Eugene

Seagate Barracuda 7200.8

Recommended Posts

Since Testbed4 is taking considerably longer to finalize than we originally anticipated, a backlog of drives has been accumulating over the months...

We've decided to reconstruct Testbed3 (the parts of which had been scattered across multiple separate systems!) to address some of the newest drives to hit the market.

First up is Seagate's Barracuda 7200.8. As the only drive to incorporate 133 GB platters, the drive stands alone with the potential to achieve new heights in performance. Does Seagate deliver? Let's take a look!

Seagate Barracuda 7200.8 Review

Share this post


Link to post
Share on other sites
In the end, those seeking quiet operation combined with Seagate's unique five-year warranty may be well served by the Barraucda 7200.8.

Second-last sentence of the article.

I'm eager to read the article about testbed 4.

Share this post


Link to post
Share on other sites

Thanks, I've been looking forward to this for a while. I was really expecting this drive to perform better. Well, now I don't feel so bad for getting the cheaper WD drives :D

Share this post


Link to post
Share on other sites

Good job, thanks for the review - especially since you had to resurrect Testbed3 to do it! I now wish I'd waited before buying mine. :(

Anyone know why the seek times on this drive are consistently so much lower than spec?

Also, what's with not using the full platter capacity? I don't remember manufacturers underclocking 266 MHz CPUs to 250 MHz to give more "marketable" performance. ;) Is there any way to reclaim this space? Especially since 1 GB = 0.93 GiB.

--

screwtop

Share this post


Link to post
Share on other sites

I've found one typo (on the last page, second-last paragraph):

the drive maintain's the line's reputation for silent operation

Thanks for the review. It confirms what has been suspected - new Seagate is not a good performer (but more likely a drive made to compete with 7K400 with baseless claims).

screwtop:

Anyone know why the seek times on this drive are consistently so much lower than spec?

I do not know for sure but I'm afraid it's because, when 7K400 came available, Seagate started marketing their "soon to come" 400GB drive. Of course the drive had to be marketed being better than 7K400 - how else would people be willing to wait for a Seagate drive of the same capacity?

I believe there's no other reason for 8ms seek time claim than the fact that 7K400 has a 8.5ms seek time (claimed AND actual).

The fact that actual seek times are over 10ms could be a compromise between noise and performance. Seagate doesn't support AAM anymore.

Share this post


Link to post
Share on other sites

Slower seek times are also affected by increased density, as precise head positioning becomes much more difficult... still... seek times are pathetic, and it shows in the benchmarks.

Maybe I can benchmark one of the RAID setups we have here...

Share this post


Link to post
Share on other sites
Slower seek times are also affected by increased density, as precise head positioning becomes much more difficult...

202792[/snapback]

You know it, I know it - Seagate doesn't. ^__^

...or possibly they know too but keep on lying anyway.

OK. Just a joke. Any manufacturer lacking basic knowledge would be out of the business in no time. Seagate definitively knows it's stuff... I just plainly believe they don't give a ####. OEMs won't probably complain about slow drives as much as they would complain about noisy, pricey or unreliable drives. End users aren't that interested, that when buying a OEM computer, they would find what HDD it uses and what are the specs and most importantly search the web for reviews to get a realistic figure of it's performance.

It's not the first time people got fooled by Seagate. Here's one review by SR from past: http://www.storagereview.com/articles/2000...T340823A_1.html

The article:

Specified seek time is quite impressive. Though all its competitors claim average seeks of at least 9.5 milliseconds, Seagate specs the U5 at a speedy 8.9 ms.
and
Our measurements place the drive's access time at 19.4 milliseconds. This figure is easily the highest access time we've ever recorded? worse than even the 4400 RPM, 12 millisecond seek Quantum Fireball lct15. Subtracting the standard 5.6 milliseconds associated with 5400 RPM rotational latency yields a measured seek time of 13.8 milliseconds? nearly 5 ms off the mark.
...and it's not just a bad sample. (I get the almost the same results: 19.9ms with AAM disabled.)

