Tea said:
Time and others with the same point of view - if it's really 6000 RPM, I'll eat one. Fair dinkum, does anyone think a drive manufacturer would under report specs? And how much extra would they have to spend on non-standard parts to make that non-standard spindle speed? Long and the short of it, I'll bet you a slab of Cascade Pale Ale to a stubbie of XXXX on it.Are you on?
See Buck's post. 6000rpm parts may indeed exist. But hey, don't shoot me, I'm only the messenger.
From
Storage Review:
AnalyzeDisk pegs the WD800AB's access time at 14.1 milliseconds, a score quite good for a 5400 RPM drive. Subtracting 5.6 milliseconds to account for the average rotational latency of a 5400 RPM spindle speed yields a measured seek time of just 8.5 milliseconds, a figure that handily beats WD's 9.5 ms claim. At 14.9 milliseconds, average write access time is equally impressive.
This brings up an interesting angle on Western Digital's recent "5400 RPM" drives. A while back in a discussion thread that has unfortunately been lost, several readers pointed out that the 60-gig WD600AB featured a "nominal" spindle speed of 5400 RPM and rotation latency that was not 5.56 milliseconds as one would expect from such a drive but rather "nominally" 5.0 milliseconds, a figure that would represent a 6000 RPM unit. Assuming a rotational latency of 5.0 milliseconds yields a measured seek time of 9.1 ms, still well below the 9.5 ms claim. Any way you slice it, the drive over-performs in this particular test.
I couldn't put it better than Eugene, but I will note that WD is now claiming 9mS rather than 9.5, which would be nearly bang on ...
Seriously, you know the shenanigans WD plays with variants within the same external model number. If I was cynical, I might suggest they submitted 6000rpm versions for review.
But no, I'm not prepared to risk a carton of Cascade on wild speculation. Maybe a carton of XXXX, but I might have trouble with takers ...
it wouldn't be the first time that 5400 RPM drives have had better access times than a Seagate 7200. In fact it's routine for the better 5400s to beat Cuda ATAs.
If you insist on going out on a limb, it's too tempting to try out that new chainsaw.
Again, from Storage Review (pity we can't just link this stuff):
Maker and Model RPM Latency Seek Access Net
Maxtor D740X: 7200 4.2mS 8.5mS 12.3mS 8.1mS
Samsung P40: 7200 4.2mS 8.9mS 13.6mS 9.4mS
Barracuda IV : 7200 4.2mS 9.5mS 13.9mS 9.7mS
WD WD800AB: 5400 5.0mS 9.0mS 14.1mS 9.1mS
Samsung V30: 5400 5.6mS 9.0mS 14.6mS 9.0mS
Maxtor D540X: 5400 5.5mS 9.6mS 14.8mS 9.3mS
RPM, Latency and Seek are as quoted by the manufacturer.
Access is as measured, and Net is the derived read seek time.
As you can see, most measurements are within 3% of the claimed figures, with the exception of the D740X (Maxtor now claims sub 8mS seeks on their site, but the datasheet still says 8.5), and the Samsung P40.
Comparing the Barracuda IV with the Maxtor D740X is interesting. Results are split four each across the eight SR tests. The Maxtor does best in the simulation of a lightly loaded server (15% ahead, although the gap closes as the load increases), and the Seagate stars in Eugene's boot drive tests, with a lead of 18%.
What does all this mean? No idea.
But it is not true to say that 5400rpm drives offer better access times than the Barracuda IV. Neither is it true to claim that Seagate wildly exaggerates their specifications. And finally, the Seagate
is competitive with its peers.
