR-800 RAM in June, as the current DDR-667 shows the AM2 socket processors being right on track with AMD's current s939 chips.

But the June release of the AM2's are still slated on the 90nm architecture... So this brings us to the release of the Conroe, which by that time will still be at a 10-15% increase over the current (at that time) AM2 systems, but you are not accounting for the AM2 fab to decrease to 65nm in Q4... Which my guess is late October... which should put the AM2 and the Conroe neck and neck performance wise, and only time will tell.

Everyone in this thread is patting themselves on the back over a next gen processor that won't be released for 6 months beating a current gen processor thats been out for 2 months... It's almost like the whole nVidia and ATi debate all over again, minus the flaming... everyone simply has skewed vision over one company's review...
 
James24 said:
Evceryone is forgetting the this is not an apples to apples comparison. Intel tested future technology against todays. The conroe is still 6 months from release, we should wait for the production verision to hit the streets before we make a conclusion.

I wonder why Intel chose not to compare it to there current EE chip? I think we all know the answer to this. There is no doubt the Conroe is going to be kick ass when it's released and should beat today technology. If it cant then that would spell trouble for Intel. So what is all the fuss over?It took Intel long enought o pick itself up, dust off, and get back into the game. Congrats Intel! I hope to hell any chip AMD plans to release in 6 month or a year beats the chips they have oput today.

The main complaint about the initial test results were that the Conroe spanked the FX. I can assure you the Intel rig was tweeked to the hilt but the AMD rig had a dated bios and dated ATI chip set. There is no doubt that the conroe would still come out on top but by how much. Screw bragging rights! If I am going to put off an upgrade based on this info, I want it to be as accurate as possible.
Click to expand...

Well it was the AMD CEO who challenged intel to a dual. When Woodcrest is released and Intel challenges AMD . Guess what Hectors reply will be.

Also when AMD releases its next generation chip in Aug. of 07 Intel will be releasing Its next generation on the 45n cpu in DEC . So you will again be cring its not fair comparison.


Also your facts are skewed go do a compare of intels latest chipset compared to the ATI chipset . Also Intel O/C the AMD64 cpu . gave it the fastest memory .

Intel ran their memory timings less than they could have also used 667 rather than the spec 800 .

All the evidence points to the fact that Intel showed AMD in the best possiable light and they held back their conroe to show less performance than whats actually there.

The evidence is there .

The reviews so far show that AM2 667 hurts performance period. Also I noted your 20% performance gain . Clock for Clock IT will be 30% . The evidence is clear.
 
$BangforThe$ said:
Well it was the AMD CEO who challenged intel to a dual. When Woodcrest is released and Intel challenges AMD . Guess what Hectors reply will be.

Also when AMD releases its next generation chip in Aug. of 07 Intel will be releasing Its next generation on the 45n cpu in DEC . So you will again be cring its not fair comparison.
Click to expand...

Just like Intel would be at 10ghz by now. I will wait and see what happens. By the way, Intel still hasnt stepped up to the plate on the challenge.
 
James24 said:
Just like Intel would be at 10ghz by now. I will wait and see what happens. By the way, Intel still hasnt stepped up to the plate on the challenge.
Click to expand...

Yes they did I just seen the preview of Conroe And Woodcrest spanked opties. Woodcrest is coming before Conroe so in 3 months Intel will throw down the gantlet
 
$BangforThe$ said:
Yes they did I just seen the preview of Conroe And Woodcrest spanked opties.
Click to expand...

OH Boy! How about current available architecture against current available archetecture? The Opty today wont be the same opty when Woodcrest finaly is released. Intel comes out 6 months in advanced to prove it can spank AMD today? Let's all high five ourselves for that....
 
Links to this new improved opty LINKS or just walk away $ months is not that long to wait for conroe reviews and I am sure that the review sites will have them out in 3 months . Intel has done a great job of damage controll here. Look at the nimber of people that said they won't buy the AMD64 but will wait on Conroe Intel was brillant.
 
$BangforThe$ said:
Links to this new improved opty LINKS or just walk away
Click to expand...

