HDD Manufacturers Moving To 4096-Byte Sectors

Luminous Coward writes "As previously discussed on Slashdot, according to AnandTech and The Tech Report, hard disk drive manufacturers are now ready to bump the size of the disk sector from 512 to 4096 bytes, in order to minimize storage lost to ECC and sync. This may not be a smooth transition, because some OSes do not align partitions on 4K boundaries."

Comments

Factors of 10

[suso]

Why not just move it to 1000 byte sectors, then we could minimize the space lost to advertising.

(Note to accuracy nazis, this is meant to be funny)

Re:Factors of 10

[Iyonesco]

If they want to use base10 the first thing they should do is respecify a byte to be 10bits.

I somehow think their enthusiasm for base10 would diminish when it would case a a 25% drop in stated capacity.

Re:Factors of 10

[rolfwind]

If they want to use base10 the first thing they should do is respecify a byte to be 10bits.

How about leaving the word byte alone and using another, distinct group of letter to do the job? Respecifying only confuses the issue, even those who know, because you're still be working with two different definitions in the same field for a long time.

Re:Factors of 10

[TheRaven64]

A byte can be 10 bits; it's an architecture-specific quantity. An octet is always 8 bits.

Re:Factors of 10

[drainbramage]

Mine goes to 11.

Re:Factors of 10

[Anonymous Coward]

What, like bite?

Re:Factors of 10

[houghi]

Stupid bits. Oh and there are only 11 types of people. Those who can read binary and those who can't.

(Yes, I did read and re-read the above and it is correct from my point of view.)

Re:Factors of 10

[Anonymous Coward]

Historically, yes, but nowadays "byte" is always used to refer to an octet in a system with an 8-bit, 16-bit, 32-bit, or 64-bit word (and so likely all the other 2^(3+n)-bit word systems to come down the pipe in the future).

Re:Factors of 10

[Aqualung812]

I'll bite. The third type is...? People that can read binary but don't understand why you say there are three types and list only two?

Re:Factors of 10

[ivan_w]

Ok.. Care to explain how you came to the conclusion it was 11 and not 10 types of people ?

--Ivan

Re:Factors of 10

[marcansoft]

Unary. The base that you use to tell the joke need not match the base mentioned in the joke. Though it does kind of break the joke.

Re:Factors of 10

[wastedlife]

I know that when rating SATA bandwidth, it uses 10 bits per byte, but I thought that was due to it using 8b/10b encoding.

Re:Factors of 10

[Anonymous Coward]

Wrong.
A word is architecture specific.
A byte is ALWAYS 8 bits.

Re:Factors of 10

[McNihil]

You really want to make it difficult in the world don't-ya ;-) Just imagine the byte ordering "woodoo" that would be needed. I don't even think an insane-mad-lunatic engineer would make something like that in the first place... talk about swimming against ALL odds.

Re:Factors of 10

[pauljlucas]

How about leaving the word byte alone and using another, distinct group of letter [sic] to do the job?

The original definition of "byte" was the number of bits used to encode a character of text and is the basic memory-addressable element in a computer. It never originally meant "8 bits".

Re:Factors of 10

[Anonymous Coward]

Oh and there are only 11 types of people

The third type is...? People that can read binary but don't understand why you say there are three types and list only two?

00
01
10
11

I count four types there.

Re:Factors of 10

[troll8901]

Do explain.

Back your explanations with web-accessible citations, please.

And no editing of Wikipedia entries to match your explanations.

Re:Factors of 10

[m1xram]

That's Grey Code. 000, 001, 011, 010, 110, 111 and the joke should read...

There are only 11 types of old timer geeks... :-)

Re:Factors of 10

[Anonymous Coward]

*ahem*, I'm sorry to tell you that byte actually stands for "binary octet"

Re:Factors of 10

[fbjon]

There is no standard that specifies that it must be any particular number of bits, however, a byte is USUALLY 8 bits. But not necessarily.

Re:Factors of 10

[clone53421]

No such thing.

Re:Factors of 10

[dissy]

Wrong.
A word is architecture specific.
A byte is ALWAYS 8 bits.

A byte can't possibly "always" be 8 bits, when a byte means a single character.
This is the definition of 'byte' from 1959. People only started getting confused recently (Recently being the past 20 years) since the IBM 360 systems which first introduced the 8 bit byte and then became a defacto standard in the 80s. Then as new computer users moved into the front, such as yourself, you assume a byte must be 8 bits because that is all you have seen a byte to mean.

There are systems that encode a single byte with 7, 8, 9, and 10 bits still today.

The only time a byte is 8 bits is when the system is structured around 8 bit units.
Hop on a PDP or Cray system and you will see a byte is 7 or 9 bits respectively.

Origins of the word 'byte':
http://www.trailing-edge.com/~bobbemer/BYTE.HTM

Re:Factors of 10

[macemoneta]

[blockquote]The original definition of "byte" was the number of bits used to encode a character of text and is the basic memory-addressable element in a computer. It never originally meant "8 bits".[/blockquote]

That is the definition of 'octet', a term frequently used in telecom. People confuse byte and octet all the time, because popular hardware architectures use an octet as a byte.

Re:Factors of 10

[Anonymous Coward]

You're doing it wrong.

Re:Factors of 10

[Anonymous Coward]

No such thing as unary? Of course there is.
http://en.wikipedia.org/wiki/Unary_numeral_system

Re:Factors of 10

[Magic5Ball]

Base 1.

Re:Factors of 10

[Mitchell314]

I'll sell you one that goes to 12.

Re:Factors of 10

[zygotic mitosis]

Because GP is one of the types that does not understand binary, of course...

Re:Factors of 10

[clone53421]

I should clarify. While a unary numeral system exists, I wouldn’t consider it a base system of the same sort as the rest, where you have place value of digits as such:

... + d[3]b^3 + d[2]b^2 + d[1]b^1 + d[0]b^0 + d[-1]b^-1 + ...

In a unary counting system, every numeral has the same significance. It’s a counting system, not a base system.

Maybe I’m just splitting hairs, but it’s at least a degenerate case.

Re:Factors of 10

[spiffydudex]

A quick google search should help that http://www.google.com/search?q=define%3A+byte&ie=utf-8&oe=utf-8&aq=t

I mean take a look at how many places say 8bits is 1byte
http://wordnetweb.princeton.edu/perl/webwn
http://en.wikipedia.org/wiki/Byte -- but the modern de facto standard is 8 bits
http://www.choicesandchallenges.sts.vt.edu/modules/evillage_glossary.htm
http://www.satech.com/glosofmemter.html
http://www.teds.com.au/www/6/1001102/displayarticle/glossary-of-terms--2104323.html
http://www.precisecyberforensics.com/glossary.html
http://www.its.strath.ac.uk/helpdesk/glossary/

There are several places including Princeton University that state 8 bits is 1 byte.

Re:Factors of 10

[Anonymous Coward]

it's funny because 10 is 2 in binary

Re:Factors of 10

[Anonymous Coward]

i propose we call it the beyete. we can use the abbreviation kb for when measuring thousands of them and mb for millions. sorry for any confusion over case, my keyboard is broken.

e.e. comeings

Re:Factors of 10

[toastar]

but 12 doesn't have any units, It's just an arbitrary scale mapping outputs.

Re:Factors of 10

[Anonymous Coward]

Its not thought. Its smacks of societies rampant idiotcy.

Re:Factors of 10

[NovaHorizon]

*clears throat* ....
WHOOSH

Re:Factors of 10

[Godji]

Whoosh, dude.

Re:Factors of 10

[CharlieHedlin]

Base 1 won't work. Your numbers would always be 0, or 00, or 000, I don't even see how to get to 1 (decimal)?

Re:Factors of 10

[digitalunity]

If unary logic is good enough for politicians, it should be good enough for anybody!

Re:Factors of 10

[Anonymous Coward]

Congratulations. You've officially made 3 = 4.

Re:Factors of 10

[Yetihehe]

For $2048 I'll make you one that goes all the way to 42.

Re:Factors of 10

[Anonymous Coward]

in that case, there would be no hardware "byte" since "byte" depends on
the character set. byte size may even be variable, as in utf-8.

Re:Factors of 10

[BikeHelmet]

4096 byte sectors are fine. It's nice that we'll get roughly 10% more usable space for no cost.

But I think it'd be nice if when I open a 4KiB file it said 4KiB. According to metric prefixes accepted in virtually every other field, 4096 bytes is 4.1KB (or 4.096KB to be exact) Being "digital" does not give the right to use the wrong prefixes and cause confusion.

It's also worth noting that this is Microsoft's fault. Other OS's are doing it properly. Microsoft only does it properly when it benefits them. HDD manufacturers have faced numerous lawsuits simply because Microsoft is using the wrong prefix, so people feel cheated out of space. And the issue will only get worse... Every time we jump from one incorrect prefix to another.

1024 bytes - 1KiB - 2.4% discrepancy.
1048,576 bytes - 1MiB - 4.85% discrepancy
1073,741,824 bytes - GiB - 7.37% discrepancy
1099,511,627,776 bytes - 1TiB - 9.95% discrepancy

When you buy a 1500,000,000,000 byte HDD (which includes free error checking bits, so really has over 1650,000,000,000 bytes - 1.5TB usable), Windows reports the capacity as 1.36TB (incorrect) - not 1.36TiB. (correct)

Can't wait for the new round of lawsuits when we hit petabytes - 1125,899,906,842,624 bytes, a 12.6% discrepancy. And the ridiculousness of this is, it's not even a real issue.

Why aren't we suing SSD manufacturers? They often give us less bytes than advertised in both metric and incorrect metric.

Does nobody care about filesystem overhead? We lose as much as 20% of our disk capacity to shitty filesystems. That's way more than the Metric debate.

What about disk performance? It's harder to measure, but doesn't I/O performance matter more than capacity? Sun proved you can design a FS that doesn't lose significant performance or require defragmenting - ZFS. Microsoft with their NTFS is probably costing us 30% I/O performance. Wouldn't it be nice to have Reiser4 available for servers or computers with UPS's?

As long as we remain fixated on something that isn't an actual issue, we'll never correct the ones that are.

Re:Factors of 10

[v1]

Your numbers would always be 0, or 00, or 000

you just answered your own question. you'd count, "0", "00", "000", "0000" etc. It'd just be really inefficient for large (in more ways than one!) numbers.

Re:Factors of 10

[the_enigma_1983]

I see four numbers there, representing the decimal numbers 0 through 3 inclusive. He's not saying that "00,01,10,11" are labels for types of people, he's saying the number of types is "11", which if read as a binary number is 3 in decimal.

Re:Factors of 10

[Pieroxy]

Have you read the first paragraph of the Wikipedia article you are linking to? Let me put it there for you, with relevant portions highlighted:

The byte (pronounced /bat/) is a unit of digital information in computing and telecommunications. It is an ordered collection of bits, in which each bit denotes the binary value of 1 or 0. Historically, a byte was the number of bits (typically 6,7,8, or 9) used to encode a character of text in a computer[1][2] and it is for this reason the basic addressable element in many computer architectures. The size of a byte is typically hardware dependent, but the modern de facto standard is 8 bits, as this is a convenient power of 2. Other factors behind this particular size may be the IBM System/360 architecture, introduced in the 1960s, and the 8-bit microprocessors, introduced in the 1970s. No formal definition exists however, and other sizes have been used in various computers historically.

Re:Factors of 10

[jbengt]

people that understand
people that don't understand
and the unmeasured majority of people that are in a state of superposition of understanding and not understanding

Re:Factors of 10

[Anonymous Coward]

A byte can be 10 bits; it's an architecture-specific quantity. An octet is always 8 bits.

isnt it: a word can be 10 bits, an byte is always an octet = 8 bits.

Ive seen different word lengths but not never different byte lengths. Maybe I havent seen enough..

Re:Factors of 10

[bzipitidoo]

There is no third. Obviously he meant Roman numerals.

Re:Factors of 10

[TheRaven64]

Look before about 1975. You'll find lots of computer manuals that differentiate between an octet, a byte, and a word. A byte is a discreet quantity, the smallest addressable amount of memory. On a lot of early computers it was 6 bits and words were variable length. I don't know of any architectures that had 10-bit bytes, but I know of a few that have had fewer than 8 bits in a byte. A word is a group of bytes which the machine can operate on with a single instruction.

Re:Factors of 10

[afidel]

Dude, MS's KB for kilo-byte predates the silly KiB definition by about two decades (KiB was only adopted as a standard in 2000 and proposed in 1998, MS DOS traces to M-DOS in 1979). Oh and as to performance, almost nothing writes in 512byte chunks, 4KB chunks are about the smallest defaults for current platforms and 8KB is becoming more common.

Re:Factors of 10

[Behrooz]

Ain't no such thang as a broke joke.

Unless you are broke, in which case you should become a comedian.

Re:Factors of 10

[drsmithy]

It's also worth noting that this is Microsoft's fault. Other OS's are doing it properly. Microsoft only does it properly when it benefits them. HDD manufacturers have faced numerous lawsuits simply because Microsoft is using the wrong prefix, so people feel cheated out of space.

I hate to rain on your anti-Microsoft parade, but back when hard disk manufactures realised they could make their hard disks look bigger than they really were, capacities were still being measured in 10s of MB, and *all* OSes were using power-of-two prefixes.

The rest of your rant is about as accurate.

Re:Factors of 10

[Glonoinha]

It appears this discussion may devolve into a KB vs. KiB, MB vs. MiB discussion.
I'll check back in a while just in case.

