Intel introduces "Core 2 Duo"

I wonder why they didn't go with "Core Two-Two"?

Company News | Reuters.com
http://yahoo.reuters.com/stocks/Quot...=INTC.O&rpc=44

"bbbl67" wrote in message
oups.com...
I wonder why they didn't go with "Core Two-Two"?

Company News | Reuters.com
http://yahoo.reuters.com/stocks/Quot...=INTC.O&rpc=44


Who cares what they call it? It is still not an efficient design for a
multi-core, much less a true dual-core proc. With this "new" chip, the two
separate, unconnected cores can communicate with each other through the L2
cache, instead of the Northbridge chip. So now you have a chip that not
only is still tied to the Northbridge chip for RAM I/O, but now the two
separate processors are consuming the lions share of the L2 cache for
concurrency and latency comms. Next they design a suite of synthetic tests
to artificially inflate the speed of the proc...but then Intel is the master
of FUD.

Bobby

In article ,
says...

"bbbl67" wrote in message
oups.com...
I wonder why they didn't go with "Core Two-Two"?

Company News | Reuters.com
http://yahoo.reuters.com/stocks/Quot...=INTC.O&rpc=44


Who cares what they call it? It is still not an efficient design for a
multi-core, much less a true dual-core proc. With this "new" chip, the two
separate, unconnected cores can communicate with each other through the L2
cache, instead of the Northbridge chip.

That alone is not necessarily a bad thing. Sharing an L2 is
certainly a decision that cuts both ways, but not necessarily
wrong. Seperate L2s can have a lot of snoop traffic between them.

So now you have a chip that not
only is still tied to the Northbridge chip for RAM I/O, but now the two
separate processors are consuming the lions share of the L2 cache for
concurrency and latency comms.

The Northbridge is the problem. A shared L2 can go either way.

Next they design a suite of synthetic tests
to artificially inflate the speed of the proc...but then Intel is the master
of FUD.

Perhaps. That's what marketeers do.

--
Keith

On 7 May 2006 23:51:02 -0700, "bbbl67" wrote:

I wonder why they didn't go with "Core Two-Two"?

Company News | Reuters.com
http://yahoo.reuters.com/stocks/Quot...=INTC.O&rpc=44

This makes me suspect that Conroe/Merom ...pardon, Core 2 is not so
much ahead of AMD performance-wise, and needs a lot of marketing help
to take off. And here is another suspition: only select few
engineering samples were released to select few reviewers to produce
rave reviews. However I bet AMD can find a few chips among the
current crop of A64 that would clock close to or even above 4GHz, at
stock voltage, air cooled, and wipe the floor with Intel pre-release
chips at that speed. Would this mean AMD has the Conroe (pardon, Core
2) killer just waiting for the launch date? (shrug) I will believe in
Core 2 superiority when it ships in quantity and gets measured up
against then-current competition - both benchmarks and $$.

NNN

wrote:
This makes me suspect that Conroe/Merom ...pardon, Core 2 is not so
much ahead of AMD performance-wise, and needs a lot of marketing help
to take off.

That makes no sense. What could giving Conroe/Merom an official name
possibly have to do with how good or bad it is?

And here is another suspition: only select few
engineering samples were released to select few reviewers to produce
rave reviews. However I bet AMD can find a few chips among the
current crop of A64 that would clock close to or even above 4GHz, at
stock voltage, air cooled, and wipe the floor with Intel pre-release
chips at that speed. Would this mean AMD has the Conroe (pardon, Core
2) killer just waiting for the launch date? (shrug) I will believe in
Core 2 superiority when it ships in quantity and gets measured up
against then-current competition - both benchmarks and $$.

I'm pretty certain, all of the advantage is due to the 4MB of shared
cache. Using just a single core will result in a single-core having
access to the full 4MB of cache for itself, which would do wonders for
single-threaded tasks. There was story today about AMD doing a shared
L3 cache for its upcoming quad-cores.

Yousuf Khan

wrote:

However I bet AMD can find a few chips among the
current crop of A64 that would clock close to or even above 4GHz, at
stock voltage, air cooled,

You're cracked.

On 9 May 2006 12:00:28 -0700, "Rthoreau" wrote:

Then there is this little bit of information [fud] [you tell me].

http://theinquirer.net/?article=31569

I'm confused. They claim the Core Duo 2500 costs $2200? I can't find
it in the pricing sheet they link to - I only see Xeons and Itaniums
in that range.

Am I missing something? That price seems a little much for the
consumer market.

max

"max" wrote in message
news
On 9 May 2006 12:00:28 -0700, "Rthoreau" wrote:

Then there is this little bit of information [fud] [you tell me].

http://theinquirer.net/?article=31569

I'm confused. They claim the Core Duo 2500 costs $2200? I can't find
it in the pricing sheet they link to - I only see Xeons and Itaniums
in that range.

Am I missing something? That price seems a little much for the
consumer market.

max



You are not missing anything at all...the new Core 2 Duo procs are terribly,
horribly expensive.

Get an AMD X2...you'll get a true dual core with cores that can
communicate directly with each other (Core 2 Duo cannot); You'll get and
on-die memory controller (Core 2 Duo doesn't); You'll get hypertransport
(Core 2 duo does not have it)...in short, if you buy the Core 2 Duo you'll
be getting a whole lot less for a whole lot more.

Bobby

On Wed, 10 May 2006 03:32:13 GMT, max wrote:

On 9 May 2006 12:00:28 -0700, "Rthoreau" wrote:

Then there is this little bit of information [fud] [you tell me].

http://theinquirer.net/?article=31569

I'm confused. They claim the Core Duo 2500 costs $2200? I can't find
it in the pricing sheet they link to - I only see Xeons and Itaniums
in that range.

Am I missing something? That price seems a little much for the
consumer market.

Maybe the Inquirer only read their email and their source was writing
in NT$ and not US$? :P

--
A Lost Angel, fallen from heaven
Lost in dreams, Lost in aspirations,
Lost to the world, Lost to myself

Rthoreau wrote:

breath deeply, hold it take another breath, hold it exhale!

Now repeat after me competition is good, competition is good!

Golly gee whiz, I did not know that!

Let's not get our panties in a twist, I think that fud is every where
on both sides, also this is somewhat an apple to oranges setup.

65 nm vs 90 nm

When AMD come out with a 65 nm part then lets see which part wins out.

(seriously) Nope. All that matters is what's available for purchase
at the moment that you are making the purchase.

In conclusion I hope a 65 nm part is cooler, and faster then a 90 nm
part don't you?

No, I want computers to get slower and more expensive with time!

In the end the consumer wins, we get faster parts, and better prices.

Then there is this little bit of information [fud] [you tell me].

http://theinquirer.net/?article=31569

Well, OF COURSE people will pay WAY more money for Intel than they
will for the equivalent AMD!

(snip irrelevant Dell stuff)