Intel Prescott P4 3.4GHz & Northwood P4 3.4GHz

Intel Pentium 4 3.4GHz "Prescott"
And Pentium 4 3.4GHz "Northwood"

By, Dave Altavilla
March 22, 2004

3.4GHz Prescott & 3.4GHz Pentium 4 Vital Signs Speeds, feeds, temps and overclocking

We took some quick temperature readings with these new 3.4GHz Prescott and Northwood based CPUs at default clock speeds and then decided to turn it up a notch or two, until we hit their respective ceilings.  We saw some dramatic differences in operational conditions for each of these new CPUs.

3.4GHz Prescott Vital Signs Open Air Test Bench No Load

3.4GHz Northwood Vital Signs Open Air Test Bench No Load

Prescott 3.4GHz @ 3.93GHz

3.4GHz Northwood @ 3.92GHz

The above tests were performed in an open air environment, rather than installed inside of a closed PC chassis.  This allowed each processor to operate under the coolest ambient air temperatures of about 21ºC (70ºF) or so.  As you can see, both processors are reporting reasonable temperatures through Windbond's Hardware Doctor health monitoring utility.  These temps were taken at "idle" with no load on the CPUs. Then, after a few trial and error reboots, we reached top end stable overclock speed for each CPU, both of which clocked in at about 3.9GHz or 230MHz times their respective 17X multipliers.  While this may sound impressive we'll caution you to remember that these tests were setup in open air environments.  Although these were what we would consider fairly stable overclocks, completing a few of our standard benchmark runs, in a closed ATX case one might in fact yield much more moderate results.

Voltage levels required to hit these speeds were 1.675V and 1.435V for the 3.4GHz Northwood and Prescott CPUs, respectively.  The temps we took while overclocked were dramatically different, however, with our Northwood core reporting in at 57ºC while Prescott hovered around 71ºC.  Although again, the core seemed fairly stable, we certainly wouldn't recommend running Prescotts for any extended period of time under these conditions.

Prescott System Characteristics Under Load At Stock Speeds One Hot Potato

The temperatures in our initial overclocking efforts were alarming enough that we moved on to stress testing the new CPU at stock speeds and again, on our open air bench test station.

3.4GHz Prescott Stock Speed
50% CPU Load

In this test, we simply fired up a client side instance of the "Folding At Home" distributed computing engine.  We set the client to run at 100% CPU load.  As you can see, in the WinXP Task Manager shot above, since the application is single threaded, we were reporting in actuality about 54% CPU utilization.  Again, what we saw for readings in Hardware Doctor, were a bit unsettling.  We confirmed with Intel directly regarding these reported temperatures and they responded that what we were seeing was typical of the new Prescott core under these conditions.  The temperature reported was pegged at 67.5ºC and this supposedly was within expectations.

At this point we're really beginning to believe that Prescott desperately needs its upcoming new platform architecture in place, to properly operate within what we feel are reasonable conditions.  Prescott's new LGA775 package will hopefully bring much better thermal characteristics, with more robust power and ground planes.  In addition, we're certain that forthcoming Alderwood and Grantsdale motherboard architectures will sport a beefier more adequately designed power array.  These new design enhancements are obviously where Prescott needs to move to, because in its current setup, the chip is quite frankly an "Easy Bake Oven", no doubt about it.

HotHardware Prescott Ready Test Systems Not all are created equal...

How we configured our test systems:
We tried to ensure that all of our test systems were configured as similarly as possible for this review.  Both of the Athlon 64 systems and the P4 system system were equipped with identical hardware, with the obvious exceptions being the motherboards and processors.  The same applied to the Athlon 64 FX-51 system, but because it required registered DIMMs, the memory was different as well.  The video cards, hard drives, driver versions (where applicable) and OS configurations were identical.  Our Pentium 4 systems were setup on an i875 Canterwood based motherboard from for optimal performance.  Before we started benchmarking, we entered the system BIOS and set each board to their "Optimized Defaults".  We then configured our RAM to run at 200MHz (DDR400), with the timings set by the SPD.  The hard drives were then formatted, and Windows XP Professional (SP1) was installed.  When the installation was complete, we hit the Windows Update site and downloaded all of the available updates, with the exception of the ones related to Windows Messenger.

Then we installed all of the necessary drivers, and removed Windows Messenger from the system altogether.  Auto-Updating and System Restore were disabled as well, and we setup a 768MB permanent page file on the same partition as the Windows installation.  Lastly, we set Windows XP's Visual Effects to "best performance", installed all of our benchmarking software, defragged the hard drives and ran all of the tests.

Alright then, let the games begin...

Synthetics - SANDRA and PC Mark 2004