In the Forums...
Posted: June 18, 2000
Written By: Keith "Farrel" McClellan
Testing a New Tweak (cont.)
The second type of tweak is the obvious performance increase tweak. These can be things like upping the sample rate on your mouse, causing you to notice smoother mouse motion, to lowering the delay on menus within Windows. These are things that don't need to be benchmarked not because they necessarily couldn't be, but because it is obvious they have been applied just by using the computer.
The third type of tweak is one that makes a significant impact on the speed of the computer but is not immediately obvious to the user. There are several ways of testing that kind of tweak. Synthetic benchmarks such as 3D Winbench, 3Dmark 2000, and SiSoft Sandra all include the necessary features and information to compare a system before and after tweaking, and certain standard game timedemos (such as the Unreal flyby loop) can also be helpful depending on the type of tweak.
Remember that these aren't necessarily hard and fast rules, and depending on the tweak, your particular system configuration, and just how long it has been since you reinstalled Windows, can all effect your results and how effective a tweak is - or how well a benchmark performs. Don't give up on a tweak just because you run into what seems like a brick wall - very often that brick wall is just a piece of glass that needs a hammer to be shattered. Perhaps you need a companion setting for the tweak to do its job, or maybe something in the BIOS or system configuration is holding you back. Whatever it is, find it - you won't regret the search, and you might learn something new about your computer along the way.
Hopefully this tweak guide will help you when you are working with our other tweak guides, or perhaps it may even help you discover your own tweak. Tweaking is an art, and as such it takes a patient hand to guide it to a new discovery. As I said earlier, there are no hard and fast rules when you are talking about a tweak - what will work on one person's computer could very well kill someone else's - a fact that is all to important to remember when messing around with your computer's internals. Experiment a bit, however, and you will be pleased with the results. And if you are pleased with the results, feel free to email me and let me know what you discovered - no use keeping your new found information bottled up inside and not sharing it with anyone. Oh, and as always, if you have any comments or questions about this guide - feel free to email me about them.
Click here for a printer friendly version of this guide