I’ve been looking at the Cavier SE WD3200JS in comparison to the WD5000KS under discussion here. Apart from the 2 dB drop in noise and half the cache size, it’s also likely to have three 107GB platters versus the four 125GB platters of the larger unit. So performance will be down and especially if they’ve tweaked the firmware to gain the lower noise floor. It seems a bit risky to buy one of these without reading a review.
The odd thing is that WD list exactly the same power consumption figures for the SE16 500GB as they do for the SE 250GB. This seems odd as even the largest capacity SE (320) should surely consume less power than SE16 500GB!

Jazz: We're not saying there is no subjective difference. We're saying it's very subtle. This quotation from the article sums it up nicely:



From the sound of things, you already have a very quiet system, and would be better off with a notebook drive if you want to minimize drive noise any more.

With respect to the subjective difference in the recordings, I agree with Mike that we probably need to think about changing the recording distance and possibly the microphone? A properly used shotgun microphone would help eliminate background noise sources with a minimal impact on frequency response. As it is right now, the near-field recording technique simply isn't working as well as it could be. Our review of the 7200.9, in which we couldn't hear the high-pitched whine in the 3" recording demonstrates this perfectly.

ROFL, this review pops up, right after I got a V80!

Changing the microphone would be too big a break from the exisiting database of sound recordings, as it would make reasonable comparisons between older and newer recordings impossible. The character of the omni mic is very open and quite neutral; any directional mic will add colorations. I think we need to experiment with recording distances, first of all.

A directional microphone would just plain be a bad idea. It would introduce the extra variable of where the mic was pointed.

I already view it as a difficult problem that most devices produce sound with directionality (especially fans). I've not complained about it since I can't think of a good solution to that problem.

MikeC and Devonavar
I know it is sort of pointless arguing about what it would possibly sound like without having both drives at hand, however I just want to say that my reasoning is this, if I can hear my 2504C inside the case, then I will probably hear the difference between Samsung and WD.

PS and I cannot use/don't want to use laptop hard drives because they are too slow compared to desktop ones.

I agree. That you would hear a difference was never at issue. How noticeable the difference is, was. And, as Mike and I have said numerous times, the difference is not as dramatic as the difference you can hear in the near-field recordings. We have both done our best to describe the difference in real-world, subjective terms. I am of the opinion that the very slight downgrade in subjective noise is not enough to outweight the huge advantage of its excellent seek noise, especially when compared to recent Samsung drives that aren't that quiet when seeking.

Sorry, having both (well, the WD2500KS and SP2504C) the WD is quieter--but I agree that sample variation (especially in a negative direction in the Samsung sense) is a very big factor.

Kudos to WD to improving quality.

I seem to remember (at least since 4GB IDE drives) Samsung generally building cheap, flaky, poor-performing drives. It is nice that their current drives perform much better, but consistent quality would be better.

Quality? I still can't get over 3200JD fiasco. It would be a while before I buy another WD hard drive and if I do it won't be going into my main rig, only into my fileserver because I don't want to risk another flop.

Was trying to decide between this and the Maxtor Pro 500GB and your review sealed the deal. Wanted something very quiet and the $220 price tag (.44/GB) was the clincher versus the $300+ for the Maxtor (.61/GB)

As to quality, I'm kind of hesitant on WD but then the drives I've been fortunate with (Maxtor and Samsung) have been slammed for reliability so what the hell do I know, besides the Raptor is a workhorse (different animal I know but maybe it'll rub off)

I just exchanged my Samsung SP2504C with this drive and i'm very happy. The noise difference from my htpc is very little. I've suspended this disk as well with rubber strings.

And now i have twice as much space, fantastic!

Is it louder or quieter?

I'd have to say it's the same, although i think it maybe vibrates a little more. I'm happy with it.

Does anyone know where to find the 00MNB0 revision in Europe ?
I've been waiting for so long for such a drive, and now I can't even find it

Three questions...

Does vertical vs horizontal orientation matter when it comes to sound?

Should I use o-rings or something to dampen the vibration?

I just bought one online for use in a NAS device... if it is not the same revision should I send it back?

Thanks for a great site, I just found you thanks to the link from WD's site.

