• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Intels Secret Revealed

Basically the article is regurgitating the same tripe being spouted by a blog of some guy with a PHD (god knows in what).

See his blog here: http://sharikou.blogspot.com/2006/04/clovertown-scores-revealed.html

Every single one of his blog entries is anti-Intel. Says it all really.

He seems to think Conroe can only offer good single threaded performance. Even though it has already been seen to offer good multi threaded performance. He seems to believe Conroe's performance will be inferior to what Intel already have with the Yonah (Core Duo) chip. The guy seems to believe that a process' working set is stored in the L2 cache as a whole, when in reality only a handful of pages from a process' working set are stored it in at any one time.

The blog is full of supposedly accurate benchmarks of Intel and AMD chips that don't even exist yet, and are years away from even sampling.

Although he does rightly point out (many times and in many different ways) that NGMA currently won't scale well past dual core because of its decrepid bus design. This is a known fact and one that Intel openly acknowledges. So far they have overcome it by adding dual FSBs to chipsets. But longer term they are working on CSI (Common Systems Interconnect) which is basically a HyperTransport clone. This is roadmap'd for a 2008 launch but personally I think it'll be late-2007.
 
Last edited:
What the hell was he driveling on about... that artical sounded so fanboi'ish...

There was no evidence of his claims, no benchmarks, no hands on testing nothing... just lots of poorly written text...

I'll wait til I see a respected review site give the Conroe a good testing.
 
HeX said:
What the hell was he driveling on about... that artical sounded so fanboi'ish...

There was no evidence of his claims, no benchmarks, no hands on testing nothing... just lots of poorly written text...

I'll wait til I see a respected review site give the Conroe a good testing.
Click to expand...
I have see the testing and it is damn impressive, even comeing from a AMD fan boy like me.
 
NathanE said:
This is a known fact and one that Intel openly acknowledges. So far they have overcome it by adding dual FSBs to chipsets. But longer term they are working on CSI (Common Systems Interconnect) which is basically a HyperTransport clone. This is roadmap'd for a 2008 launch but personally I think it'll be late-2007.
Click to expand...

Yep, will probably be timed to be released as close to the AMD's release of 65nm Quad Cores as possible. No doubt there will be another 'fight' between the two - similar to the current AM2 / Conroe situation. Am I right in saying that CSI also includes moving the memory controller onto the CPU?
 
Conroe is definatly something that AMD have to worry about. AMD has pretty much had the market with the A64's but is time about to change? Who knows.

It is known that, Intel have a very quick cpu lined up, AMD have to do something to combat it.

Dont forget about Yonah ;) My next build when I can get hold of a board.
 
NathanE said:
Yes we've all seen the VictorWang @ XS benchmarks. They've been doing the rounds for weeks :)
Click to expand...
Yea it is pretty old but I didn't know if you have seen it or not. But it is impressive at 2 GHz.
 
Probably gunner get a telling off about the Link. Also about the swearing in your first post.

They try to mod these forums a lot, due to it being a family one. Thats understandable. Might be best editing your post.
 
Bennah said:
Probably gunner get a telling off about the Link. Also about the swearing in your first post.

They try to mod these forums a lot, due to it being a family one. Thats understandable. Might be best editing your post.
Click to expand...
Can't see the sweary lol.
 
Whilst i'm really reluctant to believe any of what that guy is saying, without evidence to back it up, one has to think that Conroe might struggle when pitted with a memory-bandwidth hungry application.

I mean even with the new architecture, Hypertransport should still give AMD the memory access latencies that Conroe will lack. I've seen the benchies on Anandtech, and I think they're mighty impressive.

But, back when the first Athlon 64s found their way into market, their strength was being able to access memory faster than intel's offerings. So, one has to ask, at what point will it cause Conroe to struggle?
 
You must log in or register to reply here.
Back
Top Bottom