Nehalem goes mainstream in Q4 2009

[spoonboy]

...get your wallets out in the meantime

"According to sources close to a motherboard manufacturer, Nehalem CPU in its 1366 socket won't get to mainstream by volume or price until Q4 2009. This might sound shocking to many, as you might expect to upgrade to a quad-core mainstream Nehalem in mid 2009; and unfortunately it won’t happen, at least not for a few hundred Euros / Dollars.

Nehalem will definitely launch as an ultra high-end desktop and server part in Q4 2008 and it will take time until this marchitecture sinks to the attractive $200 to $300 price range.

Therefore, don't you worry about Intel's soon-to-be launched P45 boards and the LGA775 socket, as it will serve you good for some time to come. After all, P45 is a 65nm chipset, and hey, that is better than the 90nm P35, X38 and X48; at least it sounds like it."

http://www.fudzilla.com/index.php?option=com_content&task=view&id=7391&Itemid=1

 

[JDocs]

Correction X48 is also 65nm, its the X48's only real improvement over the X38 as pretty much any X38 can do 1600mhz fsb. Some X38s even come with manufacturer support for 1600mhz fsb.

 

[quantumsheep]

Still don't think this'll stop me jumping on Nehalem as soon as it's released. I just hope that there's a decent single card GPU out by then.

 

[quantumsheep]

YOOOOOOOOOOOOOKSHIRE!

 

[outlw6669]

Sorry JDocs but I was under the impression that the X48 is just a higher binned X38.
As such it uses the same manufacturing process as the X38.
The P45 is supposed to be the first chipset from Intel to be produced at 65nm.
If I am wrong, could you please provide me some links?

@OP,
Can't wait for the P45
Even though it is just a midrange part, it still has very nice specs!

It will be nice to have Nehalem this year also but I had heard that Intel has blocked overclocking with the new system.
I wish I could remember where I had read it at.
I believe it had something to do with the on die memory controller validating the CPU speed with the NB clock generator.
Any attempt to overclock is automatically rejected by the offended clock generator.

Anyways, it should be interesting to see what happens with it at the end of the year!

 

[spoonboy]

Don't you mean YOOOOOOOOOOORKSHIRE,

Devon is still tops though

 

[MU_Engineer]

It will be nice to have Nehalem this year also but I had heard that Intel has blocked overclocking with the new system.
I wish I could remember where I had read it at.
I believe it had something to do with the on die memory controller validating the CPU speed with the NB clock generator.
Any attempt to overclock is automatically rejected by the offended clock generator.

That is only on the LGA 1160 systems, which are the midrange and budget units. The high-end LGA 1366 units will allow overclocking. It makes some sense from a product positioning perspective as some low-end chips are simply high-end ones with a low multiplier set in their μcode. They OC the same but cost far less than higher-end parts in that case. It also makes people buy more expensive chipsets to overclock as well. I predict that this little tactic will have at best no effect on Intel's bottom line and at worst, it will lose them much of the enthusiast market. In any case, people who overclock low-end parts will go to AMD; whether or not it will hurt their bottom line depends on how fast Bloomfield is to Deneb. If Bloomfield is a bunch faster, then people will suck it up and pay Intel more for the non-crippled boards and Intel would break roughly even. If Bloomfield's slower, Intel will be hurt as the low-end overclockers that would have used the slower and (should be) cheaper Intel parts will have to go to AMD to get their OC fix.

 

[outlw6669]

Thank you for the simple explanation.
It seems I need to do some more research to prepare myself for the launch.

I agree with your assessment of on Intel's overclocking block.
Unless a stock Bloomfield is faster than an overclocked Denab at the same price point they will be bleeding enthusiasts.
I doubt that it will have much of a large impact overall as the enthusiast market makes up only a small amount of their business.

Someone seems to be getting (possibly) over confident after the success of the Core arcitecture...

 

[dvmoo7]

Just wanted to add that i just saw a Nehalem demo today and Intel is hinting that the performance jump from core 2 - > nehalem will be similar to the performance jump from p4 -> core 2 arch.. Sounds pretty confident to me...

 

[Wisecracker]

bump

 

[SpinachEater]

This 2008 Q-whatever "launch" is going to make a lot of people cry. Nehalem is out there on a desolate island when it comes to competition...so pretty much they will be so over priced even OPEC will grab their @holes.

 

[Vertigon]

"will be so over priced even OPEC will grab their @holes"

.....hehehe spot on. I can see myself upgrading in about mid 2010 when windows 7 is stable, Nehalem jumps to 16 cores and runs on a stable chipset, which I hope for godsake is a significant upgrade unlike what we've been seeing lately. I still don't think Nehalem is much of an advance in terms of frontier technology. I think whatever is post Nehalem will become the next significant landmark.

 

[amdfangirl]

Nice, but I'm going still use XP and just love the almost instant booting! (as long as they don't start using EFI)

 

[Wisecracker]

Thank you for the simple explanation.
It seems I need to do some more research to prepare myself for the launch.

I agree with your assessment of on Intel's overclocking block.
Unless a stock Bloomfield is faster than an overclocked Denab at the same price point they will be bleeding enthusiasts.
I doubt that it will have much of a large impact overall as the enthusiast market makes up only a small amount of their business.

Someone seems to be getting (possibly) over confident after the success of the Core arcitecture...

I think it's just the reality of the arch. Intel will have their own 'issues' with the transition to an IMC with multiple cores. AMD does not have the cash (certainly not like Intel R&D cash).

Heat, phase timing, electrical, EMI, etc. - and the 'unknown unknowns' - all have to be effectively managed and addressed.