Conroe - Worth the wait?

[Cozman]

Hi

Dont know if true or not ... http://sharikou.blogspot.com/2006/04/conroe-performance-claim-being-busted.html ... seems to claim Intel kinda cheating benchmarks & conroe may not be as impressive as first thought

just giving info dont flame me for it

Gotta add probably FAKE as no one has the chip to actually test as far as i know

 

[Durzel]

I don't know about that article, but 2.16Ghz Yonah chips (Conroes "son" for want of a better word) are blitzing SuperPI results lately, turning in sub 20 sec results on aircooled 3Ghz+ setups.

 

[NathanE]

The guys over on Xtreme Systems have got Conroe samples and haven't found anything fishy. If anything Intel was being modest about Conroe's performance at IDF (they weren't running the full DDR 800, but DDR 667).

Edit: Just read some of the comments to that article. The author is clearly a whack job His only complaint seems to be:

"Frankly, I am really disappointed by Intel. This gimmick of using 4MB cache to get unreasonably good scores on most simplistic tests is so cheap!"

But doesn't actually go on to say why he thinks a large cache is "cheap". Probably because he knows that the AMD64 won't be seeing any cache increases till its next die shrink?

 

[Explicit]

But doesn't actually go on to say why he thinks a large cache is "cheap".

To be fair, he does say why he thinks it's "cheap":

My analysis pin pointed the origin of some of the extreme numbers of the Conroe - 4MB cache. If you take a Super PI 1M test and conclude Conroe is 2x the speed of Athlon64, you are doped. For the same reason, if you take the default ScienceMark MolDyn test and conclude Conroe is 30% faster, you are also kidding yourself. Both these tests fit in Conroe's 4MB cache, and that's why they look too good.

I read some comments saying the 4MB cache is not a gimmick but can be a real boost for performance in some cases. In some sense, that's true, if you are running something like Super Pi 1M for fun. Even for large apps, more cache is better. But considering the complexity of today's applications, 4MB cache will only bring a small increase in performance, definitely not something like 10-20%.

All I can say is this is a waste of time. I'm not bothered with anymore benchmarks until Conroe is actually out.

 

[m3csl2004]

i see what he means, allough imo i would rather have 4mb of l2 and be considered cheap, then have the moral high ground but have 2mb

yonah still gets amazing benchmarks with only 2mb

 

[NathanE]

Oh I missed that. Probably because I ignore stuff that is wrong or doesn't make sense

"Even for large apps, more cache is better. But considering the complexity of today's applications, 4MB cache will only bring a small increase in performance"

Considering the complexity of today's apps? Yes they are very complex... So what exactly is his point here?

He also makes the statement several times that the cache is only useful (in terms of performance) if you can fit the entire program's working set into it. What tosh. The cache only stores selected _pages_ of memory, not entire processes! The guy may have a PHD but it certainly isn't in Computer Science!

Numerous people have pointed out in the article's comments section that the guy is simply ignoring history. The P6 architecture from which Yonah, Merom and Conroe are derived consistently receives a performance boost whenever more cache is added. He just seems to ignore it though. He also totally ignores the fact that Intel already has a chip (Yonah) that beats the A64 in the exact benchmarks he says Conroe won't. So what, Intel is taking a step back with Conroe are they?

The article is just satire. It'll be on the headlines of several tech news sites later today and it will have served its purpose. He will have given every AMD 'boy a couple months more lease of life.

 

[aardvark]

that article makes a lot more sense than some intel benchmarks that will obviously be biased - we are talking hundreds of millions of dollars in investment etc - of course intel will cheat as much as possible (as any company would)

as with anything we will have to wait til people actually get their hands on some real samples and test it independently - anything we say up til then is pure speculation.

 

[Pulseammo]

I'll admit I took the lazy arse route and didnt RTFA but if the guy is moaning about the cache and that the benchmarks are rigged for the stuff like the superPi tests etc, what about the gaming tests that anand did? The Conroe still won, so because of the 4mb cache or not its still doing better unless the gaming tests were somehow rigged.

 

[NathanE]

that article makes a lot more sense than some intel benchmarks that will obviously be biased - we are talking hundreds of millions of dollars in investment etc - of course intel will cheat as much as possible (as any company would)

as with anything we will have to wait til people actually get their hands on some real samples and test it independently - anything we say up til then is pure speculation.

Sigh.

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

 

[TooNice]

that article makes a lot more sense than some intel benchmarks that will obviously be biased - we are talking hundreds of millions of dollars in investment etc - of course intel will cheat as much as possible (as any company would)

Then would AMD be cheating in their AM2 sample benchies are unreliable too?

The average Joe will still buy Intel irrespective of how well Conroe perform. Intel clearly has their eye on the educated enthusiast buyers. -We- most likely would not buy the product even after a positive first impression if it was found out, through independant testing, that the Conroe is not all that. A number of people may even hold against Intel for a good few years if they find that Intel pulled a fast one. TBH, I don't care much about those benchies. Office applications + games is what I care the most. As long as the released product performs within my expectations, I will eye on their quad-core next time I upgrade (probably mid 2007). By then AMD should have a reply to the chips based on the Conroe architecture, and I'll pick the winner

 

[aardvark]

Then would AMD be cheating in their AM2 sample benchies are unreliable too?

The average Joe will still buy Intel irrespective of how well Conroe perform. Intel clearly has their eye on the educated enthusiast buyers. -We- most likely would not buy the product even after a positive first impression if it was found out, through independant testing, that the Conroe is not all that. A number of people may even hold against Intel for a good few years if they find that Intel pulled a fast one. TBH, I don't care much about those benchies. Office applications + games is what I care the most. As long as the released product performs within my expectations, I will eye on their quad-core next time I upgrade (probably mid 2007). By then AMD should have a reply to the chips based on the Conroe architecture, and I'll pick the winner

given their history i would expect intel to cheat more than amd - the problem is we just don't know - and we won't for months, so i'll wait til then to make a judgement either way.