yes obviously, but I mean will the big laptop drives be similarly faster than the small ones as the big desktop drives are than their low capacity counterparts.zzombi wrote:"I'm assuming the high transfer rates found in big drives like the F1 would also be found in big laptop drives?"
No, tracks are shorter, even if they had same rpm.
Samsung F1 series hard drives w/1TB model
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
- Patron of SPCR
- Posts: 1069
- Joined: Thu Aug 04, 2005 3:31 pm
- Location: Munich, Bavaria, Europe
STR isn't related to HDD capacity, but to platter capacity. I expect tests showing 320, 640 and 1000 F1 get about the same STR (I mean linear read speed).
Access time is related to capacity. A 1TB drive is much quicker than a 320GB. Indeed, if we use just a 320GB partition on the bigger drive we'll measure sensibly shorter access times. The extra space can then be seen just as a bonus, take it or leave it ... you'll take it of course.
Access time is related to capacity. A 1TB drive is much quicker than a 320GB. Indeed, if we use just a 320GB partition on the bigger drive we'll measure sensibly shorter access times. The extra space can then be seen just as a bonus, take it or leave it ... you'll take it of course.
I've found some nice benchmarks of the 750 and 1TB Samsung drive:
Source: http://gathering.tweakers.net/forum/lis ... es/1259539
HD753LJ:
HD103UJ:
And some other info:
Samsung will release special Raid Editions of this drive. But only for the 320gb, 750gb en 1000gb versions. The 640gb version will not be released in a Raid Edition version.
750gb = 3x250gb platter
and more information of the Raid Edition:
- Rotational vibration sensor
- MTBF 1,200,000 POH
Source: http://gathering.tweakers.net/forum/lis ... es/1259539
HD753LJ:
HD103UJ:
And some other info:
Samsung will release special Raid Editions of this drive. But only for the 320gb, 750gb en 1000gb versions. The 640gb version will not be released in a Raid Edition version.
750gb = 3x250gb platter
and more information of the Raid Edition:
- Rotational vibration sensor
- MTBF 1,200,000 POH
Also here comparing all the four 1TB drives available so far:
http://www.pcgameshardware.de/?menu=bro ... w=original
Noise for idle/seek:
Hitachi Deskstar 7K1000: 0.6/1.2
Seagate Barracuda 7200.11 ST31000340AS: 0.6/0.8
Samsung F1 HD103UJ: 0.3/0.5
WD Caviar GP WD10EACS: 0.2/0.5
The best is Samsung F1
http://www.pcgameshardware.de/?menu=bro ... w=original
Noise for idle/seek:
Hitachi Deskstar 7K1000: 0.6/1.2
Seagate Barracuda 7200.11 ST31000340AS: 0.6/0.8
Samsung F1 HD103UJ: 0.3/0.5
WD Caviar GP WD10EACS: 0.2/0.5
The best is Samsung F1
Dutch to English translation:PASware wrote:I've found some nice benchmarks of the 750 and 1TB Samsung drive:
Source: http://gathering.tweakers.net/forum/lis ... es/1259539
http://www.google.com/translate?u=http% ... en&ie=UTF8
To complement SileX's link, here's a subjective take on the drives' noise from hartware.de, comparing Samsung F1 with WD Caviar GP, Seagate Barracuda .11 and Hitachi Deskstar 7K1000.
German source | English by Google | English by Babelfish
The author says the F1 is "somewhat louder" than the GP at idle, but still well suitable for a quiet system. Under load the two even up. He does not assess vibration level, though.
German source | English by Google | English by Babelfish
The author says the F1 is "somewhat louder" than the GP at idle, but still well suitable for a quiet system. Under load the two even up. He does not assess vibration level, though.
Last edited by line on Thu Dec 06, 2007 1:58 pm, edited 1 time in total.
The non-raid have 3 years.
Hdtune is suspect, it consistently gives lower MB/s than other programs (maybe MiB vs MB issue?). It has interesting yellow points data for mechanical access though: looks like the denser platters have a harder time placing their heads accurately.
The other program is confusing: so basically after blocks get 16-32KB the HDD works full speed with only 4 ATA commands in queue (is this the queue depth?)?! The upper margin should be the average STR, no more.
And what hdd life (hdd exchange interval) is that MTBF computed for?
There's an other thread here which used to be more "up to date", thought I'd mention:
viewtopic.php?t=43466&postdays=0&postor ... bdca0951bd
Hdtune is suspect, it consistently gives lower MB/s than other programs (maybe MiB vs MB issue?). It has interesting yellow points data for mechanical access though: looks like the denser platters have a harder time placing their heads accurately.