How can you make a comment like this when you are preaching about a CPU that isn't on the market yet? Its hypocritical. Almost like your claims to the non existant x1900XTXPE...

Apples to Apples... Today vs Today, not Tomorrow vs Today, or Next Month vs Next Week...

Comparisons and Challenges can't be settled unless you are comparing two of the same generation, or One company has ceased to provide a chip for the upcoming generation...
 
BLiTzKRiEG said:
How can you make a comment like this when you are preaching about a CPU that isn't on the market yet? Its hypocritical. Almost like your claims to the non existant x1900XTXPE...

Apples to Apples... Today vs Today, not Tomorrow vs Today, or Next Month vs Next Week...

Comparisons and Challenges can't be settled unless you are comparing two of the same generation, or One company has ceased to provide a chip for the upcoming generation...
Click to expand...
While not to be rude or anything but your pointing to a cpu that doesn't exist its pure vapor and saying this will beat Conroe. LOL Keep in mind I am in an intel forum . Not an AMD forum so There's noway I can be trolling.
 
$BangforThe$ said:
The reviews so far show that AM2 667 hurts performance period. Also I noted your 20% performance gain . Clock for Clock IT will be 30% . The evidence is clear.
Click to expand...

Gee, I thought when Intel finally switched over to DDR2, people were sqwaking about how DDR2 533 doesn't have any gain over DDR400. Why don't we let AMD refine their memory controllers much the way Intel had about 2 years to get theirs right. I will admit that Conroe looks like a very promising chip, however comparing these two chips is like compairing a prop airplane with a jet and saying, see this is an accurate representation. Lets do current gen vs. current gen and keep the shit talking to that. Thanks
 
covertclocker said:
Gee, I thought when Intel finally switched over to DDR2, people were sqwaking about how DDR2 533 doesn't have any gain over DDR400. Why don't we let AMD refine their memory controllers much the way Intel had about 2 years to get theirs right. I will admit that Conroe looks like a very promising chip, however comparing these two chips is like compairing a prop airplane with a jet and saying, see this is an accurate representation. Lets do current gen vs. current gen and keep the shit talking to that. Thanks
Click to expand...

Intel P4 are bandwidth starved the AMD 64 is not . Also the higher latency of DDR2 hurts AMD64 its doesn't help. Intel really didn't refine anything . If you testest todays Intel chipsets with DDR2 400 DDr 400 would still be better. Intel really never seen an improvement untoll DDR2 667. After that the greater bandwidth helps intel . In AMD64 this will not be the case until you go Quad core than it will help. But Intel in the Quad core will have dual fsb controllers.
 
covertclocker said:
Gee, I thought when Intel finally switched over to DDR2, people were sqwaking about how DDR2 533 doesn't have any gain over DDR400. Why don't we let AMD refine their memory controllers much the way Intel had about 2 years to get theirs right. I will admit that Conroe looks like a very promising chip, however comparing these two chips is like compairing a prop airplane with a jet and saying, see this is an accurate representation. Lets do current gen vs. current gen and keep the shit talking to that. Thanks
Click to expand...

To clear this up, it's not about comparing architecture it's about comparing current competition.
 
perplex said:
To clear this up, it's not about comparing architecture it's about comparing current competition.
Click to expand...

In that case go to ATF find the yonah review in the cpu section of the forum and see how a yonah@ 2.7 bitch slaps a 2.6x2. Is that current enough
 
$BangforThe$ said:
In that case go to ATF find the yonah review in the cpu section of the forum and see how a yonah@ 2.7 bitch slaps a 2.6x2. Is that current enough
Click to expand...
No, that comparison is not "current" enough. It suffers from the same problem. Yonah is not sold anywhere near 2.7GHz.

Since you didn't get it the first time, I'll say it again: A Yonah @2.7 compared to AMD64x2 @2.6, is still comparing a future Intel CPU against a current AMD CPU.
 
$BangforThe$ said:
In that case go to ATF find the yonah review in the cpu section of the forum and see how a yonah@ 2.7 bitch slaps a 2.6x2. Is that current enough
Click to expand...