Re:Factors of 10

[geminidomino]

Tally/hashmarks are base 1, no?

||| types of people.

Re:Factors of 10

[geminidomino]

*ahem*, I'm sorry to tell you that byte actually stands for "binary octet"

Sorry to tell YOU, but it actually doesn't, since there is historical (and some contemporary) precedent of 7,8,9,and 10-bit "bytes".
Etymologies vary depending on source, including:

Byte - wordplay/pun expabnsion of "bit" ("Binary digIT")
Byte - BinarY TErm
etc...

Re:Factors of 10

[Enter the Shoggoth]

[blockquote]The original definition of "byte" was the number of bits used to encode a character of text and is the basic memory-addressable element in a computer. It never originally meant "8 bits".[/blockquote]

That is the definition of 'octet', a term frequently used in telecom. People confuse byte and octet all the time, because popular hardware architectures use an octet as a byte.

Really? I was under the impression that it was the other way around. ie. the reason it's called an octet is because it has 8 bits.

If you have some definitive evidence to the contrary I'd really be interested in seeing it.

Re:Factors of 10

[Anonymous Coward]

Windows 7 free activation here, just click Here - this will permanently activate your windows 7 free and you will get all full updates. this is free so go ahead and get activated now.

Re:Factors of 10

[Anonymous Coward]

Actually, a ``bite'' is defined to be 10 bits and the corresponding nibble is 5 bits. People generally prefer byte and nybble as they are powers of 2 rather than lacking any power at all.

Re:Factors of 10

[jonadab]

> People confuse byte and octet all the time, because
> popular hardware architectures use an octet as a byte.

Only the popular ones, eh?

Can you name a hardware architecture, for which new hardware has been manufactured within the last quarter of a century, that uses any byte size *other* than eight bits?

In 1950, your assertion would have been technically accurate, albeit pedantic. But today it's just so much anachronistic nonsense. A byte *is* an octet, period.

Ironically, however, a character is no longer necessarily representable in a single byte.

Re:Factors of 10

[macemoneta]

That's what I was saying; read the thread again.

Re:Factors of 10

[macemoneta]

4-bit microcontrollers are still made.

Re:Factors of 10

[jonadab]

> A byte can be 10 bits; it's an architecture-specific quantity.

Historically, yes, but not any time recently. Not, for instance, since people started to write software that was portable enough to be compiled for different hardware architectures. If you think codepages and endianness are bad, just *imagine* trying to port your software to an architecture with an exotic byte size! No thanks. Different word sizes and address BUS widths are bad enough.

Re:Factors of 10

[jonadab]

> According to metric prefixes accepted in virtually
> every other field, 4096 bytes is 4.1KB

In every other field, it's useful to know how many thousand of something you have.

In computing, however, it is a great deal more useful to know how many 1024-byte kilobytes you have, because space is allocated in power-of-two-sized sectors, so 1024 is either a multiple (two 512-byte sectors) or a factor (a quarter of a 4096-byte sector).

With a 512-byte sector, if I have a bunch of very small (but not zero-size) files and the disk has 128 kilobytes left, I expect to be able to store no more than 256 of my little files in that space. But if some cretinous loser programs the shell to tell me that I have 131 metric "kilobytes", I won't know how much space I really have.

If you want to use metric quantities, use kilobits, megabits, gigabits, and so on (as, indeed, is commonly done when measuring bandwidth). The byte is a power-of-two-based unit, which is what makes it useful for measuring computer storage, because computers do all their arithmetic in binary and for this reason are always going to allocate space (both in volatile memory and on secondary storage) in those terms. It's that way for a reason, and it is going to stay that way.

You'll notice they're not talking about 4000-byte or 40000-bit sectors. That would be dumb.

Re:Factors of 10

[rtb61]

I hate to break it to you, but people will read the size the "Operating System" reports and act accordingly. Complaining bitterly when the OS tells them they are being ripped off by the manufacturer. Obviously care should be taken by all involved to ensure that the customer is neither deceived nor their resources wasted. Especially when it is often done on purpose to drive the next upgrade cycle, to create patentable incompatibilities that reduce performance for the customer but for which licences can be charged and being just to cheap to fix it, even when it might cost them only tens of thousands of dollars while it costs the rest of human society hundreds of millions of dollars of waste.

What was, was, it does not mean it has to continue or that it should be tolerated. Sometimes the behaviour within the computer industry is deplorable, for a expectation of buggy products on the initial release, to lies and gross exaggerations, falsified performance testing and of course trolls all over the internet. Just to be clear I do have a very low opinion of M$.

Re:Factors of 10

[toddestan]

It's also worth noting that this is Microsoft's fault. Other OS's are doing it properly. Microsoft only does it properly when it benefits them.

Back in the old days, what MS-DOS and the harddrive manufacturers claimed matched up fine. It was the harddrive manufacturers who decided to buck convention and redefine what a megabyte so they could advertise their drives as bigger than they really were. Microsoft is just doing what they've been doing for the past 25+ years.

Besides, other OS's still use the binary definitions of kilo- and mega- when it suits them. Compare the results of free -k and free -m on Linux, for example.

Re:Factors of 10

[drsmithy]

I hate to break it to you, but people will read the size the "Operating System" reports and act accordingly. Complaining bitterly when the OS tells them they are being ripped off by the manufacturer. Obviously care should be taken by all involved to ensure that the customer is neither deceived nor their resources wasted.

Yet you don't seem to be railing on the hard disk manufacturers to present their capacity in MiB.

Especially when it is often done on purpose to drive the next upgrade cycle, to create patentable incompatibilities that reduce performance for the customer but for which licences can be charged and being just to cheap to fix it, even when it might cost them only tens of thousands of dollars while it costs the rest of human society hundreds of millions of dollars of waste.

What was, was, it does not mean it has to continue or that it should be tolerated. Sometimes the behaviour within the computer industry is deplorable, for a expectation of buggy products on the initial release, to lies and gross exaggerations, falsified performance testing and of course trolls all over the internet.

What the hell are you blathering about ?

Just to be clear I do have a very low opinion of M$.

No, you have a foaming-at-the-mouth, mindless, childish hatred of Microsoft.

Re:Factors of 10

[houghi]

There are two groups of people who can count in binary.
The comment made is clear that there are 2 groups (those who can and those who can't)
The people who can would use the correct notation : 10
Now what would the people who can't use? They would use something that is wrong : 11
So what does it show? It shows that the person who wrote it (me) is in the group of people who can't.
I could go deeper into detail aboy how I suposedly did not understood the '10' joke which placed me also in the "can't" group.

Re:Factors of 10

[TheRaven64]

You might be surprised. One of the changes that went into clang a couple of months ago was support for architectures on which a char is not 8 bits. And, yes, a huge amount of code is written assuming a char is 8 bits, although the C standard does not make this guarantee (although the combination of the C standard and POSIX / SUS make a number of constraints on the size of char that the only size that will satisfy all of them is 8 bits).

Re:Factors of 10

[fireylord]

pure religious dogma! im going to start a splinter faction that says that there should be 13 bits in a byte!

Re:Factors of 10

[Kalriath]

4096 byte sectors are fine. It's nice that we'll get roughly 10% more usable space for no cost.

But I think it'd be nice if when I open a 4KiB file it said 4KiB. According to metric prefixes accepted in virtually every other field, 4096 bytes is 4.1KB (or 4.096KB to be exact) Being "digital" does not give the right to use the wrong prefixes and cause confusion.

It's also worth noting that this is Microsoft's fault. Other OS's are doing it properly. Microsoft only does it properly when it benefits them. HDD manufacturers have faced numerous lawsuits simply because Microsoft is using the wrong prefix, so people feel cheated out of space. And the issue will only get worse... Every time we jump from one incorrect prefix to another.

Give it up. Noone wants to use stupid units of measure that sound like something you feed your cat. KB is fine.

Re:Factors of 10

[Enter the Shoggoth]

That's what I was saying; read the thread again.

That is indeed what I thought you may have meant, but I won't matter how many times I re-read it you will still be saying the opposite.

Re:Factors of 10

[jonadab]

> 4-bit microcontrollers are still made.

Yes, and they're used in control circuitry for things like washing machines and coffee makers, not for general-purpose programmable computers.

Re:Factors of 10

[BikeHelmet]

With a 512-byte sector, if I have a bunch of very small (but not zero-size) files and the disk has 128 kilobytes left, I expect to be able to store no more than 256 of my little files in that space. But if some cretinous loser programs the shell to tell me that I have 131 metric "kilobytes", I won't know how much space I really have.

If it's NTFS I wouldn't bet on more than 64 files.

But there's also filesystem overhead, so it'd be lower. And modern filesystems can't function without some space left...

Re:Factors of 10

[BikeHelmet]

It appears I'm a troll and you're insightful, because more slashdotters agree with you.

Of course, you're incorrect.

I hate quoting wikipedia, but...

http://en.wikipedia.org/wiki/Binary_prefix#Files
http://en.wikipedia.org/wiki/Binary_prefix#Hard_disk_drives

I can't find any evidence of this:

but back when hard disk manufactures realised they could make their hard disks look bigger than they really were

All the evidence I've found (including on other sites) says it was always measured in metric. At some point IBM switched to 512b block sizes just because. They could've just as easily went with 500b block sizes, except they didn't, and 512b stuck.

How that translates into some nefarious marketing fiasco is beyond me.

*all* OSes were using power-of-two prefixes.

Could you please prove that?

Because I remember hearing that *one* OS did it properly, and it was one of the early ones. Now if only I could find the link.

Re:Factors of 10

[jonadab]

If the disk has 128k free, it's unlikely to be NTFS. More likely FAT12. If we were talking about a hard drive partition, the numbers would be larger. I chose a small example so the numbers would be easy to follow, for clarity.

But my point was, disk space is always allocated in power-of-two units. Always. Even with NTFS. Consequently, it doesn't make sense to measure it any other way.

Re:Factors of 10

[BikeHelmet]

But then why give the actual filesize at all? Why not just the space taken on the disk?

What's currently shown is the actual filesize in KiB - but labelled KB. Not the actual disk space taken up. :/

I would fully support what you're suggesting, but this half-half with incorrect labelling makes no sense.

So only XP is out of luck?

[7o9]

According to the Anandtech article, only the pretty much end-of-life Windows XP is out of luck. Linux, OS X and modern Windows versions all work ... Non news?

Re:So only XP is out of luck?

[gbjbaanb]

whoooooo. WinXP is end-of-life? You'd best tell that to all the millions of users (including big businesses) out there.

What that's you say? Upgrade to Windows 7 and use its perfectly infallible XP mode?

Ah, I understand now. Hi Bill, how's Steve getting on, still a bit sweaty and concerned he's not selling enough?

Re:So only XP is out of luck?

[Paradigm_Complex]

This is news, but not because of the potential problems that could arise. It's interesting from a technological standpoint. Why/how would changing the sector size effect performance? What are the downsides - why wasn't it done before? Why is it now chosen at 4k, why not something even larger? Those questions are what make it news (for nerds). It doesn't have to break something to be newsworthy.

Re:So only XP is out of luck?

[iamhassi]

"WinXP is end-of-life? You'd best tell that to all the millions of users (including big businesses) out there."

Couldn't agree more. Hopefully I don't have to rehash how horrible Vista was, and Windows 7 came out a few months ago so it's a bit early to proclaim XP is dead when it's hopeful replacement just showed up.

I think 4096-byte sectors are Very Bad News. I have no experience with these drives but XP doesn't like them which is reason enough for me to avoid them. I hope hard drive manufactures come out with a standard naming scheme for these new drives so they're easy to identify online, like IDE, SATA, PATA, etc. Maybe AFD for Advanced Format Drive?

Re:So only XP is out of luck?

[rickb928]

Really? Go back to your fvwm desktop and stop the rumor-mongering. Ok?

Re:So only XP is out of luck?

[petermgreen]

According to the Anandtech article, only the pretty much end-of-life Windows XP
I wouldn't call XP pretty much end-of-life just yet, you can still purchase it with new systems and it's still supported until april 8 2014 (that's after desktop support expires for the NEXT release of ubuntu LTS) and I haven't seen much use of either vista or win7 in buisness/academia yet.

This isn't that bad though, the logical sectors will still be 512 byte so it's just a matter of getting the partitions aligned right and wd will apparently be supplying a tool for doing this :).

Re:So only XP is out of luck?

[iamhassi]

ah this was what I was looking for: Drobo, XP Users: Beware of 4K “Advanced Format” Drives!

Article states that not only will XP have problems but so will many other devices like media centers, USB drives, game consoles, and anything else that uses a hard drive. USB drives will be the worse though since 4k drives formatted for XP won't work with Windows 7 and vise versa. Honestly I think this is too soon, put it off another 10 years, by then we'll have OS's that would have supported 4k for 10+ yrs already and all devices should be compatible by then.

Re:So only XP is out of luck?

[AlecC]

Why wasn't it done before? Sheer inertia. 512 bytes has been the HDD sector size since time immemorial. Some HDDs in the past could be re-sectored to different sizes, and sometimes were. I did it on one generation of disks to optimise storage for a particular reasons, but it didn't work reliably on the next generation of disks, so I dropped it. Some disks had a sector of 1080 bits, I think to handle the 33rd bit on IBM System/38.

What is the advantage? Every sector has a preamble, a sync mark, a header, the payload data, ECC, and postamble. These can amount to tens of bytes, especially as you have stronger ECC for weaker signals. By having fewer sector, you recover this space from most of the sectors. This could easily add 10% to the capacity of a drive. And, as posted elsewhere, most OSes do 4K transfers most of the time.

