Western Digital will break new ground later this month when its 2TB Caviar Green WD20EADS hard drive goes on sale. Seagate packed 1.5TB into its Barracuda 7200.11 drive by using four platters and eight heads, but WD has gone a step better and squeezed 33 per cent more storage capacity into the same space. WD Caviar Green 2TB …
"2TB - 1.81TB once formatted"
Formatting has nothing to do with it. Using proper units will show why.
2TB = 2*1000*1000*1000*1000 = 2,000,000,000,000 bytes
1.81TiB = 1.81*1024*1024*1024*1024 = 1,990,116,046,274 bytes
Formatting a hard disk doesn't suddenly reduce its capacity by 190GB...
For about the millionth time
these drives do NOT vary the spin speed dynamically.
WD even had to issue a clarification because so many people were incorrectly making this assumption.
people have hooked up microphones to CRO's to measure their hum to prove the drive speed never changed.
the make a hard drive motor which can change its speed in the way suggested would be VERY difficult.
Unless by vary you mean : 5,400rpm and 0 rpm (the only variation it can do)
some of WD's other marketing material makes it clearer (last two words below), but i would have hoped for the tiniest bit of fact checking from a hardware review! peddling this kind of BS only serves to perpetuate the myth that WD have done something remarkable
IntelliPower™ — A fine-tuned balance of spin speed, transfer rate, and caching algorithms designed to deliver both significant power savings and solid performance. For each GreenPower™ drive model, WD may use a different, INVARIABLE RPM.
You're quite right Mr Anonymous that the capacity is not reduced by formating and that this is the difference between true capacity and the wretched 'decimal' Gigabytes that appear on the specification..
The point is that you only see the true capacity once the drive is installed and formatted so it appears that formatting eats capacity.
In this context 'formatted capacity' is short hand for 'you actually get'
I've found a site where you can pre-order them for those who are interested:
I would get more but I'm happy with the 3TB I've already got (RAIDed of course). Problem I find with these big drives is the cost in that you have to buy at least 2 just in case one dies and you loose all that data.
Unless the article has been modified since your reading, it states:
complete with IntelliPower motor control which means that some drives in the product range may have a rotational speed of 5400pm while others may operate at up to 7200rpm
Which clearly (to me) indicates the speed is set at the factory per unit, and does not vary at all. So your comment:
these drives do NOT vary the spin speed dynamically.
Seems to agree with how I understand the article in the first place.
We need a "defending the author" icon.
What is it with computer peoples' fascination with thinking that TiB are the "true" units? Historically, the ONLY computer-related units which have been in the power-of-two magnitudes have been memory and hard drive space. Everything else (CPU speed, network bandwidth, display resolution, etc.) have always been in the proper scientific power-of-ten magnitudes. There are very solid technological reasons for memory to be based on power-of-two magnitudes, but there aren't for disk space, and fundamentally it's not like it really matters to the end-user to begin with. And yet, operating systems still report hard drive space in powers of two instead of powers of ten, which then leads to the ridiculous "formatted capacity less" weasel words which only serve to add to the confusion.
We should stop taking hard drive manufacturers to task, when the blame lies SOLELY in the hands of the OS vendors.
It is always dissappointing to encounter data graphed in a misleading way. In this case, the obfuscation was acheived by using a non-zero crossing point for the x-axis on several comparison graphs.
A prime example of this is your comparison of Burst Speed - at first glance the Intel X25-M appears to be many multiples faster than its rotating disc rivals... but the truth is the difference is less than 20% between fastest and slowest!
Attempting to emphasize a point is one thing, but the poor choice of graphing technique only succeeds to make the review seem amateurish - more suited to a mass-market newspaper article than a report from a "respectable" tech focused website.
I had a friend who actually worked for WD. He (hi Dave) commented that often there are resonances in the drive and they vary the speed a bit to "tune these out". I suspect that on the factory line they tune the speed of the drive to have the minimum amount of resonance while the drive is running. This can lead to quieter drives since they won't shake rattle and roll. I suspect that this is the "dynamic" they are talking about. This variation in speed is just ONE of the many "tricks" they do in making large capacity drives.
Now if they made it in wide SCSI it would be even better!!
No, its not. The letter H is not silent, nor yet is it a vowel.
As for the 'formatted capacity' debate, actually the volume does decrease, with all practical file-systems, as a certain amount of space is always reserved for bad sectors, and some need overhead space for allocation tables. The effect is not as pronounced as the 1024/1000 thing for a drive this big, but it does exist.
Actually you're all wrong ;) The reason we use 'an' before certain words and 'a' before others is purely an oral one - it's difficult to pronounce the word 'a' followed by a word which begins with a vowel sound. If, for example, there was a word which began with a silent vowel followed by a consonant, (not that I can think of any) you wouldn't precede it with 'an'.
