65nm dual-core Presler has stability issue

On Wed, 16 Nov 2005 20:58:26 -0500, Tony Hill <hilla_nospam_20.DeleteThis@yahoo.ca>
wrote:

>On Wed, 16 Nov 2005 08:49:39 -0600, EdG <spam.DeleteThis@hotmail.com> wrote:
>
>> Charles Chou, Taipei; Jessie Shen, DigiTimes.com
>> [Tuesday 15 November 2005]
>>Sampling Intel’s 65nm dual-core Presler processors has revealed some
>>stability problems, according to unspecified motherboard makers, adding
>>that the problematic CPUs include the 2.8GHz 920 and 3.0GHz 930 series.
>>Launch of end products based on the Intel Presler core may be pushed
>>back from the January 2006 date originally scheduled by system vendors,
>>the makers noted.
>
>Whaddayaknow... A pre-release beta product has some bugs in it?
>Amazing! That never happens! :>
>
>Seriously though, the "stability problems" will likely be fixed either
>by an new stepping of the processor or by revisions to the power
>regulation specs for these chips, depending on just where the problems
>arise from. Either way it's unlikely to delay things by more than a
>month or two. Consider that Presler adds next to nothing to the
>existing 800-series dual-core P4s (it doubles the L2 cache, but the
>impact of that doubling seems fairly limited in real-world use) it not
>like this is going to really change things much.

I tend to agree - sampling is when you have err, samples:-) - dunno what
lead times are now but samples used to often be at 3months before
production. OTOH Intel has had a much harder time than AMD at fitting
2-cores in 90nm - one wonders......

>>Samples of Intel’s next-generation Yonah processors, on the other hand,
>>also built using the 65nm manufacturing process, have gone through tests
>>smoothly, the makers indicated. Launch of dual-core Yonah-based products
>>is expected to take place in January of next year as originally planned,
>>the makers added.
>
>Yonah is probably a much more important launch for Intel, so it's good
>for them that it is on-track.

It'll be interesting to see its life cycle..... without 64-bit. Necessary
or not there'll come a day when nobody wants to buy less - personally, for
me that was about this time last year.

--
Rgds, George Macdonald
 >> Stay informed about: 65nm dual-core Presler has stability issue 

On Thu, 17 Nov 2005 16:33:32 -0500, George Macdonald
<fammacd=!SPAM^nothanks@tellurian.com> wrote:

>On Wed, 16 Nov 2005 20:58:26 -0500, Tony Hill <hilla_nospam_20 DeleteThis @yahoo.ca>
>wrote:
>>Whaddayaknow... A pre-release beta product has some bugs in it?
>>Amazing! That never happens! :>
>>
>>Seriously though, the "stability problems" will likely be fixed either
>>by an new stepping of the processor or by revisions to the power
>>regulation specs for these chips, depending on just where the problems
>>arise from. Either way it's unlikely to delay things by more than a
>>month or two. Consider that Presler adds next to nothing to the
>>existing 800-series dual-core P4s (it doubles the L2 cache, but the
>>impact of that doubling seems fairly limited in real-world use) it not
>>like this is going to really change things much.
>
>I tend to agree - sampling is when you have err, samples:-) - dunno what
>lead times are now but samples used to often be at 3months before
>production. OTOH Intel has had a much harder time than AMD at fitting
>2-cores in 90nm - one wonders......

Presler isn't even a dual-core chip, it's two single-core chips in a
MCM. It's not even the two basically unconnected chips on the same
die like Intel's current Pentium D chips, but rather two completely
separate dies.

This kind of brings up a bit of a side question, is the problem with
the dies themselves (ie is the single-chip Cedar Mill processor also
having the same problems) or is it just with the integration of the
two chips? I'm not sure that it necessarily changes much from the
end-user perspective, just a thought I had.

>>Yonah is probably a much more important launch for Intel, so it's good
>>for them that it is on-track.
>
>It'll be interesting to see its life cycle..... without 64-bit. Necessary
>or not there'll come a day when nobody wants to buy less - personally, for
>me that was about this time last year.

I certainly would prefer a 64-bit chip these days, given the choice.
However that's certainly not the only consideration. Particularly for
a laptop I think it would come in a few steps down on the list of
features I'd be looking for. Of course, that doesn't mean that Intel
should just sit around and twiddle their thumbs for a while, but I
think they're well on the way to adding 64-bit capabilities to the
follow-up to Yonah (who's name I have since forgotten).

-------------
Tony Hill
hilla <underscore> 20 <at> yahoo <dot> ca
 >> Stay informed about: 65nm dual-core Presler has stability issue