Soldato
-
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.
Interesting view on Conroe in 2006...
- Thread starter Sumanji
- Start date
Soldato
Putting the bias to one side, it is a well put together article but he makes some odd conclusions. The Yonah is clock-for-clock equal to or more than an equivilent Athlon X2. Conroe is only going to improve upon that. That for me is enough to discredit the entire article. He waffles on about how Intel won't be able to match AMD because they aren't using a direct connection bus design, but _they already have_ matched AMD with the Yonah!
Direct connection bus would have been nice to see on Conroe, I'm sure everyone will agree, however it isn't required yet. A dual-core CPU can make do just fine with a traditional bus. It's once you get up to quad-core things start to need a proper direct connection bus. He acts as though the method Intel uses to enable cross-core communication without hitting the bus is a hack and performance hit but in reality it is not. AMD probably has an identical piece of logic on their CPU. He may have been getting confused with the shared cache topology in the Yonah and Conroe CPU's where each core has full access to all the cache. Now this does have a slight hit (2 cycles to be precise) but it is _far outweighed_ (at least in current applications) by both the cores being able to utilise the full amount of cache.
Nobody is saying the Conroe is going to be like 2x faster than what anything AMD has. I personally feel the "regular" (i.e. sub-£300) Conroe chips will pretty much exactly match what AMD has. The "Extreme Edition" and higher end Conroe's may go some lengths to extend that a little beyond.
It's not in Intel's interest to simply storm the market with a super fast chip. It makes more sense to their share holders to develop the super fast architecture first and then drag it out over the course of several years by releasing small increments in performance every quarter or so - only doing just enough to stay competitive.
Direct connection bus would have been nice to see on Conroe, I'm sure everyone will agree, however it isn't required yet. A dual-core CPU can make do just fine with a traditional bus. It's once you get up to quad-core things start to need a proper direct connection bus. He acts as though the method Intel uses to enable cross-core communication without hitting the bus is a hack and performance hit but in reality it is not. AMD probably has an identical piece of logic on their CPU. He may have been getting confused with the shared cache topology in the Yonah and Conroe CPU's where each core has full access to all the cache. Now this does have a slight hit (2 cycles to be precise) but it is _far outweighed_ (at least in current applications) by both the cores being able to utilise the full amount of cache.
Nobody is saying the Conroe is going to be like 2x faster than what anything AMD has. I personally feel the "regular" (i.e. sub-£300) Conroe chips will pretty much exactly match what AMD has. The "Extreme Edition" and higher end Conroe's may go some lengths to extend that a little beyond.
It's not in Intel's interest to simply storm the market with a super fast chip. It makes more sense to their share holders to develop the super fast architecture first and then drag it out over the course of several years by releasing small increments in performance every quarter or so - only doing just enough to stay competitive.