Performance Testing and Maintenance Vista Tips Discuss Performance Testing and Maintenance in the Windows Vista forums; Hi Friends: I have been using the PassMark Performance Test (PT) ( http://www.passmark.com/index.html ) to get an indication of when maintenance is needed. I think it is ...

[Znod]

Hi Friends:

I have been using the PassMark Performance Test (PT) (http://www.passmark.com/index.html) to get an indication of when maintenance is needed. I think it is a good idea to use a performance device, such as PT, to track the likely inevitable decline in performance over time and, thus, to indicate when performance has degraded to the point where maintenance is desirable.

Recently, my performance had decreased by about 30 or so PT points on my laptop installation of TechNet's "Real" Vista Ultimate. So, I performed a number of maintenance steps. I used disk cleanup and Perfect Disk for starters. Perfect Disk, in particular produced a good abatement of my decreased performance.

But, I also knew that I had a fair number of registry errors that I had not corrected--for fear of Registry Mechanic being too aggressive. I decided to fix the registry errors and then to gamble that I could reinstall Real Vista from within Real Vista (i.e., even though the Update from XP to Real Vista is not supposed to work). The registry repair did not change anything PT test-wise. However, the reinstall worked, and, voila, I had regained the rest of my lost performance.

I had always though that reinstalling from within to fix problems was a good idea--which has worked for me several times. But, now I am convinced that doing so can be used effectively with, or without, registry repair to regain lost performance.

[felix8406]

Znod, does using a USB drive as readyboot makes a different in your PC's performance?

[Znod]

I did not find a measureable difference when testing on my desktop--"back then." The problem is that even sequential performance tests move around enough that one can't pick up a consistent difference that might be attributed to ReadyBoost, especially when using, say, only 3 or 4 tests as a basis for conclusions. I may get around to looking for a difference on my laptop one day, but the laptop is not the best of my machines to use in testing Vista. I have been able to perceive no difference on the laptop though.

One problem with looking seriously for differences, for example, with and without ReadyBoost, is that the tests take too long to do the requisite number and not go nuts in the process. Another problem is that, one, or at least this one, can't even be sure which tests would be the best for use in assessing ReadyBoost effects.

[felix8406]

I wouldn't think that using ReadyBoost in your desktop PC will make any noticeable difference due to the amount of RAM you already have in it. I believe you mentioned many moons back that your PC has lots of RAM. You should notice the biggest difference when the PC has 1GB or less RAM.

[Znod]

Right, I still am betting that a statistically significant difference would not be discernable on machines even with 1 GB or less of RAM. Someone with such a machine needs to do some statistical tests for us. Do you have only 1 GB on your laptop?

[felix8406]

My laptop has the minimum 512MB (256x2). When I get a chance I'll test it with 512MB and 1GB USB drives.

[Znod]

Great. What test procedure will you use?

[felix8406]

I downloaded and ran PassMark, but it shows a few compatibilities issues and kept messing up my wireless connection. I'll mess around with it later and see what's going on.

[Znod]

Good plan. I have had no problem with PassMark. Hmmm???

[Snuffy]

Originally Posted by felix8406

I downloaded and ran PassMark, but it shows a few compatibilities issues and kept messing up my wireless connection. I'll mess around with it later and see what's going on.

I have had no problems with PassMark 6.1 and my wireless connection. on either Desktop or Laptop (Wireless-B BEFW11S4)LinkSys.