Which 300Gb SATA drive?

GSXRMovistar · 22 Apr 2006 at 12:49

Looking to order a new drive and my short list has the following four drives:-

Western Digital Caviar SE16 320GB 3200KS SATA-II 16MB Cache - OEM (HD-070-WD) 320 0.27(£ per Gb) 3 86.89
Seagate Barracude 7200.9 NCQ 300GB ST3300622AS SATA-II 16MB Cache - OEM (HD-066-SE) 300 0.28(£ per Gb) 5 83.37
Maxtor MaxLine III NCQ 300GB 7V300F0 SATA-II 16MB Cache - OEM (HD-096-MD) 300 0.30(£ per Gb) 5 90.42
Maxtor DiamondMax Plus 10 NCQ 300GB 6V300F0 SATA-II 16MB Cache - OEM (HD-089-MD) 300 0.25(£ per Gb) 3 76.32

Anyone got any recommendations which one I should opt for?

Thanks.

Pulsedriver · 23 Apr 2006 at 00:53

The Seagate, I just got a 250GB Sata2 16mb and they are great!

R B Customs · 23 Apr 2006 at 01:40

ive found the seagates to be a tad noisy. but i think its due to my big empty aluminium case lol.
appart from that theyre meant to be the safest, most reliable drives out there. (the only reason i bought mine)

GSXRMovistar · 23 Apr 2006 at 11:54

Cheers guys will order one shortly.

Dave2150 · 23 Apr 2006 at 14:01

I prefer WD drives - faster than the seagates and very reliable.

TooNice · 23 Apr 2006 at 14:06

Seconded. If you really want the 5 years warranty, there is the RE version.

smids · 23 Apr 2006 at 14:59

Dave2150 said:
I prefer WD drives - faster than the seagates and very reliable.

For HDDphiles yes. I challenge anyone to actually notice the difference. I'd say the Seagate - the 16MB cache should keep it ticking along nice and quickly.

TooNice · 23 Apr 2006 at 16:44

Isn't that like arguing that you won't find the difference between, say, a 3800+ and a FX60 (or some Intel CPU?)? For everyday use the difference is negligeable and there is a good chance that most perception is placebo.

Cost wise they are about the same. Warranty wise, they can be the same. So it's down to noise vs performance where there is a trade off. Both are subjective, someone can argue they feel the difference (measurable), while someone else can argue they can hear the difference (also measurable).

smids · 23 Apr 2006 at 16:54

TooNice said:
Isn't that like arguing that you won't find the difference between, say, a 3800+ and a FX60 (or some Intel CPU?)? For everyday use the difference is negligeable and there is a good chance that most perception is placebo.

Cost wise they are about the same. Warranty wise, they can be the same. So it's down to noise vs performance where there is a trade off. Both are subjective, someone can argue they feel the difference (measurable), while someone else can argue they can hear the difference (also measurable).

Nothing like that - nowhere near in fact. Slower - we are talking possibly 0.5-1ms. Are you going to be able to see that difference? No. That is in access time too, so unless you are continuously seeking data, then you won't notice it. And if you are talking about average read, then they come up pretty much the same within the margin of error.

As for CPU's your point is completely different. When you increase frequency, you increase the amount of data that can be executed etc per second. e.g. 9 IPC x 2,000,000,000 times per second or 9 IPC x 2,600,000,000 where there is double the L2 cache buffer.

ChrisLX200 · 23 Apr 2006 at 17:13

The warranty situation with HDs is more to do with a marketing ploy than it is to do with reliability. Many times manufacturers have tried to reduce the warranty period industry-wide, but competiton forces them to offer a longer warranty to retain market share. Given a choice between buying 2 drives of essentially similar spec, one offering a longer warranty, which would you buy? - exactly.

You think a drive with 5 year warranty is 'more reliable' than one with 3 year warranty? Think again. The added costs of servicing those extended warranties are offset against increased sales, it has sod all to do with additional reliability. When it goes wrong and they produce a faulty product in large numbers (like the IBM DeathStars) it causes the manufacturer real problems.

smids · 23 Apr 2006 at 17:20

ChrisLX200 said:
The warranty situation with HDs is more to do with a marketing ploy than it is to do with reliability. Many times manufacturers have tried to reduce the warranty period industry-wide, but competiton forces them to offer a longer warranty to retain market share. Given a choice between buying 2 drives of essentially similar spec, one offering a longer warranty, which would you buy? - exactly.

You think a drive with 5 year warranty is 'more reliable' than one with 3 year warranty? Think again. The added costs of servicing those extended warranties are offset against increased sales, it has sod all to do with additional reliability. When it goes wrong and they produce a faulty product in large numbers (like the IBM DeathStars) it causes the manufacturer real problems.

I don't think that it makes it more reliable having a 5 year warranty.

The fact it's a Seagate makes it more reliable in my eyes - they are renowned for being the most reliable drives. I mean I know my reliability poll isn't perfect but aren't seagate one of the longest standing HDU company's?

http://en.wikipedia.org/wiki/Seagate_Technology

In fact they made the very first 5.25" HD unit.

They have a low number on the poll - for being one of the oldest, you might have thought this to be higher, no?

As I say, the poll isn't scientific but it does make you think.

Using the words 'most reliable' is misleading in itself though - it should be 'least chance of failure' seeing as all drives WILL fail.

ChrisLX200 · 23 Apr 2006 at 18:23

smids said:
I don't think that it makes it more reliable having a 5 year warranty.