Whether to use 'an' or 'a' in front of words beginning with an H is purely down to whether you pronounce the H or not. Both can be argued as an acceptable use of standard English. This peculiarity is largely down to the fact that the letter H, while not a vowel, begins with a vowel sound: 'aitch'.
I do wish hard drive manufacturers would stop lying about the size of their drives.
We all know that all major OSes (Windows, Mac OS X, Linux (listed in no particular order)) read sizes correctly, thus:
1K = 1024 bytes.
1M = 1024K
1G = 1024M
1T = 1024G
Thus a drive of 2,000,000,000,000 is not 2TB as Western Digital would claim, but is in fact 1.81TB as the article correctly stated. Though I do like the wording "2TB, 1.81TB when formatted" - as if formatting it magically changes the definition of a terabyte.
We have laws in this country about fraud, deception, mis-selling a product, etc. How come drive manufacturers are still allowed to blatently lie about how much space you're getting, just because the average computer user doesn't know that capacities are measured in binary quantities not decimal ?
I hope one day this gets corrected and the drive manufacturers get the slap in the face they deserve. As capacities get bigger and bigger, the discrepancy between what they say you're getting and what you're really getting gets more and more. In this case, you're being cheated out of a whopping 0.19TB (194.56 GB, or 199,229 MB). Thats an entire hard drive's worth, even by todays standards.
Imagine if memory modules were sold using this system?
"The module is 2147 MB in size (2GB when installed in system)". I doubt it would wash.
Using the Queen's English, you'd pronounce the 'h' in 'horrendous' if speaking correctly, thus the printed version should assume an audible 'h' and therefore 'a' before it.
This extends out to the 'a historical' day, however it is the most contentious. However searching Google (an odd acid test admittedly) shows "a historical" has ~10m hits compared to ~4m for "an historical".
What really gets me is when people say "an historical" but also pronounce the 'h'.
And don't get me started on whether and weather... But then I'm a Scot and so pronounce it the BBC way
Just got a firewire drobo and shoved a load of spare old 500GB drives in there to act as a temporary backup for only my most important folders.
However, I've also wanted to be able to keep backups of all my dvds and animation work as well... some of the render outputs (particularly the folders containing 2500+ 1920x1080 TIFF files) are currently too big to be able to backup now that I've had to start making HD content.
Glad to hear it's quiet, too.. would rather not have 4 little rasping dinosaurs spinning away behind my desk when the backups are taking place.
1TB drives were not quite enough to offer a complete solution to all my backups needs, so I didn't bother upgrading.. 2TB does the job nicely.
I can't see myself needing a 4TB drive for a long time... unless something crazy like a new "superHD" standard comes along.
There seems to be the usual foolish confusion between RAID and backup here.
RAID mirroring (or to a lesser extent RAID5) gives you protection against a hardware failure making data unavailable.
In a full time RAID setup, what protection is there against data being unintentionally corrupted or deleted by the user, by an app, or by the OS? None.
If I had one of these, I'd want two of them, not in any kind of RAID config. And every now and then, I'd use some kind of file sync program to sync the copies of the data. In that way, there is some kind of window when an unintentionally damaged file can be restored from the backup. OK the intact copy may get overwritten if the damage isn't noticed before the next sync point, but only if the sync program doesn't ask for confirmation before overwriting the old copy...
You're right that RAID is not the same as a backup, it just provides protection against data loss due to built-in redundancy.
In some advanced file system like Sun's ZFS or NetApp's offerings, there *is* protection against data being unintentionally corrupted or deleted by the user, application or OS: they are called snapshots. Files and directories referenced by snapshots cannot be deleted until the referring snapshot is deleted -- so there is your protection against loss. Also files/dirs referenced by snapshots cannot be modified -- these file systems employ a method called 'copy-on-write' which means that if a file referenced by a snapshot is modified, the file system creates a copy. If I remember correctly, the common blocks of the two files are not duplicated, to save space, but don't quote me on that :)
A traditional 2-drive mirror is fine until you need, in this case, 2.1 TB. Also traditional mirrors often don't repair files that can't be read -- the file system/RAID controller often just returns the data on the good half of the mirror. However, ZFS also repairs the faulty file on the bad side of the mirror, as indeed it does in any redundant setup: mirror, RAID-Z1 (like RAID 5), or RAID-Z2 (like RAID 6).
For backup/syncing of files that you mention, again, ZFS offers a way of doing this -- with one important difference and huge benefit: you will never lose any data that gets deleted if you use (1) automatic snapshotting (via cron every 30 minutes or whatever), and (2) use 'zfs send/receive' to send an initial full backup and subsequent incremental backups, in which ZFS detects diffs between snapshots and sends only the differences, to another storage pool, which might be on the same machine or a different machine on the LAN/WAN.
These links might be of further interest:
I'll get me coat -- I'm off to the pub... :)
Biting the hand that feeds IT © 1998–2019