I'm liking the specs of the 300GB model, can't wait to get my hands on one.
I've been hoping Seagate would take on the 10k rpm market, but, oh well. Hopefully, this will be the last generation of the 7200.x line. I don't wanna see 7200.10 until I see 10000.x first!
On Seagate's site, all of the 7200.8 and .9 drives look like they list 11 ms for the seek time. The 7200.7 all list 8.5 ms. However, this may be a difference in measured times. I believe the 7200.7 do not include rotational latency, which is 4.16 ms at 7200 RPM. I guess we'll have to wait for the benchmarks, but I think the .8 and .9 are supposed to seek faster.
cut the guys some slack, personaly i just started seeing these show up recently and have been on the search for info on them and at the moment its fairly hard to find any. this will at least give me some new info to chew on while the reviews are in progress.
For the same reason we post tradeshows and other articles there. We really only have the short "Insider Stories" and longer "Reviews". "Reviews" can really be previews, major product releases (I think this qualifies), reports from tradeshows or other trips, etc. Buyer's Guides and Price Guides, for example, aren't "reviews" either.
The sound info from Seagate is so obvious bull. And also the comments regretfully often found here @ AT:
quote: Combine this with the hum of case and heat sink fans, and the hard drive is basically silent.
A sound being drowned out does not make it silent. And assuming that everybody uses 8 delta case fans at full speed and thus does not care about the sound from hard drives, PSUs and graphic cards is just plain stupid in my opinion. With today’s water cooling and heat pipe technologies a lot of people are turning to the silent side. A hard drive in my system is one of the noisiest components.
Don’t know what I am commenting tough, I thought I clicked an article but it turned out to be nothing more then an advert. Leave this kind of info for the news section and be an independent hardware site that you usually do so well.
Well thats somewhat obvious, and my experience with Seagate dictates they havent been as quiet as the old Barracuda IV was. Those drives *were* basically silent, you could almost never hear them... even if they were under load.
How would the Seagate's new line compare with the Western Digital's:
"Western Digital Caviar RE2 400GB 7200RPM SATA 16MB 8.7"
"Western Digital Caviar SE16 400GB 7200RPM SATA 16MB 8.9MS"
"Western Digital Caviar SE16 250GB 7200RPM SATA2 16MB 8.9MS"
I've had nothing but good experiences with Western Digital...but I might turn over to Seagate if they turn out to be more superb.
The Seagate line I think matches the Western Digital line very well.
It has less platters at 400GB capacity, 3x133 Design vs WD 4x100 design and less platters at 250GB 2x125 vs WD 3x83. Seagate has SATA2 from 80GB-500GB, where WD is limited to 120GB to 250GB for their SATA2 drives, only downfall is that to get the 16MB buffer is only available at 300GB & higher for Seagate, where WD has a single model SATA2 drive with 16MB Buffer the 250GB model, the other 120GB to 250GB drives are like Seagate limited to 8MB buffer. Seagates 80GB to 250GB for 8MB is competitive. Plus seektime is 8.5ms. Plus for Seagate their drives get a 5 Year Warranty, not like WD which only get 5 Years on it's RE and Raptors lines I believe and 3 Years on the rest.
If you look at the table of specs, you'll notice the 300GB has the same wattage as the smaller models including the 1 platter(12.4/12.8/7.2 watts). Also it's got the same acoustics as the smaller models of 2.5 idle and 2.8 seek. The larger 400/500 models have slightly higher wattage and 2.8 idle/3.2 seek. This leads me to strongly believe that the 300GB model is also a 2 platter design using 150GB a platter...and is therefore slightly short stroked on a 160GB platter. I hope I'm proven right, because assuming it's a reasonably quiet drive...it's definitely the one I WANT! Looks like a candidate for the best drive with the features of quiet, a decent amount of storage, SATA 2.5 and a 5 year warranty!
I believe it's 3x100GB. It might also be 3x133, but only using five heads (i.e. 2.5 platters). Actually, 5 heads seems more likely, but I don't know for sure.
There are a few specific platter sizes that are common. Right now, those are (as far as I know):
60GB (old and on the way out, if not gone already)
80GB (fading fast)
100GB (still in production, but diminishing over time)
120/125GB (Relatively new; different brands might go one way or the other)
133GB (only a slight bump from 120/125, really - still new. Some "failed" platters might be formatted for 120/125GB)
160GB (brand new starting with certain Seagate 7200.9)
There is variance to a degree within those sizes. I think some of the 250GB drives, for example, use 133GB platters that are downgraded to 125GB capacity.
Samsung and Microsoft are actually working on a Hybrid notebook drive which uses OneNAND flash memory currently manufacturered by Samsung itself. The drive, which is still in the development stage, is said to reduce battery use among the speed increases. Google "samsung hybrid" and you shall see.
