The First Terabyte Hard Drive Reviewed

mikemuch writes "ExtremeTech has a review and benchmarks of the Hitachi Deskstar 7K1000 1TB Hard Drive, which ushers in the terabyte age. It performs well on HDTach and PCMark benchmarks, though not as speedily as professional-grade drives. It could be just the ticket for digital media junkies. 'One of the first issues to note is that you may not see an actual one terabyte capacity on your system. First, the formatted capacity is always less than the raw space available on the drive. Directory information and formatting data always take up some space. Second, the hard drive industry's definition of a megabyte differs from the rest of the PC business. One megabyte of hard drive space is 1,000,000 bytes: 10^6 bytes. Operating systems calculate one megabyte as 2^20 bytes, or 1,048,576 bytes. Once installed and set up, Hitachi's 1TB hard drive offers up an actual formatted capacity of about 935GB, as measured by the OS. That's still a lot of space, by anyone's definition.'" Update: 05/17 21:52 GMT by Z : Adding '^s' missing from article.

Comments

Why is this still a discussion?

[Grelli]

Don't complain about the fact a megabyte isn't what you thought it was. Complain about the fact the industry still uses it for labels. But don't try and make the megabyte a mebibyte.

http://en.wikipedia.org/wiki/Binary_prefix vs http://en.wikipedia.org/wiki/Si_prefix

Re:Why is this still a discussion?

[linguae]

The reason why nobody uses "mebibyte", "kibibyte", "gibibyte", and all of these other terms are because of two reasons: they are new and relatively unknown, and they just sound stupid and unnatural (try pronouncing them). It is commonly accepted knowledge in electrical engineering and computer science circles that we use 2^10, 2^20, 2^30, etc. when describing kilobytes, megabytes, and gigabytes, respectively, except when dealing with data storage capacities (which I feel is a marketroid invention and a sales gimmick. "10^9 vs. 2^30? Who'll know the difference?"). It's been that way since the 1960s. The new terms like "mebibyte," "gibibyte," and the rest of them just sound silly, hard to pronounce, and unnatural.

Re:Why is this still a discussion?

[Mattintosh]

The difference between the two numbers is due to this:

Manufacturers have an interest in you paying more and getting less, while computers do not. Manufacturers who are successful and have the cash to spend can then lobby international standards bodies to skew "standards" in their favor, regardless of historical context and practical day-to-day usage patterns (both technical and linguistic).

Don't side with "the man" on /. unless you enjoy tar + feathers, not even if it means bucking the "standard".

Re:Why is this still a discussion?

[SEMW]

Don't try to invent a new notation to make-up for corporate marketing corrupting established and well-understood notation. I would point out the "mega" was an established and well-understood prefix for 10^6 well before the computer industry (sans hard drive makers) started to use it mean 2^20.

Re:Why is this still a discussion?

[dgatwood]

They are still wrong. The fundamental organization of computers is base 2. Ever try to design a processor that uses 1000 byte pages? Good freaking luck.

The purpose of SI units being in base 10 is because the number system that we use to measure things is ALSO in base 10. Therefore, the unit fits comfortably within mathematics associated with the relative fields. However, A base-10 numbering scheme basically does not exist in the computing world; obscure BCD hardware notwithstanding, all occurrences of base ten in computing are entirely a fiction created by the machine to try to make things more understandable to people used to base 10.

More to the point memory and storage are inherently organized in units of powers of two. Memory will ALWAYS be organized in power-of-two increments as long as computer operate based on the binary system. Why? Because this makes it possible to express divisions of memory in terms of bit boundaries. A power-of-10 memory organization would require computationally heavy division or multiplication operations throughout the memory management code, while a power-of-2 memory organization requires an extremely lightweight bit shift. For this reason, as long as we have binary-based computers, we are stuck with power-of-two units of RAM.

Similarly, a hard drive block will ALWAYS be evenly divisible by the size of a memory page or vice versa. If this were not the case, the complexity of writing an operating system would be beyond insane. Paging and memory mapping of files alone would be enough to make the engineers commit seppuku. Therefore, as long as RAM is organized into groupings based on powers of two, hard drives will always be physically laid out in blocks whose length is a power of two.

Because the fundamental organization of data in a computer is, by nature, organized into power-of-two units, describing storage in power-of-ten units makes no sense, as it will almost always be a crude approximation. There are probably exactly zero hard drives with an exact decimal gigabyte capacity. The fundamental storage in hard drives is a 512 byte block, and 512 does not divide evenly into most multiple-of-ten values. Sure, you could create a 512 decimal gigabyte drive, I suppose, but for the most part, the values just don't divide evenly by 512. Therefore, using a multiple of a power-of-ten number to describe the amount of storage will almost always be a very crude approximation, while using a multiple of a power-of-two number can be (and usually is) an exact value.

In other words, the idiots in charge of making up the SI units should have been taken out and beaten for "Gibibyte". There is one natural unit in computers, and that is the base-2-derived gigabyte. All base-10 units are inherently inaccurate, and thus a poor fit for computing. They should be summarily rejected by the industry, as they simply do not make any sense in the context of storage. Honestly, they don't make a lot of sense for networking, either for the same reason, but I'm willing to overlook that... for now....

Re:Damn...

[netscan]

I remember when I was 9 or 10 and the family computer could hold 10 gigs. That was nearly unfillable at the time. Sorry, just being nostalgic Okay, how many read this and said quietly to themselves "Man I'm old..."?

Re:106 bytes and 220 bytes, ???

[Mundocani]

Regardless of whether the original article had it wrong, someone at Slashdot should've read the posted summary text and noticed the error. You shouldn't be a "News for Nerds" editor and not immediately notice that the sentence makes absolutely no sense as written.

SI prefixes are powers of thousand.

[Zarhan]

Cue the ensuing Mebi/Gibi/Tebi vs. SI notation fights.

While it's takes a while to get used to it, I actually prefer the Bi-units now. 4,3GiB or 4,7GB is already a huge difference when talking about DVD capacity. At terabyte, it gets enormous.

Linux already uses those units.

Only place where I still see a purpose for using binary units in computing is memory - address bus is still addressed exactly with n lines so memory capacity will be 2^n. For all other cases, it's not needed. Yes, the hard drives have 512 to 4096 byte sectors, but who cares when were talking about trillions of them?

See http://en.wikipedia.org/wiki/Binary_prefix for more.

Re:WOW, 1TB

[Chandon Seldon]

What will you store on your 10 TB drive that will take up all the space?

High quality 1080p video. Animated textures for video games. A massive sample database for a voice synthesizer.

I'm not actually sure what you would do with a 10,000 TB hard disk - but 10 TB is well within the "use it up with some video" range.

Re:Zonk

[Anonymous Coward]

This is still not really correct: One megabyte of hard drive space is 1,000,000 bytes: 10^6 bytes. Operating systems calculate one megabyte as 2^20 bytes, or 1,048,576 bytes.

It should read: Hard disk manufacturing company marketing departments define one megabyte of hard drive space as 1,000,000 bytes: 10^6 bytes. Fucking reality calculates one megabyte as 2^20 bytes, or 1,048,576 bytes.

Re:kibibytes & mibimeters

[Alioth]

It's common practise to use the highest number when representing something in an ad. Like for instance Cirrus aircraft advertise their aircraft's speed in MPH, when pilots actually use knots. This is because you get a number on the ad that's 15% bigger. And I bet Cirrus advertise the speed of their planes in Europe in km/h because this yields an even bigger number.

Personally, I wish we'd just get on with it and switch to base 16. It would be so much more convenient, and I'd be back in my early 20s again!