Paulo
Member
7 posts | I'm going to flash it in a few hours, just creating a backup.
But besides the transfer rate, how does it behave with random access? If for instance it has lower PCMark scores, I'm really not sure if I should upgrade  |
Enlightenment
Administrator
 105 posts | | Posted on 9 January 2008 @ 00:34 | edited 00:37 |  |
I'll be more than interested in your results, Paulo. In fact, i think on altering the article should the new firmware really change things.
Could you post a few HDTune and ATTO-256 benchmark screenshots?
I mirrored ATTO-256 on: http://www.fluffles.net/files/ATTO-256.exe but you can also use google to find it. Take control of the input and you shall become master of the output. |
Paulo
Member
7 posts | | Posted on 9 January 2008 @ 10:40 | edited 10:49 | |
yes, it makes an hell of a difference 
http://turtlegt.planetaclix.pt/hdtune253.jpg
No capped effect, 75MB/s
http://turtlegt.planetaclix.pt/atto.jpg
Look at the 4KB value: 44000! I had ~13000 before. The drive was very slow with small block sizes, not anymore
I'm so glad I found this forum and upgraded the firmware. Thanks a lot! Now I have an HD with great performance, competitive with any other 500GB and very reliable.
|
Enlightenment
Administrator
105 posts | Looks great!
Question is: is this an attempt to hide shortcomings of the AAK-disk by firmware that does indeed deliver proper sequential performance but at the cost of realistic performance level, or is it the real deal?
It has been said that the chip used in AAK disks was slower than the chip in AAE and AAC drives. For me the question is: is a AAK-disk upgraded to AAM firmware really faster in reality than AAE?
That is something that cannot be tested with a simple program like HDTune or ATTO; those only test sequential performance (and any drive SHOULD get their physical maximum). In reality, I/O is much less 'easy' than a sequential read or write pattern. But as you say, the drive does have its 'cap' removed. I'm still very curious though. 
If i can find some time i'll test out the new firmware, if i can find an AAK disk again.  Take control of the input and you shall become master of the output. |
Paulo
Member
7 posts | Enlightenment wrote: Looks great!
Question is: is this an attempt to hide shortcomings of the AAK-disk by firmware that does indeed deliver proper sequential performance but at the cost of realistic performance level, or is it the real deal? Dunno. But what I'm going to do now is download PCMark2005 and one or 2 benchmarks you used, and test that myself
|
Paulo
Member
7 posts | | Posted on 11 January 2008 @ 00:22 | edited 12 Jan | |
Damn, I lost my previous post. Sorry!
Anyway, from the demo:
XP Startup: 12.4
General usage: 7.36
Very nice scores
I'm using Windows XP, Intel SATA drivers on a ICH8R
Edit: h2benchw results
Running application profile `swapping' ...14558.2 KByte/s
Running application profile `installing' ...19051.7 KByte/s
Running application profile `Word' ...27171.2 KByte/s
Running application profile `Photoshop' ...24995.9 KByte/s
Running application profile `copying' ...105611.5 KByte/s
Running application profile `F-Prot' ...12533.7 KByte/s
Well, way lower than yours! Since the chipsets are different, and my Hitachi disk has poor scores, I really dunno what to think about it |
kemin
Member
3 posts | My upgrade to AAM worked too!
I upgraded all four of the AAK drives in my RAID 10, without a problem. I ran QuickBench 4 before and after, and noticed a definite improvement.
However, my numbers won't be directly comparable to anyone with a different setup, so I'll just give the relative changes.
Sequential reads improved overall by 7%, but small files improved the most (41% with 4K files).
Random reads also improved by 9%.
I was hoping for a bigger performance gain, but I'm still very happy with the results! |
mercurymercury
Member
3 posts | Hi, i have been using seagate HDD products for quite some time now n never had any problems, till recently i bought 2 320GB drives.
Yes, AAK firmware on both (7200.10 both 320GBs 16mb). i am also experiencing below par performance.
i wonder if this new AAM firmware will also apply on my drives (both of them P/N ending with 308).
Anyone tested it yet? or does it only apply to the 500GBs (like all tested in this thread). i've been hesitating to upgrade cause there hasn't any official statement from seagate n the firmware itself can't be found on their site, also those who successfully upgraded are using 500GBs not 320GB like me?
Any help here? thanks.
|
Enlightenment
Administrator
105 posts | | Posted on 19 January 2008 @ 02:19 | edited 31 Jan | |
I think AAM will work also on 320GB drives. Just be sure to have a copy of the data that it holds! If it is important, anyway.
Update: i was wrong, the AAM firmware update only applies correctly to 500GB models. Pity  Take control of the input and you shall become master of the output. |
jack|ass
Member
5 posts | Does applying the new firmware trash the data on the drive? |
daviasdf
Member
3 posts | Hmm... I just got 2 320gb drives from Fry's this week and they were both AAK. Haven't installed them yet as the mobo they came along with had a broken nb hs mount.