The other program is confusing: so basically after blocks get 16-32KB the HDD works full speed with only 4 ATA commands in queue (is this the queue depth?)?! The upper margin should be the average STR, no more.
And what hdd life (hdd exchange interval) is that MTBF computed for?
There's an other thread here which used to be more "up to date", thought I'd mention:
viewtopic.php?t=43466&postdays=0&postor ... bdca0951bd
It is now in stock in the UK.
http://www.overclockers.co.uk/productli ... &subid=940
Komplettare still stating the 12th at a much nicer price.
http://www.komplett.co.uk/k/kl.aspx?bn=10096
Currently no other UK sites I know of list the 1TB product, but they look like they will be in stock in most places soon.
Andy
http://www.overclockers.co.uk/productli ... &subid=940
Komplettare still stating the 12th at a much nicer price.
http://www.komplett.co.uk/k/kl.aspx?bn=10096
Currently no other UK sites I know of list the 1TB product, but they look like they will be in stock in most places soon.
Andy
-
- Posts: 76
- Joined: Mon Oct 29, 2007 2:02 pm
-
- Posts: 20
- Joined: Sat Oct 30, 2004 1:45 pm
- Location: Brooklyn, NY
Presumably the RAID edition is tuned for better performance and longevity while in a RAID setup than a non-RAID version.
As an example, you'd likely see a measurable difference in performance from a Seagate 7200.11 vs an ES2, despite the two drives being close to identical physically.
Ideally, I'd also wait for the RAID edition, but I want to get my SATA RAID 5 setup going sooner rather than later, so I'm taking a chance on the F1 now. I'd been putting the build off hoping this drive would become available, and it's looking like that will pay off.
Oh, and I received the 1TB F1 from Newegg on Thursday.
It's a bit heavier than I expected, but that's fine.
No tests besides power on and BIOS detection yet, but I can tell you that I could not tell when the drive spun up, or that it was even on other than to place my hand on it.
Ordinarily, I can tell what brand a drive is just by listening to it. The F1 may prove problematic because so far I can't hear it. However, if it's really as quiet as it seems, I can live with that.
As an example, you'd likely see a measurable difference in performance from a Seagate 7200.11 vs an ES2, despite the two drives being close to identical physically.
Ideally, I'd also wait for the RAID edition, but I want to get my SATA RAID 5 setup going sooner rather than later, so I'm taking a chance on the F1 now. I'd been putting the build off hoping this drive would become available, and it's looking like that will pay off.
Oh, and I received the 1TB F1 from Newegg on Thursday.
It's a bit heavier than I expected, but that's fine.
No tests besides power on and BIOS detection yet, but I can tell you that I could not tell when the drive spun up, or that it was even on other than to place my hand on it.
Ordinarily, I can tell what brand a drive is just by listening to it. The F1 may prove problematic because so far I can't hear it. However, if it's really as quiet as it seems, I can live with that.
Samsung SpinPoint F2 1.5 TB by 2008:
Original German:
http://www.hartware.de/news_43643.html
Translated to English:
http://translate.google.com/translate?u ... n&ie=UTF-8
Expect 2TB by 2009.
Wow!!!
Original German:
http://www.hartware.de/news_43643.html
Translated to English:
http://translate.google.com/translate?u ... n&ie=UTF-8
Expect 2TB by 2009.
Wow!!!
-
- Posts: 1
- Joined: Sun Dec 09, 2007 5:49 pm
-
- Posts: 160
- Joined: Sat Dec 01, 2007 4:25 am
"Presumably the RAID edition is tuned for better performance and longevity while in a RAID setup than a non-RAID version."
The key word here is presumably. From tests I've seen so far there are close to no differences. Until we know there are mechanical differences, like materials or better bearings, they're unlikely to last longer. The vibration sensor may correct some tracking errors when many hdds are present, but who here still mounts them rigidly? That sensor isn't enough to make them last longer. They usualy have slower seeks too, for that same reason. In the end they do have longer warranty, but higher price.
"Ordinarily, I can tell what brand a drive is just by listening to it. The F1 may prove problematic because so far I can't hear it. However, if it's really as quiet as it seems, I can live with that."
That's good news, provided it is actually spinning From my limited experience even same model hdds may sound different, some having platters better balanced mechanically, so I guess you weren't unlucky.
If I had an F1 I wouldn't downgrade to an F2 While the hdds progress is not bad, ssds progress twice as quick, so I guess the F2 name will become the right match.