Re:So only XP is out of luck?

[kill-1]

The new hard drives will have a compatibility mode. It will be slower though because it has to read-modify-write behind the scene.

Re:So only XP is out of luck?

[pslytely psycho]

How about FUBAR?

Re:So only XP is out of luck?

[Anonymous Coward]

Why not something even larger? I guess it goes to average file size. Hard drives can only read and write whole sectors. If the file is smaller than a sector, this means extra data is read; and when writing, the entire sector must be read, the portion the file occupies is updated from memory (the hdd's cache), then the entire sector is written out.

Re:So only XP is out of luck?

[lorenlal]

Eventually, you have to put a line in the sand. If you push off the deadline, manufacturers will still take their time, and they'll be in the same place 9 years and 11 months from now.

Example: IPv6.

Re:So only XP is out of luck?

[alen]

MS has a clear support policy. Maybe you like Apple's 3 year support policy better than Microsoft's 10 year 7/3 policy?

Re:So only XP is out of luck?

[alen]

and it's not like corporations will buy all the new 2TB drives to use as a OS drive in their ancient XP workstations? they will just buy a new PC with Windows 7 installed or use a 7 corporate image

Re:So only XP is out of luck?

[Anonymous Coward]

Except that pretty much every OS in use now has IPv6 support. The problem isn't the installed base it's the fact that so long as NAT holds out there's no pressure to switch to it.

Re:So only XP is out of luck?

[drinkypoo]

Except that pretty much every OS in use now has IPv6 support.

Except that name resolution is broken for IPv6 on Windows XP, which is the operating system not supporting 4k sectors that people are complaining about... so IPv6 was a super shitty example for you to try to defend.

Re:So only XP is out of luck?

[LOLLinux]

Or the Linux policy of barely a year.

Re:So only XP is out of luck?

[conureman]

Who's gonna install old XP on one of these new HDDs? HIBT?

Re:So only XP is out of luck?

[Surt]

Consumer OS, and not XP, which is in widespread use.
(why is consumer OS important to clarify? Because all that really matters for ipv6 adoption is the router OSs.)

Re:So only XP is out of luck?

[BenoitRen]

Comparing this issue to IPv6 is not fair, because we've known about its existence for longer, and there is more support for it out there.

Not to mention that there is no real need to switch to 4096-byte sectors. Sure, it's nice, but not necessary to keep going.

Then again, there are people who doubt the need for IPv6 as well, saying it's a solution in search of a problem.

Re:So only XP is out of luck?

[Sir_Lewk]

That's ok, the linux policy of free upgrades more than makes up for that for me.

But hey, you're a known troll, logic doesn't have much to do with this does it?

Re:So only XP is out of luck?

[Jeffrey_Walsh VA]

Isn't the purpose of some of these overhead items error correction? Will this lead to more unrecoverable read errors?

Re:So only XP is out of luck?

[bamf]

Or as in our case, buy new PCs and load an XP image. I suspect there are a huge number of companies doing this. We've no fixed plans to go to Windows 7 any time soon.

Re:So only XP is out of luck?

[AlecC]

The ECC (Error Check and Correct) is used for error correction. But generally speaking, the number of bits needed to check a block of data rises slower than the number of bits in the data - probably as the log of the number of bits, though I don't know. So grouping up sectors and providing a slightly longer ECC will save a significant number of the ECC bits. Of course a sector having eight times as many bits is eight times as likely to get corrupted, simply because of its size. But such faults are rare, though not rare enough to ignore. Of course, it will be a fraction less reliable. But the manufacturers do reliability/performance trade-offs all the time, and this is only one more of them. Presumably they reckon they have the reliability under control. If you want greater reliability, you need RAID anyway - to protect against drive failure as well as localized corruption. The probably reckon that anybody with really valuable data will be RAIDed anyway,

Re:So only XP is out of luck?

[Anonymous Coward]

ECC costs are non-linear, so with larger blocks you can get more reliability with less waste.

Re:So only XP is out of luck?

[A beautiful mind]

Not to mention no DHCPv6 on Windows XP either...

Re:So only XP is out of luck?

[moonbender]

ECC is more efficient for 4k blocks. Apparently, 100 bytes of ECC for a single 4k block are as reliable as 320 bytes of ECC for eight 512 byte blocks. See http://www.anandtech.com/storage/showdoc.aspx?i=3691

Re:So only XP is out of luck?

[pclminion]

ECC means "error correcting code." It isn't just a way of detecting errors, it's a way of operating even in the face of them. As such, the length of the ECC chunk is proportional to the sector payload size. So going to a larger sector does not reduce ECC overhead, unless you switch to a more efficient ECC code (and you can do that without changing sector size). The other stuff, like sync, is fixed size and your argument does apply to that. The real reason for using a 4-kilobyte sector is because it exactly matches the size of a VM page on most consumer architectures, therefore guaranteeing at a hardware level that a single page worth of information cannot be fragmented.

Re:So only XP is out of luck?

[Anonymous Coward]

Another reason to upgrade.

If you haven't bought a new OS in 9 years, but you're still buying new hardware, it's your own fault for shit not working. If you've got money to spend on top of the line hardware, you've got money to spend on a new OS.

Re:So only XP is out of luck?

[Anonymous Coward]

Bad Example mod! 4096 sector size actually has a use. IPv6 on the other hand...
 
Just hoping that HD manufacturers CLEARLY LABEL THEIR INCOMPATIBILITIES and also keep the standard HD's on the market for a good long while. Maybe something faster than the raptor drives? I know those have gotten a bad name (at least among the people I know) for mechanical failures, but there is something to be said about having the fastest HD for a couple years.

Re:So only XP is out of luck?

[kbielefe]

Sidestepping your ignorance or deliberate deception on periods of typical Linux support contracts, it still amazes me that comments touting Microsoft support periods continue to appear on articles like this. Who cares if support goes out 10 years if you can't buy a new hard drive that will work with the OS? It's articles and comments like this that give me difficulty discerning what exactly Microsoft "support" entails. A warm fuzzy number you can call where they say you have to upgrade to Windows 7 for that hardware to work?

Re:So only XP is out of luck?

[bonch]

Windows XP is the most-used version of Windows, if not the most-used operating system, period.

Re:So only XP is out of luck?

[Leebert]

That's ok, the linux policy of free upgrades more than makes up for that for me.

You know, in a production computing environment, the cost of the software is pretty darn close to the least significant part of the costs of an upgrade, right?

Re:So only XP is out of luck?

[obarthelemy]

I beg to differ.

Ok, the upgrades are free, but it's hard to find docs and make some things work.

I'd rather have an older OS that works and has good documentation, than get on an upgrade - find out things don't work - hunt for elusive doc hell cycle.

Re:So only XP is out of luck?

[bonch]

I know I do, considering how bad 10-year-old Microsoft operating systems are.

Re:So only XP is out of luck?

[Antique Geekmeister]

They'll be used on external USB drives, with the features not well-documented on the box or the advertisement. And your average non-Geek buying a new drive for their old system will also run into this.

Buyer, beware.

Re:So only XP is out of luck?

[gbjbaanb]

I don't know, if the drive fails, the support contract will supply a new HDD, they're all the same right?... not.

Also, you'd be surprised at the number of large customers who have just upgraded *to* XP. 2 of our customers have a choice of the OS to run, keep NT4 (which is not such an option anymore) or upgrade to Vista (no, didn't think so), 7 (possubly, but they're risk-averse and Win7 might turn out to be as bad as Vista really), so that leave XP. So XP gets installed on thousands of PCs and will stay there for years.

Re:So only XP is out of luck?

[moosesocks]

Ubuntu LTS releases are supported for 3 years on the desktop, and 5 years on the server.

The project has also been doing an admirable job of reigning in bloat, and new releases generally run fine on old hardware (particularly with the Xubuntu/Xfce combination).

Re:So only XP is out of luck?

[crustymonkey]

That is exactly what the company I work for does. For that matter, that is what the last 2 companies I worked for did. The current is a very large business, the previous a small company. Maybe some smaller companies on XP will bump to Win 7, but the "lumbering behemoths" are still at least a year or two out and you know for sure there will be many a boxen replaced in that time. This could also end up lighting a fire under the big shops to make the plunge to Win 7.

Re:So only XP is out of luck?

[m1xram]

I didn't spend any money on the OS, been running versions of Fedora since Win98se. If you want brand new hardware with free OSes, contact System76 or ZAreason or even Dell.

Re:So only XP is out of luck?

[AllynM]

The 'compatibility mode' you speak of will be no slower than the same drive being used under a newer OS. All it does is shift the mapping so that a non-aligned XP partition functions in an aligned manner as far as the physical sectors go.

http://www.pcper.com/comments.php?nid=8113

Allyn Malventano
Storage Editor, PC Percpective

Re:So only XP is out of luck?

[wastedlife]

Each of the distros has the freedom to set their own price, support, and update policies. Choose the one that has the policy that meets your needs. Here are a couple examples:

Red Hat Linux has a 7 year lifecycle and you pay for it. If you want free but no tech support, get CentOS or another distro built from Red Hat sources.

Ubuntu Linux has the following lifecycle and the releases are all free:

Ubuntu Server Edition has a dual release cycle. Every 6 months a version is released that contains the latest packages and updates. This is maintained for 18 months from release with an upgrade path to the next 6 month release version. For greater robustness and easier planning, every two years an LTS version is released that is maintained for 5 years.

I could not find a quote, but I believe the Desktop version has 18 months of updates for regular versions and 3 years of updates on LTS versions.

Red Hat includes tech support as well as maintenance upgrades(phone or web support depends on the package you buy). Canonical (the company that makes Ubuntu) sells support contracts for tech support, while updates and community support are free. Microsoft charges a hefty sum per support call, although OEMs often support the Windows install included with the product for whatever the warranty period.

Re:So only XP is out of luck?

[fprintf]

So some enterprising young person will be the first to invent IPv6 for Windows XP just like Trumpet Winsock was cobbled onto Windows 3.1 to provide Internet access way back before Win95. If there is some compelling reason to join in IPv6 network then such capability will be built.

Re:So only XP is out of luck?

[man_of_mr_e]

You only get free upgrades (as opposed to updates) for Free (as in beer) distro's.

Re:So only XP is out of luck?

[Antique Geekmeister]

Most used or not, it's 8 years old, and the update cost of a newly purchased machine with a plain OS installation disk includes roughly 2 Gig of downloaded data, and at least 5 reboots. (Measured last week on a clean installation of Windows XP Pro.) Even popular games that are shipping now do not run under it: that tells me it's obsolete.

Re:So only XP is out of luck?

[man_of_mr_e]

Well, you should. XP is now in "extended" support, which means that in less than 2 years it will achieve full end of life status. When that happens, there will no longer be any security patches issued, which means when new flaws are found your computers will stay vulnerable to exploitation of those flaws forever.

Re:So only XP is out of luck?

[Anonymous Coward]

That's what I wanted to say: the reliability factor is unchanged, but they have gained over 10% write space. That's not to mention the increase in ECC calculation speed (it's performance is improved at least 33%), overhead write time, or even the fact that most HDD controllers and OSs are capable of the 4096 byte DMA transfers already. They would be fools not to implement this, but it should be noted they would also be fools to take the 512 byte sector sized HDs off the market any time soon.

Re:So only XP is out of luck?

[Idiot with a gun]

I've never understood this long living love for XP. The longer I work with it (I'm a support tech), the more I hate it. It genuinely has the feeling of an OS that was organically grown, without any fore planning. Wireless control often ends up in the hands of a user-space program instead of in the OS (wtf?), and updates are done through a god awful activex webpage. Blech. The long term (and even short term) stability of XP these days is poor at best, and I have no clue why everyone claims to love it.

On the other hand, most people I've met who make fun of Vista, never used it. My dad was slamming it earlier "Did you ever use it?" "... No". The vast majority of complaints about it stemmed from 2 problems:

• The so called "power users" always complain about any change, regardless of whether or not it's good.
• Underpowered machines were marked as "Vista Capable" when they were not.

And to honest, 7 is quite good. This is coming from a die hard Linux user (who actually liked Gentoo).

Re:So only XP is out of luck?

[clone53421]

It isn't just a way of detecting errors, it's a way of operating even in the face of them. As such, the length of the ECC chunk is proportional to the sector payload size. So going to a larger sector does not reduce ECC overhead

It increases as the log of the payload, so yes, a larger payload does reduce ECC overhead.

Re:So only XP is out of luck?

[m1xram]

4k sectors are logical when you look at increased drive capacity, especially with multi-bit per position technology. New stuff is coming, you can't stop it. We had ISA, VESA, PCI, and ePCI. We had IDE, EIDE, SATA1, 2 and 3. I won't mention all the different types of memory we've gone through. Hardware is a moving target of innovation.

Eventually XP users will be forced to do something like the Win9x'ers had to do. Ether you can get a new box that supports the latest MS OS or you can install a version of Linux that works on your current hardware and implements the new features.

I've found that the amount of transition-to-Linux pain is dependent on the peripherals you use. For instance, my scanner gives me a hard time whenever I upgrade, but my digital camera never does.

Re:So only XP is out of luck?

[m1xram]

I love the Linux policy of free upgrades forever. Have not paid a penny for pre-broken in a box software since Win98se.

Re:So only XP is out of luck?

[Anonymous Coward]

Or, like mine, keep buying new desktops and use the corporate XP image on them...

Re:So only XP is out of luck?

[BenoitRen]

We had ISA, VESA, PCI, and ePCI.

What's VESA doing in that list? That's a display standard, not a motherboard slot.

