DailyTech - "Conroe" To Account for 40% of Intel Desktop in Q1'07

Submit News

Hardware "Conroe" To Account for 40% of Intel Desktop in Q1'07
Kristopher Kubicki (Blog) - April 17, 2006 3:50 PM

21 comment(s) - last by Viditor.. on Apr 22 at 11:40 PM

If you're buying an Intel CPU a year from now, you're buying a dual core

Intel plans to put multi-core processors in 85% of new Intel-based desktops by 2007, at least according to the most recent roadmaps from the company. Three quarters after the launch of Intel's upcoming Core processor, dubbed Conroe, is expected to account for a staggering 40% of all Intel-based desktops. 65nm Pentium D processors will account for the multi-core desktop processors not allocated for Conroe -- roughly 45% of all Intel desktop sales.

Intel also released its guidance for retail processor sales of Conroe, and the trend is identical to the OEM forecasts; Conroe will account for 40% (or more) of retail Intel desktop chip sales in Q1'07. The 65nm Pentium D processors still in the retail and OEM channel by 2007 will mostly be used in business platforms and low cost systems. Pentium 4 processors will still hold a small percentage of the Intel market, but these chips are for the absolute lowest end systems and include Celeron single core forecasts.

This is one of the most aggressive ramps for a new technology outlined by Intel on the desktop. Prescott, Intel's 90nm single core CPU, was only expected to consume 30% of OEM desktop sales three quarters after launch. Even a year after Intel's Pentium D launch, dual core technology still accounts for less than 35% of all desktops.

Conroe gave enthusiasts a bit of a double take after benchmarks from the Intel Developer Forum put the CPU 20% faster than AMD's best offerings in head to head benchmarking. AMD has plans of its own to counter Conroe. By the middle of 2007, AMD is expected to launch its K8L architecture. AMD's Henri Richard claims K8L is more "evolutionary than revolutionary," but even minimal improvements on the tried and true K8 could cascade into large gains in performance.

Comments

Threshold

Username
Password
remember me

This article is over a month old, voting and posting comments is disabled

New brand name or...?

By zsdersw on 4/17/2006 3:56:01 PM , Rating: 2

So what's Conroe going to be officially named? I hope it's not "Core" with Exxxx after it.. that's lame.

RE: New brand name or...?

By KristopherKubicki

(blog) on 4/17/2006 3:59:16 PM , Rating: 2

That's it.

Parent

RE: New brand name or...?

By zsdersw on 4/17/2006 8:06:01 PM , Rating: 2

Really? Then why isn't everyone calling it that at this point?

"Conroe" isn't really much shorter than "Core E6400" or whatever, especially when you consider that everything mentioning what Conroe is has to include an accompanying sentence which tells us that Conroe is the desktop version of "Core" being sold in notebooks.

Parent

RE: New brand name or...?

By TomZ on 4/17/2006 8:58:37 PM , Rating: 1

The model name needs numbers in it because there will be so many versions to keep straight. As with other processors in the past, the Conroe name will probably stay around in the market along with the new "official" names from Intel.

Intel has pretty much picked the term "Core" as their new processor brand name.

Parent

RE: New brand name or...?

By zsdersw on 4/17/2006 9:49:58 PM , Rating: 2

I know about the need for numbers.. but it still seems lame and underwhelming to me for everything to have the plain and boring name "Core".

I hope you're wrong about what the brand name will be.

Parent

RE: New brand name or...?

By DallasTexas on 4/21/2006 1:17:53 PM , Rating: 2

What would you like to call it.
Why the F is that important? Core is as good a name as any. How about we call it Opteron-Killer, is that good for you?

Parent

RE: New brand name or...?

By Viditor on 4/22/2006 11:28:20 PM , Rating: 2

quote:
How about we call it Opteron-Killer

Ironic that you mention Opteron as it's the one chip that is in no danger from the new core...

Parent

40% for 2007 actually sounds low to me

By hstewarth on 4/17/2006 4:05:12 PM , Rating: 1

Does 40% sound low for 2007. Especially if Intel comes out with some low end models. I would expect by 2007 Pentium 4/D will be pretty much gone by the waste side. Except for companies that use same configuration all the time.

I am actually planning on not getting a Conroe, but take the extreme step and get a dual Woodcrest system. I feel that the time has came to take the big plunge because up and coming Intel processors is a signficant event.

RE: 40% for 2007 actually sounds low to me

By KristopherKubicki

(blog) on 4/17/2006 4:09:34 PM , Rating: 2

For the most part, Pentium D and Pentium 4 (and the Celeron derivatives) will be the low end. There is *no* Conroe low end yet.

Parent

RE: 40% for 2007 actually sounds low to me

By trivik12 on 4/17/2006 4:42:39 PM , Rating: 2

I think the 1.6GHZ (E4200) part would cost only aroun $150. That's lowend for me. I thought Conroe single core (Millville) would release by Q1'07 and would replace celeron (or will be called celeron as well).

