• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Comprehensive Speed List?

EvilPokemon

Weaksauce
Joined
Jul 1, 2009
Messages
84
I was wondering if there's a comprehensive speed list of hard drives? Not burst speeds, as those are useless :p Seek times.
Preferably ones that date back several years.

Looking around the closest I can find are a few threads on other forums with a few dozen member tests. Mostly in Raid0.
 
Last edited:
Um... http://www.storagereview.com/

BTW This is basically how the world works in terms of hard drive performance comparing all hard drives multiply list below by x where x is the number of drives in your RAID 0 array:
Green SATA = Slow
Non Green SATA = Slightly less slow
10K/ 15K = Less slow
Hybrid Drive = middle of the road
Low-end SSD = Fast
High-end SSD = Fastest

If you compare all latest gen drives, that is basically the breakdown. The difference between 120MB/s and 150MB/s on a spindle disk is nothing (due to seek times) versus a SSD. Add more drives, you can get 800MB/s and still not feel faster (in non-sequential tasks) than a SSD. Having lots of spindle disks and quite a few SSDs (now over a dozen SSDs) brand/ model of spindle disks informs me more about the RMA service quality.
 
I think I was unclear in my first post. Rather than absolute top speed, I'm looking for decent speed from older drives.
Edited my first post to be more clear.
 
Do you have these drives already and are trying to compare performance without benchmarking them, or are you looking to purchase 3 year old drives to use? If the former, why not benchmark? If the latter, with current 2TB drives <$100 and 1TB drives at around $55 it seems like buying older drives (especially just given age) would not be a winning proposition.
 
If the former, why not benchmark? If the latter, with current 2TB drives <$100 and 1TB drives at around $55 it seems like buying older drives (especially just given age) would not be a winning proposition.

I would not bother with 3 year old drives for this either.
1. The drives would have 1/2 of their expected (5 to 6 year) lifetime used up.
2. The drives would be significantly slower than current drives.
3. The drives will use almost 2 times the power of current generation drives.
 
Random access times for drives from 2008.

Naturally the Raptor/VelociRaptor are the fastest.
 
2. The drives would be significantly slower than current drives.

He was asking about seek times. Are you sure seek times have improved? I had the impression (without carefully looking at specs) that while sequential rates had improved, that seek times were either the same or a little slower for the higher density drives. I'm talking about 3.5" 5400 and 7200rpm drives, not 10K or 15K rpm drives.
 
Generally seek times have not changed much. However with higher density disks performance can be better for the same data set (if properly defragged) because the data will be closer together thus the seeks will be shorter. Remember shorter seeks are faster than full disk seeks.

This is one reason I have faster VM performance with a 100GB virtual disk on a 2TB green than a 320 or 750 GB 7200 RPM drive from 2008.
 
Generally seek times have not changed much. However with higher density disks performance can be better for the same data set (if properly defragged) because the data will be closer together thus the seeks will be shorter. Remember shorter seeks are faster than full disk seeks.

This is one reason I have faster VM performance with a 100GB virtual disk on a 2TB green than a 320 or 750 GB 7200 RPM drive from 2008.

Okay, I did not realize that you were assuming that the seeks would usually be within a smaller portion of the platter with the newer drives. I suppose that can be a decent assumption in some cases. As long as the metadata is either cached, or near the data being accessed.
 
Back
Top