Am I out of luck regarding any updated firmware since my P/Ns end in 570? I'm considering returning them since I'm RMAing the board anyway... |
Paulo
Member
7 posts | jack|ass wrote: Does applying the new firmware trash the data on the drive? It didn't on mine. I'm 100% happy with the result.
I'm considering buying another AAK 500GB and set up RAID, with Intel Matrix |
jack|ass
Member
5 posts | Hrm... So on mine, it says "invalid data" after "attempting to read binary file dld00.bin." It then reads dld01.bin and says "completing firmware download, resetting device." After a few seconds, "Model ST3320620AS SN <serial> FW 3.AAK" pops up, making me think this wasn't successful. It does appear to still be bootable, since Grub comes up... Anyone seen this? This is a -308 drive, before anyone asks. |
jack|ass
Member
5 posts | Alright, figured it out. My cd drive was also SATA and causing the updater to stall after resetting the drives (presumably because the program that was running went away. :)) So I flashed and rebooted and it seems to be ok. However, the drive still idenitfies itself as a 3.AAK in linux's boot messages. Is this just a identifier string that the flasher doesn't change? Is there a way to determine if I'm actually on the AAM firmware? (other than hopefully my array running at the expected speeds...) |
Steamer
Member
2 posts | I've tried the update both on a 320gb -308 drive and a 500gb -308 drive.
It only works on the 500gb -308 drive, and the firmware indeed changes from AAK to AAM on that drive.
On the 320gb drive, it fails, and the firmware stays at version AAK. The message that update utility reports is not clear about the failed update.
Emile |
tombquick
Member
2 posts | Great article and follow-up discussion. Seagate will have to listen if we continue to push them, or it will eventually catch up to them in the buying public.
Thanks! |
Tripps
Member
2 posts | | Posted on 30 January 2008 @ 23:29 | edited 31 Jan | |
Great work...
Hey wondering if you have done benchmarking on speeds of new SATAII drives vs IDE for speed? (Not benchmarking software where SATAII is faster but some real world benchmarks).
I have been LOOKING for days to find some real world testing which puts them head to head, but I have yet to find some affirmative proof that SATAII is actually faster and not just capable of being faster.
CAN YOU HELP?
Thank you  |
Enlightenment
Administrator
105 posts | Tripps wrote: Great work...
Hey wondering if you have done benchmarking on speeds of new SATAII drives vs IDE for speed? New SATAII drives are not fitted with PATA connectors anymore, harddrive manufacturers are pushing for SATA now, not releasing PATA equivalents for new products.
Parallel ATA, which is part of the IDE philosophy just like Serial ATA is, is a shared access interface, so with 2 drives on one cable one drive would have to wait for the other device.
Serial ATA 3Gbps is faster than PATA because of lower transfer latency, because the interface can send and receive the data in a shorter time. Note that sequential I/O does not care about latencies much, thanks to read-ahead. But real applications which sit on theit bums until they receive the data, will be faster, although the impact will be limited. Factors like data density, rotational speed and firmware/filesystem optimizations will determine overall performance of a harddrive. Take control of the input and you shall become master of the output. |
daviasdf
Member
3 posts | daviasdf wrote: Hmm... I just got 2 320gb drives from Fry's this week and they were both AAK. Haven't installed them yet as the mobo they came along with had a broken nb hs mount.
Am I out of luck regarding any updated firmware since my P/Ns end in 570? I'm considering returning them since I'm RMAing the board anyway... So I sent back the 2 AAK drives and ordered 2 news ones and they turned out to be AAE  Awesome. |
Tripps-
Member
2 posts | Enlightenment wrote: ...harddrive manufacturers are pushing for SATA now, not releasing PATA equivalents for new products. ??????
Serial ATA 3Gbps is faster than PATA because of lower transfer latency, because the interface can send and receive the data in a shorter time. Note that sequential I/O does not care about latencies much, thanks to read-ahead. But real applications which sit on theit bums until they receive the data, will be faster, although the impact will be limited. Factors like data density, rotational speed and firmware/filesystem optimizations will determine overall performance of a harddrive. Yes, but have you ever done or know any URL/links of reputable sites that have done tests of the latest 500GB, 16MB cache, 7,200RPM PATA for speed (with REAL WORLD not benchmarking simulations) against the same manufacturer's 500GB, 16MB cache, 7,200RPM SATAII drives?
I've been attacked on a number of sites that state SATA although capable of faster speed...when compared to a single PATA drive on it's own IDE channel is no faster than a single SATAII. I find benchmark programs like HDTune/HDtach that say SATA is faster.
But respected user guru's on those sites say that those programs are not comparable to Real work use of HD's today.
I'd appreciate some help.
Tripps
|