Eventually XP users will be forced to do something like the Win9x'ers had to do.

Not all "Win9x'ers" feel forced to migrate. :)

Re:So only XP is out of luck?

[sjames]

4K is good because most OSes already use 4K as the minimum allocation unit. It matches the memory page size neatly. Because of that, they can go from 512 to 4096 without causing any losses at all. If they went to 16K (for example), many systems would suffer from having a great deal more slack space wasted.

Re:So only XP is out of luck?

[sjames]

There surely IS some justification for considering an OS "pretty much end-of-life" when it's vendor has made several efforts to kill it off and only reluctantly relented when it's replacement proved to be a genuine turkey.

Re:So only XP is out of luck?

[pz]

The ECC (Error Check and Correct) is used for error correction.

ECC stands for Error Correcting Code, as per the original derivation from number theory. Only recently has the more breathy, marketing-friendly version come into use.

Re:So only XP is out of luck?

[Magic5Ball]

TFA explains the change and its context in a much better way, without simultaneously assuming that the reader is both technologically illiterate and sufficiently knowledgeable about the concept of low-level storage units.

But thanks for the link spam.

Re:So only XP is out of luck?

[lorenlal]

VESA as in the VESA Local Bus:
http://en.wikipedia.org/wiki/VESA_Local_Bus

This was a motherboard slot, it was basically an extended ISA slot.

Re:So only XP is out of luck?

[m1xram]

This is news, but not because of the potential problems that could arise.

XP is a huge installed base of computers. A new high capacity device that won't work on them is big news. It may be big enough to embarrass MS to do something about it. Both Vista and Windows 7 have forced MS to keep XP going for netbooks. Neither of the new OSes will run with any decent speed on a 10inch netbook. Even XP is a bit sluggish.

It's interesting from a technological standpoint. Why/how would changing the sector size effect performance? What are the downsides - why wasn't it done before?

Those questions were covered quite well by other posters.

Why is it now chosen at 4k, why not something even larger?

In addition to what other posters wrote, I would add that, it, sector size, will get larger as drives increase capacity. When we see 128Tbyte drives become common watch for another change to reduce total sectors to something more manageable. We don't know what the limit will be because research continues. Currently, work progresses on storing bits in the state of a single electron.

Those questions are what make it news (for nerds). It doesn't have to break something to be newsworthy.

Sure sure. But, for the millions of XP users this is news.

Re:So only XP is out of luck?

[Martin Blank]

It's fairly rare that some critical piece of hardware comes out that currently-supported versions of Windows will never support. Besides, no one is talking about banning the older drive architecture. If Microsoft draws the line in the sand on this one, I think we'll be able to forgive them for it, especially since it will be easy to label new drives appropriately, and the older drive architecture will continue to be available. One can still get new PATA drives, after all, and they've been out of primary preference for a good number of years.

Re:So only XP is out of luck?

[Anonymous Coward]

VLB, VESA Local Bus. Available on most 486 motherboards

Re:So only XP is out of luck?

[Anonymous Coward]

The concept of MicroSoft's "support" policy is murky at best. IE6 was embedded into my version of Windows XP. Yet repeated attempts to get them to fix problems with it go unanswered. So what's the point of extending XP's support cycle if they aren't going to fix problems with it?

Re:So only XP is out of luck?

[hairyfeet]

It also says in the article you linked to that there are edge cases where Windows 6.x (Vista and 7) will choke on 4Kb sectors and that furthermore just about every imaging software out there will misalign and screw things up in 4Kb sector schemes. Ummmm...no thanks? Really do NOT want to risk having every image I make be completely worthless and having those of us that perform backups be in the same boat as those that don't. Of course than the imaging software manufacturers can gouge us for the "new improved" version that will work...ummm no thanks again?

Re:So only XP is out of luck?

[kill-1]

The alignment is another thing. But the read-modify-write extra accesses can still happen with the right alignment so the drives will be slower in compatibility mode.

Re:So only XP is out of luck?

[GasparGMSwordsman]

According to the Anandtech article, only the pretty much end-of-life Windows XP is out of luck. Linux, OS X and modern Windows versions all work ... Non news?

According to many estimates that is 3 out of 5 computer users have XP as their OS. That is pretty big. The lowest estimate that I have seen is that XP is only running on 1 out 5 daily use computers. Another way of looking at it is that there more XP users than there are, Unix, Linux, Mac OS (any version) and any other non-Windows OS combined.

Sources: (62.2%): http://www.w3schools.com/browsers/browsers_os.asp

(69.05%): http://marketshare.hitslink.com/operating-system-market-share.aspx?qprid=10

(48.47%): http://gs.statcounter.com/press/encouraging-start-for-windows-7

(70.48%): http://gizmodo.com/5398689/reality-check

Also the expected end of life is no earlier than April 8, 2014, four years from now...

Re:So only XP is out of luck?

[LWATCDR]

Microsoft was still selling XP as of October of this year. In fact XP is still for sale new from New Egg. http://www.newegg.com/Product/Product.aspx?Item=N82E16832116515&cm_re=XP-_-32-116-515-_-Product
Microsoft better start to patch this issue or the drive makers better mark them as not compatible with Windows XP.

Re:So only XP is out of luck?

[WuphonsReach]

It's the "devil you know". XP has many pitfalls and issues, but it's well known, works well enough in a multitude of environments. It can run on systems from 256MB boxes with a 1GHz single-core CPU up to a 3.5GB unit with quad-core 3GHz.

Vista's design and launch was a complete disaster. Both from the technical side and from the PR side. Naturally, people are going to choose XP over Vista if given a choice if only for familiarity. Vista also launched right when the economy was headed into a recession - really bad timing. Not to mention that Vista came out ~2 years after it was supposed to and changed/broke a lot of things.

I suspect we'll see security updates for XP through at least 2014 or 2015. Microsoft will be forced to do so by the OEMs and large corps (and possibly the government).

I plan on setting up our first Win7 test box in the spring, and if it goes well, any machines purchased next summer/fall will be Win7 and we'll consider upgrading older PCs. (All of our "modern" PCs are dual-core with 2GB of RAM, so we're in good shape.) I believe that holds true for most other corps, test now, deploy sometime in 2010 at the earliest.

The other issue at hand is that PCs are no longer on a 3 year upgrade cycle. Most corps are now on a 4-5 year cycle or longer, and small business typically work on a 6-8 year cycle (skipping every other generation). Modern multi-core machines are going to have very long legs because the multi-core lets them stay responsive under heavy load, unlike the older single-core designs. Most users won't care if it takes a minute or two to print something, as long as the UI stays responsive (they can go check their email or look up something else).

Re:So only XP is out of luck?

[plague3106]

Why do you expect a newer harddrive with this feature to work? Do you complain too that you can't get an integrated radio + gps in a car that never had the option?

Re:So only XP is out of luck?

[plague3106]

Sounds like a great way to ensure higher support costs. Oh.. you're the support?

Re:So only XP is out of luck?

[poetmatt]

where do you come up with this? People spend millions to billions on software every year, even at small companies that sport their own batch of developers. Having that software cost removed simply removes a cost.

Re:So only XP is out of luck?

[petermgreen]

USB drives will be the worse though since 4k drives formatted for XP won't work with Windows 7 and vise versa
Do you have a source for this claim? it seems very unlikely to me.

From the operating systems point of view these look, it's just that they will perform badly if the partitions are misaligned so I don't see how it will cause drives formatted on one system to fail to work on another.

Re:So only XP is out of luck?

[alen]

for the first 7 years they release patches, service packs, tools, add ons like Powershell and you can call support for issues. they will also provide developers with API and other support in developing applications for the OS

for the last three years it's only security hotfixes

new features like reading these new drives will probably never be added and are for a new version of the OS which is completely fair

Re:So only XP is out of luck?

[petermgreen]

Sidestepping your ignorance or deliberate deception on periods of typical Linux support contracts
He didn't say if he was stating lengths from release or length of overlap (to me the latter is the more important figure)

Who cares if support goes out 10 years
It's 10 years (5 mainstream, 5 extended) minimum from release, 7 years (2 mainstream, five extended) minimum overlap between releases and 2 years (all extended) minimum overlap if you skip a release. IIRC XP will have exceeded all of those.

if you can't buy a new hard drive that will work with the OS?
These "advanced format" drives will work fine with XP, they just require a little extra effort (either using a third party paritioning tool, fitting an extra jumper to change the sector mapping or using the WD tool to realign the partitions after setup) if you want maximum performance. Besides I can still by PATA drives so I doubt these drives will be the only ones on the market any time soon.

Similarly if I go to almost any major vendor I can still get computers and computer parts that are supported with XP, some of the consumer crap isn't but virtually every buisness machine and seperately sold peice of hardware i've seen lists XP as supported.

It's articles and comments like this that give me difficulty discerning what exactly Microsoft "support" entails.
For most of us the most important part of the support is continuation of security updates (though they have occasionally refused to release one that they really should have released by claiming that it's not nessacery in a default environment), I would be very uncomfortable running exposed systems (and I coun't any machine used to browse the web as exposed) on an OS that was no longer getting security updates.

There is also problem support and non-security hotfixes (free if created while in mainstream support, pay for if created during extended support) but for most of us these are fairly irrelevant.

As I alluded to above though what really matters is support from third party vendors, I can still buy the latest hardware and run XP on it with no problems, just try doing that with a comparable aged linux distro (e.g. debian woody).

Re:So only XP is out of luck?

[wtfbill]

I feel your pain, but if you're not slipstreaming the stuff in before the install (writing a batch file for this is TRIVIAL), you're doing it the hard way. No reboots beyond the standard installation reboots, either. MUCH easier. Just sayin...but yeah, if that's what you use it for, that's a pretty good definition of obsolete.

Re:So only XP is out of luck?

[hkmwbz]

Who cares if support goes out 10 years if you can't buy a new hard drive that will work with the OS?

How is it Microsoft's fault if new hardware isn't working with XP?

Re:So only XP is out of luck?

[iserlohn]

You mean I can get free access to detailed docs on the NT kernel?

Re:So only XP is out of luck?

[VGPowerlord]

The concept of MicroSoft's "support" policy is murky at best. IE6 was embedded into my version of Windows XP. Yet repeated attempts to get them to fix problems with it go unanswered. So what's the point of extending XP's support cycle if they aren't going to fix problems with it?

Odd, I recall Microsoft releasing IE7 and IE8 for XP. It's your problem if you choose not to install the new versions.

Re:So only XP is out of luck?

[VGPowerlord]

MS has a clear support policy. Maybe you like Apple's 3 year support policy better than Microsoft's 10 year 7/3 policy?

I know I do, considering how bad 10-year-old Microsoft operating systems are.

Yes, it would be nice if they'd finally kill support for ME. Actually, haven't they? I was under the impression Windows ME came out in late 1999, which would mean it's no longer supported.

Seriously, though, XP is 8 years old now (late 2001 to late 2009).

Re:So only XP is out of luck?

[petermgreen]

Wireless control often ends up in the hands of a user-space program instead of in the OS (wtf?)
Yeah I can see that the mess of different wireless configuration systems would be a PITA for a support tech, for most of us it's not that big a deal though.

and updates are done through a god awful activex webpage
Again not ideal but once you turn on automatic updates you can mostly forget about that.

The long term (and even short term) stability of XP these days is poor at best
Maybe it's something with your particular hardware or software selection, i've generally found XP more than stable enough.

and I have no clue why everyone claims to love it.
IMO an operating systems job is to handle my hardware, run my applications and otherwise stay out of my way and use as little resources as possible. XP IMO does a better job of this than vista (though I will admit that my only experiance with vista was pre-sp1). I haven't had a chance to try 7 yet.

Re:So only XP is out of luck?

[VGPowerlord]

Strangely enough, Windows will let you format NTFS drives with 512-byte allocation units (as well as 1k and 2k). I guess this option will have to go.

Maybe it already did. It's been a while since I formatted an NTFS drive after Windows itself was installed.

Re:So only XP is out of luck?

[obarthelemy]

No, I mean you can get remote desktop, directory synching, and VLC to work, and get docs on how to do it. The first 2 of those I failed to do nor find current docs about in Linux, and the 3rd one took me about one hour of trial and error, which is too much.

But, I'm a lowly user, not a roxxor kernel haxxor.

Re:So only XP is out of luck?

[Leebert]

where do you come up with this?

I've been around.

Having that software cost removed simply removes a cost.

No doubt. The point I was making was that there is a major cost in short upgrade cycles. Just because the software is free does NOT mean the upgrade does not cost.

I assert that removing the software cost in most production enterprise cases does not remove the majority of the cost of an upgrade. Testing, migration, training, smashing inevitable resulting bugs, and similar activities make up the majority of the cost, in my experience.

Re:So only XP is out of luck?

[Anonymous Coward]

The fix is to upgrade to IE7 or IE8, which have been available for quite some time. It's not their fault you continue to use a broken and non-compliant browser when they have supplied a better one at no additional cost.

Re:So only XP is out of luck?

[yuhong]

plain OS installation disk

With what Service Pack integrated?

Re:So only XP is out of luck?

[aztracker1]

I remember similar issues with drives larger than 16-32GB when they came out, Also had an issue with my 2.25TB raid originally... all I can say is big f-ing deal. You're using XP, stick to smaller/older drives.

Re:So only XP is out of luck?

[Anonymous Coward]

XP's replacement came out way back in 2002. It was called Mac OS X 10.2.

Re:So only XP is out of luck?

