I agree on most points mentioned by Kevin Hopefully you guys will test out the 1 TB as well. But I strongly disagree with these points as mentioned above:
- the advantage in power draw and vibration matters if you're running many of them
If the spindle speed jeopardized the performance due to vibration - then all enterprise models would have been 5400 RPM disks but this is not the case. Infact Seagate's forthcoming Enterprise Turbo HDD will spin at 15K with 128 GB of Nand Cache designed specifically for NAS/Enterprise/SANs. And if you are talking outside of Enterprise (desktop), most consumers don't go beyond 4 disks so the point is irrelevant.
- for "cold data" the higher speed of the 7.2k rpm drives won't gain you anything
Completely the opposite actually. The mechanical disk drives perform the worse when the cold data is spread across different sectors (not on the same cynlinder). Infact, this is exactly the scenario where a 7200 RPM/10k RPM helps because the data is not in the cache hence we cannot take advantage of spatial locality due to obvious reasons. This is actually very easy to prove mathematically. I am even going to go as far as saying I don't think you know what "cold data" actually means.
- if your disk system is I/O limited by network or USB the higher speed of the 7.2k rpm drives won't gain you anything
This is bunkers to say the least. Even with a 30 MB/sec bandwidth cap of USB 2.0 in half duplex mode, no mechanical disk drive in the market today (not even 15k Cheetahs) can saturate this bandwidth in 512/4k Random seek for sectors that are not on the same cynlinder.
- you can reach higher platter densities at lower rpm, i.e. make bigger drives or achieve the same capacity cheaper
Negatory. Platter densities and spindle speed are actually mutually exclusive. Areal densities depends on head technology and the recording method (longitudinal/perpendicular etc). A fairly practical counter proof exists with WD Velociraptor drives. 1 TB version Raptor reaches sequential speed in excess of 200 MB/sec yet spins at 10K RPM as well as Seagate's latest generation SAS Cheetah units.
I agree that it probably should have been called "Blue"
This says it all. The fact is that 1 TB black is no better than the blue yet is more expensive. Infact, WD is simply choosing to ignore this issue altogether on their official forums despite numerous complains (please check the link provided by OP?). Some are even going as far as saying the Black is just a rebadged Blue in the case of 1 TB version. The performance numbers don't disagree with that statement so IMHO WD deserves the hate.
The idea of sending a flagship model (4 TB Black) to the press is a classical technique that manufacturers use to create a halo affect. In a sense that the top performing unit will propagate to finding it's way in making people think that the entire lineup must be superior. Storage review - as experts in storage should ALWAYS watchout for this IMHO.
I'm not sure I understand the hate towards WD compared to Seagate, especially considering they both have nearly identical product lineups not counting the hybrids.
You say Green, Blue, Red, Black, Xe, Re, Se, etc as if its a bad thing to have purpose built drives for certain scenarios.
Kevin, I totally get where you are coming from. But I think you took my point to the extreme. I never said that purpose built drives are bad. Ofcourse not. All I was saying this is the same reason why GM couldn't afford themselves by developing different cars of same type with different brand that took them down. (Pontiac, Saab, Saturn).
What I meant to say is they need to simplify their product lineup. Otherwise, we risk of having Yellow, Pink, Magenta and Golden (etc). as well in their product lineup.
Why not have simple product lines. Mainstream, Enthusiast, Enterprise
Drives like Blue and Black can goto Mainstream. Drives like SSHD and Raptors can goto Enthusiast and Enterprise can have upper 10k and 15k offerings. A hard drive is a hard drive and it should just work. What's exactly wrong with that? (simplicity) If we look back 10 years, we still had NAS, Enterprise storage, SANS - but at the time we didn't have a million different product lineups. As an infrastructure manager, I don't recall hard drives blowing up just because they didn't have a specific category for each application? Infact, most of those drives we have are still working today :/
Edited by cppguru, 13 February 2014 - 12:04 PM.