Fastest CPU in April

[George Macdonald]

Now that I read aobut it, it actually makes perfectly good sense. The
Core 2 Duo/Core 2 Quad can hold up the high-end with quad core and
higher-end dual core systems, these "Pentium E2000" series chips can
take up the middle ground of dual-core chips and Celeron will replace
the "Core Solo" brand for single-core chips. Fairly neat market
segmentation actually.

But Core Solo/Duo is the new naming within the Centrino and Viiv & VPro
platform strategies which Mr. Otellini enunciated back in early summer.

I can't imagine that Intel is in anything other than phase-out mode
with the old Netburst core. Certainly they weren't about to switch
100% of production immediately to the new Core architecture, bu surely
they must be moving in that direction! Assuming they've got all the
bugs worked out of the process it gives them a faster processor with
lower power consumption and a smaller die to boot.

And *yet*.... they just released a whole new Pentium D series with low
power characteristics in August(?) or so. Someone just couldn't bear to
trash the new masks after all the months of work?

The 90nm Pentium D (the 800 series) were a single die, the 65nm
Pentium D (the 900 series) were two dies.

Hmph I missed that - they went backward... from a twin die to two separate
dice glued together?

All I see indicate that the E2000 will still be a dual-core chip. That
leaves lots of room for a single-core Celeron. The trick will be in
pricing. The E2000 series will have a small price/performance niche
to fill between the Core 2 Duo (which are already under $200 for the
lowest cost E6300 model) and the Celeron.

Keep in mind thought that the Celeron brand is mostly selling for
$50-$75. The most expensive Celeron carried at Newegg is the Celeron
356 (3.33GHz, 533MT/s bus, 512KB cache) that they list at $65. There
is definitely room for pricing between the $75 Celeron and the $150+
Core 2 Duo, but it's not huge.

Well it goes against the new "announced" strategy, AIUI, but anything's
possible I suppose.

 

[wizzywiz]

Horse pucky..........I don't suppose you'll get around to telling us what it
might be, That takes 'days' to 'execute'.

You're correct.

 

[Tony Hill]

But Core Solo/Duo is the new naming within the Centrino and Viiv & VPro
platform strategies which Mr. Otellini enunciated back in early summer.

Things look like they may have changed, on the desktop front at least.
I'm no longer seeing any plans for any "Core Solo" desktop chips,
though again Intel hasn't published a roadmap recently so I might just
be missing them.

And *yet*.... they just released a whole new Pentium D series with low
power characteristics in August(?) or so. Someone just couldn't bear to
trash the new masks after all the months of work?

Phase-out mode doesn't happen overnight. In August Intel was JUST
releasing their Core 2 Duo chips and was primarily producing Netburst
core chips. We're now 4 months on an I suspect that production has
shifted, but still probably more Netburst core than Core-based chips.
Changes in production don't happen overnight, especially at Intel.
They've got a lot of fabs producing a whole lot of chips and they like
to make sure that things are working *well* before they switch over.
Traditionally it's taking them at least a year to transition all their
lines over. This is why I'm expecting the Pentium E2000 series to
come out first to displace the Pentium D line, but the Core-based
Celerons probably won't be out until next summer.

Hmph I missed that - they went backward... from a twin die to two separate
dice glued together?

Yup, and then back and forth again with the Core 2 Duo (1 die) and the
Core 2 Quad (2 dies). I guess there's a certain flexibility in that,
easier to adjust production to meet current demand. The market for
single-core Pentiums seems to be basically non-existent these days, so
rather than spending money to change production lines over to
dual-core dies they can just glue a pair of the dies together.
Alternatively if the market demands it they can sell the single-core
dies on their own with some cache disabled and they have a Celeron
chip.

Well it goes against the new "announced" strategy, AIUI, but anything's
possible I suppose.

Intel seems to have been a bit quiet on what they're announcing ever
since Core 2 Duo was released. Lots of rumors and little hard facts.
This suggests to me that the plans aren't quite as firm as they have
been in times past.

 

[krw]

You're correct.

So you'll remain a troll?

 

[George Macdonald]

Things look like they may have changed, on the desktop front at least.
I'm no longer seeing any plans for any "Core Solo" desktop chips,
though again Intel hasn't published a roadmap recently so I might just
be missing them.

I noted the lack of a desktop Core Solo one post earlier than quoted
here... though I tend to doubt the viability of such a chip. Then again
they'll want to do something with the err, "rejects".