[Jonner]

Since "Linux" is not one operating system as Windows is, I'll stick to a Linux-based OS that is easy to install, very popular and that I know best: Ubuntu.

With respect to your tasks, Google instantly found me: http://www.howtogeek.com/howto/ubuntu/allow-remote-control-to-your-desktop-on-ubuntu/ , http://www.ubuntugeek.com/unison-file-synchronization-tool.html , and
http://www.videolan.org/vlc/download-ubuntu.html

I'm no kernel hacker (hax0r implies cracker) but I do know how to type search terms into Google. I'm genuinely curious about how one would quickly figure out how to do those things on Windows without using a web search engine.

Re:So only XP is out of luck?

[Anonymous Coward]

Maybe AFD for Advanced Format Drive?

BFD (Big Format? Drive)

Re:So only XP is out of luck?

[Bigjeff5]

... and that furthermore just about every imaging software out there will misalign and screw things up in 4Kb sector schemes.

That's why I'm in love with ImageX, it basically gives you high-tech zip files at a fraction of the size of sector-imaging software. It doesn't matter what format the original and new drives are, and by the same token sector size is also a non-issue - it has completely eliminated that problem. Unfortunately, because it is file based and a Microsoft tool, you can't use it on a filesystem that MS can't read or write to natively. I'm not sure if you can get around that limitation with the various filesystem drivers and disk imaging tools or not. That is one area in which sector copying always works no matter what. New versions of Ghost are similar to ImageX and are capable of handling at least ext2.

Re:So only XP is out of luck?

[TooMuchToDo]

My god man! Are you not slipstreaming SP3, all the hotfixes, etc into your install disk? That should cut your install time down to 30 minutes.

Re:So only XP is out of luck?

[Bigjeff5]

XP can't natively format anything larger than 32gb, but that's just a limitation of their formatting tool. Third party tools (HP has one that is very good) don't have this limitation and XP reads the disks just fine, obviously.

More than likely these new drives would just need an XP driver, just like the old days. It isn't common now, since most hard drives operate almost identically and so a few drivers cover almost every manufacturer, but it used to be that you got a driver disk included whenever you bought a new hard drive. We'll just start to see that popping back up since XP still makes up a large portion of the market.

Re:So only XP is out of luck?

[Mr. DOS]

...once you turn on automatic updates...

By this and your previous remark, it looks like you've never done tech support. If so, then you've no idea how broken Automatic Updates can be (and make things).

      --- Mr. DOS

Re:So only XP is out of luck?

[Anonymous Coward]

What the hell are you still using a 10 year old operating system on the latest hardware for anyhow?

Re:So only XP is out of luck?

[Antique Geekmeister]

_I_ use OS images, by preference. Not everyone has the leisure to do so, nor does that OS image necessarily work on all hardware. Leading edge aptops purchased with company money by salespeople who insist on the latest advertised features,in particular, have some rather odd chipsets and can require vendor provided media to provide basic network and touchscreen functionality.

In this case, it had been just such a laptop from a vendor who makes it difficult to download drivers for slightly out of date hardware. So we used the vendor's installation media, XP with service pack 1.

Re:So only XP is out of luck?

[Antique Geekmeister]

When the size of the patches, slipstreamed or not, approaches the size of the OS, it's obsolete.

Re:So only XP is out of luck?

[hairyfeet]

But it doesn't sound like it'll make disc images, does it? I frankly couldn't give a shit about backing up my files, as I have them stored on at least 2 drives and the important ones also have off-site backups. What I do NOT want to have to do is spend hours reinstalling and setting everything back up the way i like it.

I have been using a really nice software from Paragon* that does pretty much anything you'd want to do with a disk, along with the Windows 7 built in disk imaging, so I've been having good luck so far in that regard, but the LAST thing I want is to have my images borked by some new sector size BS. Not to mention i dual boot XP and Win7 so i have no idea if having both XP and 7 on the same drive will cause problems. Does anybody know if using the WD XP tool will slow down or Bork Win7?

*-For those that don't want to pay the $35, which it is WELL WORTH BTW, you can keep an eye out on Giveaway of the day as Paragon is really good about placing their tools on there when a new version comes out. Between that and Primewares (great site, lousy name) and its excellent freeware only search engine you can pretty much completely outfit a new PC for zero $. But for disk imaging I've found you really can't go wrong with paragon. Really solid software IMHO.

Re:So only XP is out of luck?

[Bigjeff5]

Case in point, the very large company I work for spent millions of dollars just determining how much it would cost to upgrade to Vista from XP. Given the fact that volume licensing costs are in the neighborhood of $30-50 per license, I can tell you that the cost of the software itself was insignificant. The real cost was in upgrading all of the custom software that would not work with Vista. Given the scope of a global upgrade project, I imagine this cost was in the billions of dollars range to be large enough to cancel the project. The licensing itself would have only been in the millions of dollars range.

What free software advocates often neglect to consider the costs incurred by the free upgrade. A change in one system sometimes necessarily makes another system incompatible. Managing the changes to the system to ensure compatibility within upgrade cycles cost time and effort. Time and effort both cost significantly more money than the upfront price of a piece of software. Linux and Windows are both expensive in this way, it is the nature of software, but Linux tends to be much more expensive than Windows. Often the difference is more than enough to make up for the cost of the paid software. To expand my previoius example, a move from Windows to Linux has no hope of being cheaper than a move to Vista, and the move to Vista was far too expensive.

Because of the enhanced XP compatibility mode, a move to Windows 7 would likely be significantly cheaper than a move to Vista, which is exactly why Microsoft put it in the business and ultimate editions.

Re:So only XP is out of luck?

[Grishnakh]

I think they're great news. Linux users can plug them in, and they'll Just Work (tm), and the same for OS X users. WinXP users will have all kinds of problems, but that's par for the course for using Windows. Hopefully, with so many people still using XP, this will become a giant problem, showing how bad Windows really is.

Re:So only XP is out of luck?

[Bigjeff5]

What about when networking doesn't work? I can tell you from personal experience, it's a bitch. The man pages in Ubuntu suck, and having all your good documentation online is not necessarily a good thing. I had to rely on a windows machine to figure out how to fix it.

Also, Linux is not one operating system the same way Windows is not one operating system (there are ten major variations of Windows). In fact, Windows fits the distinction better than Linux, as each new version of Windows far more different than the various flavors of Linux. For Linux distros using the same kernel, the differences are generally similar to the differences between Windows Home, Professional, and Ultimate editions. Yeah, not as many flavors, but they all interact with each other quite well.

I've played with Linux off and on over the years, but my primary experience is recent - I switched to Ubuntu from Vista after Vista hardware issues pissed me off. I eventually switched back to Vista after about a year due to the exact same problems the GP was talking about.

The truth of the matter is everything is easier in Windows than in Linux. Linux was made by developers for developers, and that simply results in a poor non-developer experience. I've never had to edit an install script in Windows to install a program properly, I definitely can't say that for Linux.

Oh, and the term "hacker" has been used (incorrectly) in place of cracker for at least a decade now. They even call script-kiddies hackers. Hax0r is just leet speak for hacker. So, of course hax0r implies cracker, because hacker implies cracker even though there were originally distinct differences between the two.

Re:So only XP is out of luck?

[Bigjeff5]

You realize that's less than 1/3 of Microsoft's standard support cycle? Windows 2000 is only just going out of support this year, and XP will have another 5 years or more of hotfixes left on it, due to Vista's late (and horrible) release. That puts an Ubuntu LTS support term at less than 1/4 of Windows XP's. Microsoft doesn't sell XP any more, but they are a long way from stopping support for it.

The server picture is about the same.

Re:So only XP is out of luck?

[Bigjeff5]

Not really, most business machines come with a smaller than 200gb hard drive. It will be a long, long time before they begin coming with anything greater than 1.5tb, and all of the current drives up to that level will still be available when that happens, just like PATA drives are still available today. By then, I'm sure the lumbering behemoths will have switched to Windows 7, especially since the business edition will be a far less painful switch than the Vista switch was.

Re:So only XP is out of luck?

[Bigjeff5]

I don't know, if the drive fails, the support contract will supply a new HDD, they're all the same right?... not.

Except that that particular problem will be easy to avoid, since XP compatible hard drives up to 1.5tb will not be going anywhere for the next decade or so. Just look at PATA drives, you can still buy those if you need to, and up until a few years ago a lot of business machines still shipped with them.

Hardware manufacturers know how to play the game, they will make sure they keep their customers happy.

Re:So only XP is out of luck?

[Anonymous Coward]

IPv4 4 life! ipv6 just looks ugly

Re:So only XP is out of luck?

[raynet]

XP can be installed on partitions that are properly aligned and you also get nice speed boost. After aligning XP to 64kB stripe size of my Intel ICH9R raid5 block size, the write speeds went up from 1-5MB/s to 50-100MB/s. The partition just has to be created in CLI tool before installing XP, so majority of people cannot do so, but maybe they should be installing a more modern Windows.

Re:So only XP is out of luck?

[Bigjeff5]

By this and your previous remark, it looks like you've never done tech support. If so, then you've no idea how broken Automatic Updates can be (and make things).

Based on this and your previous remark, you've obviously never used WSUS. If so, then you've no idea how easy it is to avoid unnecessary or ill advised updates that can interfere with your current operating environment.

And if it's the Automatic Updates itself that is broken, shit man, System Restore! Or just re-image the drive. You do use imaging in your environment, right*?

* I am, of course, assuming you are in a corporate environtment. If you are dealing with consumers and not employees, god help you. You simply cannot blame the stupid shit people do on the OS. "I deleted the Windows directory, to save some space, but now my computer doesn't work!" *facepalm*

Re:So only XP is out of luck?

[raynet]

That formatting limitation in XP only applies to FAT32, you can make much much bigger NTFS formatted partitions.

Re:So only XP is out of luck?

[dbIII]

Unless you use stuff like MS Exchange - then software costs start to exceed hardware. Of course time costs to run such stuff rise as well, but just assume a perfect install that requires no more support than competing products because that is how it is advertised.

Re:So only XP is out of luck?

[Mr. DOS]

God help me, then :P I haven't seen a deleted system32 directory yet, but you're right, what I'm talking about can't be entirely blamed on the OS.

      --- Mr. DOS

Re:So only XP is out of luck?

[Anonymous Coward]

If you were running Debian woody you can still get all the packages that were there (you have to modify sources.list) Also you could upgrade woody to a later Debian release free of charge... You could dist-upgrade to Lenny which will have support fo the new drive standard (Etch possibly would also have) (O.K this plan is not exactly easy, I have done the equivalent various times on Ubuntu servers successfully)

Re:So only XP is out of luck?

[obarthelemy]

Thanks for the Unison link, I'll try it. I was hoping to just use rsynch, but it seems there are issues with NTFS timestamps, right now it's re-synching everything everytime, which takes way too long (1.5TB drive).

For VLC, there is an issue with X on ATI IGP chipsets, I was only getting a slideshow. I got that fixed, though I don't remember how.

For remote control, I find VNC extremely slow and ugly when controlling a Linux PC from a Windows client. I was trying to get the nomachine RDP server to work to see if the result was better, I couldn't. I'm sure I'm missing something fairly basic, but the doc didn't really help, either.

These 3 functions work well right away in Windows, so I haven't even had to look for doc. It did take me a while to figure out that the right way to look for docs for ubuntu was google, not ubuntu's site ^^

Re:So only XP is out of luck?

[dbIII]

but Linux tends to be much more expensive than Windows

In nearly all cases the old software will still run on the new system so I really don't know where you get that from. For example "xv" still runs on new systems and it's got nothing but a few patches for new graphics formats since 1994. The only thing I've hit problems with is macrovisions obscentity of a licence manager "flexlm".
Now while migrating from MS windows to linux may be expensive upgrading linux systems is not. If you meant the migration is expensive please say so instead of misleading readers. If you didn't mean that please elaborate with a real example.

Re:So only XP is out of luck?

[afidel]

Just like the HP EVA and some other enterprise arrays do for Windows 2003 and other non-aligned OS's running on VMWare.

Re:So only XP is out of luck?

[Bigjeff5]

Don't get me wrong, corporate employees are just as bad as average joe customers, you just don't often get the same control over average joe's computer initially, allowing you to make things like standard images or have all the software your customers will need on a restore available for install. Without that control things start to get hairy.

Re:So only XP is out of luck?

[moosesocks]

I'm not arguing that Microsoft doesn't have an unusually good long-term support system -- they're second to none.

However, it's important to consider that Windows 2000 was pretty much only supported on paper for the past few years.

It's also notable that Microsoft's release cycle has been extremely slow since the release of Windows 2000. XP was an unnecessary upgrade for many business customers, and was really only adopted in the enterprise as Win2k-era hardware became obsolete (it also helped that the two operating systems were extremely similar -- most of XP's extra features could easily be disabled)

As we all know, Vista was crap, and Win7 only barely made it into the decade (still missing a handful of features that were originally slated for Vista).

Ubuntu, on the other hand, has made a dozen stable releases since 2005, most of which have been fairly painless upgrades. In fact, I'd call a Ubuntu distribution upgrade fairly comparable to a Windows service pack update in terms of time, difficulty, and compatibility.

In that regard, Ubuntu and Windows occupy a fairly level playing field (AFAIK, Microsoft don't support old service pack releases). Ubuntu releases are timed roughly equivalently to Service Pack updates.

Re:So only XP is out of luck?

[Bigjeff5]

That only caused MS to continue SELLING it and create odd programs like their free downgrade option; their EOL terms are practically set in stone - XP's terms are a few years beyond the norm because of the late release of Windows Vista (5 years after XP, instead of the usual 2-3).

