Frisbee

Distinguished
Oct 22, 2002
87
0
18,630
Hello;

I would like to see a test at Tomshardware which shows at what point 5400rpm drives with high data-density per platter become faster than 7200 rpm drives with lower data-density.
To test this, we should make sure that all drives tested have the same amount of platters and heads. The total storage capacity is of no importance to test this.

For example:
If we would test 4 drives:
1. a 7200rpm drive with ONE 20Gb platter and 2 heads
2. a 7200rpm drive with ONE 40Gb platter " " "
3. a 5400rpm drive with ONE 60Gb platter " " "
4. a 5400rpm drive with ONE 80Gb platter " " "

Would the 5400rpm drives perform better than either of the 7200rpm drives which have lower data density?

I like quiet machines. That's why I would like to know.
I currently have an IBM 13,5Gb 7200rpm harddisk, which is rather noisy.
I was thinking: Current 5400rpm drives have data-densities of 40, 60 or even 80Gb per platter. COULD IT BE THAT THESE DRIVES ARE FASTER THAN MY CURRENT 7200 RPM DRIVE? Although they have a lower rpm, their data density must make up for this at one point sooner or later, right?

Cheers from Belgium;
 

Napoleon

Distinguished
Aug 17, 2002
331
0
18,780
Well, density (BPI) increase only helps with sequential access. Another thing to pay attention to is the increased seek time caused by rotational latency. For a 5400rpm drive, one spin would take 60s / 5400 = 11.1ms. For 7200rpm drive, one spin would take 60s / 7200 = 8.3ms. Assuming that on the average the latency adds half a rotation to the seek time, "real life" seeks would be about 1.4ms slower with the 5400rpm drives. If you have lots of random access going on, the increase in rotational latency becomes a significant factor.

To my understanding, only drives offering 80GB/platter density are the 5400rpm Maxtor drives. If you set them against 60GB/platter 7200rpm drives, the simple ratio comparison would be 80:60 vs 7200:5400, that is, 4:3 vs 4:3. No gain in transfer rate, but slower seeks.

Check out this <A HREF="http://www6.tomshardware.com/storage/02q3/020920/samsung-06.html#drive_idle_noise" target="_new">THG noise chart</A> about 5400rpm Samsung drive and compare to 7200rpm Barracuda IV. Difference isn't all that huge, especially if the Barracuda IV in question is the 3 platter 120GB version (the 120GB Samsung drive has 2 platters). Too bad it only says "Barracuda IV" in the chart. Informative... not. Anyway, if you want peace and quiet at a decent performance, you might want to wait for a 60GB Barracuda V. Pricewatch ETA is late November. You'd get 60GB/platter, 7200rpm, SATA if you want it and 8MB cache. No 3-year warranty, though. Or a huge performance advantage over Barracuda IV. :frown:

To my understanding, Barracuda V will be about as quiet as the Barracuda IV. Got a 60GB Barracuda IV myself, and IMO it is very quiet. Also, seek noise (in Quiet Seek mode) is very low-key too. Downside is, at least Barracuda IV gets <A HREF="http://www6.tomshardware.com/storage/02q3/020920/samsung-05.html#temperature" target="_new">quite hot</A>.


<font color=red><b><i>You want WHAT on the [-peep-] CEILING?!</i></b></font color=red> -Michelangelo
 

lhgpoobaa

Illustrious
Dec 31, 2007
14,462
1
40,780
Interesting test to do, though very hard to implement in practice.
Besides, the test will really only be valid for sustained transfer rates, and we all know that 7200rpm drives have lower rotational latencies, faster seeks (and more seeks per unit time).

<b>And if you gaze for long into Toms Hardware Forums, The Forum gazes also into you! :eek: </b>
 

Frisbee

Distinguished
Oct 22, 2002
87
0
18,630
Thanks for the answers folks.
The REAL reason why I am interested in 5400rpm drives is their quietness. A month ago I installed a Maxtor D540X 5400rpm drive in an old computer of a friend, and I was amazed: you simply don't hear it at all!
The second reason may be that I think that higher rpm causes a little less reliability.

So I just want a drive to be quiet. I don't care if it takes Outlook or Word 3 seconds to load. I wouldn't even care if that would happen on a 10Ghz cpu.

About that Barracuda drive you mentioned: I had already heard that they are quiet for a 7200rpm drive. But I installed a Maxtor D740X 7200rpm drive in my mom's pc, and although it is a lot quieter than older drives, I still can hear the wining noise, and I really hate that noise, more than fan-noises. I don't think the Barracuda is THAT much quieter than that Maxtor drive, right?
On the day that they wil make 7200rpm drive which are as quiet as today's 5400rpm models, I will buy one. But as of now, I have never ever seen any 7200rpm drive which is as "noiseless" as the Maxtor D540X.

The only question I still may have is: does anybody know if these newest 5400rpm drives with 60 to 80Gb per platter should be fast enough to record from tv to Mpeg2?
My cpu should be fast enough (Celeron 1.2Ghz), and I am thinking about buying a tv card/encoder.

