Diagnostic / test software?

[SnakeCharmer]

My PC just doesn't seem to be running as good as it should, and honestly hasn't since I've owned it.

Is there any diagnostic or test software or site that I can go to or buy that can help pinpoint where my problem might be?

Just for reference, I just ran 3dMark2006 and got a score of 5532 3DMarks. 
Processor: AMD Athlon 64 X2 5200+ Windsor 2.6GHz Socket AM2 89W Dual-Core Processor
Memory: Kingston ValueRAM 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400) Dual Channel Kit Desktop Memory
Vid card: EVGA 512-P2-N635-AR GeForce 7950GT 512MB 256-bit GDDR3 PCI Express x16 HDCP Ready SLI Supported KO HDCP Video Card
HD: SATAII Western Digital 3.0 gb/s 250 gb (I think that's the one)
Motherboard: GIGABYTE GA-M55SLI-S4 AM2 NVIDIA nForce4 SLI ATX AMD Motherboard
PSU:  Antec SmartPower 2.0 SP-500 ATX12V 500W Power Supply

My old system smoked it using the same program at 8188 3DMarks.
Processor: AMD Athlon 2400+ XP single core
Memory: 2048MB RAM
Vid card: ATI V700 Pro 256 MB AGP

I've been tempted to wipe the HD and reinstall XP to see if that would help...

[Trippy]

Did you save a detailed listing of your old benchmark? How do the CPU benchmarks compare vs. the GPU benchmarks on the new vs old computers? I.e. which part seems to be dragging down the score?

Download PC Mark 2005 and then go to here:

http://www23.tomshardware.com/cpu_2007.html

and see if your scores are close.

[Engels]

Couple of questions first:

What AV are you running?
How much data on your HD?
Last time you defragged?
Have you run any anti-spyware programs, such as Adaware or Windows Defender?

If you're convinced nothing above addesses the issue, download and burn a copy of Ultimate Boot CD. Boot to that CD, and there you'll find any number of disgnostic programs, from CPU and Memory tests to hard disk diagnostics. For a fast ram test, I recommend Memtest86.

[Morfiend]

check download.com for sisoft sandra, its a decent diagnostic tool.

[SnakeCharmer]

Did you save a detailed listing of your old benchmark? How do the CPU benchmarks compare vs. the GPU benchmarks on the new vs old computers? I.e. which part seems to be dragging down the score?

Download PC Mark 2005 and then go to here:

http://www23.tomshardware.com/cpu_2007.html

and see if your scores are close.

Will the basic (free version) be good enough, or should I put the cash out for the full version?

What AV are you running?

Not sure I know what you mean...  Doh. Just hit me.  AVG is my antivirus.  Completely up to date.  Scans every 2 days.

How much data on your HD?

As in how much free space?

C: is 25 GB partition, 16.6 GB used, 7.4 GB unused.  Basically all that is on the C drive is Windows XP.  Everything else gets put on D:
D: is 225 GB partition, 70.1 GB used, 138.0 GB unused.  Hrm.  Not sure where the remaining 12 GB is....

Last time you defragged?

Couple of weeks ago.

Have you run any anti-spyware programs, such as Adaware or Windows Defender?

I run AdAware and Spybot about once every two weeks.

If you're convinced nothing above addesses the issue, download and burn a copy of Ultimate Boot CD. Boot to that CD, and there you'll find any number of disgnostic programs, from CPU and Memory tests to hard disk diagnostics. For a fast ram test, I recommend Memtest86.

Thanks for the input, I'll give it a shot.

[Trippy]

Did you save a detailed listing of your old benchmark? How do the CPU benchmarks compare vs. the GPU benchmarks on the new vs old computers? I.e. which part seems to be dragging down the score?

Download PC Mark 2005 and then go to here:

http://www23.tomshardware.com/cpu_2007.html

and see if your scores are close.

Will the basic (free version) be good enough, or should I put the cash out for the full version?

It's been a while since I've looked at PC Mark but the free one should be okay to give you an overall number (like 3DMark).

[SnakeCharmer]

OK.  Will do

Hrm.  Thought I'd put up a screenshot of system processes running.  Oh well.

