Hello,



Is that the original Solo suspension cords? That have not broken -- yet?

WinXP on FAT32 or is there a way to offload the NTFS journal file?

I have no WinXP pagefile and all my WinXP Temp directories are on the RAM drive. Don't know about the XP journal file. There does seem to be some writes going on in XP that I can't move.

Let me know if I can move anything else.

I have no common drives between WinXp and Ubuntu. If I want to move anything I use the NAS as a temp store.

For Ubuntu, both /tmp and /var are on the Raptor. PLUS the / is ext2 formatted - has to be on the Mtron. /tmp also ext2 but /var is ext3.

About the NTFS metafile frenzy fetch Sysinternal's Process Monitor, watch file access and delete all default filters. Lot's of acesses to $LOGFILE and $MFT and so if I remember correctly (on Linux currently).
About ext2 have a look at the noatime and nodiratime mount options.

Already have noatime - no need for nodiratime (subset of noatime).

Thanks for the info re Sysinternal's Process Monitor. WinXp is now no more than 10% of my use - mainly iTunes and the odd game.

I hope the Bare Velociraptor fits the 2.5" Scythe enclosure with minimal modding. I'd then put the 2.5" Scythe into a 3.5" NoVibes. Heheh. Overkill, but why not?

I would like to know if the production Velociraptor will have some kind of "Void Warranty" sticker or something that would prevent one from removing the drive from the Icepack Noise Amplifier.

Because then I'd just wait for the OEM version to be available.

The Icepack is a completely boneheaded move on Western Digital's part.

I haven't seen a retail box version of the velociraptor listed anywhere. My OEM from newegg does not have any void warranty sticker.

I disagree. While the actual cooling functionality can be questionable, it would be nearly impossible to market a 2.5" drive for desktop PC's that wouldn't fit into a standard drive bay.

MikeC: Can you give us some insight into why on earth does removing the enclosure make it quieter, and how is the situation affected when comparing those 2 ways when they are actually mounted to a case in any of the common ways?

I just don't do warranty-killing mods anymore and would really like to understand this better.

I haven't mounted the drive in a case, this is not normal tes procedure. But I can try it and report back.

The change in acoustics was explained briefly in the review. The close coupling of the HDD to the HS allows all the vibration to go into the HS -- literally, to "sink" into the aluminum frame. You may have noticed that vibration with the HDD in the frame is extremely low, the lowest we've ever seen. This is good in a way, but it also isn't good in another way -- the heatsink itself has lots of relatively unsupported parts: The ends of all the fins. Those ends are where the audible noise is coming from, AFAIK. I haven't experimented enough to know whether this is true. I can try sticking the whole HDD + frame into a NoVibes and see what the end result is... and maybe appiy pressure with the eraser end of a pencil on all the various parts to see which ones are actually making the resonating noise.

OK, after another 45 mins of playing the with stuff, I've concluded that the entire aluminum frame ends up resonating at all frequencies, and there's no easy way to stop it once it starts. Higher pitches noises are especially bad -- the whole range 1.5~10kHz gets full of tonal peajs you can easily hear. I'll post those up into the main article later along with a new MP3 file -- no time right now. Putting the who thing into a novibes barely did anything. It needs to be inside one of Scythe's rubber boxes, methinks.

But take it off the frame and it's as smooth and sweet as a baby's cheek.

What about filling the fins in with some silicone caulking (in the event that there is a warranty sticker attached to the IcePack?

No warranty sticker on mine.

That is good news for me since the 2.5" NoVibes is nearly impossible to find for sale in the USA.

Hoping it fits in the 2.5" Scythe Box. Then put the Scythe Box in the 3.5" NoVibes III.

So do I understand this correctly now: the heatsink is just badly designed and flimsy, resonating just at those frequencies that the drive generates the most? But wouldn't a regular computer case be much much worse yet for those resonations?

It's not flimsy at all. It weighs 320g, more than the drive itself. But it does resonate.

Here's a screen capture of the freq spectrum w/ the drive in the frame:



Note the peaks centered at 700Hz, 1.5~2.2kHz, and 7kHz. These are all audible from a meter away.

Now here's the freq spectrum w/ the drive by itself:



Note all the reductions in the range 1.5~8kHz.