If you were to begin using a directional mic - and, you're right, it wouldn't be a good idea - securely mount a laser to the microphone in a manner that won't strongly effect the sound [at these SPLs, it's not that much of a concern, so long as you're not stupid about it] and direct the laser at a given point on each drive, for instance, the hub. [We do this for repeatable drum micing in some situations.] Then make sure you're recording from a distance great enough to minimize the coloration of omission.

But, you know, just don't. That'll be simpler.

Are there any differences between the 2500KS and the 5000KS besides capacity? i.e. Is the 5000ks newer and fix flaws in the 2500ks, or were they released simultaneously and made the same (excepting capacity)?

I've been reading SPCR for quite some time and it has always been the best information source I could find on HDD noise. Each day I'm using many HDDs in several PCs, some of them built to be the quietest I can get; obviously using many hints found on SPCR. At home I'm using Nexus Drive-a-Way enclosures for 2.5" and 3.5" drives (the best enclosure I found; SPCR review sadly missing), undervolted Nexus fans, modded Antec and Thermaltake cases, acoustic foam, fanless Yesico PSUs, fanless motherboards and GPU cards etc. By the way I sometimes spend a night taking sound recordings with pro/semi-pro hardware when I'm modding a PC. (..and, yes, girlfriend is asking me to enter therapy)

The review about the WD5000KS says "The 500 GB Caviar SE16 is the quietest 3.5" desktop drive that we know of on the market today".
I guess this doesn't take platter count nor gigabytes into consideration, only talks about 3.5" 7200rpm drives, and takes into account some average on measured idle/AAM/no-AAM noises, plus vibration rating.

Listening to the mp3 samples, I agree the WD5000KS seek noise is amazingly better than the 3.5" competition; however I prefer to focus on idle noise level alone, simply because it's what I'm hearing most of the time when using a 3.5" drive. I only hear worst seek noises when searching a file within a 20,000 files collection and a that time I'm ok (to some extent) to hear seek noise for a few seconds time.

I made AB comparison listenings of the WD5000KS versus the Samsung SpinPoint P120 SP2504C (cause I'm quite used to SP2504C and SP2514N noise signatures, having several of them; it's been my favorite 200-300 GB for the last two years).

Well, drive capacity not being taken into consideration, I clearly prefer the SP2504C idle noise. The WD5000KS has _much_ higher presence on mid-high frequencies, and this is where I'm most sensitive regarding HDD idle noise. I think this doesn't show enough in dB/A values. I made some freq graphs printscreens that you can see here http://idlenoise.ifrance.com/old.

So, as some other people here I found the WD5000KS idle noise mostly average versus the competition. It would be interesting to have some SPCR review of a 3-platters Caviar SE16 (250 or 320 GB); maybe buying it anonymously in a store could help too.

alfred
Interesting analysis. I'll be sure to listen to the sound samples when I get home, but your graph seem to support spcr recordings that WD500 has higher pitch idle component. This definitely is a important as some frequencies are more annoying than others and this was my concern that I would hear the harsher signature of WD500 over smoother noise of 2504c that I voiced to MikeC and Devonavar on the first page of the thread.

The question is how much more audible that higher frequency component of WD500 drive is going to be once the drive is inside the case.

So I have couple of questions for you. First, you say you much prefer the 2504c sound over WD500, but how did you compare them? Did you compare them in an open case 0.5 meter away from you, or did you compare them mounted inside the case (and what case at that, were the drives softmounted, etc, etc...) from your normal sitting position. In other words can you describe your setup? Two if you compared them inside the case, how exactly did you find the difference between them? Was the WD500 really that much louder than 2504c? And third and the last question, how did you record the sound samples? Were the drives inside the case or not, and what distance the microphone was from the drives?

Hi JazzJackRabbit,

I'm sorry I didn't make this clearer -- I didn't compare these two drives at home, the AB listening test I wrote about was based on the mp3 files from the SPCR reviews. The flac and wav files you can get on the link simply are 6-seconds audio files that alternate 1 second Samsung (left channel), then 1 second WD (right channel), three times; the goal was to help anyone to compare idle noises. BTW a simpler method to compare two audio tracks is to use the free Foobar 2000 audio player (get it here http://www.foobar2000.org/), it has an ABX tool built in.