The fact it's a Seagate makes it more reliable in my eyes - they are renowned for being the most reliable drives. I mean I know my reliability poll isn't perfect but aren't seagate one of the longest standing HDU company's?

They are. Maybe your poll should have been restricted to failures within the last 6-12 months to retain relevance to the drives on sale today? Look at my sig - I've got a big investment in Seagate on this particular machine, but I've got lots of other drive makes in other machines too.

smids said:
http://en.wikipedia.org/wiki/Seagate_Technology

In fact they made the very first 5.25" HD unit.

They have a low number on the poll - for being one of the oldest, you might have thought this to be higher, no?

Oh they did make some of the first consumer HDs, I think I had one of them

It was a 10mb (10 megabyte) 5-1/4" drive that died from terminal seek errors within 2 years, but that has no relevance to today's offerings I think.

smids said:
As I say, the poll isn't scientific but it does make you think.

Using the words 'most reliable' is misleading in itself though - it should be 'least chance of failure' seeing as all drives WILL fail.

I didn't use those words - I said 'more reliable', as in a relative term. You are right - they will all fail eventually, and lets hope we get plenty of warning first! The main point of my message is the cynical (to my mind) use of extended warranty periods being used as a way of suggesting the drive on offer is 'more reliable' than a competitor product. The manufacturers hate this situation - it costs them a lot of money to service warranties they would rather not offer (and which we pay for in the end).

TooNice · 23 Apr 2006 at 18:43

smids said:
Nothing like that - nowhere near in fact. Slower - we are talking possibly 0.5-1ms. Are you going to be able to see that difference? No. That is in access time too, so unless you are continuously seeking data, then you won't notice it. And if you are talking about average read, then they come up pretty much the same within the margin of error.

As for CPU's your point is completely different. When you increase frequency, you increase the amount of data that can be executed etc per second. e.g. 9 IPC x 2,000,000,000 times per second or 9 IPC x 2,600,000,000 where there is double the L2 cache buffer.

The extra clockspeed only benefits if the application is intensive enough. Word is not exactly going to be noticeably faster (we are also talking in ms) because you have another 600Mhz and twice the cache to play with. If you talk about video encoding or something along those lines, then your CPU will show some benefit. Likewise, do some I/O intensive applications i.e. video processing etc. and a faster HD will show some benefits. Considering that the WD is priced similarly to the Seagate, the only reason to pick the later is if you value noise more highly than performance. The difference is performance may not be huge, but neither is the difference in noise.

I avoid talking about reliability, because I think that is subjective (basically, if one drive fails on you, you'll probably remember it for a long time - I know plenty of people avoiding Hitachi's because od the IBM's Deskstar fiasco - but TBH, the Hitachi Deskstar are still very fine drives that I could recommend in this thread, if performance is more important than noise). I am not saying Seagate's drive aren't solid, they are. My 2nd gen Cheetah was still working the day I retired a 7 years old PC. But I also don't have any reason to believe their IDE drive would be better than it's competitors. And no, them being the oldest does while maintaining a low value in the poll doesn't say much. We don't really know how many people here have bought a Seagate drive before do we?

There is a long time running reliability survey at the Storagereview. Like the poll here, it has it's limitations, but right now, the WD3200 series is doing pretty good for what that's worth.

GSXRMovistar · 25 Apr 2006 at 13:44

I went for the Western Digital in the end, purely for the fact that the Seagates were out of stock so made my decision easy.

Mercutio · 26 Apr 2006 at 02:15

Just to throw another card in - the Hitachi's lead the pack in the speed area around that size last time I looked at the storage review site. Almost nothing in it but again, its subjective

GSXRMovistar · 26 Apr 2006 at 19:35

Well drive turn up today and I think it's knackered.

For the last week I've had an older Western Digital SATA drive (WD2500JD) to play with and all I had to do was hook it up to my Abit motherboard (AX8-3rd Eye) with sata power and data cables, instantly recogonised by the BIOS and XP.

But now when I connect the new drive (WD3200KS) instead of the old one it spins up but isn't seen by either the BIOS or XP.

Anyone got any ideas before I have to go through all the hassle of getting it replaced?

smids · 26 Apr 2006 at 20:18

GSXRMovistar said:
Well drive turn up today and I think it's knackered.

For the last week I've had an older Western Digital SATA drive (WD2500JD) to play with and all I had to do was hook it up to my Abit motherboard (AX8-3rd Eye) with sata power and data cables, instantly recogonised by the BIOS and XP.

But now when I connect the new drive (WD3200KS) instead of the old one it spins up but isn't seen by either the BIOS or XP.

Anyone got any ideas before I have to go through all the hassle of getting it replaced?

Make sure the 150MB/s jumper is set when connecting to a non-SATA-II interface.

GSXRMovistar · 26 Apr 2006 at 21:33

smids said:
Make sure the 150MB/s jumper is set when connecting to a non-SATA-II interface.

You sir are a star!!

Thanks that sorted the problem, looks like my motherboard isn't SATAII then, does that effect the performance much?

smids · 26 Apr 2006 at 22:10

GSXRMovistar said:
You sir are a star!!

Thanks that sorted the problem, looks like my motherboard isn't SATAII then, does that effect the performance much?

No, it will not have any impact whatsoever except in burst reads where you are limited to 150MB/s rather than 300MB/s - you really won't see it!