Tom's Hardware > Forum > Storage > General Storage > Interpreting Hard Drive benchmarks

Interpreting Hard Drive benchmarks

Forum Storage : General Storage - Interpreting Hard Drive benchmarks

Tom's Hardware: Over 1.4 million members in 6 different countries available to answer all your high-tech questions. Sign up now! Its free!
Word :    Username :           
 

I have a few hard drives, two of which I'm testing to see which would be more appropriate for my OS. A WD Raptor from around 3-4 years ago, and a new WD 750GB drive. I'm having a hard time being convinced that the raptor that was $300+ at the time is now slower than the comparitvly cheap new drive. So I ran some benchmarks but I am unsure as to what to make of them. Bearing in mind that the 750 drive is my current drive for basically everything and that there is nothing on the old raptor, how important is seek time compared to transfer rate and burst rate, and which do you guys recomend for a faster startup?

http://i67.photobucket.com/albums/h302/scylerules/HDTune_Benchmark_WDC_WD7500AAKS-00R.png
http://i67.photobucket.com/albums/h302/scylerules/HDTune_Benchmark_WDC_WD740GD-00FLA0.png

Sponsored Links
Register or log in to remove.
- 0 +

The result you got from HD Tune only gives you sequential read speed, they are only useful for gauging speed of large file transfers. The easiest way to get actual desktop pattern performance results is to use PCMark Vantage's HDD tests, those will give you what you'll actually experience under typical usage.

------------------------------ GTL Ref Tweaking Guide - PM for detail
Brand is for the weak-minded, only product matters.
Resilient to marketing.
http://farm3.static.flickr.com/2520/3850950213_aea6bbbcfe_o.jpg
Reply to wuzy
- 0 +

Since you already have both drives, why don't you clone your main drive and see if you notice a difference with the raptor? Considering that it is an older model, and not the velociraptor, I think you will notice little difference. In the mean time, look at some application suite benchmarks for a number of drives from www.storagereview.com:
http://www.storagereview.com/php/b [...] h_sort.php

My take is that startup is not as important as what you do after startup. Then, fast random access is more important. Startup happens once, and you can minimize that hit by using sleep to ram instead of shutting down.

Reply to geofelt

Generally as said above, Speed in megabytes are only needed for sequential transfers or reading files that are perfectly fragment free

 

The access times are actually more important for most people as those are how fast the drive can get(Move the read/write heads) from one place to another. So even if you have a 90mb/ sec transfer if the head has to move to 30 spots you will loose time doing that. This can leave the raptor ahead.

 

The degree of performance will vary with what you are doing. A drive with faster access times(raptor) will work much faster when reading files that are fragmented, but if you can get all of your files most commonly used near the start(fast) of the drive and have little to no fragments you will get to take advantage of higher throughput.

 

There are times when the hard drive is NOT the bottleneck. Some games that use heavy compression for meshes/textures may actually tax the cpu enough to cause it to not be able to accept data as fast as the drive can give it. This is why some apps get no real boost in load times from a faster drive or better access times.

 

Last off, burst rates, they are a joke. Since a drives buffer is small (16-64MB) they will never hold enough data to be useful IMO.

 

I would almost say to use the new drive only because its newer and should last longer. If you do you can maximize performance by partitioning it to only what you need. This keeps all data on the start(fast) section on the drive and the section with the lowest(good) access times.

 

A short stroked raid volume better acess times, still, no raptor
http://img32.imageshack.us/img32/806/fastttt.jpg
The full volume. With windows and all the good in the first 500 gigs i get good speeds and fairly good access times
http://img197.imageshack.us/img197/7356/fullsized.jpg
images hosted on http://imageshack.us

 

Some info on short stroking can be found here
http://www.tomshardware.com/review [...] ,2157.html
Partitioning gives near the same results and is more easy. It does not waste space, the rest of the drive can still be used just without the advantage when its used at the same time as the main partition is working.

 

I hope this helps to clear it up a bit


Message edited by nukemaster on 06-22-2009 at 06:03:41 AM
------------------------------ http://i33.tinypic.com/sw3a5y.png
http://tinyurl.com/26uxxb - C2/i7 Temp? http://tinyurl.com/cj3pw - VGA power?
http://tinyurl.com/5v55wk - C2 Mem performance? http://tinyurl.com/6pmbke - SLI/Xfire?
http://tinyurl.com/yfmxdc9 - Part Guide?
Reply to nukemaster
Tom's Hardware > Forum > Storage > General Storage > Interpreting Hard Drive benchmarks
Go to:

There are 1288 identified and unidentified users. To see the list of identified users, Click here.

Sponsored links
  • Ask the community now
  • Publish
Ad
They won a badge
Join us in greeting them