They should add some sort of suffix to all reviews that are really product briefings that just lets you know immediatly that it is mainly MFG information.
I appreciate that this is manufacturer info rather than independent review but I for one would prefer to know that such a drive is just around the corner and be able to wait rather than shell out for a 7800.8 drive and regret the timing of the purchase. Don't worry, I'll take the manufacturer info & stats with a grain of salt... I can handle it...
"In the next few weeks we will be looking at the performance of the 500GB 16MB SATA model of the 7200.9. We'll put it through our usual synthetic, simulated, and real world tests, and we'll compare it not only to the 400GB 7200.8, but also to some older drives like the 120GB 7200.8 as well as Maxtor's and Hitachi's higher capacity drives."
Why didnt they release this article after they did this testing? Obviously they have this drive in their hands already, and theres no NDA. This article should be a flash advertisement on the side of the page instead of a full fledged article.
"Introducing Seagate's 7200.9: The New Generation"
Introducing at least to me implies that it's not a review. We (Purav) have the parts on order and will be testing ASAP. The NDA lifted today, and we did not have product in house beforehand. To me, at least - since I've bought quite a few Seagate drives in the past year - the new lineup is important enough to warrant an article of its own.
I think Seagate is comfortable, in thier position of tied for first place, we will probably see a Hitachi 600GB based on the 5 platter design again of 120GB Before we see Seagate match that with a 4x150GB design. Also the other Drive manufacturers need to catch up to this level, namely Wester Digital and Maxtor.
The tech to run 160GB platters is new. Right now, they 160GB platter is only being used in the 80GB and 160GB drives, which is telling. I would have thought a 240GB and 320GB would also use the new platters, but that's probably a future product.
Also you got to not that Seagate may not want to dilute thier product line with intermidates such as 240GB & 320GB drives. Since noe one has made 240GB drives and only WD to my knowledge has 320GB drives.
but what about performance? The slowwest part of a system is the hard drive. I need speed, not 500gigs that only a few will really use daily. I have a pile of shows and movies but that is what DVD's are for. I need speed to make my system run faster. I guess I will stick with my 15K 36gig master boot drive and another one for backup.
Maybe hard drive makers will get it one day. Seems WD is the only one to address that for desktops so far?
is there some kind of raid array that will allow 3,5,10 drives to be merged as one?
seek times wouldnt improve but transfer rates might if the drives could intelligently split up singular file transfers into chunks for each drive... or is that some crazy idea i just dreamed up? ;) (4x 200gb drives stiped would be relativly cost effective compared to 4x or more raptors)
also 16mb buffers seem a bit low for the latest gen drives, 32/64 would be nice for the 300-500gb models.
and I think large cheap drives is where the moneys at now, cant blame seagate for attacking that front.
However, RAID50 exists. Now, this might be backwards, but: 6 drives. Make RAID5 array w/ 3 drives; make another RAID5 array w/ other 3 drives; make a RAID0 array with 2 RAID5 arrays. So, if you wrote 4 blocks of data, 2 go to RAID5-A, 2 go to RAID5-B.
As far as fault protection, if one drive goes, it's ok. If two drives go it matters from which RAID5 array. From the same array, the whole thing is gone, from different arrays, it's no problem. 3 drives is certain doom, though.
quote: also 16mb buffers seem a bit low for the latest gen drives, 32/64 would be nice for the 300-500gb models
More seems tempting. The test with tha 250GB Maxtor showed that streaming likes big buffers.
What kind of RAM is that buffer anyway? Is it ordinairy SDRAM? What would a buffersize > 16MB do? Is there a theoretical maximum and/or sweet spot in relation to the read/write capacity?
I think hdd's use a one chip ram solution, around 5ns is common
the thing you have to note is that a normal ram is an 8 chip solution for single sided ram
that's why 64mb is probably the highest they could go economically
and isnt the problem with using flash ram in a hybid the problem of lifespan? nand flash rates at 1 million writes/reads as mtbf? thats probably not enough for a hdd?
If you have a ton of HUMONGO filed, say, the complete season of everything volumes 1-100, in Divx, partition the drive in NTFS with a 16mb allocation size, and your data will FLY, especially if the allocation size matches the buffer (or am I spouting off crackpot science?)
It's not a review, it's an announcement of a new product which we feel many people will find interesting. If I were about ready to spend $100 on a 250GB drive, I'd sure like to know that Seagate has just released their 7200.9 series. This is an introduction to the technology, and an explanation about what has changed. The 16 MB cache models are certainly noteworthy, and all indicators are that the new drives will at worst match the old 7200.7/.8 Seagate models.