40% with intel market is huge numbers.

Parent

RE: 40% for 2007 actually sounds low to me

By killerroach on 4/17/2006 4:11:45 PM , Rating: 2

That number is for Q1 '07, which is a very aggressive target considering they're only launching around mid- to late Q3 '06. The question that remains, though, is whether or not that means they expect the yields to be that good, or that they're just going to shelve older parts. Of course, it could be a mutually beneficial solution, being that they move to shelve the older parts by slashing the prices on the currently existing ones...

Parent

RE: 40% for 2007 actually sounds low to me

By defter on 4/17/2006 5:49:59 PM , Rating: 3

Do you remember what was Athlon64's share of AMD's desktop CPUs two quarters after launch? I recall it was something like 10%.

40% in Q1-2007 is very impressive ramp.

Parent

RE: 40% for 2007 actually sounds low to me

By yzhu92 on 4/18/2006 9:00:46 PM , Rating: 2

I'm only gonna be buying Intel processors if they're cheaper than AMD's.

Parent

% of Performance/Mainstream...

By Furen on 4/17/2006 10:53:07 PM , Rating: 2

Am I the only one who notices that this may not include the "Value" segment? 40% would be insanely high if this does include the whole desktop segment, but I'd doubt it.

RE: % of Performance/Mainstream...

By hstewarth on 4/18/2006 9:57:13 AM , Rating: 1

Actual 40% for all Desktop segment is not that unreasonable. Even by AMD own documents Intel has 82% of the market. So 40% or the deskop market for Conroe would not be too hard for Intel.

But I think this means 40% of Intel's desktop segment.

I just think once the Conroe is out, the Pentium 4/D will have no more market.

Parent

RE: % of Performance/Mainstream...

By Viditor on 4/22/2006 11:40:32 PM , Rating: 2

Furen is quite correct that the high-end is far less than 40% of Intel's or AMD's desktop shipments...however Conroe will be both value and high-end markets so 40% does make sense.

Of course that 40% isn't till next year, so at 10% I don't expect to see many people being able to buy Conroe at NewEgg till late Q4.
Many people still quote Intel's 80% of global x86 sales but forget that the number includes everything (game consoles, embedded systems, servers, etc...). Intel's current marketshare of retail desktops in the US is actually only 18.5%, and 10% of THAT is 1.85% of the volume that they are currently shipping...of course that doesn't include business desktops, but it will take ~6 months for the platform to be qualified for SIPP platforms.

Parent

Intel 2007

By crystal clear on 4/18/2006 8:27:14 AM , Rating: 2

This will be a good year for Intel in general.The tatgets they set for themselves are a question of wait & see,if they can achieve them.Timing is an important factor-link it with Vista & the going is good for both.
If Intel can come up with a trade in program-wherby old CPUs can be traded in.The old CPUs can be resold to many underdeveloped countries.This could boost their sales/revenues.
AMD is able & can surprise all in the future-did somebody hear about REVERSE HYPERTHREADING?What are its feaseability

"Reverse hyperthreading"

By Spoonbender on 4/18/2006 9:13:06 AM , Rating: 3

Total nonsense would be my guess. It makes no sense to have multiple cores run the same thread. A single, wider core would be able to do that much more efficiently.
I think it's just some rumour spawned by an AMD fanboy going "Intel Sux. Intel does hyperthreading, so hyperthreading sux. AMD rocks, so they must be doing the opposite! Which means reverse hyperthreading! Man, I'm a genius"

Parent

Units or Dollars?

By TomZ on 4/17/2006 4:11:21 PM , Rating: 1

Are the above forcasts in terms of quantity of CPUs, or in sales revenue?

RE: Units or Dollars?

By stupid on 4/17/2006 4:27:47 PM , Rating: 2

I would suspect it's only unit's sold. The article focus more on market share than on revenues. If the performance numbers are anywhere near the teaser benchmarks then Intel should be able to sell Conroe pretty well to those people who wants to upgrade to a newer PC. Of course the die hard AMD fanboys will wait for K8L to come out and simply skip Conroe.

Me? I'm neither a fanboy of AMD or Intel. I simply want the best CPU that I can afford that will give me power and low heat since I want to build quiet/silent PC. If Conroe turns out to be a heat radiator, then I'll settle for an AM2 Athlon 64 X2. I don't want to wait 'til next year to upgrade since my the time Conroe comes out my Athlon XP will be almost 5 years old, and it is showing it's age.

Parent

RE: Units or Dollars?

By hstewarth on 4/17/2006 6:04:34 PM , Rating: 1

I think this #'s are on Intel desktop #'s only and have no relationship to AMD sales at all.

I do except that Intel new sales will increase because of this chip however but that is just my personal opinion. It just a feeling but I feel this generation will signficantly change the cpu market.

Parent

"There is a single light of science, and to brighten it anywhere is to brighten it everywhere." -- Isaac Asimov