So.. my first post simply rise a question : why do I get this ±1.8dB gap in computed average RMS power, based on ±7sec samples from SPCR's own mp3s, when SPCR displays idle noise at 21dB for both drives ? I know there are many technical parameters to take into account and that these could explain some differences in the values we get, but my point is : a ±1.8dB gap is clearly significant and it explains the perceived amplitude difference between the two idle noises from SPCR's mp3s.

However
Today I just bought a WD3200KS (manufacture date 08-JUN-2006); this is the 3-platters Caviar SE16. I'll soon make some recordings of this one and a P120 in the same environment to get some comparison point.
And I'll get a Samsung HD401LJ (400 GB, T133 series, 3-platters) delivered friday or next monday, so I'll test this one too. I'll try to setup two or three different test environments (free air, inside case, inside Nexus Drive-a-Way enclosure..) and give some pics and technical details (PC and audio hardware specs, etc).

This is excellent. Maybe in many ways we (the SPCR community) have become lazy because of the high-quality of SPCR articles and we rely too much on the editorial team to spoon-feed us quiet components; this is like back to the early days of the site when individual forum members took more of the initiative to find quiet components; anything which adds to the pool of knowledge on the site is to be applauded.

alfred
OK. That was funny. You didn't clearly state that you just analyzed spcr sounds samples and I got confused even more by your flac sound sample (I couldn't listen to them at work so I assumed you recorded your own samples). Heh, I hoped there would be someone else who could support my position.

Well, in any case I'm looking towards your own comparisons.

If you read through the earlier comments in this thread, you'll find that the issue of the mp3 recordings has already been discussed in detail. Long story short, the mp3 recordings are not representative of what Mike and I heard when listening to the drives themselves. The difference was enough that we have re-examined the methodology that we use to make recordings. Look for an article with details (and new recordings of the WD5000KS) in a couple of days.

The short answer is very simple: A weighting, measured at 1m, not flat (unweighted) dB at 3" . Devon has already stated that the recordings do not correctly depict the relative noise from a "normal" position/distance. Analyzing those sound files can only give you more detailed info about this artificial inaudible difference.

This is good news indeed
But does it mean you now think there was something misleading in the mp3 recordings of this review ? And maybe older mp3 recordings too ?

Misleading only if you isolate the MP3 recordings and use them exclusively for assessment. The recordings are one of many components of the reviews that you should pay attention to. As I have said (repeatedly!), the close mic position replicates our own listening methodology when we are trying to listen for essential differences, and it emphasizes all kinds of details you would not necessarily hear from farther away. It is not at all misleading if you understand what the 3" recording distance means in terms of perspective, and if you take in all the rest of the review -- especially the analytical narrative -- which I have to insist are the best and most essential part of any SPCR review. The rest are tools to support our interpretation of what we hear.

MikeC I agree with you on all these. I've always taken the review comments as the main evaluation information. When reading the WD5000KS article I was somewhat intrigued that my listening of your mp3 didn't entirely correlate to what you described; I now understand the higher-than-real-life mid/high frequencies presence is solely due to the fact your mike is being placed so close above the HDD.

By the way I've started doing some HDD noise personal audio recordings and up until now my results entirely correlate with your tests. I'll post all details about my tests on http://idlenoise.ifrance.com/recordings

Just finished testing SP2504C, all details on the aforementioned link. It really does sound less harsh on mid-high frequencies, compared to my WD3200KS. Listen for yourself I guess selecting one drive over the other is a question of taste, maybe not everyone is most sensitive in the same frequency band as me. Also we have to remember that the WD3200KS gets close to the low noise floor of the SP2504C with 3 platters instead of 2.

I'll be testing the Samsung HD401LJ as soon as it gets delivered.

Alfred, so the WD3200KS is a quiet HD? because my 300GB Seagate 7200.8 is KILLING me, especially the unbelievably high idle noise and seeks that sound like someone ripping apart cardboard.

shimq1 -- my WD3200KS is nice yes, I've put it in a Nexus Drive-a-Way now and it quietens somehow the slightly harsh idle noise. It does heat more than a SP2504C however, I'd say about 2°C or 3°C. And I still prefer the smooth low idle noise of the Samsungs.