Price is the only thing I would worry about. If the 250GB 7200.9 costs $150 vs. $100, I probably wouldn't bother. Right now, http://labs.anandtech.com/search.php?q=seagate+160...">prices on the 160GB models are in favor of the new 7200.9. The remaining http://labs.anandtech.com/search.php?q=7200.9">7200.9 drives in our RTPE may or may not be worth consideration. $380 for the 500GB is more than I'm willing to pay for a single drive.
argh forget it. i guess i just want to see the reviews perfect for the sake of keeping anandtech the best it can be, so even though a typo doesn't affect the real content of an article (but it still devalues it)
This is true. If this was my first visit and I hadn't read articles here before, the typos really just make the author and the site appear ignorant. People in general, atleast ones that know the correct way, tend to stop reading things if they are written improperly; they just look like a child wrote it.
You know what. Anandtech has never posted any news that has ever been considered "out of touch". They're reviews are as unbiased as reviews can be without being written by cold, unfeeling machines. If Anadtech decided that these segate drives were good enough to give a free comercial too, then thats the best kind of marketing isn't it? I remember back in the day when a 40gig was huge, I called a local dealer and asked them for the cheapest 40gig they had. The guy said "We'll, that would be a Ninja [something], but I wouldn't recommened that unless you want to come back tomorrow and call in the warranty." Ok, that was a but off topic. Long live the Queen, free tibet, and all that jazz.
I would actually call this an article detailing Seagate's official introduction of the 7200.9 line. As far as purchasing these drives go, they have shipped in the last few weeks and are in stock at many places already.
As we mentioned in the article, we will be providing benchmarks of the 500GB version of the 7200.9 within the next few weeks.
"The lowest amount of sound audible by human ears is 26 decibels, so the idle noise output is borderline inaudible to our ears"
This seems to be taken from the info Seagate has published, unfotantly it is not correct. The db(p)-scale has it reference at the human hearing threshold. In other words, the human hearing threshold is 0 db. (although there obivously are individual differences, and background noise rarly is under 15 db).
2.6 bels is of course the same as 26 decibels. (deci = 1/10)
That said, people claiming that 26 dB is not quite probably are going by advertising numbers rather than reality. I have an SPL (Sound Pressure Level) guage, and 26 dB is extremely quiet. I can still hear it, but a room with no noise in it ranges from 24 to 30 dB. A car driving by outside makes about 40 to 45 dB of noise.
The problem is that if you trus the noise levels from computer hardware manufacturers, you might actually believe that your PSU, fans, HDD, etc. only put out 26 dB of noise. I've got a case with three "26 dB" fans, plus a "28 dB" PSU. Anyone want to guess at the real noise level? I won't keep you in suspense; at 1 meter form the case, the actual measured noise level is 48 dB, FAR higher than the rating of any of the components. Multiple fans plays a part, but even a single "26 dB" fan still came up with a 38 dB rating on my SPL.
I would agree mostly with Seagate's claim that 26 dB (2.6 Bels) is below the threshold of human hearing. A device putting out 26 dB at a distance of one foot would be inaudible at a range of 10 feet, that's for sure. (Unless you have dog hearing....) Whether or not that's how Seagate rated their drives, though, I can't say.
First of all, a room with "no noise" is much more quiet than 24-30 db, it just that you are used to the background noise. a 26 db fan at one feet is around 15-16 db at 10 feet, which is a clearly adible sound that too (this i have measured in an anecoic room). However, in most rooms the background noise would mask it. 26 db is quiet, but its far from inaudible.
We don't live in sound-proof environments, though. While it may not be truly silent, an empty house without any noise from neighbors or traffic is still going to have a background noise level above 20 dB. If something is below the background noise level for a house, I'm willing to call it "silent" - even if it isn't technically dead silent. I guess that's a difference of opinion, though.
That of course, can be discussed, but most people would easily detect 26 db, or 20 for that matter. And saying that 29 db is "barly audible" is a far strech IMHO.
Agreed. At 2AM (when I am sleeping with my computer idle 10 feet away) I can hear its 28dB and its not very quiet. I would say in the dead of night, a computer would have to be roughly 18dB to be considered unhearable above typical background noise.
However, there seems to be a mix-up between Sound power and Sound Pressure - two totally different entities.
The best would be if they reported sound power, as sound pressure depends the enviroment, and the power is constand for a given sound source (and you can calculte the pressure at a given distance in a given room). But it's almost the pressure who is given in computer-components. And the ear "hears pressure" so for the hearing its more useful to talk about the pressure.
Because the SI-system (which is centuries old), clearly defines Giga as 1 000 000 000. The fact that some software doesn't follow the standard can't be blaimed on the HD-manufactureres.
The modern version of the metric system is from the 1060s, so i guess i was abit unclear, however, the metric system itself, which the SI-system is built upon is fomr the 18th century.
Basically, the problem is that KB, MB, and later GB, TB, etc. all came from ealy computing days, where 2^10 was close enough to 1000 that they abbreviated it to KB. Later on, SI came into being and really got pissy about the use of "Kilo" for "1024" rather than "1000". The hard drive manufacturers are of course using the multiples of 1000 because it makes their product look better. Why say 93.13 GB when you can say 100 GB?