The key word here is presumably. From tests I've seen so far there are close to no differences. Until we know there are mechanical differences, like materials or better bearings, they're unlikely to last longer. The vibration sensor may correct some tracking errors when many hdds are present, but who here still mounts them rigidly? That sensor isn't enough to make them last longer. They usualy have slower seeks too, for that same reason. In the end they do have longer warranty, but higher price.
"Ordinarily, I can tell what brand a drive is just by listening to it. The F1 may prove problematic because so far I can't hear it. However, if it's really as quiet as it seems, I can live with that."
That's good news, provided it is actually spinning From my limited experience even same model hdds may sound different, some having platters better balanced mechanically, so I guess you weren't unlucky.
If I had an F1 I wouldn't downgrade to an F2 While the hdds progress is not bad, ssds progress twice as quick, so I guess the F2 name will become the right match.
-
- Posts: 606
- Joined: Wed Oct 13, 2004 4:03 pm
- Location: Albany, GA USA
What I know about pro vs. consumer network equipment may be valid for "raid editions" vs. normal versions of HDs, too:PsychoStreak wrote:Presumably the RAID edition is tuned for better performance and longevity while in a RAID setup than a non-RAID version.
The pro versions come with a much longer burn in test time. Typically equipment has high infant mortality rate. But after this critical phase failure rates decline, only rising again at the end of expected lifetime. The extended burn-in/testing sorts out those pieces most likely to fail, increasing the effective MTBF in field. It's this considerably longer testing that accounts for the higher prices.
Then the non-raid will last longer, provided they make it past infancy.
Some even say MTBF discounts infant failures...
Infancy is covered by waranty anyway, so I guess the more people buy raid, the cheaper the rest can get their plain vanilla longer lasting faster non-worn el cheapo non-raid. Quite an irony.
In conclusion the RAID-Edition rocks!
Some even say MTBF discounts infant failures...
Infancy is covered by waranty anyway, so I guess the more people buy raid, the cheaper the rest can get their plain vanilla longer lasting faster non-worn el cheapo non-raid. Quite an irony.
In conclusion the RAID-Edition rocks!
Yes, but the normal version is more likely to fail. Raid editions cater for a different market, it's for those running critical applications where failures must be avoided at all cost (warranty doesn't matter much here as it wouldn't cover any business losses incurred).zzombi wrote:Infancy is covered by waranty anyway, so I guess the more people buy raid, the cheaper the rest can get their plain vanilla longer lasting faster non-worn el cheapo non-raid.
and that's why they are called raid-edition: you have to use it in raid to prevent losing of dataRaptus wrote:Yes, but the normal version is more likely to fail. Raid editions cater for a different market, it's for those running critical applications where failures must be avoided at all cost (warranty doesn't matter much here as it wouldn't cover any business losses incurred).
And you would agree that those critical aplications exclude home users. Realistically, how much can those hdds been run in? 1-2 days, or 1 week at most. We waited half a year for this model, we could give it 1 more week. And except a minority of cases (say 1% premature failures) you get a better drive.
Unless there's more than this, or they're somehow binned, I see little reason for a home user. But to each his own.
Unless there's more than this, or they're somehow binned, I see little reason for a home user. But to each his own.
Though I don't have figures, I believe it's longer than that.zzombi wrote:And you would agree that those critical aplications exclude home users. Realistically, how much can those hdds been run in? 1-2 days, or 1 week at most.
In the first 3 months of usage drives have ~10% average failure rate if under high average load, according to this study (figure 3, page 5): http://labs.google.com/papers/disk_failures.pdfAnd except a minority of cases (say 1% premature failures) you get a better drive.
Very interesting study, btw, well worth the read.
I agree. One could still do his own burn-in testing before putting the drive into real use That would've saved me the trouble a few years back when my 2 week old 250GB hitachi corrupted my root folder...Unless there's more than this, or they're somehow binned, I see little reason for a home user. But to each his own.
-
- Posts: 81
- Joined: Mon Dec 11, 2006 3:25 pm
- Location: Lyon, France
The HD103UJ is available at Newegg, but does someone know something about the HD102UJ ? (1TB, 16MB cache instead of 32, thus cheaper). Why isn't it available yet, and WHEN will it finally be available ?
Being cheaper than the already cheap HD103UJ, I believe the HD102UJ will be a real killer in the 1TB market segment.
Thanks.
Being cheaper than the already cheap HD103UJ, I believe the HD102UJ will be a real killer in the 1TB market segment.
Thanks.