[SnakeCharmer]

Here's the results from PC Mark 2005:

Detailed Test Results

System Test Suite
HDD - XP Startup   9.63 MB/s
Physics and 3D   180.38 FPS
Transparent Windows   Test failed
3D - Pixel Shader   415.63 FPS
Web Page Rendering   3.17 Pages/s
File Decryption   45.17 MB/s
Graphics Memory - 64 Lines   2310.58 FPS
HDD - General Usage   6.41 MB/s
Multithreaded Test 1 / Audio Compression   N/A
Multithreaded Test 1 / Video Encoding   Test failed
Multithreaded Test 2 / Text Edit   146.59 Pages/s
Multithreaded Test 2 / Image Decompression   29.66 MPixels/s
Multithreaded Test 3 / File Compression   5.46 MB/s
Multithreaded Test 3 / File Encryption   24.56 MB/s
Multithreaded Test 3 / HDD - Virus Scan   67.92 MB/s
Multithreaded Test 3 / Memory Latency - Random 16 MB   11.6 MAccesses/s

[Trippy]

Does it give you a composite CPU benchmark score like what's shown on the TH page?

[SnakeCharmer]

No, because it 'failed' the Transparent Windows test - whatever that is.

[Trippy]

Then use SiSoftware Sandar like Morfiend suggested and run those CPU benchmarks.

[Ookii]

I would also be concerned about how many and what kind of processes you are running in the background.

Doing 'msconfig' and checking the startup applications also helps a bunch as well as looking at all your services and seeing what's running.

[SnakeCharmer]

Here's a screenshot of the processes running as per the task manager:

Background processes

I can't post the msconfig until I get home.

Edited to remove massive screenshot...Click the link

[bhodi]

I see a lot of "unnecessary" processes there that you could pair down if you realllly wanted.

NM* (nero does not need to be memory resident. when you want to burn a cd, run the software)
antivirus AVG stuff (you don't need memory resident virus scanners unless you're really paranoid)
aawservice adaware stuff (again, doesn't need to be memory resident -- scan periodically instead)
ituneshelper (this checks for updates, keep the rest but ditch this one)
LCD* (it's a monitor. What do you need memory resident programs to do, exactly? How often do you change resolution? Wasted cycles here.)

This is a way better program; systernals is an outstanding company with a number of great tools and they got bought my microsoft a few years ago. Also use their autoruns tool to remove all that crap from your startup.

[Ookii]

Well you do have a ton of processes running (51!), but none of them look that bad, most are just little programs for some of your peripherals/random hardware (like LCDMon.exe LCDPop.exe LCDClock.exe LCDMedia.exe, your keyboard seriously needs 4 different processes running?).

I would suggest trimming the fat persay and getting your processes below 40, half that shit you don't really need.  I would also only run AVG on-demand rather than all the time (some people might argue differently, it worked for me for a decade) whereas it doesn't slow your computer down as much as Norton, it will still slow you down quite a bit (depending on how you have it set).

Then there's all that defragging nonsense which you did a couple months ago, I'd see what Diskeeper Pro says about your HD as it seems about time to do that again.  You could also try downloading Tune-Up Utilities and have it scan around and find all sorts of things that can be fixed in all sorts of secret places on your computer, which amounts to mainly the registery.  There are also most likely countless other programs/better solutions to the whole 'computer cleanup' thing that I am missing.  And then again maybe it's a hardware thing and that 500W PSU isn't putting out enough juice to run your Video Card effectively causing your score in a 3D benchmarking program to be low.

It's troubleshooting time!

EDIT: Bhodi types with more intensity.

[Trippy]

For the processes it doesn't matter how many are running. What matters are how much memory they are using (if they take you into swapping territory everything is going to feel sluggish) and how much CPU time they are using up.

[Engels]

I'm wondering if those failed categories indicate a video card failure.

[Ookii]

For the processes it doesn't matter how many are running. What matters are how much memory they are using (if they take you into swapping territory everything is going to feel sluggish) and how much CPU time they are using up.

And we are saying the same thing, less processes equals a faster computer.

