SanDisk SSD significantly underperforming standard benchmarks

[ssdtester]

I installed a new SanDisk Ultra II 480GB SSD (SDSSDHII480G), and I ran two PassMark 8.0 tests (2,336; 2,558) -- my average score was 2,447. According to harddrivebenchmark.net the overall average score for the same drive is 3,908. That's 60% higher than what I'm getting.

I'm running on a Dell Precision T7600 with an Intel Xeon E5-2687W and 64GB of DDR3 RAM. I reset the computer right before running the tests to prevent the Windows cache from interfering. The firmware is up to date (X31200RL); write caching is enabled; and it's AHCI. I also tried CrystalDiskMark and the results were way below the manufacturer claims -- that's not surprising though.

How do I know if I got a bad drive, or if is something not configured correctly? Anything else I should try?

 

[SBMfromLA]

When you installed the drive... did you make sure you connected it to a SATA 3 port instead of a SATA 2?

 

[ssdtester]

When you installed the drive... did you make sure you connected it to a SATA 3 port instead of a SATA 2?

Yes it is SATA 3. Source: I have the SanDisk SSD Dashboard installed.

I'm currently updating the BIOS to see if that has any effect.

 

[ssdtester]

I originally posted an update to the question, but I think it's the closest thing I'm going to get to an answer.

---

I updated the BIOS from A09 to A12. While looking at the BIOS settings more closely, and doing a little research, I learned that while the drive is connected to a SATA 3 port inside the case, the connection then goes through a hard drive controller (maybe the PERC H710?) into a SAS port on the motherboard. Because most people recommend SATA + AHCI for SSDs, it seemed likely that this could be the issue.

This had me stuck for a while, but I noticed that there are two SATA connections on the motherboard that go to the disk drive slots (DVD drive, not hard drive). I had to remove a couple of screws to get at the connectors, but I was able to plug the SSD into the "pure" SATA connection. I couldn't mount the hard drive that way, but since it doesn't have any moving parts, I gingerly set it inside the disk drive slot and booted up.

I re-ran the PassMark disk test and... 2,895. It's higher, but the average score being 3,908, it's still a ways from what it should be. The SanDisk SSD dashboard claims that it's still a SATA 3 (6 Gb/s) connection, but it's lied to me about that once, so maybe it's wrong again; I don't know.

I also re-ran the CrystalDiskMark test and the sequential read/write speeds were significantly faster. While I'm still not getting close to what I expected out of this drive, it doesn't seem likely at this point that it's a bad drive.
Here's the difference: http://www.screencast.com/t/gGtanheqGR

Please post any other solutions if there's something else I should try!

 

[Dereck47]

Your SSD's advertised speeds are "up to" 550MB/s Read and 500MB/s Write.
http://kb.sandisk.com/app/answers/detail/a_id/16376/~/sandisk-ultra-ii-ssd-specifications

Your "SSD through SATA Connection" benchmark results show that you are getting 81.6% of advertised Reads and 92.5% of advertised Writes.

Probably best to leave it connected where it is; just buy a longer SATA cable so that you can mount the drive properly.

Are you using a SAS to SATA cable to connect the drive to the PERC controller?
If so, double-check that it's rated at 6Gb/s. Maybe that's why your SAS benchmark speeds are lower.