You can also hear this difference here in one MP3 recording -- 10 seconds with frame, 10 seconds without.

All this is now on page 3 of the article.

To repeat -- I could not subdue the additional noise of the frame with a NoVibes, or even with my two hands holding the drive+frame. (well mostly it was still there... just changed.)

In my opininon these SSD drives are coming very fast. Maybe next year the price will decrease to that level, when you consider buying a 16/32GB very fast SSD with 150MB/sec read/write, 0,1ms access time as system drive.
An example configuration:
-16GB Mtron 2.5" SSD (now approx 300$)
-Storage hard disk like a 1TB Green Power (now approx 160-180$)

16/32GB SSD-s are enough for system drive in most cases so the only question is the price drop pattern of these drives next year.

Thanks!

I really wanted one of these drives, but when the time came to put my money where my mouth was, I picked up one of the new WD 640AAKS drives.

Is it really worth the money?

Mike: Thanks for the clarification. So to make it silent, I have to take it out of the frame and soft mount it? I mean, if the thing touches any metal even without the frame, it would be really bad right?

Are you talking back to WD about this issue? Maybe they fix it.

One other thing I don't understand:

How come the average latency can be so much worse? Isn't it supposed to be a linear function of the spindle speed?

Seek time tells you how long it takes for the drive head to move to the correct position, and then the latency tells how long it takes for the spindle to spin to the correct position afaik.

For every 10krpm drive it should therefore be 3ms average (half rotatation), 6ms max (full rotation)?

regarding latency: I'm simply citing what's on WD's website. Their press briefing only highlights the improved seek times, doesn't mention latency. All the other specs appear to be substantially improved.

I have been talking with WD, and forwarded all the info in that last post about the noise w/ and w/o the frame. The rep expressed great surprise, and the info has been forward to their engineers. We'll see if anything comes of it.

As for how best to use the WD -- that's certainly the way I'd use it, but you may be perfectly fine w/ it in the frame. Not everyone hears the same way...

Went and bought a Velociraptor just now. This review certainly made the sale for me.

That picture is from http://www.silentpcreview.com/article272-page4.html Submitted by Mike Chin on Tue, 2005-10-11 so I don't think they'll ever break as they are frozen in time by the camera

I'm surprised you didn't notice the SPCR watermark on the image.

ok.. oddball question here.

What is the seek noise like on this thing?

It sounds like you can get rid of the whine by removnig it from the heatsink -- which is good.

But when/if I get one i want to mount it in such a way that I HEAR the seek noises but no annoying whine.

You can hook up the motherboard's hard drive light header to the PC Speaker, if you want.

If AAM is kept off, the seek noise is audible even when suspended -- if your ambient is low enough and the PC is close enough. This is true of 99% of 3.5" drives... tho a few may be slightly better than the VR.

Mike,
Great review. I've always wanted to go 2.5'' to save space, especially now since I'm using an NSK1300, but I've been tentative to give up the quickness of a raptor drive. The new VR's sound great, but I'd rather not spend so much. Did WD give you any indication of a release date for the 150GB version of the VR? There's a PDF on their website that includes a 150GB VR model in their product lineup, but no indication of price/availability.

What I find interesting about the two attached charts (SPL with and without the heatsink) is the mass dampening effect.

As was pointed out, the high frequency (700+ Hz) dropped by removing the mass, while the low frequency (300- Hz) went up with the removal of the same mass. It's probably easier to hide the low frequency noise so it makes more sense, but as a pure exercise in showing real world mass dampening effects it's an interesting tool.

It's more revealing if you watch the freq. spectrum live. There's less boost in the lower freq range than those snapshot images indicate. Remember they are moments frozen in time. The next moment will not show exactly the same graph because HDD noise is not constant. Try running the MP3 file on a software that lets you analyze it -- actually a 30-day trial copy of SpectraPlus would work nicely -- and you'll get more insight into it all. Also, just listening to the sound file, you can't hear much difference in the low end, but you sure can hear the effect on the mids and highs. That's the way it is live, too, as I hear it.

Mike - well someone in WD was listening, according to this article about WD's work on a 20,000 RPM raptor: