Benchmarking - before and after tuneups

[sorcerer]

I searched the forum but only came up with two references - one to CustomPC Benchmark, which seems old (Google comes up with hits that refer to 2009/2010, and even they refer to the 2007 version of the software), and also to Geekbench, which is $79.99 for the standalone version - so I'm not inclined to use either at the moment.

I just want to get a 'before and after' measurement for my own benefit (maybe even show customers if they ask) when a slow, sluggish machine comes in and after performing a tune-up.

Anyone know of any free (or at least cheaper than $79.99), easy to use (as in not having to set a gazillion parameters), modern benchmarking software out there?

 

[SilverLeaf]

Honestly, I've never bothered with it. Most benchmark programs that I am aware of test things like CPU performance, memory throughput, graphics performance, disk read/write speeds, etc. These are irrelevant to my tune-up process.

If, after I perform a tuneup, the O/S isn't noticeably faster; either I didn't do a good job, or it didn't need a tuneup in the first place. Hardware issues are ruled out before hand during the diagnostic process.

 

[sorcerer]

Honestly, I've never bothered with it. Most benchmark programs that I am aware of test things like CPU performance, memory throughput, graphics performance, disk read/write speeds, etc. These are irrelevant to my tune-up process.

If, after I perform a tuneup, the O/S isn't noticeably faster; either I didn't do a good job, or it didn't need a tuneup in the first place. Hardware issues are ruled out before hand during the diagnostic process.

To an extent, I agree, but I would like to see it quantified sometimes. "Noticeably faster" is rather subjective, just like some chefs class a steak as 'well done' even when there's still blood coming out of it but others don't; just like I could go into three different indian restaurants and order the same meal, but the 'hotness' of the curry would range from mild to head-blowing depending on which chef cooked it, even though they're all meant to be the same, so I want to actually 'see' the difference.

 

[SilverLeaf]

To an extent, I agree, but I would like to see it quantified sometimes. "Noticeably faster" is rather subjective, just like some chefs class a steak as 'well done' even when there's still blood coming out of it but others don't; just like I could go into three different indian restaurants and order the same meal, but the 'hotness' of the curry would range from mild to head-blowing depending on which chef cooked it, even though they're all meant to be the same, so I want to actually 'see' the difference.

What parameters would you propose to measure....other than boot time and maybe browser startup time? What do you do during a TuneUp that would cause hardware benchmarks to be relevant?

 

[CLC]

What parameters would you propose to measure....other than boot time and maybe browser startup time? What do you do during a TuneUp that would cause hardware benchmarks to be relevant?

+1 It's not going to speed up your hardware.

 

[sorcerer]

What parameters would you propose to measure....other than boot time and maybe browser startup time? What do you do during a TuneUp that would cause hardware benchmarks to be relevant?

To be honest, I haven't got a clue. I'm only a beginner in the ways of computer repairs which is why I'm asking, but as you said earlier, after a tune-up a machine should be "noticably faster" so if that's the case then some parameter or other should be measurable. Something has changed for the better in order to make the machine faster, so what is it and can it be measured by something?

 

[SilverLeaf]

To be honest, I haven't got a clue. I'm only a beginner in the ways of computer repairs which is why I'm asking, but as you said earlier, after a tune-up a machine should be "noticably faster" so if that's the case then some parameter or other should be measurable. Something has changed for the better in order to make the machine faster, so what is it and can it be measured by something?

It's all very subjective. Think about why a customer would bring in a computer for a tuneup in the first place. Usually, its something along the lines of "its been running really slow...and it takes forever to load Facebook". It's been my experience that it has to get pretty bad before they go through the trouble to call for service.

So, (after you rule out hardware and malware issues), you clean up autoruns, toolbars, BHO's, uninstall crapware, check services, and so on....whatever is part of your tuneup service. Inevitably, this will have a dramatic impact on how responsive the UI is. Boot up time is reduced, the desktop becomes available and usable quicker, applications start faster, and so on. Sure, you could "measure" the improvement if you feel the need. But, virtually every time the difference is going to be noticeable. But again, its the O/S that's more responsive and unencumbered by all the garbage you removed. So, running a hardware benchmarking suite will be of little value. The important part is that your customer notices the difference. If you really wanted to quantify the results, I suppose you could measure things like boot time, time to desktop, IE startup time, and such.

 

[sorcerer]

It's all very subjective.

That's exactly my point!

Sure, you could "measure" the improvement if you feel the need.

Again, exactly what I want and exactly the reason why I posed the question. If we, as humans, can percieve an improvement in the running speed of the machine, surely there should be some software out there that can make it tangible and put it into figures for us?

What actually made me think of this is that I've just built a new desktop machine for someone and, although it's got 8GB of DDR3 RAM and lots of other fast, modern stuff, it doesn't 'feel' as fast as my 3-year old machine that's only got 4GB of DDR2 RAM and is a much lower spec machine. I just wanted to compare the two using something a bit more scientific than 'my perception' of things.

 

[ComputerRepairTech]

I dunno I suppose you could try passmark it may help you determine a computerA to computerB comparison but as far as before and after tuneup I think thats a waste of your time. Sure theres quite a few cases that I can think of where it would show a difference such as processes taking up excessive resources, bad hard drive performance due dma/driver issue, etc.

However there are also tons of situations when improvements wouldnt show up. Think about how the client is going to feel when they see this score that shows no improvement.

 

[HFultzjr]

What I usually do is show them the boot-up times, before and after. I do it manually, but I'm sure there is a frees version of something out there. I then show them via task manager the number of processes running, before and after, the amount of ram being used before and after. On very slow running machines, I can usually show positive results in boot time, programs loaded at start, and memory usage. You could also show "disk activity" before and after. Then a simple 5 minute "test drive" and you have shown your results.

The biggest performance issues, other than malware, seems to be start-up programs, not enough ram, and something constantly pounding the hard drive. Of course, rule out hardware issues.