Re:So only XP is out of luck?

[Tensor]

So we should wait only 10 years to have HDDs larger than 2TB ?

Re:So only XP is out of luck?

[Bigjeff5]

Try four more years before end of life, and that means it will still be supported after the next Ubuntu LTS version is EOL.

Re:So only XP is out of luck?

[Bigjeff5]

Ubuntu, on the other hand, has made a dozen stable releases since 2005, most of which have been fairly painless upgrades.

I completely disagree. If you did not make every incrimental upgrade in order before the next small upgrade came out, the upgrade process was a nightmare. Nothing like installing Windows SP2 on a plain jane Windows OS, skipping SP1, or doing the same with SP3 skipping 1 and 2 both. MS hotfixes generally took care of themselves with auto-update, even if you were missing a lot of updates. With Ubuntu, if you didn't dilligently fix each small error as they showed up in an upgrade cycle you'd end up with dozens of problems, even hundreds if you waited more than a few months. It was frustrating to no end, you're locked into the upgrade cycle even tighter than you are with Windows updates, it's ridiculous.

I admit I didn't stick to the LTS releases, but Linux changes so fast you end up with an even worse experience if you do use only the LTS anyway.

Re:So only XP is out of luck?

[geminidomino]

but Linux tends to be much more expensive than Windows

In nearly all cases the old software will still run on the new system so I really don't know where you get that from.

It depends on the distro. Since Ubuntu seems to be the current darling (and the one I have personal experience with), that's not as much of a given as you may think because of their LTS policies. Suppose there is a necessary communications package that connects to 3rd party systems, and those systems change their protocols. The package is updated to work with the new protocols and gets rolled out in a new version. The new version won't show up in the LTS, ever, because of their policies on it, making things a little weird (or a lot weird, depending on the number of systems running it).

I am, of course, referring specifically to pidgin in my example (this has come up many times, every time one of the big 3 decides to screw with 3rd party clients), but it's not hard to see that its not impossible for an actual IMPORTANT app to fall into the same trap.

Re:So only XP is out of luck?

[adolf]

You forgot some other somewhat cold and dead things: MCA, EISA, PCMCIA, PCI-X. And ESDI, parallel SCSI, ST-506 MFM, and the similar RLL. And, let's not forget the current Newfangled Great Idea that Nobody Gives a Fuck About, BTX.

Why should I assume that 4k sectors won't just be a small and dismal blip on the technological radar like http://en.wikipedia.org/wiki/Micro_Channel_architecture>Microchannel or EISA was, as just another seemingly good idea that nobody ever actually wanted badly enough to buy it?

Re:So only XP is out of luck?

[wtfbill]

Right there with ya, friend; I don't use XP myself. It's a pain to fresh install something that heavily patched. But if you're forced to install XP for granny or something, like your original post seems to indicate, slipstreaming is the way to go. Saves a LOT of headaches. Another approach that's worked for me in the past is to nlite the thing and rip out a lot of the heavily patched subsystems and replace those with OSS equivalents. Of course, don't do THAT for granny. That helps a lot if you want to get rid of about 25-50% of the updates. BTW, you can do that for Vista and 7 too w/vLite...built a version of Vista that boots in 30 seconds on my laptop and installs from a thumbdrive in 12 minutes. Still not as useful as most Linux distros outta the box, but it's a real improvement over stock MS installs. No cruft (well, for MS anyway) and rock stable. All drivers and updates there from the get-go and a wee bit over 700 megs for the install iso fully patched. If you gotta run windows, it's nice...

Re:So only XP is out of luck?

[jonadab]

The problem with IPv6 is that it tried to do too much all at once, most of it completely unnecessary, and some of it arguably not even a good idea. The only part of it we really can't live without is the increase in address size, which could have been handled much more simply with a lot less other change.

Re:So only XP is out of luck?

[Jonner]

Well, I'm glad you figured out that the best way to search for solutions is to use Google. Remember to generalize that. Many times I've looked for something on Microsoft's site before I realized it was much faster just to use Google even if I added "site:microsoft.com".

You say that Windows XP was able to synchronize NTFS filesystems well right away, which seems reasonable. However, how well does it do synchronizing EXT3 or other native Linux files ystems? Hopefully Unison on Ubuntu will eventually handle NTFS as well as Windows does natively, but it's hardly fair to expect that when XP can't do anything with native Linux file systems.

I have also encountered many problems with video drivers on Linux-based OSes and that's certainly a weak area. However, did your ATI driver work perfectly as soon as you installed Windows? It's never worked that way for me the many times I've installed Windows; I've always had to install drivers from the manufacturer. OTOH, when I installed Ubuntu on my laptop with ATI IGP, it worked very well with no input from me. It certainly doesn't happen that way all the time, but I've never seen that on Windows.

You're certainly right that VNC is far from ideal, but it is very easy to get working.

Re:So only XP is out of luck?

[m1xram]

You forgot some other somewhat cold and dead things: MCA, EISA, PCMCIA, PCI-X. And ESDI, parallel SCSI, ST-506 MFM, and the similar RLL.

I didn't forget anything I left them out.

Why should I assume that 4k sectors won't just be a small and dismal blip on the technological radar like http://en.wikipedia.org/wiki/Micro_Channel_architecture>Microchannel or EISA was, as just another seemingly good idea that nobody ever actually wanted badly enough to buy it?

Don't assume anything, see what the market decides like other intelligent shoppers. Tell me you didn't buy an EISA or Microchannel system. lol

Re:So only XP is out of luck?

[Anonymous Coward]

You guys have to go to work every day whether you're upgrading or not - the 'expense' then, is that you might actually have to do something.

Re:So only XP is out of luck?

[adolf]

Ok, I'll tell you: I went from an XT, to a short series of 386SX's, to a VLB 486, and then into the PCI-equipped P5 era that we still seem to be unable to totally escape. None of these machines had EISA or MCA.

Re:So only XP is out of luck?

[Anonymous Coward]

What? Of all the proposals that came out of the ROAD/IPng process the SIPP proposal -- which is what IPv6 is -- was the most minimal change.

It consisted of: 128-bit addresses with minimal class/AFI semantic, elimination of the header checksum and IP fragmentation, a useful clean-up of MTU issues, elimination of the option field (replaced with header extensions), an attempt to make existing higher-order optional configuration systems mandatory (multicast, stateless autoconfiguration, and network layer security -- all of which were defined to use essentially the same mechanisms as IPv4 was already using), and that's essentially it.

Oh wait, "ttl" was renamed to "hop limit" and the timer-based decrement requirement formally expunged.

"Most of it completely unnecessary" -> address length was considered necessary, the other features were certainly useful if not strictly necessary. The timer-based decrement of ttl had already been abandoned because of implementation complexity and the reality that queues as long as a full second became exceptionally rare in the early 1990s, so this change just codified existing practice. The unnecessary stuff was the optional->mandatory featureset and the changes to the options processing; the first divides into "ignored" and "already in place in practice anyway" and the second was aimed at a fastpath/slowpath router architecture where the fastpath was used for optionless packets (i.e., most packets), which was popular at the time. It also added room for expansion of the semantics encoded into a packet header.

Things that were not done: any meaningful change to IP hop-by-hop longest-match routing (for unicast/anycast), any separation of the overloading of network layer address and host identifier, any meaningful change to multicast forwarding or routing, anything more than a "gigantic address space makes collision less likely" approach to address collision avoidance, any consideration at all of address ephemerality, any support for interoperation within a heterogeneous catenet (i.e., dual stack was inevitable from day 1, which killed rapid widescale deployment, because v6-only hosts could not talk to v4-only hosts without complicated application layer gatewaying and so forth),

Consider the other proposals:

http://tools.ietf.org/html/rfc1752#section-7

CATNIP would have created an extensible "superheader" which would allow for a heterogeneous catenet, preserving most semantics among the differently-protocoled internetworks concatenated together; the "superheader" could be used natively. Superheaders grow "upwards" so if new semantics were required in parts of the global big-I Internet -- to support multimetric routing, for example -- that would be straightforward to do "in the middle" involving one or more networks (ISPs, corporates, whatever; it does not have to be the backbones). Modern NATs evolved out of (or at least very strongly informed by) CATNIP; even IPv6/IPv4 translation is being explored, which is a very CATNIP-esque idea. In particular, CATNIP did not require hosts to talk CATNIP, or change themselves at all. No dual-stack.

TUBA would have used an OSI profile and take advantage of the work already done for OSI. There was a substantial overlap of people involved in OSI and Internet routing work at the time, and pooling resources seemed like a good idea for people working on e.g. DECNET phase IV (and IS-IS) and the future of EGP/BGP protocols. OSI was also mandated by the US federal government (they recently did this for IPv6 too) so router vendors could move the packets around with existing routing systems reasonably well. CLNP as a packet format had some attractive features, and the addresses were extensible and "automatically" (a) split into network-routing-part and host-identity-part and (b) the network-routing-part had level-hierarchy specified already. TUBA would also have been dual-stack. On the other hand, TUBA *may* have been able to inh

Re:So only XP is out of luck?

[DrXym]

It does remove a cost, but if support is not free (and in the case of most commercial Linux dists it is more expensive than Windows), then you're removing one cost and getting another one. Yes paid support is optional, but I bet most Red Hat / Novell customers pay for it anyway. Linux always offers the "go it alone" option of course which is great for individuals but the reality for enterprises is they still have to pay somebody to find & build patches and have the technical wherewithal to apply them.

Re:So only XP is out of luck?

[Seahawk]

Even popular games that are shipping now do not run under it: that tells me it's obsolete.

I would guess that is only because DirectX 10 is not supported on XP - and the reason, I believe, is because MS wanted to force gamers to Vista(It was kind of problematic that people that were first movers did not really move to Vista)

Re:So only XP is out of luck?

[alfielee]

more likely Steve has his eye on dancing lessons for his next job. In the last job he danced & pranced about the stage woo-hooing like a trained seal on drugs & he's still reeling from the poor scores he got.

Re:So only XP is out of luck?

[Antique Geekmeister]

I can't speak to all the games for which this is true. Halo 2, from Microsoft, is an example of them deliberately blocking its use on XP, without the DirectX issue, and there are published hacks to make Halo 2 work on XP. (I ran into that when testing Halo 2 problems for an eager relative who lacked a Vista box to play it on.)

But this is the sort of thing that occurs for all older operating systems.

Re:So only XP is out of luck?

[Leebert]

You guys have to go to work every day whether you're upgrading or not - the 'expense' then, is that you might actually have to do something.

Two points.

1.) I do things other than running on a constant upgrade cycle.
2.) If my job is constant upgrading, and my company can get rid of the constant upgrade cycle, then they can save the money they are spending on me.

Corollary to #2 above: If your job is threatened by process efficiency, automation, etc., make absolute certain that you are the one to suggest and implement the process efficiency, automation, or whatever. Instead of losing your job, more likely you'll find yourself suddenly indispensable.

Re:So only XP is out of luck?

[rytier]

I admit I didn't stick to the LTS releases, but Linux changes so fast you end up with an even worse experience if you do use only the LTS anyway.

This might be a mistake on your side - AFAIK the Ubuntu team plus their beta-testers are mostly focused on upgrades from one previous version, PLUS previous LTS version of Ubuntu. So upgrades between LTS versions are usually well tested and discrepancies documented, if not solved before release.

So to get out of the cycle, just wait for next LTS version expected in April 2010, and then wait for another LTS version... unless they change their release process in the meantime :-)

Re:So only XP is out of luck?

[denis-The-menace]

Your issues with NTFS timestamps are related to NTFS using UTC (Greenwich time) for timestamps.
Therefore every time you go to (and from) Daylight Savings Time (DST) all your timestamps get changed by one hour!

You create a CD/DVD/tape backup and the timestamps are frozen but on the HD NTFS will change them. If you move your server to a new timezone ==> NTFS will changes timestamps when you change the timezone setting on the server.

Q: Is NTFS the only file system with this wonderful work-creating feature?

Re:So only XP is out of luck?

[Anonymous Coward]

Games shipping that don't run on XP? Pray tell, enlighten me as to which of the current crop of releases... for the PC... don't run on XP?

No, Halo 2 for Windows honestly doesn't count either. Tell me one that wasn't put out by a monopolistic vendor of operating systems.

Re:So only XP is out of luck?

[wampus]

No one was talking to you, dipshit.

Re:So only XP is out of luck?

[nuckfuts]

For cripes sake. Does every comment here have to trigger an anti-MS response? He's not "defending" anything, and made no mention of XP in his post. The example was that manufacturers may choose not to implement a new technology when the old one is still available and working. IPV6 is a perfect example of this.

Re:So only XP is out of luck?

[Bigjeff5]

The cost of a piece of software involves a great deal more than just whether or not it will still run on a new system. Things like whether or not it must be configured for each individual machine/user, or if everyone can use the same configuration, how often it breaks, when it breaks how easy it is to fix, etc. Things like that can add a great deal to the cost in support dollars, which are far more expensive than all but the most expensive pieces of software. If it takes more than an hour to tweak a free program versus an equivalent non-free program, you've almost certainly blown your savings by going with the free software.

This is why Linux is almost always more expensive than Windows, at least in my experience. When something breaks in Windows, a system restore or re-installing/repairing the application fixes it, whereas I have yet to come accross a fix for a problem in Linux that was simple and straightforward. It took me a week to fix a sound issue I was having once - had I been on the clock I could have put a sizeable chunk of cash aside for a new computer with the time it took to fix that one little issue - one I've never come accross while using Windows, I might add.

