[snowboarder33]

So I recently purchased the hardware for my eventual HTPC build and wanted to test it out to see if it's up to par.

Since passmark scores seem to be widely used and allow for easy comparability, I decided to use it as the benchmark.

First the CPU, which is the Intel Celeron G540 2.5GHz. I thought this component would be the most straight forward since I believe there shouldn't be much variation within the same model.



Not the case - it looks like my CPU is performing at 50% of the average!

Next was the GPU - the Radeon HD 6570. This test I know can produce many variances since there are so many different variations of the same card, but I thought my card could get somewhere near the average.



Again, not the case - performance is near 50% of the average. Those results are all with the Radeon HD 6570 card.

How can I be getting such poor results? I thought at least my CPU would have no problem comparing with the average, but it's not even close. Is it, by chance, that I just purchased all faulty hardware? Why must I suck at this/should I just give up?

[Bunto Skiffler]

just off the top i would check the mb fsb in bios. also check the report with another tool like everest.

[snowboarder33]

Okay, I downloaded Everest, what exactly am I looking for?

One thing I noticed right away is that after I ran the Everest, it shows my MOBO as name and chipset unknown. Same thing with BIOS.

So would that signal there is something wrong with my MOBO's driver?

[outatouch0]

The test systems in reviews are going to optimized (high end component) systems. They may be water cooled, etc. Results will vary from setup to setup.
Make sure you have your BIOS and drivers etc fully up to date: don't use the disks that come with the components. For more on this you can google 'tctc' (tweak guides tweaking companion) to learn how to optimize your system.

[snowboarder33]

I installed this driver here, from the Asrock website (http://www.asrock.com/mb/Intel/H77M/?cat...&os=Win764):


Wouldn't that be the most up to date?

---

The test systems in reviews are going to optimized (high end component) systems. They may be water cooled, etc. Results will vary from setup to setup.
Make sure you have your BIOS and drivers etc fully up to date: don't use the disks that come with the components. For more on this you can google 'tctc' (tweak guides tweaking companion) to learn how to optimize your system.

Pretty sure these results are just user submitted, so unlikely it would be a test system. Also, I doubt a few tweaks to the G540 could increase the performance by 50%, I could be wrong though.

[snowboarder33]

Windows' System Information says:

BIOS Version/Date: American Megatrends Inc. P1.00, 3/6/12
SMBIOS Version: 2.7

[snowboarder33]

Oh man...

So I downloaded that INF driver and reran the CPU test and got 2614.8! Now that seems a bit more usual.

I was hoping for the GPU score to increase too, but unfortunately it remained at around 520. I don't know how much the decreased 1000MHz memory clock speed of this model affects the performance, but that score is well below the GT 430 scores, which I though it would do better against. Would it be reasonable for a GeForce 430 @ 1333MHz to outperform a Radeon 6570 @ 1000MHz?

[Bunto Skiffler]

Here's a video I did about Everest reports and such, something to start with..

[snowboarder33]

Thanks for the video. I would like see details of my system while watching it though, but my motherboard's identity still isn't showing up in the Everest report, do you know why that would be?

Hmm PC Wizard detects the MOBO, maybe it's just something with everest

[Bunto Skiffler]

Ok, don't worry about the mb not being detected. All that means it that is a newer version than the Everest you are using.

Steps:
====
1. In Everest go to "Reports" and execute a HW report and save it as a TXT.
2. Open that up and scroll down to "Motherboard"
3. Let's see what the 'Bandwidth' is there. Bigger the better.

See what we got here.