I have a Western Digital Sata-1 Raptor 36 GB with 10000 rpm and

a Samsung Sata-2 HD501LJ 500 GB with 7200 rpm.

Which one is faster if I use the harddisk only as a System disk running the OS on it?


The 10k RPM HDD is likely to be faster. Mechanical Hard disks don't saturate SATA bandwidths as of now, so SATA1/SATA2 is irrelevant and it comes down to 7.2k/ 10k RPM HDDs, in which case the 10k RPMs are faster.


Use HD tach to benchmark both drives and compare them, but that raptor is old enough that a current generation drive is likely faster. WD's 3 platter 1GB Black edition drives performed at about the same level as an older 150GB raptor. I'm not sure if your seagate is 1x500GB or 2x250BG platters, but even the latter and not configured at the factory for performance is probably faster than a raptor that's 2 generations older than the one that the WD drive was beating. Even though it spins faster the lower platter density bottlenecks the transfer rate badly.


I would say the 10K RPM is faster. SATA1 transfers data up to 1.5Gbps. SATA2 allows for burst to be transfered up to 3.0Gbps, but after the burst I've heard many drive can't sustain that rate and are no faster than SATA1.

The 10K RPM will help with the rotational latency seek times and getting the data faster.

I think the only place the SATA2 drive will win is if you can get the data and burst it faster than the 10K can get it can transfer it via SATA1.

Also, if the drive is going to be used over the network, like in a NAS. The only use for hard drive speed and spindle speed is to give you an idea how fast the NAS could scan/rebuild the drive. The network speed is so much slower than a direct connect SATA I don't now if it really matters. For example, my ReadyNAS NV+ came with SATA1 drive and is only wired for SATA1. I don't think that matters too much in the day to day because SATA1 (1.5Gbps) is still faster than Gigabit Ethernet (1Gbps) even without TCPIP overhead.

  • FWIW, the 10K RPM does NOT help with seek times. Seek times would probably be about the same. A 10K drive reduces the rotational latency to read a sector on a track once the seek to that track has completed. It should also improve the throughput when reading sectors on the same track. No? – irrational John Sep 26 '10 at 16:28
  • Sorry for using the wrong term. What I loosely meaning by seek time was the time the drive took to get to a new track's sector it needed. Which is correct it doesn't make the head get to the new track faster but the rotation latency would be less and sector it needed would arrive under the head sooner. I should have used used Rotation Latency. Thanks – Scott McClenning Sep 26 '10 at 16:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.