Can I see a similarity on these cases? Yep. I sure can. When U5 came, Seagate claimed seek times lower than most competitors on ATA market... still the drive had seeks much slower than even previous generation of drives: slower than IBM, slower than Maxtor, slower than 3.5 inch Quantums, even slower than previous Seagates!

Share this post


Link to post
Share on other sites

Thanks for the review Eugene!

One additional note on the new Barracuda (Tonka) 7200.8 is:

The discs are no longer written (some call it low-level-formatted) within the drive by slow big expensive servo-track-writers, but externally by new high-presicion disc writers and then assembled into the drive later.

This saves a lot of production space, time and logistics.

You note this, that there is no more seal on the side of the drive where usually a big hole is covered. On former generations the head of the servo-track writer was feed in there.

I find this, together with a 133GB/platter capacity, quite a remarkable technology step.

Regards,

Share this post


Link to post
Share on other sites

The Seagate site makes it hard to find, but there is a 200 GB PATA version of this drive. I have one in front of me. The model number is ST3200826A.

I'm really disappointed to find out here it apparently doesn't use the new 133 GB platters. I'm actually pissed off at Seagate now. I bought it on special after Tom's Hardware said the 300 and 400 GB drives had good performance. I'm sure SR is correct when they say that isn't the case. The drive is also made in China, which I'm not too keen on.

I hope the reliability at least lives up to the standard of previous Seagate drives, as that's ultimately the most important factor for me.

Share this post


Link to post
Share on other sites
As its designation implies, the Barracuda 7200.7 incorporates the 7200 RPM spindle speed associated with the brand since its introduction over 13 years ago. With this newest 'Cuda,...

Shouldn't that read 7200.8?

Share this post


Link to post
Share on other sites
Finally, in the SR Gaming DriveMark, the 7200.8 ekes out a small lead over the 7200.7 with a showing of 529 I/Os per second.

But the chart reads 539. Which is right?

Share this post


Link to post
Share on other sites

I'm surprised that the article made only passing reference to reliability issues. Googling 7200.8 and "Delayed Write Failure" turns up many references to what sounds like a widespread problem - the drive locks up under stress. (Some of those Google results point back here to SR)

I purchased the 300GB model hoping to put it into a USB enclosure and using it for backing up a few of my servers. Every night, it would lock up after transferring between 25-50GB of data. I spent about a week trying to figure out what the problem was.

Seagate received my shipment nine days ago and still hasn't shipped a replacement. The warranty doesn't do much good if you're forced to buy another drive while waiting for the first!

Share this post


Link to post
Share on other sites

well I guess I bought into the hype of the seagate 400gig drives and ended up buying three of them and a 300 7200.8 drive. For me they seem alot faster then the 7200.7 200 sata drive I had, the ncq makes a big difference with a nf4 board, but that didn't seem to be teted in the review. I do notice the 400gig drives are really hot my raptor seems to be 7c cooler then them, and I thought that was suppose to be a hot drive. I don't really trust maxtor drives so I went with seagate, since I figure if the drives has 5 year warrenties they should last for the 1.5-2 years I plan to keep them for without any issues. Benchmarking the 400 seagate doesn't seem to do it justice in most reviews, when I actually use them, they seem alot faster. I think there are two different firmwares 3.01 and 3.02. The 3.01 I have on my 300 gig drive seems to perform alot worse then the 400 but that maybe just do to the lower platter density. I found 400 to be really heavy compared to my other drives, so the heavy platter probably have something to do with the slow seek, or seagate may just want them to be more realiable to not have alot of returns during thier 5 year warrenty period. If I trusted maxtor I would have bought thier 300gig drives, but they didn't seem large enough for my need, and seem to have quality control issues for alot of people.

Share this post


Link to post
Share on other sites
I found 400 to be really heavy compared to my other drives, so the heavy platter probably have something to do with the slow seek

203118[/snapback]

Platter weight has nothing to do with seek times. Think again.

