Posted on 2 November 2007 @ 05:20

While testing the AAK disk, i did not notice any weird sound. I do not think AAK disks are any less reliable, just less fast for desktops.

Take control of the input and you shall become master of the output.

Posted on 7 November 2007 @ 17:37

http://www.fotosik.pl/pokaz_obrazek/d83764b19c99b571.html

2 x Firmware 3.AAK, model ST3320620AS - RAID0 on Intel ICH9R
S/N 6QF1TAL8
S/N 9QF551TQ
The odd thing. Is it different a Firmeware AAK ?

DFI LanParty UT P35-T2R / C2D E6300

Posted on 15 November 2007 @ 18:46

edited 18:52

Well, I just received a 3.AAM firmware update from Seagate tech support, and the performance problem is fixed! I highly suggest contacting them and asking for this firmware update if you have a 3.AAK drive, as the performance difference is quite noticable, as you can tell by looking at the 3.AAE performance versus 3.AAK firmware. Looks like they found whatever the problem was! If whoever you end up contacting doesn't know about the 3.AAM firmware update yet, ask them to contact another person there going by "Alan TS49" as that's who I had success with. He was apparently in direct contact with the engineers that write their firmware, and should be able to help get you straightened away.

Note: I don't have an SATA-2 controller, so that's why my burst rates look to be SATA-1 speeds, because it's an SATA-1 controller.

Posted on 16 November 2007 @ 12:53

edited 12:56

Ok, I just shot an Email to Seagate Singapore (where I'm based} and they bluntly told me, no, they don't send out firmware upgrades to end users. I do respect their decision though, who knows how many more drives would have to be sent in due to failed flashing. Not too sure how next I should approach this problem, anyone care to comment?

Posted on 16 November 2007 @ 13:15

Well, I would be a little persistent about it. Describe the performance problems, and say that you're aware of at least one end user that received the 3.AAM firmware and it solved the problem for him. I was told that they'll be releasing it once it completes all their rigorous tests, and I was actually kind of a guinea pig. It took me over two months of back-and-forth emails with their tech support before I finally received this firmware in their last response. I had been pestering them since I first saw this article here, linked from Slashdot. Eventually I got in contact with this Alan fellow, who took the ball and ran. He was talking with the engineers there for a few weeks and then yesterday I got the good news and the firmware to try. Hopefully it won't take them long to publicly release it. But, as I say, that is what I would do. Just keep at it until you reach someone that will help you, and I would make a point of mentioning that you know "Alan TS49" has helped another customer with this very issue and that this 3.AAM firmware does in fact exist and solve the problem. Unfortunately I had to agree not to release it myself to get it for my drives.

Posted on 17 November 2007 @ 04:56

Any chance you can post the new firmware? I've got a 4-disk RAID5 array with two drives that are AAK and two that are AAE. It's been underperforming and I couldn't figure out why until I saw this page.

Posted on 17 November 2007 @ 13:56

cherrytwist at gmail dot com

Posted on 4 December 2007 @ 09:14

Interesting discussion. It is quite possibly related to the issues people have been having using the eSATA interfaces Seagate FreeAgent Pro drives.

I just blogged about this topic here:

http://blog.noegruts.com/2007/12/seagate-freeagent-pro-esata-problems.html

Hopefully this will help anybody that comes here wondering why their external Seagate drive doesn't work so well over eSATA...

Posted on 13 December 2007 @ 00:14

Well, I just got off the phone with Seagate tech support. I got escalated to a 2nd level tech support guy, and I told him I wanted the AAM firmware. He looked up the serial numbers of my four AAK drives, and basically told me I'm out of luck - the firmware on my drives cannot be upgraded.

I was persistent, as _Shorty recommended, and I asked specifically for Alan (TS49). I was told that Alan doesn't do telephone support, and that I should send an email, which may or may not ever reach Alan.

So that's where I am now. I'm not sure if my tech support guy was being unhelpful, or if my drives are really not upgradeable.

Posted on 13 December 2007 @ 20:47

hagbard wrote: Great article. Thanks

But you should now have a look at the newer Western Digital WD5000AAKS drives that they've been sending out in the past couple weeks. There are a bunch of us finding that the access time on these drives have gone from 13.x ms to 19.x ms. No one is reporting any timing in between, which tells me they've changed some component.Could be. One friend of mine who is a high-profile IT freelancer, suggested the use of a slower SoaC-chip (System-on-a-Chip) to be the cause of lower throughput. Because the chip had a consistent higher low-level latency it would not be able to keep up with the data flow possible with the AAE disk, so in short: it had a performance-cap. I have not been able to verify this.

What you describe with the Western Digital could be the same kind of scenario. But i would need more information to judge upon that. Any chance you could send me an URL about this?

I have drives five years old that have better access time than that.Please do remember that access time in itself is not directly relevant for performance. In real scenario's it is the number of I/O operations per second, also known as IOps, which actually indicate performance.

One important exception is blocking I/O, which more strongly is bottlenecked by seek times. This is why the WD Raptor is faster for some workloads than the newer generation 7200rpm disks. It loses sequential performance to the 7200rpm disks however, because these have far greater data density than the WD Raptor.

Take control of the input and you shall become master of the output.