Phase-out mode doesn't happen overnight. In August Intel was JUST
releasing their Core 2 Duo chips and was primarily producing Netburst
core chips. We're now 4 months on an I suspect that production has
shifted, but still probably more Netburst core than Core-based chips.
Changes in production don't happen overnight, especially at Intel.

ISTR the C2Ds were available in July. Then about a month later they
announced the new 65nm low power Pentum Ds - this was a new iteration of
the chip... more a phase-in than a phase-out, which was the puzzling point.

They've got a lot of fabs producing a whole lot of chips and they like
to make sure that things are working *well* before they switch over.
Traditionally it's taking them at least a year to transition all their
lines over. This is why I'm expecting the Pentium E2000 series to
come out first to displace the Pentium D line, but the Core-based
Celerons probably won't be out until next summer.

From the info they made public, I figured they had 2.5 fabs (Ireland was in
ramp-up) running 65nm when they announced the new Pentium Ds - it made no
sense to me at all.

Yup, and then back and forth again with the Core 2 Duo (1 die) and the
Core 2 Quad (2 dies). I guess there's a certain flexibility in that,
easier to adjust production to meet current demand. The market for
single-core Pentiums seems to be basically non-existent these days, so
rather than spending money to change production lines over to
dual-core dies they can just glue a pair of the dies together.
Alternatively if the market demands it they can sell the single-core
dies on their own with some cache disabled and they have a Celeron
chip.


Intel seems to have been a bit quiet on what they're announcing ever
since Core 2 Duo was released. Lots of rumors and little hard facts.
This suggests to me that the plans aren't quite as firm as they have
been in times past.

It could also mean lots of err, internal conflicts at the marketing *and*
technical levels - asses are being kicked, jobs are being stripped... the
night of the umm, "long knives".

 

[Tony Hill]

I noted the lack of a desktop Core Solo one post earlier than quoted
here... though I tend to doubt the viability of such a chip. Then again
they'll want to do something with the err, "rejects".

Of course, the rejects will go where Intel's rejects have been going
for the past 6 or 7 years, into the Celeron bin.

ISTR the C2Ds were available in July.

They were released July 27th, pretty darn close to August
Technically though the core first came out in June in the form of Xeon
5300 series processors.

Then about a month later they
announced the new 65nm low power Pentum Ds - this was a new iteration of
the chip... more a phase-in than a phase-out, which was the puzzling point.

I'm not actually sure just what "low power" Pentium Ds you are
reaferring to, to tell you the truth. Intel doesn't have anything
marketted as such. My only guess is that you're thinking of the
initial B1 stepping of the Pentium D 900-series, which had a TDP of up
to 130W and many power saving features disabled vs. the later C1 and
D0 steppings which had a TDP of "only" 95W and the power saving
features re-enabled. However the C1 stepping was released by in March
or August of this year.

August may well have been when they released the D0 stepping of these
chips, and probably that stepping may have released power consumption
slightly, but certainly there was no significant change vs. the C1
stepping. There are no press releases or anything of that sort either
(not that there often are any for a new stepping).

It could also mean lots of err, internal conflicts at the marketing *and*
technical levels - asses are being kicked, jobs are being stripped... the
night of the umm, "long knives".

Could be. Intel is in the midst of some fairly major restructuring
efforts, so that might explain some of the lack of communication.

 

[Trent]

I noted the lack of a desktop Core Solo one post earlier than quoted
here... though I tend to doubt the viability of such a chip. Then again
they'll want to do something with the err, "rejects".

Or maybe not:
http://www.xbitlabs.com/news/cpu/display/20061115223825.html

 

[George Macdonald]

Of course, the rejects will go where Intel's rejects have been going
for the past 6 or 7 years, into the Celeron bin.

IFF the Celeron moniker survives which is the point at discussion. Intel's
stated "platformization" strategy certainly indicates/allows the
possibility that Celeron would not survive.... one would assume based on
the success or otherwise of Viiv, Vpro et.al.

They were released July 27th, pretty darn close to August
Technically though the core first came out in June in the form of Xeon
5300 series processors.


I'm not actually sure just what "low power" Pentium Ds you are
reaferring to, to tell you the truth. Intel doesn't have anything
marketted as such. My only guess is that you're thinking of the
initial B1 stepping of the Pentium D 900-series, which had a TDP of up
to 130W and many power saving features disabled vs. the later C1 and
D0 steppings which had a TDP of "only" 95W and the power saving
features re-enabled. However the C1 stepping was released by in March
or August of this year.

