Which Barracuda .10 is faster?

[Justintime]

Yes, i got 250s as well and i can't tell the difference.

 

[ByteJuggler]

yes, except some have more platters and heads than others dont they

http://www.trustedreviews.com/article.aspx?art=3042

look at the 750gb compared to the 320gb - twice the platters and read heads.

Yes but that makes nowt difference since generally only one head is active reading or writing at a time. That is why having more head/platters doesn't normally double/triple/quadruple a drive's throughput rate. Generally, one sector/block is written in it's entirety on a single platter. (To put it differently, HDD's don't normally internally "stripe" accross multiple platters.)

That is why the only figure you need to look at to get an idea of the sustained transfer rate for a hard drive is the areal density on a single platter, and why all the 7200.10 drives have nearly identical linear sustained transfer rates (since they reportedly have the same areal density for the entire range.)

 

[Messiah Khan]

I wonder why drives don't offer the option to 'stripe' across multiple platters, or at least read from different platters at the same time. Surely it would mostly only require independant motors on the read head.

 

[james.miller]

Yes but that makes nowt difference since generally only one head is active reading or writing at a time. That is why having more head/platters doesn't normally double/triple/quadruple a drive's throughput rate. Generally, one sector/block is written in it's entirety on a single platter. (To put it differently, HDD's don't normally internally "stripe" accross multiple platters.)

That is why the only figure you need to look at to get an idea of the sustained transfer rate for a hard drive is the areal density on a single platter, and why all the 7200.10 drives have nearly identical linear sustained transfer rates (since they reportedly have the same areal density for the entire range.)

why dont you look at the benchmarks and reconcider? its there in black and white, afterall. note how some drives using short stroked platters changes the performance of the drives? number of platters does have an impact. The use of those platters does also.

http://forums.storagereview.net/index.php?showtopic=22660

For the record, i never said anything about platters acting like internal raid. I never mentioned striping, doubling of performance or any such nonsence. I only said the performance is affected.

 

[ByteJuggler]

why dont you look at the benchmarks and reconcider? its there in black and white, afterall. note how some drives using short stroked platters changes the performance of the drives? number of platters does have an impact. The use of those platters does also.

http://forums.storagereview.net/index.php?showtopic=22660

For the record, i never said anything about platters acting like internal raid. I never mentioned striping, doubling of performance or any such nonsence. I only said the performance is affected.

Allright, but that (that multiple platters neccesarily substantially changing a drives performance) was what you seemed to imply/get at. I apologise I misread your comment. For the record (since we're putting things on record ) I also never meant to imply that a drives linear transfer performance is neccesarily precisely and unambiguously defined solely by areal density (nor that it is constant across the disk for that matter, in case I gave that impression), since as I'm sure you'll agree there are many factors which affect what actual speed you'll observe in the real world from a hard drive, including whether it's short stroked or not. However, it (coupled with rotational speed) remains a very good indicator for fundamentally what sustained transfer rates a drive should be capable of.

Seagate has stated that the 7200.10 range (somewhat unusually) supposedly all use the same areal density for the platters used in all the 7200.10 disks. In their official product brochure for the range they also state the official maximum sustained transfer rate _for the entire range_ (by implication for every drive) to be 78Mb/sec, which implies most of the drives will have very similar performance. And this was essentially all that I was originally trying to point out. Benchmarks do tend to report roughly a similar picture. It's swings and roundabouts to a degree, but that's roughly where most of the drives in the range appears to has their transfer capability pegged at.

However, your comment did make made me wonder, and I've thus looked up Seagate's official technical specifications for the entire range (each drive listed seperately), particularly areal density and sustained transfer rate. It turns out that there are actually a slight differences between the different models, although the max sustained transfer rate is nevertheless reportedly the same for all the drives (72Mb/sec) apart from the 750Gb which has a marginally higher maximum at 78Mb. So really they fib slightly by suggesting the entire range does 78Mb/sec peak. (Interestingly though, my 7200.10 320Gb does seem to get very much 80-ish Mb/sec peak...<shrug>) The average areal density likewise is the same for all the disks (114.4Gbit/square inch) apart from the 750Gb which has an areal density of 128.2Gbit/square inch.) The track density is reportedly identical on all the disks. These stats and a lot else is from p.29 here. The disks are thus in my opinion very similar in terms of max sustained transfer rate, although obviously not the same.

So apologies once again, I didn't mean to upset you. We just seemed to have had a bit of a miscommunication... I'll try to be clearer in future and try to read more carefully.