600GB Velociraptor MIA??

[hugodrax]

They are out of stock everywhere I look. Newegg,macmall etc..

Is there a new model coming out and this is why they are out of stock.

the 600GB seems like a nice fast drive and plenty of storage for my needs, but MIA it seems.

[Hellhammer]

WD might have issues with its production. Anyway, why VelociRaptor? It's hotter, noisier and more expensive.

[hugodrax]

WD might have issues with its production. Anyway, why VelociRaptor? It's hotter, noisier and more expensive.

faster, look at the iobench scores. Pretty significant difference especially with random I/O, even over the 2TB WD drive.

[Major Reeves]

WD might have issues with its production. Anyway, why VelociRaptor? It's hotter, noisier and more expensive.

It's faster, not hotter actually it generates less heat than a normal 7,2k rpm hd, more reliable and they can actually "be" used in raid configs.

Anyways, iMac board that way ->

[300D]

The only selling point is has over an SSD is $/GB.

[nanofrog]

The only selling point is has over an SSD is $/GB.
The Velociraptors are Enterprise grade disks, and are suited to high write environments, unlike MLC based SSD's.

[300D]

Which changes nothing about what I said.

[JesterJJZ]

A SATA3 Velociraptor would be very welcomed. Maybe they can even bump it to 15k. Don't care much for capacity as these would be used for boot. 300GB is more than perfect.

[hakuryuu]

Which changes nothing about what I said.

Longevity in a high write environment is a huge advantage over SSD.

[nanofrog]

Which changes nothing about what I said.
The use of "only" in your statement does change things, as it's an innacurate statement with the inclusion of that specific word.

MLC isn't meant for high write environments. MLC on it's own, is only good for 1E4 writes from the manufacturers, such as Samsung, Intel,... It's just the nature of the technology. SLC is a bit better, at 1E5 writes (intended for enterprise SSD's). These are minimum write cycles BTW.

Wear leveling improves matters by remapping other cells when one dies, but there's limitations (available unused capacity, and it can't actually improve the write cycles of the cells themselves, as that's fixed by the technology used to create them). It's a rotation scheme. Capacity decreases when cells are remapped. Minor per instance (assuming there's not a massive cell count lost in a short period of time, which can happen, such as an entire chip), but it will add over time. If an entire chip does go, then the usable capacity will shrink by the capacity of the Flash chip that's died.

Now what you need to understand is, that with the specs listed by SSD drive makers, is that:
1. The statistics only use the best 90% of all cells, not all of them (100%).
2. Statistics are performed on empty drives, which isn't real world conditions in almost all cases (there's usually data that remains on the drives, such as the OS and applications).

SSD's are still rather new, and need the bugs worked out of the technology (i.e. better Flash chips to become available, such as FeRAM (http://en.wikipedia.org/wiki/Ferroelectric_RAM)), and OS's need to be optimized for SSD's as well (i.e. Windows is, OS X isn't as of yet, given the information published on 10.6.4). OS X will eventually support SSD's, but we don't know when.

SSD's are great for OS/application disks as it exists currently, but there are limitations as to what it's usable for.

Ultimately, this has been covered before in other threads, and IIRC, you participated in a couple of them, so this shouldn't be new information to you.