March or August??Ô_õ

August may well have been when they released the D0 stepping of these
chips, and probably that stepping may have released power consumption
slightly, but certainly there was no significant change vs. the C1
stepping. There are no press releases or anything of that sort either
(not that there often are any for a new stepping).

Yes that's it and you're correct there was no end-user "marketing" - just a
PCN, which The Inquirer cottoned on to:
http://intel.pcnalert.com/content/eolpcn/PCN106404-01.pdf and which was
discussed here... in your absence. What this PCN covered was the use of
a "Mainstream FMB" for the revised processors vs. a "Performance FMB"...
quite a drastic change in fact, enough to need a BIOS change. The *max*
current was dropped from 125A to 100A so more than slightly changed... and
this was for the C1 -> D0 step change according to the docs. If the power
saving was re-enabled for the C1, then this is more than a minor tweak for
D0... apparently done "quietly".

 

[Tony Hill]

IFF the Celeron moniker survives which is the point at discussion. Intel's
stated "platformization" strategy certainly indicates/allows the
possibility that Celeron would not survive.... one would assume based on
the success or otherwise of Viiv, Vpro et.al.

I don't think either VPro or Viiv are likely to have much impact on
the Celeron brand, those are both targetted somewhat upstream of the
Celeron's usual target market.

Of course, given the relative lack of success of both programs, the
whole point might be moot.

March or August??Ô_õ

Sorry that should have read "March or April of this year". Problem
between brain and fingers on that one!

Yes that's it and you're correct there was no end-user "marketing" - just a
PCN, which The Inquirer cottoned on to:
http://intel.pcnalert.com/content/eolpcn/PCN106404-01.pdf and which was
discussed here... in your absence. What this PCN covered was the use of
a "Mainstream FMB" for the revised processors vs. a "Performance FMB"...
quite a drastic change in fact, enough to need a BIOS change. The *max*
current was dropped from 125A to 100A so more than slightly changed... and
this was for the C1 -> D0 step change according to the docs. If the power
saving was re-enabled for the C1, then this is more than a minor tweak for
D0... apparently done "quietly".

The biggest change was probably in the specification rather than the
chip itself. When the initial Pentium D 900 series was released they
used the maintream rating of 95W TDP for only the low-end 920 (2.8GHz)
and 930 (3.0GHz) chips, using the "performance" rating of 130W TDP for
their higher end 940 (3.2GHz) and 950 (3.4GHz) chips. Then, with the
release of the C1 stepping they moved all of those chips to the
"mainstream" rating of 95W but introduced the new 960 (3.6GHz) chip
with their 130W rating. The D0 stepping just bumped that 960 chip
down to the 95W rating.

Now obviously these chips aren't going to exactly match their rating,
a 3.6GHz C1 stepping isn't going to use 35W more power than a 3.4GHz
chip each and every time. There is always some variation to these
things and as the process matures and is tweaked power consumption
will tend to come down. The D0 stepping might have made a few more
tweaks to the chip to improve power consumption further, but it's
unlikely to be a major change. These new steppings are pretty normal
even as a chip is past it's prime. As metnioned previously, I expect
that the Pentium D will continue to be produced in reasonably large
quantities for several months to come, even as it is being phased out.

 

[krw]

IFF the Celeron moniker survives which is the point at discussion. Intel's
stated "platformization" strategy certainly indicates/allows the
possibility that Celeron would not survive.... one would assume based on
the success or otherwise of Viiv, Vpro et.al.

Deja Vu all over again! Intel learned nothing from IBM's failure
in the '90s? "Platforms" was the keyword then too. This is
classic politics over customers. DEC died of it, IBM barely
escaped that ugly death. INTC surely seems to be at least as
poisoned from within.

<snip>

 

[George Macdonald]

I don't think either VPro or Viiv are likely to have much impact on
the Celeron brand, those are both targetted somewhat upstream of the
Celeron's usual target market.

Of course, given the relative lack of success of both programs, the
whole point might be moot.


Sorry that should have read "March or April of this year". Problem
between brain and fingers on that one!


The biggest change was probably in the specification rather than the
chip itself. When the initial Pentium D 900 series was released they
used the maintream rating of 95W TDP for only the low-end 920 (2.8GHz)
and 930 (3.0GHz) chips, using the "performance" rating of 130W TDP for
their higher end 940 (3.2GHz) and 950 (3.4GHz) chips. Then, with the
release of the C1 stepping they moved all of those chips to the
"mainstream" rating of 95W but introduced the new 960 (3.6GHz) chip
with their 130W rating. The D0 stepping just bumped that 960 chip
down to the 95W rating.