Share this post


Link to post
Share on other sites

Anandtech recently posted a review as well. They draw a much more favourable conclusion about the 7200.8

TweakTown also seemed to think that the 7200.8 is a great performer.

Is this drive going to be re-run on TB4 or has it already been? Any indications of how it performs there?

Share this post


Link to post
Share on other sites

Well, honestly, I'm quite surprised with the overtones of the review because in my opinion they're way too positive. I've been using 250 GB 7200.8 SATA NCQ drive for a month and this model replaced my previous 7200.7 200 GB drive. I decided to take a risk and purchase the drive without seing any prior reviews because my previous experiences with Seagate have been rather positive. Well, I lost the game. I'm very disappointed and I'd rather not buy any new Seagate products anytime soon. Surely, not without prior deep digging.

The 7200.8 model is simply slow (in real-world applications sometimes slower than my 2-years old 7200.7) and noisy, very noisy (I mean seeks - the motor is whisper quiet). The latter is especialy painful because I decided to buy 7200.8 mostly due to the fact that 7200.7 IDE drive was such a silent harddrive. 7200.8 is not such a case, furthermore - Seagate decided to disable option to control the AAM of the drive and they are NOT going to enable it back. I've checked that out with their support team. Instead, they have built in some kind of internal acoustic management which turns the performance mode on if the drive's working in NCQ mode (means - the chipset and SATA controller are compatible with NCQ and NCQ's been turned on) - or turns it off and switches to accoustic mode if the platform does not support NCQ (or NCQ has been disabled). To me it's simply not acceptable because I want to be able to control and choose how the hardware is expected to be working. I'm going to sell this poor product as soon as new Samsung P120 harddrives are available here.

Seagate, wake up! You cannot relly on your past achievements only. Care more about what end users think about your products or... (sorry).

Share this post


Link to post
Share on other sites

Too little too late.

And I dont mean the 7200.7

I do honestly understand Storage Reviews problems sorting out the new testbed/methods, but I have already read loads or reviews on this drive and decided that I dont want one.

But nothing I read in Storage Reviews review swayed me in the slightest, not one little bit, SR still dont do any kind of realistic drive noise measurements, nearly all hard drives on the market make a similar amount of "Spin noise", but SR doesnt measure the "Seek noise".

Directly related to the "Seek noise" testing, it would be a good idea to do 2 tests, one with AAM enabled, and one disabled.

I am now looking forware to Storage Reviews review on the new Samsung P120 range of drives, with "Seek noise" listing with AND without AAM enabled.

Other than the above grumbles, yet another superb review from SR.

Andy

Share this post


Link to post
Share on other sites
I'm surprised that the article made only passing reference to reliability issues.  Googling 7200.8 and "Delayed Write Failure" turns up many references to what sounds like a widespread problem - the drive locks up under stress.    (Some of those Google results point back here to SR) 

203083[/snapback]

Great. I have just replaced a bunch of Maxtor MaxLine Plus IIs with these disks. The Maxtors would not do the bonnie benchmark without locking up (RAID 5 on an ICP Vortex controller). The Barracudas caused no problems so far.

Share this post


Link to post
Share on other sites

As I expected, Seagate is no good! So I'll buy two Hitachi T7K250 of 250GB each for my RAID 0 on a nForce 4-SLI.

I agree with Andy, SR should measure seek noise.

Share this post


Link to post
Share on other sites
I agree with Andy, SR should measure seek noise.

203341[/snapback]

I would have reservations about having SR measure seek noise without the proper equipment. The range of measurements with seek noise is larger than with idle noise (usually), so unless SR measured a large quantity of drives, it'll be easy to draw erroneous conclusions about how Drive X is louder than Drive Y or vice versa.

Share this post


Link to post
Share on other sites

So what 400 gig drive do you guys recommend? I am going to use it as an extra drive (not boot drive) for sound fx files and samples that I need to stream and access for recording work. I am getting a new G5 2.3 or 2.7 and this would be the 2nd internal drive.

Would the Hitachi Deskstar 7k400 be a better choice, or is it not a good idea to go to a 400 gig drive?

Thank you.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now