SI later proposed the "kibi, mibi, gibi, etc." prefixes to get around the discrepancy. In reality, few companies are using these terms at present. That may change in the future, but in truth most people don't care. We mention this discrepancy more so that people are aware of why the difference is there than to place blame.
I have one 7200.7 SATA, so far no problems but seek noise is loud, I was waiting for 7800.9 to get one 500GB drive.
Please make some acoustic tests when u review 7800.9.
Also with NCQ ON/Off, the 7200.7 sometimes was slower with NCQ ON.
Bench with Maxtor, WD, Hitachi and Samsung drives would be nice.
Hopefully this will bring down the price of the 500GB drives a bit. Running out of space for TV shows, and I already have a couple 400's and 250. Anybody find a RAID-5 external enclosure yet that holds 3 or more drives? I've been looking for one for a while now. I'm getting nervous with all my stuff not being RAID or backed up.
Um, the hotplug/hotswap feature sounds interesting, is it supported on any of the integraded sata controllers? I know my nforce4 has some "safely remove hardware" icon for my drive, but im quite sure the hdd doesnt support such a maneuver. Its quite confusing...
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
72 Comments
Back to Article
PitBrz - Monday, October 17, 2005 - link
7200.9 New King !!!HDTach average read: 63,6 MB/s
RAID0 2x160GB TecBench: Average 115 MB/s, Max 143 MB/s, min 68 MB/s
http://forums.anandtech.com/messageview.aspx?catid...">http://forums.anandtech.com/messageview...RTPAGE=7...
mariush - Wednesday, October 12, 2005 - link
What's with all the ©Seagate Confidential texts on the images ?blwest - Wednesday, October 12, 2005 - link
I didn't really find this article useful. No benchmarks. I could have gotten this information from seagate's website....SnowyEnigma - Tuesday, October 11, 2005 - link
I just bought 3 400g 7200.8!flatblastard - Tuesday, October 11, 2005 - link
I'm liking the specs of the 300GB model, can't wait to get my hands on one.I've been hoping Seagate would take on the 10k rpm market, but, oh well. Hopefully, this will be the last generation of the 7200.x line. I don't wanna see 7200.10 until I see 10000.x first!
techspeed - Tuesday, October 11, 2005 - link
why the avg seektime 11ms compared to 8.5ms in early models?JarredWalton - Tuesday, October 11, 2005 - link
On Seagate's site, all of the 7200.8 and .9 drives look like they list 11 ms for the seek time. The 7200.7 all list 8.5 ms. However, this may be a difference in measured times. I believe the 7200.7 do not include rotational latency, which is 4.16 ms at 7200 RPM. I guess we'll have to wait for the benchmarks, but I think the .8 and .9 are supposed to seek faster.Adamv1 - Monday, October 10, 2005 - link
cut the guys some slack, personaly i just started seeing these show up recently and have been on the search for info on them and at the moment its fairly hard to find any. this will at least give me some new info to chew on while the reviews are in progress.Mumrik - Monday, October 10, 2005 - link
Nice advert...Why the hell did you post it in the "Latest Reviews" section?
JarredWalton - Monday, October 10, 2005 - link
For the same reason we post tradeshows and other articles there. We really only have the short "Insider Stories" and longer "Reviews". "Reviews" can really be previews, major product releases (I think this qualifies), reports from tradeshows or other trips, etc. Buyer's Guides and Price Guides, for example, aren't "reviews" either.ksherman - Monday, October 10, 2005 - link
and i just bought a 400GB Seagate... oh well.Live - Monday, October 10, 2005 - link
The sound info from Seagate is so obvious bull. And also the comments regretfully often found here @ AT:A sound being drowned out does not make it silent. And assuming that everybody uses 8 delta case fans at full speed and thus does not care about the sound from hard drives, PSUs and graphic cards is just plain stupid in my opinion. With today’s water cooling and heat pipe technologies a lot of people are turning to the silent side. A hard drive in my system is one of the noisiest components.
Don’t know what I am commenting tough, I thought I clicked an article but it turned out to be nothing more then an advert. Leave this kind of info for the news section and be an independent hardware site that you usually do so well.
segagenesis - Monday, October 10, 2005 - link
Well thats somewhat obvious, and my experience with Seagate dictates they havent been as quiet as the old Barracuda IV was. Those drives *were* basically silent, you could almost never hear them... even if they were under load.andyc - Monday, October 10, 2005 - link
How would the Seagate's new line compare with the Western Digital's:"Western Digital Caviar RE2 400GB 7200RPM SATA 16MB 8.7"
"Western Digital Caviar SE16 400GB 7200RPM SATA 16MB 8.9MS"
"Western Digital Caviar SE16 250GB 7200RPM SATA2 16MB 8.9MS"
I've had nothing but good experiences with Western Digital...but I might turn over to Seagate if they turn out to be more superb.
coldpower27 - Monday, October 10, 2005 - link
The Seagate line I think matches the Western Digital line very well.
