Silicon Investor (SI) -- The First Internet Community

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.

Pastimes : Dream Machine ( Build your own PC ) -- Ignore unavailable to you. Want to Upgrade?

To: Cheeky Kid who wrote (2153)	8/25/1998 1:06:00 AM
From: Zeuspaul	Respond to of 14778

>>tests done with windows?<< I have not done any yet but I will come up with something. The contour benchmark will work in the Windows version of the program and should tell us something. I can add the dual 333 NT machine for additional info. I will not be able to run the macro based benches in the Windows environment. I should be able to come up with something similar. In order to program the program in Windows I have to learn C++ ? language. Zeuspaul

To: Cheeky Kid who wrote (2153)	8/26/1998 12:19:00 AM
From: Zeuspaul	Read Replies (2) \| Respond to of 14778

Off Topic...Additional testing for a specific application across 3 OS's This work is application specific. Your mileage may vary. I find some of the results interesting..particularly the dual 333 NT system. Sire, I modified the application specific tests to include Win 95 and NT. The NT and Win 95 versions of the program are identical and are 32 bit code. The DOS version is similar. (the same file opens in DOS and NT and Win95) The tests include five computers and four tests. The same project file is opened in all tests. Benchmark Description Contour #reply-5579575 except the correct amount of data points is 33667. Link I believe this is all mathematics. Nothing is written to the screen. This process creates the TIN. Redraw replaces the Pan/Zoom bench. This is 10 redraws of an Identical complex screen ( the contour map + 33667 data points ) Orientation No equivalent in Win95 or NT but it is included as it is the item that originated the speed question. ____________________________________________________________ The test computers..Two machines are identical but are tested in two OS's. Dual PII 333 512 MB RAM Appian dual monitor card (learned about it on this thread) NT PII 333 384 MB RAM Matrox MII DOS PII 333 same machine with Win95 boot PII 266 128 MB RAM Imagine DOS PII 266 same machine Win95 boot P133 32 MB RAM DOS P66 64 MB RAM DOS ATI ________________________________________________________ Results dual 333 PII 333 PII 333 PII 266 PII 266 P133 P66 NT DOS Win95 DOS Win95 DOS DOS Contour 15.047 12 12.029 14 15.256 40 44 Link 3.203 3 2.962 3 3.694 8 14 Redraw 30 27 34 33 46 80 156 Orientation 54 19 19 19 __________________________________________________________ Comments All results are seconds. Seconds to three decimals are reported by the program. Whole number seconds are with a stop watch or are reported as whole numbers by program. The identical results in the Orientation test for three machines lend credence to the keyboard repeat rate theory. That old VESA driver seems to be faster in DOS than the new Matrox driver is in Win95. The NT task manager indicated that both CPU's were at 100 percent utilization in the Contour and Link test. I find it interesting that a single processor outscored the fully utilized same speed processors in two tests. So far I have found no benefit from the dual processor system. I still do not fully understand the graphs in the task manager. There is an option to show one graph for both processors or one graph per processor. I chose the two graph option and one graph for each processor appeared. I do not understand why there are two lines on each of the CPU graphs. The green line is above the red line. The text indicates 100 percent utilization when the green line maxes out. The red line follows a similar pattern to the green line but it is below the green line. Zeuspaul