If you have to explain a statement like that it proves that the user does not have the adequate knowledge to troubleshoot their own problem and is not trusted to make that sort of a judgement call as to which processes are the nasty ones.  Thus the statement shouldn't even be said at all and the user should be advised to disable any processes they aren't using.  This is also a good practice from a practical point of view as even processes that use no memory nor CPU cycles would be silly to keep running if they don't do anything useful in the first place.

[Trippy]

For the processes it doesn't matter how many are running. What matters are how much memory they are using (if they take you into swapping territory everything is going to feel sluggish) and how much CPU time they are using up.

And we are saying the same thing, less processes equals a faster computer.

No, we are not.

If you have to explain a statement like that it proves that the user does not have the adequate knowledge to troubleshoot their own problem and is not trusted to make that sort of a judgement call as to which processes are the nasty ones.  Thus the statement shouldn't even be said at all and the user should be advised to disable any processes they aren't using.  This is also a good practice from a practical point of view as even processes that use no memory nor CPU cycles would be silly to keep running if they don't do anything useful in the first place.

Diabling processes willy-nilly is the wrong way to go about figuring out why his computer seems to be benchmarking slowly.

[Trippy]

I'm wondering if those failed categories indicate a video card failure.

More likely a driver issue, assuming it's a problem at all.

[Krakrok]

HDD - General Usage   6.41 MB/s

This is total shit I think. Should be ~36MB/s.

Run PC Pitstop and see what it says for your hard drive speed.

[Trippy]

36 MB/s is sustained transfer rate, not general random access disk usage:

General Hard Disk Drive Usage: This trace contains disk activities from using several common applications.
These are:
- Opening a Microsoft® Word document, performing grammar check, saving and closing
- Compression and decompression using Winzip
- Encrypting and decrypting a file using PowerCrypt
- Scanning files for viruses using F-Secure® Antivirus™.
- Playing an MP3 file with Winamp
- Playing a WAV file with Winamp
- Playing a DivX video using DivX codec and Windows® Media Player
- Playing a WMV video file using Windows® Media Player
- Viewing pictures using Windows® Picture Viewer
- Browsing the internet using Microsoft® Internet Explorer
- Loading, playing and exiting a game using Ubisoft™ Tom Clancy’s Ghost Recon

[Trippy]

I'm running at 1920x1200 on a 24 inch widescreen (damn you, Schild, damn you to hell!!! - from my wife), and it seems to be taxing it pretty hard on CoD4 and a couple other games.  The native on the 24" is 1920x1200, and as I've always understood it that it was best to run the native resolution of the monitor you're using for best image quality.

I think, though, that something is just up with my system somewhere.  5400 3dmarks on 3dmark06 seems awfully low, and after a format, reinstall, updated drivers and everything it's not scoring any better.  I'm thinking about getting a new 8800GTS and seeing what happens, and if I'm not satisfied, build a new quad core.

I'd blame your motherboard (settings) before I'd blame your video card.

[SnakeCharmer]

I'd blame your motherboard (settings) before I'd blame your video card.

Oo

Elaborate?

FWIW, this is my mobo.

[Trippy]

Didn't we already discuss this in another thread? Did you ever run those SiSandra CPU benchmarks?

[SnakeCharmer]

Yeah, sort of.  Merge/split or whatever if you want.  Downloading the Sandra thing.  See what happens.

[Trippy]

If your CPU tests look okay then almost certainly your problem is you either misread your old system's benchmark or it's not the same benchmark, which could happen since the X700 (I'm assuming that's what you had) doesn't support SM 3.0 while the 7950 does. In other words your old system did not have better graphics performance despite what the number said.

You need to research your current 3Dmark2006 score and see how it compares to comparable systems. Here's an example:

http://forum.pcvsconsole.com/viewthread.php?tid=17997&page=8

If you scroll down to this system: Athlon X2 4800+, Geforce 8800 GTX, 2GB Ram

you can see that he got a score comparable to your old system which means one of you is wrong. He has a slower CPU but a much faster GPU so the 5.5K you got may be perfectly reasonable.

[Nerf]

I get right around 5k with an 8600GT, 2.66ghz core2duo, and 4gb of ram.  If I bump my CPU up to 3.45ghz, my score goes up about 100 points, 3dmark doesn't give a shit about cpu past a certain point.