It has less platters at 400GB capacity, 3x133 Design vs WD 4x100 design and less platters at 250GB 2x125 vs WD 3x83. Seagate has SATA2 from 80GB-500GB, where WD is limited to 120GB to 250GB for their SATA2 drives, only downfall is that to get the 16MB buffer is only available at 300GB & higher for Seagate, where WD has a single model SATA2 drive with 16MB Buffer the 250GB model, the other 120GB to 250GB drives are like Seagate limited to 8MB buffer. Seagates 80GB to 250GB for 8MB is competitive. Plus seektime is 8.5ms. Plus for Seagate their drives get a 5 Year Warranty, not like WD which only get 5 Years on it's RE and Raptors lines I believe and 3 Years on the rest.
coldpower27 - Monday, October 10, 2005 - link
Is the Seagate 7200.9 16MB 300GB 3x100 or 2x150 with the advent of 160GB Platter technology from Seagate.
tomoyo - Monday, October 10, 2005 - link
If you look at the table of specs, you'll notice the 300GB has the same wattage as the smaller models including the 1 platter(12.4/12.8/7.2 watts). Also it's got the same acoustics as the smaller models of 2.5 idle and 2.8 seek. The larger 400/500 models have slightly higher wattage and 2.8 idle/3.2 seek. This leads me to strongly believe that the 300GB model is also a 2 platter design using 150GB a platter...and is therefore slightly short stroked on a 160GB platter. I hope I'm proven right, because assuming it's a reasonably quiet drive...it's definitely the one I WANT! Looks like a candidate for the best drive with the features of quiet, a decent amount of storage, SATA 2.5 and a 5 year warranty!coldpower27 - Tuesday, October 11, 2005 - link
Yeah I hope so to, I am so far fond of the 2 platter design drives. I don't really want to a larger number of platters if unecessary.JarredWalton - Monday, October 10, 2005 - link
I believe it's 3x100GB. It might also be 3x133, but only using five heads (i.e. 2.5 platters). Actually, 5 heads seems more likely, but I don't know for sure.There are a few specific platter sizes that are common. Right now, those are (as far as I know):
60GB (old and on the way out, if not gone already)
80GB (fading fast)
100GB (still in production, but diminishing over time)
120/125GB (Relatively new; different brands might go one way or the other)
133GB (only a slight bump from 120/125, really - still new. Some "failed" platters might be formatted for 120/125GB)
160GB (brand new starting with certain Seagate 7200.9)
There is variance to a degree within those sizes. I think some of the 250GB drives, for example, use 133GB platters that are downgraded to 125GB capacity.
semo - Monday, October 10, 2005 - link
why don't some hdd maker sticks 2-3gb of flash memory to a 200gb something hdd.i don't even have to list the possibilities of such a solution, do i?
PuravSanghani - Monday, October 10, 2005 - link
Samsung and Microsoft are actually working on a Hybrid notebook drive which uses OneNAND flash memory currently manufacturered by Samsung itself. The drive, which is still in the development stage, is said to reduce battery use among the speed increases. Google "samsung hybrid" and you shall see.Thanks,
Purav Sanghani
semo - Monday, October 10, 2005 - link
no, thank you.semo
Spacecomber - Monday, October 10, 2005 - link
I guess I was expecting something more in an Anandtech article than what apparently is just a summary of a press briefing.Maybe this article needs to have a title that clearly designates it for what it is.
cryptonomicon - Tuesday, October 11, 2005 - link
They should add some sort of suffix to all reviews that are really product briefings that just lets you know immediatly that it is mainly MFG information.JNo - Monday, October 10, 2005 - link
I appreciate that this is manufacturer info rather than independent review but I for one would prefer to know that such a drive is just around the corner and be able to wait rather than shell out for a 7800.8 drive and regret the timing of the purchase. Don't worry, I'll take the manufacturer info & stats with a grain of salt... I can handle it...ss284 - Monday, October 10, 2005 - link
Agreed. The article states:"In the next few weeks we will be looking at the performance of the 500GB 16MB SATA model of the 7200.9. We'll put it through our usual synthetic, simulated, and real world tests, and we'll compare it not only to the 400GB 7200.8, but also to some older drives like the 120GB 7200.8 as well as Maxtor's and Hitachi's higher capacity drives."
Why didnt they release this article after they did this testing? Obviously they have this drive in their hands already, and theres no NDA. This article should be a flash advertisement on the side of the page instead of a full fledged article.
