From <b><A HREF="http://www.aceshardware.com" target="_new">Ace's Hardware</A></b>
<font color=red>Thanks to Florian and Palm for writing in about these <b><A HREF="http://www.tecchannel.de/hardware/740/index.html" target="_new">Northwood and Brookdale benchmarks (German)</A></b> published by tecChannel.de. As you know, Northwood is the 0.13-micron Pentium 4, which also uses a new package, <b><font color=blue>mPGA-478</b></font color=blue>. The smaller Northwood Pentium 4 should run faster and cost less to produce due to its smaller size. Interestingly, the chip is said to contain the same amount of cache as Willamette, which means an 8 KB L1 (D-CACHE) and a 256 KB L2.
The review includes a number of benchmarks of a 1.5 GHz Northwood coupled with the Brookdale chipset and PC133 SDRAM. Unfortunately, the limited memory interface does not seem to play to the chip's advantages, as it performs rather poorly (generally about the speed of a 1 GHz Pentium III) and falls behind the 0.18-micron 1.5 GHz Pentium 4, which is paired with the i850 chipset. The Northwood/Brookdale/PC133 combination is roughly 10% slower than an equivalently clocked Pentium 4 with PC800 DRDRAM in most benchmarks. Will Brookdale be a success given this level of performance? If the price is right, perhaps. As Northwood's introduction draws closer, we should see more results with faster memory, like PC800 DRDRAM and PC2100 DDR SDRAM.</font color=red>
<b>Edit</b>
<b><A HREF="http://www.aceshardware.com/" target="_new">Ace's Hardware</A></b> has put up a correction on this article. It's actually is <font color=blue>Willamette Pentium 4</font color=blue> core in new mPGA-478 packaging.
Good or Bad have no meaning at all, depends on what your point of view is.
Nice bullsh!t post you put there you moron. No one has a Northwood chip yet, and for one they ain't gonna be realeased at 1.5Ghz, and they got 512K of L2 cache.
CYRIX RULES!
<b>The smaller Northwood Pentium 4 should run faster and cost less to produce due to its smaller size. Interestingly, the chip is said to contain the same amount of cache as Willamette, which means an 8 KB L1 (D-CACHE) and a 256 KB L2. </b>
Good or Bad have no meaning at all, depends on what your point of view is.
There are a ton of the older ES "engineering samples" running around now, they are supposed to be returned to Intel or destroyed. the oldest samples are slower with less cache, without actually knowing the build date. its hard to say if thats one of the first P4 northwoods built.
Im sure it makes AMD owners spooge onthemselves but those benchmarks show the processor they are benching is not even as fast as a wilmette not counting the PC133 SDRAM. The cache size is also an indication of the age of the ES.
Just ignore those trolls. The results you cited are perfectly reasonable. The 0.13 micron technology enables P4 to go to higher clock speed, but the chips rated at the same speed and of the same core design should run about the same, regardless processing technology. Naturally, the one with PC133 memory will underperform compared to the one with dual channel RDRAM. However, I would assume the 0.13 micron P4 will have more headroom for overclocking if Intel had not locked the multiplier.
**Spin all you want, but we the paying consumers will have the final word**
So why does'nt Intel allow the release of the brookdale using DDR? We all know the chipset supports it, yet they (intel) choose to delay the DDR launch for it? Something sure smells fishy in Denmark here..... Still trying to slam that RDRAM down its users throats. Users should not even bother with Brookdale until it is released with DDR support, then this chipset will show us what it is realy capable of and why the motherboard makers are calling it the next I815. One last question, is the p4 after the die shrink still slated to get back its missing FPU?
Raystonn has been playing "good cop" to his friends' "bad cop," or more accurately rational spin vs irrational spin. Kidding aside, there is nothing wrong to champion your employer's cause (in fact I think it is natural) as long as you do it by rational argument and solid data. Kudos to Raystonn who consistently delivers that. It's indeed nice to see him back.
**Spin all you want, but we the paying consumers will have the final word**
It appears that it actually is <b><font color=blue>Willamette Pentium 4 core</b></font color=blue> not a <b><font color=blue>Northwood</b></font color=blue> (<b><A HREF="http://www.aceshardware.com" target="_new">Ace's</A></b> and <b><A HREF="http://www.tecchannel.de/hardware/740/index.html" target="_new">the German site</A></b> had corrected this article).
>><b>The original site responsible for the review has published a correction indicating the chip is indeed a standard Willamette Pentium 4 in the new mPGA-478 packaging and not a true Northwood. This stands to explain a great many things, such as the smaller (256KB) L2 cache.</b><<
Good or Bad have no meaning at all, depends on what your point of view is.
. The results you cited are perfectly reasonable. The 0.13 micron technology enables P4 to go to higher clock speed, but the chips rated at the same speed and of the same core design should run about the same, regardless processing technology. Naturally, the one with PC133 memory will underperform compared to the one with dual channel RDRAM. However, I would assume the 0.13 micron P4 will have more headroom for overclocking if Intel had not locked the multiplier.
Tuatulin 512 kb L2 is the fastest cpu for clock for clock and without ASUS mobo.
Tuatulin have the same design that that copermine except the 512 L2.Coppermine is second in clock for clock strangely 1 now that he got 0.13 micron.
(not sure)
If the size of the cpu is smaller with the same number of transistor the path for data transfer is shorter so faster in the same way ?????
You are about to answer a thread that has been inactive for more than 6 months. If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.