New Precott 2.8A DO VCore only 0.956V-1.052V

[stan03]

intresting i wonder what they are going to say. i hope its not a typo for the voltage

 

[theflyingrat]

Why is it that this CPU only has a bus speed at 533 MHz? when the Prescott available now has a bus speed at 800mhz.
Also can the temperatures be true? At 1.052Vcore can the CPU generate 69.1c? Seems like a lot of heat.

I can answer both of these for you.

The 2.80A already exists! They've been on the market for a couple months, almost always in the OEM channel. That's not a typo - it's just another CPU to fit into a specific performance range for OEMs to build their PC lines around. Prescotts have always been in both 533 and 800 FSB configurations.

Again, 69C does NOT mean that the CPU WILL run at that temp! That's the max rated temp before thermal management has to kick in to cool it down, and prevent it from being damaged. Look at Intel's spec finder yourself - you'll see that even the lowly, much cooler-running 2.4A is also specced at 69C. Even the 3.4E, which should run much hotter than the 2.4A, still carries only a 73C rating. That's a spread of only 4C for a 1GHz span of speed. Heck, look at the Northwood ratings - the 2GHz Northwood is rated to a whopping 76C!!!! That doesn't mean that it will produce that much heat; just that if it creates that much heat, it can still continue to run.

 

[NookieN]

The bus speed is correct. The voltage probably is not.

I e-mailed the Prescott PDT and they're looking into it now.

 

[Petr]

I can not tell for sure about the voltage. Prescott's data are no longer in Specification Update and the new Spec. Update (specific for Prescott) is not yet out. But I think this is a typo, no CPU can run with this voltage at that frequency.

And yes, there are Prescotts 2.8A since launch on 2nd February. They do not have HTT and run on 133 MHz FSB. Those are targeted for OEMs. Initially there should not have been Prescott 2.4A which is still not listed in any datasheet.

 

[theflyingrat]

http://processorfinder.intel.com/sc...cFam=483&PkgType=ALL&SysBusSpd=ALL&CorSpd=ALL

Spec sheet for 2.4A.

As for the voltage of the new stepping - guess we won't find out until we actually see them on the market, eh? It is a funny 'coincidence' how that voltage matches some of the Pentium-M spec, isn't it? Someone copy/pasting in the database too fast?

 

[Petr]

http://processorfinder.intel.com/sc...cFam=483&PkgType=ALL&SysBusSpd=ALL&CorSpd=ALL

Spec sheet for 2.4A.

No, that's not full specification. I need to know especially Icc.

 

[theflyingrat]

Well, it does give basic details, but if you need more specific ones, perhaps you could get them directly from Intel?

 

[batboy]

I think Flyingrat is onto something there. I'm guessing the vcore is a typo.

The 533 bus is easy to explain, this allows an upgrade path to a Prescott (which is cheaper to build than the Northwood due to it's smaller core, so Intel makes more profit) if you only have an older motherboard that don't support HT or 800 bus CPUs.

 

[brint]

You are probably right.

 

[theflyingrat]

I think Flyingrat is onto something there. I'm guessing the vcore is a typo.

The 533 bus is easy to explain, this allows an upgrade path to a Prescott (which is cheaper to build than the Northwood due to it's smaller core, so Intel makes more profit) if you only have an older motherboard that don't support HT or 800 bus CPUs.

You know, Batboy, I was thinking on the lines of that before - on the no-HT/no-800 capability. That would be the first logical explanation I would think of. But any motherboards I could think of that would fit this description are the 845PE, 845E, and 850E for these. And of those, only the 845E (I think) did not have HT capability. But as far as I know, none of these chipsets or motherboards built with them ever had any support for the voltage spec that the Prescott carried - they wouldn't seem to work at all! And if you did somehow get them to work, chances are good that the VRM circuitry would be bludgeoned to death by the Prescott within days!

My only guess after that was so OEMs could conduct a bit of "PC line engineering" with it.

You see, the 2.8A, as I know first hand, isn't particularly fast. Even the old 2.8B beats it handily, and the 2.66 gives it a good run for its money. All while making less heat (mine does 38C idle, 54C full load with a Zalman 7000 AlCu at stock speed) and wearing on the motherboard less. But we all know Intel wants to make as many .09 micron CPUs as they can; they're cheap to build, and they have to pay for those fabs somehow! So, making as many Prescotts as possible, even low-end ones allow them to make more return. OEMs then have a low-performance option to put in their PCs. Intel and the OEM are both happy.

The best example of this I can think of is Dell's Precision 360 line. The baseline CPU is the 2.80A. The 2.8E is a $120 upgrade. It's a rip-off upgrade, but from a performance standpoint, the 2.80A is bordering on abysmal for its clock speed, so to some folks, it'd almost be worth it! The upgrade continues on up the line - 3.0E, 3.2C/E, 3.4C/E, and P4EE models.

Fortunately, this price gouging allows me to buy the CPUs off my company for cheap, while upgrading from an external source saves the company money. It's about the only good that comes out of it, really.

My 2.80A is currently running 3.33 at stock voltage, and it seems to perform well with a bit of its bus restriction alleviated. My dumpy MSI motherboard with no voltage or memory divisor controls is what's holding me back; the CPU at 3.33 runs at 42C idle and 58C full load. Enough to heat my room pretty easily.

As stated before, I'll let everyone know once my buddies get their 2.80As up and running and how well they tweak.

I'm beginning to think I'm the only person on this forums that has or ever will have 2.80As!