You didn't understand what I was trying to say. I was talking in the pro-Intel way, in other words trying to say: It's perfectly reasonable to compare Conroe with X2 Because X2 will be the Conroe's competition upon release.
 
Man those things are impressive, It's going to be hard for AMD to counter that. Good job Intel, now we may have some better competition to bring down prices. (not likely)
 
perplex said:
You didn't understand what I was trying to say. I was talking in the pro-Intel way, in other words trying to say: It's perfectly reasonable to compare Conroe with X2 Because X2 will be the Conroe's competition upon release.
Click to expand...

Don't worry, they get it, they just don't want to admit defeat. As if they are AMD LOL!

[
 
s_s256 said:
Man those things are impressive, It's going to be hard for AMD to counter that. Good job Intel, now we may have some better competition to bring down prices. (not likely)
Click to expand...

Yes, AMD will have to cut prices a whole lot.
 
visaris said:
...
Since you didn't get it the first time, I'll say it again: A Yonah @2.7 compared to AMD64x2 @2.6, is still comparing a future Intel CPU against a current AMD CPU.
Click to expand...
You're right, you can't compare a notebook chip to a full-blown desktop chip. When Conroe is released, it'll be a fair comparison. :)
 
Donnie27 said:
Yes, AMD will have to cut prices a whole lot.
Click to expand...
Thank God.... Ever since the took the performance crown their prices went up. Oh well, I guess I will have to get used to paying more for a chip.
 
D4hPr0 said:
I cant believe this future and current debate is still going on.
Click to expand...

It's kind of fun to keep fueling fire ;) Who really gives a crap! As long as Intel and AMD continue to duel each other we will continue to reap the benefits. I personlay will continue to use AMD even if they fall a bit behind Intel just as the intel fans have stuck with the P4. As long as the performance is close, I will remain loyal.
 
1c3d0g said:
You're right, you can't compare a notebook chip to a full-blown desktop chip. When Conroe is released, it'll be a fair comparison. :)
Click to expand...

What in the hell you guys mean, future Processor?

Yonah

Newegg has 3 models.

http://www.pricescan.com/item.asp?category=misc&itemno=970125

http://www.pricescan.com/item.asp?category=misc&itemno=969686

Sure they're not massive amounts of these notebook processors.

http://www.pricescan.com/item.asp?category=misc&itemno=785448

Though I'd easily wait for Conroe and do much better.
 
James24 said:
Thank God.... Ever since the took the performance crown their prices went up. Oh well, I guess I will have to get used to paying more for a chip.
Click to expand...

Not me and that goes for Intel or AMD. $259 to $310 is my limit=P I don't care what Generation or etc... That E6600 at just over $300 the one that I have my eye on.
 
Donnie27 said:
What in the hell you guys mean, future Processor?

Yonah

Newegg has 3 models.

http://www.pricescan.com/item.asp?category=misc&itemno=970125

http://www.pricescan.com/item.asp?category=misc&itemno=969686

Sure they're not massive amounts of these notebook processors.

http://www.pricescan.com/item.asp?category=misc&itemno=785448

Though I'd easily wait for Conroe and do much better.
Click to expand...

They werent debating on availability but rather that they arent out yet at 2.7ghz.
 
Im really impressed so far with what Intel has brought to the table with Conroe. Looks like I may step over to Intel for the first time in over 12 years. My last intel chip was a Pentium 1 66MHz. :eek:

Way to go Intel :cool:
 
pandora's box said:
Im really impressed so far with what Intel has brought to the table with Conroe. Looks like I may step over to Intel for the first time in over 12 years. My last intel chip was a Pentium 1 66MHz. :eek:

Way to go Intel :cool:
Click to expand...
I am impressed as well and am glad to see them come back. I will stop short of saying I will switch unless there is a huge performance defference. I do feel AMD has dragged it's feet a bit when they could have pushed the performance higher.
 
Definitely impressive IMO. Looking forward to seeing what happens between now and its release. And this is coming from an AMD guy.
 
You must log in or register to reply here.
Back
Top