DougLite
10-17-2005, 10:58 AM
Does anybody remember the 8.0ms seek time claim for Barracuda 7200.8, when it was first launched? That was a figure that no one got anywhere close to in a review. SR noted a figure closer to 11ms, even after subtracting the 4.16ms latency of a 7200RPM spindle. Well, Seagate has decided to revise the seek time figure, ratcheting it up 3ms to 11ms :eek: Here's the data (http://www.seagate.com/cda/products/discsales/marketing/detail/1,1081,630,00.html) on the flagship 400GB drive.
I know that in the past, seek time claims have been little better than rough estimates. It was not uncommon for actual measured performance to fall short of the maker's claim by half a millisecond. However, missing an estimate by 3ms is unaccpetable. I am glad that Seagate has corrected this...but it should never have happened in the first place. We all knew as soon as the reviews came out that 7200.8 has seek performance issues. This also explains 7200.8's much better showing in the SR High-End drive mark (with shorter seeks and more/longer linear transfers that take advantage of the drives high density) than the more random (yet still localized of course) Office DriveMark.
Hopefully, Seagate has corrected this in 7200.9. However, I doubt it. I would not be at all suprised if 7200.9 delivers sub-far seek performance just like its predecessor. Seagate needs to go back to the drawing board and correct this deficiency immediately, if they have not already done so.
I know that in the past, seek time claims have been little better than rough estimates. It was not uncommon for actual measured performance to fall short of the maker's claim by half a millisecond. However, missing an estimate by 3ms is unaccpetable. I am glad that Seagate has corrected this...but it should never have happened in the first place. We all knew as soon as the reviews came out that 7200.8 has seek performance issues. This also explains 7200.8's much better showing in the SR High-End drive mark (with shorter seeks and more/longer linear transfers that take advantage of the drives high density) than the more random (yet still localized of course) Office DriveMark.
Hopefully, Seagate has corrected this in 7200.9. However, I doubt it. I would not be at all suprised if 7200.9 delivers sub-far seek performance just like its predecessor. Seagate needs to go back to the drawing board and correct this deficiency immediately, if they have not already done so.