Cheers from Belgium;
Carl
 

Napoleon

Distinguished
Aug 17, 2002
331
0
18,780
<i>Editing a little. Some links were bad, and God forbid, there were typos...</i>

That reminds me, we had a thread about <A HREF="http://forumz.tomshardware.com/hardware/modules.php?name=Forums&file=viewtopic&p=57175#57175" target="_new">Quietest HD</A> a while back (<i>arr, seems that I can't get the "show all" link to really work, no matter what I try).</i>

StorageReview is a good place to look for HD noise measurements. Their results are only comparable within the scope of Storagereview articles, but those folks are very <A HREF="http://www.storagereview.com/articles/200111/20011109Renaissance_10.html" target="_new">clear and methodical</A> about their noise level measurements.

D740x is hardly a good representative of a <i>quiet</i> 7200rpm HD. Take a look at various SR charts, Barracuda V (preview, 120GB, two platters) measures at <A HREF="http://www.storagereview.com/articles/200210/20021014ST3120023A_6.html" target="_new">37.8dBA</A>, Barracuda IV (80GB, two platters) is 41.3dBA whereas D740x (80GB, two platters) scores a whopping 46.3dBA. Set against the Barracuda V, d740x is loud as a whistle!

For reference, Seagate U6 (80GB, two platters, 5400rpm) scored 38.8dBA and 536DX (100GB, 3 platters) scored 41.5dBA. With two platters, it probably would score something about 40dBA (my guesstimate). D540x has 40GB/platter, whereas 536DX has 33GB/platter.

All things considered, chances are that if you want a <i>really</i> quiet drive, some single-platter 80GB 5400rpm Maxtor drive would probably be ideal for you. Assuming that Maxtor is/will be manufacturing such a drive. If not, I'd look/wait for single-platter 60GB, 7200rpm, SATA, 8MB cache Barracuda V for a quiet performer.

Of course, if you're really sensitive to high-pitched whine, (Storagereview's) A-weighed dB measurements might not be putting quite enough emphasis to the highly annoying 1kHZ - 5kHz band. Then again, if you're <i>that</i> gung-ho about noise and reliability, take a look at 2.5 inch laptop drives and 2.5 -> 3.5 mounting brackets. :smile:


<font color=red><b><i>You want WHAT on the [-peep-] CEILING?!</i></b></font color=red> -Michelangelo<P ID="edit"><FONT SIZE=-1><EM>Edited by Napoleon on 10/23/02 04:04 PM.</EM></FONT></P>
 

Frisbee

Distinguished
Oct 22, 2002
87
0
18,630
Woah! Thanks for all the info.
Maybe I'll wait for the Barracuda V, and read some tests about it. I'll check that storagereview-site too.

Hmm, I only hope the Cuda V will be less hot than the IV, I'm not really fond of "hot" electronics. Tends to be less reliable, and I don't want no harddisk that will go bananas within the first 5 or 6 years. You've probably read about the new 1-year warranty too. Therefore, the 3-year warranty of Western Digital's "Special Edition" drives attracts me.

So in short, I haven't really made up my mind yet. I'll keep on reading some tests, and see what that Barracuda V really is all about.

Thanks!
 

lhgpoobaa

Illustrious
Dec 31, 2007
14,462
1
40,780
Heh. As with evrything in life, there are allways trade off's between heat, noise, reliability, capacity and speed.

The JB series are very zippy, however like any 7200rpm ball bearing drive they do emit their fair share of noise.
To me its not a problem as it blends intot he background and really isnt too bad. What i dont particularly like are seek noises, and the seeks of my maxtor are very loud and scratchy compared to the nearly inaudable seeks of the 800JB i have.

P.S. heat shouldnt be a problem if you arrange your case right.

And if you expect your hard drives to all last 5-6 years i think thats expecting a little too much for an IDE drive. They simply arnt designed to live that long. Why do you think the warantee's were only ever 3 years?


<b>And if you gaze for long into Toms Hardware Forums, The Forum gazes also into you! :eek: </b>
 

piasabird

Distinguished
May 16, 2002
140
0
18,680
Take 2 5400 drives and put them in a raid configuration. I dont know if raid supports 5400 drives or not. I think I heard someone did this with 2 2.5" laptop drives.

when you see smoke is that a good sign?
 

piasabird

Distinguished
May 16, 2002
140
0
18,680
I think at the IBM hard drive site they had some utility software for their hard drives to adjust the characteristics of the drive for better accoustics.

when you see smoke is that a good sign?
 

Ncogneto

Distinguished
Dec 31, 2007
2,355
53
19,870
There is one other item that needs to be addressed when dealing with seek times. That would be the size of the media ( platter). For instance, modern 15k drives have 2.5" platters while 10k drives have 3.0" platters. Thus, not only is the platter spinning faster on a 15k drive but the distance that needs to be traversed for a full or even half rotation is less.

It's not what they tell you, its what they don't tell you!
 

Napoleon

Distinguished
Aug 17, 2002
331
0
18,780
Yes, IBMtool has that feature. Look up the link for "Quietest... HD" in my previous post(s) in this thread, I posted about it in that thread. I have downloaded the utility and used it on my 60GB Barracuda IV with the (old) 3.10 firmware. Worked like a charm. Basically, I had the option to use either Quiet Seeks (default of my 3.10 at least) or Performance Seeks. Perf. seeks were about 0.8ms better in my benchmarks, but the difference in seek noise was <i>clearly</i> audible, at least to me.

___
<font color=green><i>Feet are frozen, hair's on fire, on the average, everything's fine.</i></font color=green>