Performance on E6600 w/2Gb RAM - Help Please?!?!

[jonmcguffin]

I have what I thought was a pretty good performing system until I did a little benchmarking and it seems my system is about 30% slower than comparable systems with the same processor/memory configuration. My systme spec's according to CPU-Z.

Runnong on a Gigabyte DS3 i965 motherboard

Processor
Core 2 E6600 @ 2.40Ghz
Core Speed: 2400.1Mhz
Multiplier: x 9.0
Bus Speed: 266.7 Mhz
Rated FSB 1066.6Mhz

Memory (Channels# Dual)
2 - 1Gb Sticks = 2Gb Memory
Corsair PC2-6400 (400Mhz) XM2X1024-6400
Frequency: 400.0Mhz
FSBRAM: 2:3
CAS# Latency: 5.0 Clocks
RAS# to CAS# Delay: 5 clocks
RAS# Precharge: 5 clocks
Cycle Time (Tras): 18 clocks
(the rest is greyed)

Obviously I don't overclock my system and the systems I'm comparing too aren't overclocking either. The drive subsystem is not a problem. The performance issues are solely somehwere between CPU and Memory. It seems the issues are CPU/Memory related. The SPD of the memory in CPU reports the above matched settings too.

Any ideas as to what my problem might be? I really appreciate anybodys help..

Thanks!!

Jon

 

[Accord99]

Can you provide more details on the benchmarks, your scores and the comparable system scores?

 

[jonmcguffin]

I do video editing work and I'm comparing my results rendering a file to many other users in a forum who are posting their timed results. They are able to render the same particular file in about 240 - 260 seconds, while my machine is taking in the 300 second range. There are about 20 or so people who are all getting pretty similiar results with a similiar configuration and I'm the only guy getting the slow times.

I was wondering if perhaps my memory timings weren't correct or maybe the default multiplier wasn't right, etc.

 

[orion23]

Are their systems running @ stock speeds? or are they overclocking the C2D's?

You ram timings could probably be adjusted to 4-4-4-12

But what about your Hard Drive?

Are you encoding to the same drive as the OS? If you have a second drive, try using that one for the encoding!

 

[jonmcguffin]

All systems are running at stock speeds, that has been confirmed. My drive subsytem should not be an issue as this "task" is not drive intensive, it's CPU and RAM intensive though. I have 3 - WD hard drives, all 7200rpm 16Mb drives and two of them are in a RAID 0 configuration.

 

[lopri]

Try running Cinebench? It's very hard to diagnose your problem without more details and/or some numbers that are more or less standardized for easy comparisons.

http://www.maxon.net/pages/download/cinebench_e.html

And here is a reference point.

http://www.techreport.com/revi...q3/core2/index.x?pg=13

 

[Blain]

"I'm comparing my results rendering a file to many other users in a forum who are posting their timed results."
Comparing your system against other forum member benchmarks is pointless.
But that's what we love to do here at AnandTech...

"I do video editing work"
It's pointless to worry about "benchmarks" while not using your drive subsystem. It's simply not your real world.

"I was wondering if perhaps my memory timings weren't correct"
Your memory timings will have very little effect on your tests.

Bottom Line:
1. Make sure your components are running at their correct speeds.
2. Make sure your software is updated and running correctly.
3. Make sure your system isn't bloated with applications running in the background.
4. Be as happy with your rig as possible while lusting in your heart for others.

 

[HeatMiser]

Can you post a CPUz screenie?

 

[jonmcguffin]

The CPU-Z scores are all listed in my original post. There's no point to do a screen print and post because I stated them exactly as above.

Jon

 

[jonmcguffin]

Originally posted by: Blain
"I'm comparing my results rendering a file to many other users in a forum who are posting their timed results."
Comparing your system against other forum member benchmarks is pointless.
But that's what we love to do here at AnandTech...

Not sure if I follow you here Blain. I think if I want to make sure my system is performing up to or near a "standard" comparing my results versus others with very similiar configurations is in fact a very WISE thing to do, isn't it???

Originally posted by: Blain
"I do video editing work"
It's pointless to worry about "benchmarks" while not using your drive subsystem. It's simply not your real world.

Rendering times on this particular test doesn't involve any read disk activity at all. Only write activity which isn't really a factor either because I would be writing this file that is being created on a 5400rpm notebook drive and that would be plenty fast. We're talking about creating a 60gb file in about 4 minutes. Not a problem. This is why I was saying, don't worry about my disk drive setup, it's not important here and not a reason as to why my performance numbers seem to be so much slower than everybody elses.

Originally posted by: Blain
"I was wondering if perhaps my memory timings weren't correct"
Your memory timings will have very little effect on your tests.

If memory timings don't mean much, why does everybody make such a big deal about theM?

Originally posted by: Blain
Bottom Line:
1. Make sure your components are running at their correct speeds.
2. Make sure your software is updated and running correctly.
3. Make sure your system isn't bloated with applications running in the background.
4. Be as happy with your rig as possible while lusting in your heart for others.

I have the first three covered, won't be happy though until my render times match those with comparable systems.

Thanks for the reply though..

 

[jonmcguffin]

Thanks everybody for the interest. Here is the link to the original forumn in which I am talking about...

http://www.sonycreativesoftwar...mID=4&MessageID=529827

 

[yuchai]

Check temps and CPU frequency while the test is running? Perhaps heatsink is not making a proper contact and the CPU is throttling under load.

 

[jonmcguffin]

Originally posted by: yuchai
Check temps and CPU frequency while the test is running? Perhaps heatsink is not making a proper contact and the CPU is throttling under load.

Ooohh..

Excellent suggestion.. I hadn't thought of this but I did check the "power saving" settings in control panel and everything is turned off there so I woulnd't think the CPU would be doing any throttling activity.

 

[HeatMiser]

Originally posted by: jonmcguffin
The CPU-Z scores are all listed in my original post. There's no point to do a screen print and post because I stated them exactly as above.

Jon

edit: thought about it again...i won't bother posting. good luck figuring this out!

 

[jonmcguffin]

I must have offended you somehow HeatMiser.. It wasn't my intention and I'm sorry. I have no idea on how to post a screen cap and like I said, I typed everything that shows up there in my original post. Sorry..

Jon

 

[HeatMiser]

No worries. I should have been more specific about what I was asking for. What does the voltage say in CPUz? Also, does CPUz say 9x for multiplier, or is that your setting in the bios? It is possible that some of the intel throttling/power saving features are still turned on in your bios. This can manifest itself as a reduced multiplier or reduced voltage (which you can see in CPUz), both of which can reduce performance. Throttling shouldn't be active when you're under load, but it might be worth checking.

Also, is your corsair memory C4 or C5 (usually their part numbers will say that at the end, so 6400-C4, or in the case of C5, there may be no indication, so it will just say 6400). If it is C5, then you have the correct stock timings. If it is C4, you should be able to run them at 4-4-4-12 timings. That can make a difference, but I'm not sure 20-30% difference.

Do you have the latest bios for your mobo?