Intel Corporation (INTC) Message Board

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC)

INTC 36.91

-1.1%

Dec 31 3:59 PM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: Raymond Thomas who started this subject	7/12/2001 1:41:49 PM
From: fingolfen	Read Replies (2) of 186894

I was looking at Anand's review of the P4 1.8GHz, and some things really struck me as interesting... the P4 seems to be scaling with frequency MUCH better than the K7. I ran some quick calculations using the 1.2GHz nd 1.4GHz K7 numbers as compared to the range of P4 speeds (1.3, 1.5, 1.8, and 2.0, though I separated out the 2.0 calculations as the bus speed changed too!). The calculations are somewhat simplistic, but the comparison is valid. I first calculated the increase in frequency (High-Low)/Low and the increase in performance ((Perf at high)-(Perf at low))/(Perf at low) (the absolute value was always taken as some benchmarks are "higher is better," and others are "lower is better"). I then made a simple ratio of the (Increase in performance in percent)/(increase in frequency in percent). In this case, a higher number is ALWAYS better. Granted this is a limited data set for the K7, and I dropped the MP version entirely, but it is very interesting... Content creation Winstone 2001 K7: 0.43 P4 (1.3-1.8): 0.62 P4 (1.3-2.0): 0.66 Business Winstone 2001 K7: 0.40 P4 (1.3-1.8): 0.71 P4 (1.3-2.0): 0.58 Internet Content Creation SYSMark 2001 K7: 0.78 P4 (1.3-1.8): 0.91 P4 (1.3-2.0): 0.90 Office Productivity SYSMark 2001 K7: 0.42 P4 (1.3-1.8): 0.69 P4 (1.3-2.0): 0.71 SysMark 2001 - Overall K7: 0.18 P4 (1.3-1.8): 0.50 P4 (1.3-2.0): 0.47 OBench 01 - Load Level 1 K7: 0.44 P4 (1.3-1.8): 0.61 P4 (1.3-2.0): 0.70 OBench 01 - Load Level 2 K7: 0.78 P4 (1.3-1.8): 0.92 P4 (1.3-2.0): 1.13 Studio Max - Anand version K7: 0.79 P4 (1.3-1.8): 0.84 P4 (1.3-2.0): 0.88 Truespace K7: 0.79 P4 (1.3-1.8): 0.79 P4 (1.3-2.0): 0.94 Quake III Arena K7: 0.35 P4 (1.3-1.8): 0.59 P4 (1.3-2.0): 0.70 DroneZ K7: 0.52 P4 (1.3-1.8): 0.59 P4 (1.3-2.0): 0.82 The implications of these numbers moving forward are pretty interesting. The P4 has a lot more frequency headroom at a given technology node than the K7. The P4 is also getting more out of those frequency boosts than the K7. A 1.8GHz P4 gives a 1.4GHz K7 all it can handle on most benchmarks... a 1.5GHz K7 will have even more trouble with a 2.0GHz P4... and Intel will be moving to Northwood at 2.2GHz+ in Q4 (with extra cache)... if you extrapolate this trend over time, it's easy to see how stories like this: theinquirer.net ... are starting to crop up...

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next