• Welcome to Valhalla Legends Archive.
 

Remove Array() by index

Started by CrAz3D, March 22, 2004, 01:47 PM

Previous topic - Next topic

Stealth

Creates a set of sample data for use in testing the speed of the array/collection.
- Stealth
Author of StealthBot

CrAz3D

Oh, heh.  Silly me.  Thanks
rebundance - having or being in excess of sheer stupidity
(ré-bun-dance)
Quote from: Spht on June 22, 2004, 07:32 PMSlap.
Quote from: Adron on January 28, 2005, 09:17 AMIn a way, I believe that religion is inherently evil, which includes Christianity. I'd also say Christianity is eviller than Buddhism (has more potential for evil).
Quote from: iago on April 19, 2005, 01:06 PM
CrAz3D's ... is too big vertically, at least, too big with ... iago ...

Adron

Quote from: o.OV on March 25, 2004, 02:13 AM
Adron, I need you to confirm that my changes were made correctly.

list of changes:
Timer is now GetTickCount
100000 is now 1000
command_1 has been modified

Changing it from 100000 to 1000 is a huge change. I don't think you'll get very testable results from that because times should decrease too much, as in a collection going from 2 seconds to 0.02, which is close to how small times VB can measure.

o.OV

Quote from: Adron on March 25, 2004, 05:31 PM
Quote from: o.OV on March 25, 2004, 02:13 AM
Adron, I need you to confirm that my changes were made correctly.

list of changes:
Timer is now GetTickCount
100000 is now 1000
command_1 has been modified

Changing it from 100000 to 1000 is a huge change. I don't think you'll get very testable results from that because times should decrease too much, as in a collection going from 2 seconds to 0.02, which is close to how small times VB can measure.

I only decreased it like this because memory
may have been a problem.
About the timing..
milliseconds seem to work well enough.
The results are consistent in the order of ranking.

command_1 25ms
command_2 1406ms
command_3 374ms
command_4 122

I'm still puzzled as to why command_3
still isn't performing to expectations for me.
the memory usage for the collection should be
about 100 times smaller..
so it can't still be a problem with memory..
can it?
If the facts don't fit the theory, change the facts. - Albert Einstein

Adron

Yes, that's strange... I'd love to see some results from other people - try the code and post whether my or o.OV's results are odd.

CrAz3D

QuoteNative Fast:
   com1:.171
   com2:12.283
   com3:.968
   com4:5.015

Native Small:
   com1:.171
   com2:12.656
   com3:1.015
   com4:5.171

Native No:
   com1:.171
   com2:12.437
   com3:.984
   com4:5.796

PCode:
   com1:.203
   com2:20.640
   com3:1.031
   com4:9.687

All seem to work for me

(com4 is o.Ov's)
rebundance - having or being in excess of sheer stupidity
(ré-bun-dance)
Quote from: Spht on June 22, 2004, 07:32 PMSlap.
Quote from: Adron on January 28, 2005, 09:17 AMIn a way, I believe that religion is inherently evil, which includes Christianity. I'd also say Christianity is eviller than Buddhism (has more potential for evil).
Quote from: iago on April 19, 2005, 01:06 PM
CrAz3D's ... is too big vertically, at least, too big with ... iago ...

Adron

Hmmk, so, the odd results are o.OV's. Now I wonder why he's getting so bad results for collections...

o.OV

*gives computer a mean kick*  >:(

I'm gonna go check MSDN for anything relevant to my odd results.
If the facts don't fit the theory, change the facts. - Albert Einstein

drivehappy

#38
http://www.experts-exchange.com/Programming/Programming_Languages/Visual_Basic/Q_20873546.html

Doesn't really account for the time differences between Adron's results and O.o's but it could possibly be the removal of the collections elements causing some delay on lower memory systems.

|