JarredWalton - Monday, October 10, 2005 - link
"Introducing Seagate's 7200.9: The New Generation"Introducing at least to me implies that it's not a review. We (Purav) have the parts on order and will be testing ASAP. The NDA lifted today, and we did not have product in house beforehand. To me, at least - since I've bought quite a few Seagate drives in the past year - the new lineup is important enough to warrant an article of its own.
bhtooefr - Monday, October 10, 2005 - link
If 4 133GB platters give 532GB, then why not use 4 160GB platters, and get 640GB, and leapfrog Hitachi?You can use four platters.
You have 160GB platters.
Why not do it, and say you've got 600GB before Hitachi?
coldpower27 - Tuesday, October 11, 2005 - link
I think Seagate is comfortable, in thier position of tied for first place, we will probably see a Hitachi 600GB based on the 5 platter design again of 120GB Before we see Seagate match that with a 4x150GB design. Also the other Drive manufacturers need to catch up to this level, namely Wester Digital and Maxtor.
Calin - Tuesday, October 11, 2005 - link
They are now the king of the size, and they will wait until Hitachi launch something bigger until they will launch the 640GB drive.Calin
JarredWalton - Monday, October 10, 2005 - link
The tech to run 160GB platters is new. Right now, they 160GB platter is only being used in the 80GB and 160GB drives, which is telling. I would have thought a 240GB and 320GB would also use the new platters, but that's probably a future product.coldpower27 - Tuesday, October 11, 2005 - link
Also you got to not that Seagate may not want to dilute thier product line with intermidates such as 240GB & 320GB drives. Since noe one has made 240GB drives and only WD to my knowledge has 320GB drives.With 200GB to 250GB, & 300 to 400GB.
Marlin1975 - Monday, October 10, 2005 - link
but what about performance? The slowwest part of a system is the hard drive. I need speed, not 500gigs that only a few will really use daily. I have a pile of shows and movies but that is what DVD's are for. I need speed to make my system run faster. I guess I will stick with my 15K 36gig master boot drive and another one for backup.Maybe hard drive makers will get it one day. Seems WD is the only one to address that for desktops so far?
xsilver - Monday, October 10, 2005 - link
is there some kind of raid array that will allow 3,5,10 drives to be merged as one?seek times wouldnt improve but transfer rates might if the drives could intelligently split up singular file transfers into chunks for each drive... or is that some crazy idea i just dreamed up? ;) (4x 200gb drives stiped would be relativly cost effective compared to 4x or more raptors)
also 16mb buffers seem a bit low for the latest gen drives, 32/64 would be nice for the 300-500gb models.
and I think large cheap drives is where the moneys at now, cant blame seagate for attacking that front.
joex444 - Tuesday, October 11, 2005 - link
it's a dream. you can't merge RAID arrays.However, RAID50 exists. Now, this might be backwards, but: 6 drives. Make RAID5 array w/ 3 drives; make another RAID5 array w/ other 3 drives; make a RAID0 array with 2 RAID5 arrays. So, if you wrote 4 blocks of data, 2 go to RAID5-A, 2 go to RAID5-B.
As far as fault protection, if one drive goes, it's ok. If two drives go it matters from which RAID5 array. From the same array, the whole thing is gone, from different arrays, it's no problem. 3 drives is certain doom, though.
ceefka - Monday, October 10, 2005 - link
More seems tempting. The test with tha 250GB Maxtor showed that streaming likes big buffers.
What kind of RAM is that buffer anyway? Is it ordinairy SDRAM? What would a buffersize > 16MB do? Is there a theoretical maximum and/or sweet spot in relation to the read/write capacity?
xsilver - Tuesday, October 11, 2005 - link
I think hdd's use a one chip ram solution, around 5ns is commonthe thing you have to note is that a normal ram is an 8 chip solution for single sided ram
that's why 64mb is probably the highest they could go economically
and isnt the problem with using flash ram in a hybid the problem of lifespan? nand flash rates at 1 million writes/reads as mtbf? thats probably not enough for a hdd?
Xenoterranos - Monday, October 10, 2005 - link
If you have a ton of HUMONGO filed, say, the complete season of everything volumes 1-100, in Divx, partition the drive in NTFS with a 16mb allocation size, and your data will FLY, especially if the allocation size matches the buffer (or am I spouting off crackpot science?)geoff2k - Monday, October 10, 2005 - link
Googling ST3300622AS already gets a few hits, but no one appears to have any in stock -- when can we expect these at retailers?(Prices so far for the 300 GB SATA are 243 CAD$, 211 US$, 203 US$).
cryptonomicon - Monday, October 10, 2005 - link
"as you cab guess from the name, "first paragraph
ElFenix - Monday, October 10, 2005 - link
"in their 400GB version drive at 133GB per square inch. "unless platers are 1/2 of a sq. inch per side, which i doubt, this statement is wrong.
i'm still having trouble understanding why this isn't in news, rather than a front page article.