In a corporate environment, Linux makes sense when you need expensive support personnel anyway, and going Linux can save you thousands of dollars in initial costs. That's why there are so many Linux web servers and mainframes and the like. For the desktop though, it doesn't make a lick of sense to use Linux, it's just way too expensive.

Re:So only XP is out of luck?

[m1xram]

Oh no a troll! What should I do now?

Re:So only XP is out of luck?

[dbIII]

As I thought - you are not comparing what you say you are comparing and are instead talking about MIGRATION FROM ONE ENVIRONMENT TO A COMPLETELY DIFFERENT ONE.
Please do not mislead the readers.
You are correct that farting about with unique individual desktops with unique settings will always be more time consuming than collectively managing identical systems - the problem is you are blaming it on the platform and not the unique individual desktop situation.

Re:So only XP is out of luck?

[mr_mischief]

That would be true if some ISPs weren't still supplying just a bridge to part of their customers. Those ISPs would at least have to move everyone to routers or tunnel the customer's IPv4.

Re:So only XP is out of luck?

[mr_mischief]

That really depends on how many applications you're supporting on top of the OS, whether or not they are also OSS, what outside support you have through contracts, the automation and centralization you have in place for administration of the desktops, the quality of your distro, and the technical sophistication of your users.

Imagine a scenario in which you have a solid, stable distro with good vendor support or in which your own support staff is expert and you only have to support a handful of fairly straightforward and well-implemented apps. A customer service rep often needs no more than a browser and maybe a simple word processor, for example. In this scenario, you're not likely to see more support costs for Linux, Firefox, and OpenOffice Writer than for Windows, IE, and MS Word. If you can give them AbiWord or KWrite you may get lower costs.

Re:So only XP is out of luck?

[mr_mischief]

Mandriva vows 5 years minimum of package updates for server editions. They only offer 12 months on GUI thingys and 18 months for the base system for the desktop editions, but a new version comes out twice a year. You can get the new versions of the desktop every year for a few tens of dollars with tech support or free with on-your-own support (although their forums answer pretty much anything their knowledge base doesn't). Most, but not all, software not part of the distro just works after an upgrade. Per-incident support can be bought for downloaded versions starting at $33 per incident and getting cheaper if you buy bundles. I've never needed to contact their support department, so I can't comment on the quality of their support staff.

Red Hat charges a good deal for support, but you get really good support. At least you did a decade ago.

I've not dealt with the paid support for Novell or Canonical, makers of Suse and Ubuntu. I can't comment on them.

AppleCare is expensive, but I've known people to have new Macs overnighted as loaners so they could ship whole systems to Apple for diagnosing application crashes and still keep using their other apps. Apple's support for PPC suffered near the end of the end-of-life period IMO, though. Beware of Apple products that barely precede a major shift in lines.

Try Microsoft's support on for size when you buy a PC loaded with OEM Windows from a manufacturer that has a one-year warranty. The only thing you'll get for nine of your ten years are updates and headaches. Guess how most people get Windows?

I'm not saying Microsoft's support isn't worthwhile. I'm just saying it's not as outstanding compared to everything else as some are saying. Most other OS vendors offer their support, such as it is, even when the OS is sold installed. MS's OEM pricing with the PC vendor required to do the customer contact part of support means MS isn't even responsible for most copies of Windows. You don't know how good your Windows support is until you have tried contacting your PC vendor.

Re:So only XP is out of luck?

[Trogre]

How about security updates?

Re:So only XP is out of luck?

[jonadab]

> Of all the proposals that came out of the ROAD/IPng
> process the SIPP proposal -- which is what IPv6 is
> -- was the most minimal change.

If the other proposals were even further removed from IPv4, then everyone involved in the process was an idiot.

Do you want evidence that IPv6 tried to change too much? Here's some: in a couple of days it'll be 2010, and pretty much everything is still running on IPv4 more or less exclusively. At this point there is no reason to believe IPv6 will *ever* be widely adopted, in the sense of actually replacing IPv4.

IPv6 hasn't "made it" for a variety of reasons, and they all boil down to this: too many changes, and not enough backward compatibility.

For one thing, an IPv4 address is not automatically a valid IPv6 address, and it should have been. The next-gen address space should have been constructed in a way that would make it easy and natural to reserve corresponding IPv6 addresses for all existing IPv4 addresses and allow the software to automatically make the translation. The most obvious way to do that is to divide the 64-bit address into four 16-bit words, and if the most significant halves of all four words are empty, then it belongs to the same system that has the corresponding 32-bit IPv4 address. We could even have express the new addresses in human-readable dotted-quad notation, at which point the rule would be, if all four numbers are less than 256, then it's also valid as an IPv4 address that points to the same system. For backward compatibility, systems using the new protocol could have been programmed to "fall back" TCP connection attempts to the old protocol when contact cannot be established, but only if both systems have addresses that are valid for IPv4 (i.e., no bits set in the top half of any of the four sixteen-bit words). Updated systems could have been talking to one another in the new protocol and still talking to legacy systems using the old protocol, without any extra work at all on the system administrator's part. Network administrators wouldn't even have had to *learn* anything new. DNS wouldn't have had to change significantly (except to allow numbers larger than 255).

But no, if you want to use IPv6, you have to learn a whole new addressing system, assign separate IPv6 addresses to each system that don't correspond in any meaningful way to the IPv4 ones, have separate DNS records for the IPv6 addresses of each and every system... Why is a network administrator who has IPv4 addresses available going to go to the extra effort? Where's the benefit? And if the guys who run established networks don't adopt IPv6, then there's no incentive for anyone else either, because you can't use it to connect to established network services.

And all that nonsense about trying to replace DHCP with mechanisms built into IPv6, that's just gravy on the cake. Why on *earth* did anyone think that was going to fly?

It's like the IPv6 designers had never heard of inertia.

Re:So only XP is out of luck?

[marcosdumay]

"What that's you say? Upgrade to Windows 7 and use its perfectly infallible XP mode?"

Is Wine available for Windows? Dosemu is quite sucessfull on that ninche...

Re:So only XP is out of luck?

[marcosdumay]

Hum... MS SQL server costs as much as a just-not-top-of-line server. Or maybe more now, since the crisis pushed hardware costs down (someday I'll look into buying both again, or just the server if I get luck).

Anyway, at the same price you contract the useless Microsoft support you can hire some quite usefull Linux support. Unless you simply doesn't use support and can live with the tickets MS throws at you when buying (ok, nobody really uses MS support, but what is the point then?), the only difference in pricing is on the license.

Re:So only XP is out of luck?

[wampus]

I'd suggest seeing if someone ported the troll kernel module from 2.4 so you can try it yourself. Maybe 2010 will be the year of desktop linux trolling.

Re:So only XP is out of luck?

[aztracker1]

My original reference was back in the Win2K days, when earlier releases didn't support drives of a certain size, I honestly don't recall, it may have been as low as 8GB, because of the drive specifications change. IIRC it was SP3 or 4 for Win2k that included the fix (though the fix was available before said SP).

Re:So only XP is out of luck?

[Anonymous Coward]

Eventually, you have to put a line in the sand. If you push off the deadline, manufacturers will still take their time, and they'll be in the same place 9 years and 11 months from now.

Example: IPv6.

The reason IPV6 has not taken control of the market or shown much progress is when the idea was thought up, IPV4 was running out of addresses. At the same time NAT was also thought up which negated the use of IPV6. Organizations used to need a public IP for everything.. now they just need a few and then they forward onto a private network. A lot of the US postal service (as for 7 years ago) computers still had public IP's for every computer in every branch, but Im sure they have changed to NAt for security reasons and that opened up a whole bunch of addresses for the world.

Re:So only XP is out of luck?

[Anonymous Coward]

Oh, I totally agree that the process was full of idiots.

The outcome was maximally idiotic though, precisely because of inertia.

In order to get people to transition you don't want to focus on whether there is a minimal change or a big change, but whether the new system provides real, tangible, immediate benefit to someone who converts in isolation from everyone else. Two proposals were designed to do that right away -- CATNIP and NIMROD -- both of which would provide some utility even in small networks (plausibly even on a single two-host not-publically-networked LAN).

TUBA was designed to take advantage of the fact that isolation was not strictly necessary as there was native routing of OSI CLNP across some of the Internet at the time (MITRE was a substantial user, doing actual FTAM traffic and developing its implementation profile; GOSIP was "under active development" too), and people were experimenting with using TUBA telnet to log into their Cisco routers. IS-IS was already in place, and there was scope in early BGP4 to carry OSI NLRIs (in fact AlterNet (UUNET) had at least two peer networks doing interdomain OSI CLNP with them).

The amount of IPv6 traffic on day one: zero. It was at least six or seven years before there was more IPv6 traffic on the underlying public Internet infrastructure than there was CLNP traffic (the latter died off), and it was even longer before IPv6 traffic in aggregate everywhere likely exceeded OSI CLNP traffic everywhere (several SONET/SDH management systems used OSI protocols (born in ETSI and CCITT/ITU(T) in the control channels).

IPv6 still has nearly no utility (considering that zeroconf et al.works for IPv4) advantage for small isolated subnets except some futureproofing against numbering collisions (which can in principle happen using RFC 1918 private addressing, and which can also happen -- with lower probability -- with IPv6 autoconfiguration).

You still will have big connectivity gaps (and require application layer gatewaying ("proxying" if you like)) if you run an IPv6-only node. CATNIP was designed to hunt for an acceptable frame format and so all CATNIP hosts were expected to be able to handle talking to non-CATNIP IPv4, TUBA, IPX, XNS and even SIPP/SIP/PIP hosts (i.e., compatibility with all the other IPng candidates for v6, useful if they got deployed faster or worked better in the end); CATNIP's API would have avoid lots of problems interacting with aggressive NAT, and saved developers many headaches. TUBA had internetworking in the core before the proposal was standardized. NIMROD was OK with map-and-encap systems, and like CATNIP somewhat presaged NAT (and more the current LISP interdomain proposals) and was designed to make that less painful at the API too.

One "advantage" of IPv6 was preserving the API. Of course, anyone whose gone into the guts of a server program that needs to know its own identity (for management purposes for instance) will know that it didn't even do that well. In fact, you enumerate many significantly different things one has to know when tackling IPv6. However, the argument was "inertia" suggested a minimal change *to the protocol stack* rather than a big change that produced a more compelling and more featureful API that could also manage a compatibility mode for talking with existing IPv4.

too many changes, and not enough backward compatibility

I think we're almost in full agreeement on this.

Almost, because it's too many changes to allow for casual "upgrade", and it's worse for nontrivial small-i internets.
It's also too few changes in that the new features are not compelling enough to make the pain of "upgrade" worth it unless it is absolutely necessary (e.g., you really cannot get more v4 addresses and cannot use v4 address space "compression" technologies like NAT etc.)

So, you get IPv6 propaganda saying "doom is at hand! convert now!" and "convert now! it's not THAT hard!" and "convert now! there are gre

Re:So only XP is out of luck?

[Jonner]

Ubuntu has lots of documentation that can be installed locally, including man pages, info, and GNOME documentation in addition to the Ubuntu-specific stuff. However, it's often inadequate by itself. I've often tried to troubleshoot things on Windows and Linux-based OSes using the documentation available on the local machine and I've usually not gotten very far. The reality is that usually the fastest way to figure out any problem with any OS is to use the Web, whether it's to access official documentation or forums. I don't think there's much difference between Windows and Ubuntu in that regard. Even if Windows local documentation is better organized than Ubuntu's it's often missing essential details.

If you've never had any trouble installing programs or drivers on Windows, count yourself extremely lucky. I've run into plenty of problems on every type of system I've ever used and I doubt that will change any time soon.

I made the distinction about hacker vs. cracker because I was replying to someone who claimed "But, I'm a lowly user, not a roxxor kernel haxxor." Clearly, that person meant a hacker in the traditional sense and not a cracker.

WD is already shipping them

[daha]

There are certain models of the Western Digital Caviar Green drives that are already shipping with a 4K sector size, such as this one: http://www.newegg.com/Product/Product.aspx?Item=N82E16822136490

Re:WD is already shipping them

[Pedrito]

There are certain models of the Western Digital Caviar Green drives that are already shipping with a 4K sector size, such as this one: http://www.newegg.com/Product/Product.aspx?Item=N82E16822136490 Where do you get the 4K sector size from? From here: User Sectors Per Drive 1,953,525,169 1.9 billion * 4K sectors = 7.6 GB 1.9 billion * 512 byte sectors = 972 MB Or am I missing something?

Re:WD is already shipping them

[Anonymous Coward]

Where is that indicated in any of the specs? How would someone determine if a drive has a 4K sector size?

http://www.wdc.com/wdproducts/library/SpecSheet/ENG/2879-701229.pdf

Re:WD is already shipping them

[noidentity]

There are certain models of the Western Digital Caviar Green drives that are already shipping with a 4K sector size

I think you mean 4.096K sector size. It's a hard drive, after all.

Re:WD is already shipping them

[jc42]

... a 4K sector size

I think you mean 4.096K sector size.

Heh. You're fighting a losing battle. 99% of the human population is unable to handle numbers with more than one significant digit. The tech crowd is a little better, i.e., most of us can handle two digits most of the time.

Yes, you might observe that 4.096 rounds to 4.1, but that doesn't matter, because computer types always truncate. So 4.096 truncates to 4.0 and the final ".0" is dropped because it's the default.

If you want people to distinguish numbers that differ in the 3rd digit, you have to talk to engineers. They're less than 10^-5 of the population, and you can't expect such fine distinctions in anything outside of technical journals.

