Intel’s next generation processor micro-architecture development seems to proceed ahead of track. Based on claims of certain Asian-based media, the world’s largest chipmaker will release its highly-anticipated Conroe processors in July, 2006. The commercial introduction of a product that can potentially offer improved performance and some other features may significantly affect the market of microprocessors.
Citing sources among mainboard makers, Taiwanese web-site DigiTimes reports that Intel will push up the launch of its desktop code-named Conroe central processing unit (CPU) and delay the debut of its Intel 965-series chipset so that both hit the market at the same time in July, 2006. Earlier it was expected that the Conroe chips will be launched in the fourth quarter, 2006, while the new breed of chipsets was expected to be released in the second quarter of the year.
Earlier than expected commercial launch of any product may indicate that its development is proceeding smoothly and its yield rate is expected to be mature enough when production starts about a quarter before the commercial launch.
According to the publication, the acceleration of Conroe roadmap was performed in a bid to boost demand for desktop computers manufactured by Intel Corp. The commercial launch in July may indicate that actual products are likely to be displayed at Computex Taipei 2006 show in early June. Given that release of the chipset was planned at an earlier date, mainboard makers are unlikely to face any difficulties with designing infrastructure for the new Conroe chip.
Intel Conroe processor will feature two processing engines with 4MB cache in total, as well as Intel’s next-generation micro-architecture, which combines high-performance and energy-efficiency. The chip will be produced using 65nm process technology.
Intel Corp. did not comment on the news-story.
Comments currently: 9
Discussion started: 12/21/05 05:52:23 PM
Latest comment: 08/14/06 04:00:01 AM
Frankly Conroe and Merom can't come soon enough. And if Intel doesn't smell the bad news on the wind before those cpu's come, they need their noses checked.
Oh well we'll see if it's true in a few more months
12/21/05 05:52:23 PM]
The problem with Yonah will be the speed bins they get them to run at, which while competitive with X2's on a speed for speed basis, don't run at the speeds that X2's run at now. A compounding problem is that 64bit is really going to be a serious marketing push for Vista. Vista, lacking AS many features to give it marketing oomph, is likely to say that with it you can finally get the power of 64 bit. They'll do this because they've not really pushed XP64, and they need some serious marketing push to get Vista out the door to more than first adopters.
That means, simply because of the marketing push, that there will be pressure on hardware vendors to be up to spec for Vista. We all know 32 bit will work, but the average Joe is going to believe that you can get more with 64. And when that effect hits the market, anything that doesn't say 64 bit on the outside is going to get avoided at least some of the time on the market, IF there is a 64 bit machine alongside it. Right now, there aren't many 64 bit laptops, but if that changes, the market dynamics change bigtime.
So Yonah isn't a bad thing, but it might very well run into the same thing that AMD faced in the heyday of the P4. In those days if you couldn't say you ran at 3+ghz you weren't worth a customers money, and that was ridiculous but it was a fact of selling. This is going to be the same thing in another area, but with the tales reversed. On the desktop, no change, but everyone is already beginning to get the word that P4 isn't something they want. So mobile sales propped up Intel in the last 2 quarters because the Pentium M is a good chip.
So Yonah is a nice cpu, but I think the forces around it are going to do damage to it. It is rather funny to see such a dual line coming from the same company. On the desktop Intel says "64 bit is the next wave!" and then with Yonah they are going to say "the newest mobile dual core solution!" and quietly hope no one notices the missing phrase. In a few more months, people are going to notice...
12/22/05 05:03:14 AM]
Hmmm...So they push Conroe forward, eh?
Well, Yonah is already on the level with A64 (just a bit behind, but close enough), as Anandtech.com have demonstrated. The only thing missing is 64bit.This is addressed by Conroe/Merom/etc.
So Intel is giving Yonah a relatively short life. I guess its a worthy sacrifice, given that they need to respond to AMD.
I know the infrastructure (mobo/chipset/etc) they've setup is compatible with both Yonah and Conroe, so the transition for manufacturers should be Ok for the most. (There's always some company that try to "save a buck", and end up killing compatibility with something).
The problem with Vista is that Microsoft is facing a "Why should I upgrade to Vista, when my existing or alternative solution does the job?" crowd. People don't like to continually spend money every few years on basically the same thing over and over again.
Microsoft knows this, that's why they've "prettied up" the interface to make it look like its a brand new thing, when it isn't...Its Win2k3 code, last time I checked...This is especially true for businesses, some still use Win2k because its still perfectly fine for what they use it for. Its often difficult to encourage an employer to move to a different platform when the existing does the job.
I do agree with TA152H. People who buy 64bit CPUs aren't the average Jane/Joe PC user. They're engineers/researchers/etc. They are more likely to use Linux or some other low cost, but flexible solution, rather than Windows.
If you build a relatively low cost cluster, you don't want to pay license for every node in your setup!
64bit is gonna take a lot longer to adopt on the desktop. The benefits it brings, doesn't really improve a typical desktop user's requirements. Most often, it does jack all for them.
Think about it...What does 64bit offer to the desktop user?
Improved performance? => Not on desktop apps it doesn't. Improvement is often negligible. (Sometimes below current 32bit versions)
Address for more RAM? => Any typical desktop users need more than 4GB of RAM? Not alot.
Accuracy? => Does the desktop user need 64bit precision in calculations? No. Folks like engineers/scientists need that kind of accuracy.
So what's left? Even if Intel, AMD, and Microsoft make a hard push for 64bit, it ain't gonna work. Why? Average PC users will ask why they need 64bit. How is it gonna benefit them when it really doesn't?
What's worse is the lack of wider hardware support, namely, drivers. If you have tried AMD64/EM64T version of Windows XP, you soon realise its a waste of your money as you find out a number of your hardware doesn't work with it. (since hardware manufacturers don't provide 64bit versions of their drivers).
The situation is slighlty better on Linux, as you can try an compile the open-source ones. (But of course, thanks to hardware manufacturer's reluctance, hardware support for Linux is not as big as WinXP...But still much better than the WinXP 64bit one!)
Its like AGP to PCI-Express Graphics transition.
Did you know 70%+ of PC users still have AGP solutions?
(Despite what mobo manufacturers and ATI/Nvidia have done).
The ultimate decider for an adoption for the standard is not a company (no matter how hard they push), its the customer. Look at VHS vs Beta. Beta was technically superior, but VHS was chosen.
The real question is to ask how one can justify the customer spending money on something that isn't really gonna provide major benefits to them.
As for confusion with naming? A typical person is ALREADY confused by Pentium 4's number scheme! It took me an hour to figure out what exactly is the difference between ALL Intel CPUs.
To explain it to the average person requires alot longer!
Intel has released multiple lines of CPUs in the past, and they'll do it again. Its a question of the marketing department trying not to screw up the naming scheme and cause confusion.
12/22/05 04:48:03 PM]
Add your Comment
Enter your username and e-mail address. Password will be sent to you.