JarredWalton - Monday, October 10, 2005 - link
It's not a review, it's an announcement of a new product which we feel many people will find interesting. If I were about ready to spend $100 on a 250GB drive, I'd sure like to know that Seagate has just released their 7200.9 series. This is an introduction to the technology, and an explanation about what has changed. The 16 MB cache models are certainly noteworthy, and all indicators are that the new drives will at worst match the old 7200.7/.8 Seagate models.Price is the only thing I would worry about. If the 250GB 7200.9 costs $150 vs. $100, I probably wouldn't bother. Right now, http://labs.anandtech.com/search.php?q=seagate+160...">prices on the 160GB models are in favor of the new 7200.9. The remaining http://labs.anandtech.com/search.php?q=7200.9">7200.9 drives in our RTPE may or may not be worth consideration. $380 for the 500GB is more than I'm willing to pay for a single drive.
cryptonomicon - Monday, October 10, 2005 - link
argh forget it. i guess i just want to see the reviews perfect for the sake of keeping anandtech the best it can be, so even though a typo doesn't affect the real content of an article (but it still devalues it)joex444 - Tuesday, October 11, 2005 - link
This is true. If this was my first visit and I hadn't read articles here before, the typos really just make the author and the site appear ignorant. People in general, atleast ones that know the correct way, tend to stop reading things if they are written improperly; they just look like a child wrote it.SpaceRanger - Monday, October 10, 2005 - link
I wouldn't call it a review. Read more like a commercial for Seagate.Xenoterranos - Monday, October 10, 2005 - link
You know what. Anandtech has never posted any news that has ever been considered "out of touch". They're reviews are as unbiased as reviews can be without being written by cold, unfeeling machines. If Anadtech decided that these segate drives were good enough to give a free comercial too, then thats the best kind of marketing isn't it? I remember back in the day when a 40gig was huge, I called a local dealer and asked them for the cheapest 40gig they had. The guy said "We'll, that would be a Ninja [something], but I wouldn't recommened that unless you want to come back tomorrow and call in the warranty." Ok, that was a but off topic. Long live the Queen, free tibet, and all that jazz.PuravSanghani - Monday, October 10, 2005 - link
I would actually call this an article detailing Seagate's official introduction of the 7200.9 line. As far as purchasing these drives go, they have shipped in the last few weeks and are in stock at many places already.As we mentioned in the article, we will be providing benchmarks of the 500GB version of the 7200.9 within the next few weeks.
Thanks,
Purav Sanghani
ATWindsor - Monday, October 10, 2005 - link
"The lowest amount of sound audible by human ears is 26 decibels, so the idle noise output is borderline inaudible to our ears"This seems to be taken from the info Seagate has published, unfotantly it is not correct. The db(p)-scale has it reference at the human hearing threshold. In other words, the human hearing threshold is 0 db. (although there obivously are individual differences, and background noise rarly is under 15 db).
PuravSanghani - Monday, October 10, 2005 - link
Looking at the charts, we meant to say 2.5, 2.6, 2.8, and 3.2 bels. The numbers have been fixed to match the slides from Seagate.ATWindsor - Monday, October 10, 2005 - link
I see you also added "according to seagate" which is good.JarredWalton - Monday, October 10, 2005 - link
2.6 bels is of course the same as 26 decibels. (deci = 1/10)That said, people claiming that 26 dB is not quite probably are going by advertising numbers rather than reality. I have an SPL (Sound Pressure Level) guage, and 26 dB is extremely quiet. I can still hear it, but a room with no noise in it ranges from 24 to 30 dB. A car driving by outside makes about 40 to 45 dB of noise.
The problem is that if you trus the noise levels from computer hardware manufacturers, you might actually believe that your PSU, fans, HDD, etc. only put out 26 dB of noise. I've got a case with three "26 dB" fans, plus a "28 dB" PSU. Anyone want to guess at the real noise level? I won't keep you in suspense; at 1 meter form the case, the actual measured noise level is 48 dB, FAR higher than the rating of any of the components. Multiple fans plays a part, but even a single "26 dB" fan still came up with a 38 dB rating on my SPL.
I would agree mostly with Seagate's claim that 26 dB (2.6 Bels) is below the threshold of human hearing. A device putting out 26 dB at a distance of one foot would be inaudible at a range of 10 feet, that's for sure. (Unless you have dog hearing....) Whether or not that's how Seagate rated their drives, though, I can't say.