I dunno where you're getting your info but the doc above and the 900 series
data sheet clearly show the *large* reduction in max power with the C1 to
D0 transition for the 940 thru 960.

Now obviously these chips aren't going to exactly match their rating,
a 3.6GHz C1 stepping isn't going to use 35W more power than a 3.4GHz
chip each and every time. There is always some variation to these
things and as the process matures and is tweaked power consumption
will tend to come down. The D0 stepping might have made a few more
tweaks to the chip to improve power consumption further, but it's
unlikely to be a major change. These new steppings are pretty normal
even as a chip is past it's prime.

I have to believe the data sheet - a 20% reduction in *max* power current
draw is more than a tweak.

 

[Tony Hill]

I dunno where you're getting your info but the doc above and the 900 series
data sheet clearly show the *large* reduction in max power with the C1 to
D0 transition for the 940 thru 960.

I'm getting my info straight from the horses.. umm.. mouth:

http://processorfinder.intel.com/list.aspx?ProcFam=2112

All C1 stepping Pentium D chips that Intel sold had a TDP of 95W
except for the SL9AP, which was a Pentium D 960 C1 stepping rated for
130W.

I have to believe the data sheet - a 20% reduction in *max* power current
draw is more than a tweak.

A 20% reduction in actual max power is is more than a tweak, a change
in specification by 20% might not be much more though. Intel, much
like AMD, is not longer providing an exact spec for the power of
individual chips but rather a couple maxmum power consumption points.
This is entirely targetted at OEMs so that they know what they need to
design their systems for and know exactly what chips they can and
cannot use. This is why you don't see any power rating between 95W
and 130W because Intel simply hasn't defined such any in-between
ratings for their Pentium D line.

Basically this means that the Pentium D 960 C1 stepping might have
consumed only 90W max (or less) most of the time. However there is
going to be a range of power consumption (most likely some sort of
bell curve), and if the top 1% of the chips consumed 96W then Intel
would need to spec the entire sSpec at the next higher rating point,
and that is 130W. In this example if Intel were to bring their
worst-case power consumption down by just a watt or two they could fit
it within their 95W spec. It's quite possible that average power
wouldn't even change much, just a tightening of manufacturing could
reduce the variation between average and worst-case.

 

[George Macdonald]

I'm getting my info straight from the horses.. umm.. mouth:

http://processorfinder.intel.com/list.aspx?ProcFam=2112

All C1 stepping Pentium D chips that Intel sold had a TDP of 95W
except for the SL9AP, which was a Pentium D 960 C1 stepping rated for
130W.


A 20% reduction in actual max power is is more than a tweak, a change
in specification by 20% might not be much more though. Intel, much
like AMD, is not longer providing an exact spec for the power of
individual chips but rather a couple maxmum power consumption points.
This is entirely targetted at OEMs so that they know what they need to
design their systems for and know exactly what chips they can and
cannot use. This is why you don't see any power rating between 95W
and 130W because Intel simply hasn't defined such any in-between
ratings for their Pentium D line.

I think it's fair to assume that, with D0, the chips which already met the
"mainstream" FMB specs, on C1, dropped down even further and were well
below the reqts; IOW yield distribution moved upwards. I note that the 960
also transitioned from a fixed voltage, with C1, to variable with D0...
pre-empting the OCers?

Basically this means that the Pentium D 960 C1 stepping might have
consumed only 90W max (or less) most of the time. However there is
going to be a range of power consumption (most likely some sort of
bell curve), and if the top 1% of the chips consumed 96W then Intel
would need to spec the entire sSpec at the next higher rating point,
and that is 130W. In this example if Intel were to bring their
worst-case power consumption down by just a watt or two they could fit
it within their 95W spec. It's quite possible that average power
wouldn't even change much, just a tightening of manufacturing could
reduce the variation between average and worst-case.

Without a statistically significant set of measurements, it's impossible to
know what the 960 was running at, worst case, with C1, and your bell curve
is an awfully big assumption - given the highly non-linear behavior here
one can easily assume the other way. Of course one would assume that Intel
*does* the measurements and decides which chips can be pushed on (a fixed)
voltage and hit the higher performance... and which have to be
"downgraded".

 

[Gremenbulin]

Or a better program/programmer.

dMn

Quad is only a giant leap for the ordinary desktop. Some scientific and
commercial applications require far more power, and have done for
years.