Black Box IEEE 1284 Betriebsanweisung PDF herunterladen (Seite 26)

Page 26 of 98 The Black Box Toolkit

7.1.1. Multi-processor systems and Hyper-Threading (Intel Pentium 4)

During testing it has become apparent that dual, or multiprocessor, systems that are used

to host the BBTK may display strange calibration results. Newer Intel Pentium 4’s with

Hyper-Threading may also display similar results as they emulate a dual processor

system. Testing has suggested that such systems whilst having a very high clock rate, e.g.

a 3.0GHz P4 with Hyper-Threading, may actually sample at half the expected rate as each

processor receives alternate clock ticks and a given piece of software typically runs tied

with its affinity to one processor. To improve sampling rates you have the option to disable

one or more processors, turn off Hyper-Threading in the BIOS or use another PC with a

single processor.

You should bear in mind that a sample rate anywhere above 1kHz will offer sub-

millisecond sampling rates. If you are concerned about the sampling consistency of your

system you have the option of carrying out an external calibration check. This simple

process is outlined below.

7.2. External calibration

If you are concerned by your host PC’s performance as a result of a calibration test you

should carry out further external checks. One such method involves using a second PC

with a standard CRT running at a known refresh rate, e.g. 100Hz. Hook up an opto-

detector to the CRT and position it over a constant white area of the screen so that the

opto-detector is activated. Then run Digital Stimulus Capture (DSC) for say 10 seconds.

Finally analyze the collected RTL file using the data analyser. You should see a regular

blip every 10ms which lasts for about 2-3ms. 10ms is due to the refresh rate and 2-3ms is

due to the phosphor decay time. If you are using a different refresh rate you will need to

calculate the display re-draw time using the following formula.

1000 / refresh rate = redraw time in ms e.g. 1000 / 70Hz = 14.29ms,

As a final check you can copy and paste the lower sheet from the data analyzer into

Microsoft Excel and enter a simple formula which will show any deviation from the

expected redraw time for any sample. An example Excel sheet is shown below both with

and without formulas exposed. Note all timings are in milliseconds. You could also make

use of an external signal generator should you have one to hand.

Event No L3 Onset L3 Offset L3 Duration Redraw

1 633.2375 635.5425 2.305 -

2 643.2348 645.5323 2.2975 =B3-B2

3 653.2313 655.5789 2.3476 =B4-B3

4 663.2217 665.5396 2.3179 =B5-B4

5 673.2403 675.5842 2.3439 =B6-B5

6 683.2268 685.5424 2.3156 =B7-B6

7 693.2468 695.6041 2.3573 =B8-B7

8 703.2436 705.5534 2.3098 =B9-B8

9 713.2409 715.6021 2.3612 =B10-B9

10 723.2335 725.5662 2.3327 =B11-B10

11 733.2524 735.5957 2.3433 =B12-B11

12 743.2389 745.5517 2.3128 =B13-B12

13 753.2477 755.6005 2.3528 =B14-B13

14 763.2562 765.5548 2.2986 =B15-B14

15 773.2455 775.6226 2.3771 =B16-B15

16 783.2588 785.5683 2.3095 =B17-B16

17 793.2584 795.6201 2.3617 =B18-B17

18 803.244 805.5786 2.3346 =B19-B18

19 813.2576 815.6067 2.3491 =B20-B19

20 823.258 825.5714 2.3134 =B21-B20

21 833.2565 835.6093 2.3528 =B22-B21

22 843.2614 845.5698 2.3084 =B23-B22

23 853.2788 855.623 2.3442 =B24-B23

24 863.2659 865.5759 2.31 =B25-B24

25 873.2694 875.6219 2.3525 =B26-B25

Min =MIN(F3:F26)

Max =MAX(F3:F26)

Mean =AVERAGE(F3:F26)

SD =STDEV(F3:F26)

Var =VAR(F3:F26)

1 2 ... 21 22 23 24 25 26 27 28 29 30 31 ... 97 98

Kommentare zu diesen Handbüchern

Keine Kommentare

Black Box IEEE 1284 Betriebsanweisung Seite 26

Kommentare zu diesen Handbüchern

Verwandte Produkte und Handbücher für Vernetzung Black Box IEEE 1284