ATWindsor - Tuesday, October 11, 2005 - link
First of all, a room with "no noise" is much more quiet than 24-30 db, it just that you are used to the background noise. a 26 db fan at one feet is around 15-16 db at 10 feet, which is a clearly adible sound that too (this i have measured in an anecoic room). However, in most rooms the background noise would mask it. 26 db is quiet, but its far from inaudible.JarredWalton - Tuesday, October 11, 2005 - link
We don't live in sound-proof environments, though. While it may not be truly silent, an empty house without any noise from neighbors or traffic is still going to have a background noise level above 20 dB. If something is below the background noise level for a house, I'm willing to call it "silent" - even if it isn't technically dead silent. I guess that's a difference of opinion, though.JarredWalton - Monday, October 10, 2005 - link
"26 dB is not quite" should be "26 dB is not quiet". Still need that edit function.... grumble.DRavisher - Monday, October 10, 2005 - link
Indeed. Though I doubt anyone can accurately detect even a 1dB sound :).ATWindsor - Monday, October 10, 2005 - link
That of course, can be discussed, but most people would easily detect 26 db, or 20 for that matter. And saying that 29 db is "barly audible" is a far strech IMHO.DRavisher - Monday, October 10, 2005 - link
I agree. You can certainly hear 20dB+ sounds. Especially if you were to set up a completely closed environment (no background noise that is).Xenoterranos - Monday, October 10, 2005 - link
OK, there's a period there. they said 2.6 thats "2 point 6"DRavisher - Tuesday, October 11, 2005 - link
They said 2.6 bels, which equalts (does it not?) 26 decibels (dB).Tiamat - Monday, October 10, 2005 - link
Agreed. At 2AM (when I am sleeping with my computer idle 10 feet away) I can hear its 28dB and its not very quiet. I would say in the dead of night, a computer would have to be roughly 18dB to be considered unhearable above typical background noise.However, there seems to be a mix-up between Sound power and Sound Pressure - two totally different entities.
http://en.wikipedia.org/wiki/Sound_power_level">http://en.wikipedia.org/wiki/Sound_power_level
http://en.wikipedia.org/wiki/Sound_pressure_level">http://en.wikipedia.org/wiki/Sound_pressure_level
ATWindsor - Monday, October 10, 2005 - link
The best would be if they reported sound power, as sound pressure depends the enviroment, and the power is constand for a given sound source (and you can calculte the pressure at a given distance in a given room). But it's almost the pressure who is given in computer-components. And the ear "hears pressure" so for the hearing its more useful to talk about the pressure.jkostans - Monday, October 10, 2005 - link
What do HD manufacturers call 1,000,000,000 bytes a GB?ATWindsor - Monday, October 10, 2005 - link
Because the SI-system (which is centuries old), clearly defines Giga as 1 000 000 000. The fact that some software doesn't follow the standard can't be blaimed on the HD-manufactureres.rendezvous - Monday, October 10, 2005 - link
Because the SI-system (which is centuries old)Centuries as in from 1960?
http://en.wikipedia.org/wiki/SI">Le Système International d'Unités @ Wikipedia
ATWindsor - Tuesday, October 11, 2005 - link
The modern version of the metric system is from the 1060s, so i guess i was abit unclear, however, the metric system itself, which the SI-system is built upon is fomr the 18th century.JarredWalton - Monday, October 10, 2005 - link
Basically, the problem is that KB, MB, and later GB, TB, etc. all came from ealy computing days, where 2^10 was close enough to 1000 that they abbreviated it to KB. Later on, SI came into being and really got pissy about the use of "Kilo" for "1024" rather than "1000". The hard drive manufacturers are of course using the multiples of 1000 because it makes their product look better. Why say 93.13 GB when you can say 100 GB?SI later proposed the "kibi, mibi, gibi, etc." prefixes to get around the discrepancy. In reality, few companies are using these terms at present. That may change in the future, but in truth most people don't care. We mention this discrepancy more so that people are aware of why the difference is there than to place blame.
jkostans - Monday, October 10, 2005 - link
So you're saying windows reports "gibibytes" not gigabytes?ATWindsor - Monday, October 10, 2005 - link
Correctgeoff2k - Monday, October 10, 2005 - link
So they can make smaller drives?Well, that and:
http://physics.nist.gov/cuu/Units/binary.html">http://physics.nist.gov/cuu/Units/binary.html
Zar0n - Monday, October 10, 2005 - link
I have one 7200.7 SATA, so far no problems but seek noise is loud, I was waiting for 7800.9 to get one 500GB drive.Please make some acoustic tests when u review 7800.9.
Also with NCQ ON/Off, the 7200.7 sometimes was slower with NCQ ON.
Bench with Maxtor, WD, Hitachi and Samsung drives would be nice.
Lifted - Monday, October 10, 2005 - link
Hopefully this will bring down the price of the 500GB drives a bit. Running out of space for TV shows, and I already have a couple 400's and 250. Anybody find a RAID-5 external enclosure yet that holds 3 or more drives? I've been looking for one for a while now. I'm getting nervous with all my stuff not being RAID or backed up.Visual - Monday, October 10, 2005 - link
Um, the hotplug/hotswap feature sounds interesting, is it supported on any of the integraded sata controllers? I know my nforce4 has some "safely remove hardware" icon for my drive, but im quite sure the hdd doesnt support such a maneuver. Its quite confusing...