(Actually, I just made up that 10^-5 number. I wonder what the real order-of-magnitude estimate is. Yeah, google knows, but it knows several estimates that differ by several OoMs. ;-)

Re:WD is already shipping them

[daha]

Excuse me: 4 KiB sector size.

Re:WD is already shipping them

[mister_playboy]

Heh. You're fighting a losing battle. 99% of the human population is unable to handle numbers with more than one significant digit.

This is just plain retarded. Phone numbers? Addresses? Weekly/monthly incomes? Prices in the store? Even mindless TV watching most likely involves at least 2 significant digits worth of channels to choose from.

(Actually, I just made up that 10^-5 number. I wonder what the real order-of-magnitude estimate is. Yeah, google knows, but it knows several estimates that differ by several OoMs. ;-)

Pulling numbers out of your ass is even less accurate than dropping significant digits.

Re:WD is already shipping them

[Anonymous Coward]

Prices in the store

You mean the .99 part of $x.99 that nobody pays attention to? Yeah, thought so.

Re:WD is already shipping them

[jc42]

Well, yeah; you'd expect that people have better memories than that. But consider: The marketers long ago learned that a price like $29.95 sells a lot better than $30.00, and it's because almost everyone truncates the former to $20, i.e. to one significant digit. People shown prices and then asked to compare the prices of things from memory will judge $29.95 and $21.95 products as being about the same price, while $29.95 and $30.95 are remembered as having a large price difference.

There are lots of examples of people behaving this way. The TV-channel one is interesting, but I've actually watched people looking for shows, and it usually becomes obvious from their search pattern that they usually don't remember the channel numbers. A few people do remember a handful of channels that they watch a lot, true, but in most cases, they just remember things like "It's 60-something" and have to search for the second digit.

I wish I were being overly cynical, but there's a lot of market-psychology research saying that it really is this bad.

OTOH, there is a fairly wide range in how human brains work, and we know pretty well that "intelligence" isn't a 1-dimensional quantity that's totally characterized by a single "IQ" number. There are people who routinely handle 4, 6, or 10-digit numbers without error. There are others who can't do this, but can remember the names and faces of hundreds of people. There are lots of intelligence "modules", and many people have one or more that works well. So the real problem is talking about a single such module, and treating it as a person's "intelligence". We're really not as dumb as that might indicate; we're each sorta dumb at a lot of things and sort smart at others. A forum like this would tend to attract people with good number and logic modules. Though I do get dubious about this at times when reading at 0 or -1.

Re:WD is already shipping them

[wastedlife]

You missed some zeroes or something. I got 7.6 TB using 4K sectors, and 931 GB using 512 byte sectors. My guess is that it is using 512 byte sectors.

Re:WD is already shipping them

[Paracelcus]

Iv'e got two of them in my dinky SOHO NAS unit.

Great place for movies and reruns of Sanford and Son!

Re:WD is already shipping them

[noidentity]

They don't all end in 99. Last visit to store (from memory), oranges were I believe $0.97 per pound, cereal I got was $2.79, frozen vegetables were $1.79 and $1.99, yogurt was $3.33, 6 eggs $2.39, frozen orange juice $1.25.

Care to provide examples?

[bogaboga]

"...This may not be a smooth transition, because some OSes do not align partitions on 4K boundaries..."

In cases like these, it always helps to provide examples. Care to do so? Thanks.

Re:Care to provide examples?

[DeHackEd]

I just checked my system. /dev/sda1 is /dev/sda + 32256 bytes, which is 63 512-byte sectors. /dev/sda2 is also on an odd-numbered sector alignment.

Fedora 11 fresh install, which is less than a year old.

Re:Care to provide examples?

[andrewd18]

I realize this is Slashdot, but both of the articles linked talk about the affected operating system. Hint: It shares an ending with a colloquial name for urine.

Re:Care to provide examples?

[tepples]

I realize this is Slashdot, but both of the articles linked talk about the affected operating system. Hint: It shares an ending with a colloquial name for urine.

Wii? PSP?

Re:Care to provide examples?

[BobMcD]

That's a terrible hint. Now I have to RTFA. THANKS!

Re:Care to provide examples?

[Gulthek]

It took me longer than it should've to answer this riddle. Shortcut for the similarly caffeine deprived: andrewd18 means "P" as in Windows XP.

Seriously, I was like "Win...dows?" "U...nix?" "Micro...soft?" "OS...X"? "BS...D"?

Re:Care to provide examples?

[Anonymous Coward]

Solar... isssss

D:

Re:Care to provide examples?

[Anonymous Coward]

It's not that other OSes aren't affected but that there's not enough of a usebase for the affected versions that anyone cares.

From the AnandTech article:

Notably, Linux and Mac OS X are not affected by this issue. Western Digital has tested both of these operating systems, and officially classifies them as not-affected. Ultimately we suspect that if you went back far enough you could find older versions of these OSes that are affected, but unlike Win 5.xx, there’s not a significant legacy user base to worry about.

Re:Care to provide examples?

[Anonymous Coward]

OS X-crement?

Re:Care to provide examples?

[mathfeel]

The riddle would be funnier and clearer if it says ...colloquial name for 10 days old urine.

intelligent interfaces

[FuckingNickName]

Why does the sector size presented by the interface have to reflect anything about the hardware? isn't this like the CHS/LBA conversion done under the hood? What about the ability to request a particular sector size, with the default being 512 bytes and the recommended amount being the hardware amount for optimisation purposes? Memories of 512 versus 2048 in the CD booting of older versions of VMS...

Re:intelligent interfaces

[tepples]

Why does the sector size presented by the interface have to reflect anything about the hardware?

If the OS clusters aren't aligned to physical sectors, the hard drive's controller has to read-modify-write all the time.

Re:intelligent interfaces

[SharpFang]

But almost the same conversion is done already!
Do you really believe your hard drive has 256 heads?

Re:intelligent interfaces

[petermgreen]

It doesn't, and indeed these WD drives will still have 512 byte logical sectors so there will be 8 logical sectors to one physical sectors.

The problem is if the partition is misaligned the OS is likely to make a load of unaligned writes. Those unaligned writes will force the drive to do a read-modify-write (which afaict will mean waiting for a complete rotation in the middle of the operation)

Add this to the fact that some systems (most notablly XP) have a habbit of aligning partitions on the boundries of cylinders* and that a typical cylinder is 63 sectors and you have a pretty much guaranteed misalignment.

It's easilly fixed if you know about it and have the right tools (WD supply one) but if you aren't aware of it you will likely get poor performance for no obvious reason.

*cylinders don't really exist on modern drives but bioses emulate them since the traditional real mode hard drive access mechanisms use them.

Re:intelligent interfaces

[Thanshin]

Do you really believe your hard drive has 256 heads?

It had only six, at first, but we didn't know the thing about burning the stumps.

Re:intelligent interfaces

[Anti_Climax]

Well played, sir... Well played

Re:intelligent interfaces

[Giant Electronic Bra]

Better not be playing 4e when you burn those stumps, might stun your drive and I don't think they get saving throws...

Isn't this just a firmware change?

[SharpFang]

It doesn't sound like the 512 bytes per sector is tightly bound to hardware. More like a low-level reformat plus change of some #defines in the firmware to transform from one to another type. Which would mean there could be i.e. a jumper setting for sector size, allowing for backward compatibility.

Also, the fact an OS doesn't enforce partition alignment doesn't mean it won't respect a disk formatted to aligned partitions. Just provide a 3rd party partitioning tool that aligns the partitions right, and install the OS on pre-made partitions. If your business depends on WinXP so much, your IT dept should be capable of doing it.

disable ECC?

[Anonymous Coward]

I heard some talks from the ZFS folks at Sun about how they were floating the idea to HD mfgr's of just disabling ECC on the drives. ZFS checksums every block, and in a RAID configuration, it would be able to transparently correct any checksum errors. I think this may have also been the motivation behind bringing triple-redundant RAID to ZFS.

The motivating idea was that this would reduce the overhead involved on ECC and gain extra space.

Thoughts?

Re:disable ECC?

[MobyDisk]

That wouldn't work with existing file systems that assume the drive does this. That's like deciding to remove the checksums from TCP and IP because a few protocols provide their own checksums. Might work in specialized cases. Probably just adds risk though for no benefit.

Re:disable ECC?

[Waffle Iron]

It doesn't seem like a great idea to me. There are a lot of different ECC algorithms and implementations. It seems to me that it would be better to let the hard drive manufacturer select one that closely matches the expected signal and noise characteristics of a particular disk drive rather than some generic algorithm in the filesystem.

Re:disable ECC?

[Jeff DeMaagd]

I can see this working for drives made specifically for RAIDs. Lose ECC on single drive configurations and you're asking for trouble. At least for RAIDS, a controller would need to be aware of this and do the remapping themselves, in the end, I don't know if it's worth doing this at all. If some enterprising RAID controller company could prove it works better to do it this way, then I can see it happening.

Re:disable ECC?

[Abcd1234]

Well, presumably the idea would be to add an ATA command which allows one to disable ECC on a drive on-the-fly. Or, at minimum, a hardware switch of some kind.

Re:disable ECC?

[Junta]

That's like deciding to remove the checksums from TCP and IP because a few protocols provide their own checksums.

Funny you should mention IP checksums, that's one feature removed from the IP layer in IPv6 precisely because the 'important' protocols do it themselves anyway (i.e. TCP).

Re:disable ECC?

[TheLink]

If they really did that, I'd say they were clueless. Such a feature would increase the odds of error.

ZFS might checksum every block. But what happens when ZFS is not everywhere? Does the BIOS or whatever equivalent support ZFS checksumming for reading the boot sectors? So those sectors better be 100% or you better be turning it off for boot drives. You have to use ZFS everywhere and for everything. For example, if you ever try to image a 1TB disk without ECC, the odds of bit errors will be high. Even if ZFS can repair it - you'd only find out much later (too late?) and likely after another error prone write.

Such a feature would just be creating more opportunities for people to get things wrong.

And for what benefit?

> The motivating idea was that this would reduce the overhead involved on ECC and gain extra space.

I think the people who'd want ZFS or RAID would rather have better reliability than the 10% or so extra space.

Even if they don't know it at first ;).

Re:disable ECC?

[butlerm]

That's insane. ECC at the hardware / firmware level corrects the vast majority of bit errors transparently in a manner that is invisible to the operating system. If you took out sector level ECC, the drives would be useless in anything other than a ZFS RAID configuration, and even then performance would drop in the presence of trivially ECC correctable errors, due to the re-reads and stripe reconstructions at the filesystem level.

Drive performance would probably drop because the heads would have to stay in closer alignment without the ability of ECC to correct data read errors caused by small vibrations and electrical noise. In addition, sector relocations would probably increase because tiny flaws that do not impair the ability of a drive to write an ECC correctable sector would force the drive to remap that sector to another part of the disk.

It is a similar issue with various wire level data transmission schemes. If DSL connections did not use error correcting codes, they would suffer much higher packet loss rates than they do now, especially at distance. Most those packets would generally get retransmitted due to transport level checksum errors, but why resort to performance impairing fall back measures when the problem can be largely eliminated at a lower level?

Re:disable ECC?

[Izmunuti]

Ugh. Sounds like a bad idea. Hard drive channels are noisy. How will ZFS fare if lots and lots of sectors read from every drive have at least a couple of bits in error? With no ECC in the drive, errors would be common.

Re:disable ECC?

[beelsebob]

The point though is that zfs does this *anyway* even if the disk does it too, so why have the disk do it too and lose more space.

Re:disable ECC?

[Anonymous Coward]

One of the reasons there is so much BS and hype surrounding ZFS...

Drives already do and have for years done error correction and most revector badblocks dynamically and the OS never even knows. It's part of how drives are so reliable as it is. Pushing that logic up in to the filesystem just introduces more complexity to the filesystem and reduces overall reliability.

Talk to any drive engineer, what they want the most is for filesystem engineers to stop trying to outsmart the drive guys. THere is not a guaranteed relationship between a sector and it's location on the disk, the drive and the firmware it has will try to figure that out the best way it can.

What they really should spend their time on is filesystems and their relationship with cache will have to change in the next decade, solid state medias don't need caching the same way disks do, I fully expect disks to start coming as hybrid devices with solid state storage and disk based storage as a singular device. There are some relatively complex problems to solve to provide media awareness to all the storage algorithms.

Re:disable ECC?

[Anonymous Coward]

An ECC should be implemented as close to where the problem occurs as possible. For a hard drive, this means on sectors, not (abstract) blocks. Otherwise you'll see the OS rewriting clean sectors that are parts of unclean blocks (if the block size is greater than the sector size), and gain nothing if the block size is smaller than the sector size.

Re:disable ECC?

[drinkypoo]

If you were going to eliminate ECC in one place or another, it wouldn't be on the drive. The drives have to operate in the real world of analog states, while the filesystem works in the virtual world of "whatever the disk actually feeds me". Disks have to have correctable ECC just to reliably give you accurate data from magnetic media at these densities. It would make more sense to upgrade the on-disk ECC and give the filesystem better access to the disk's ECC.

Re:disable ECC?

[Anonymous Coward]

Possibly not the best source (citation needed!), but according to Wikipedia (emphasis mine):

o The IPv6 header is not protected by a checksum; integrity protection is assumed to be assured by *BOTH* a link layer checksum and a higher layer (TCP, UDP, etc.)

As I seem to recall this is what I remember the rationale being - layer 2 is, generally speaking, so reliable and performs its own checksumming in hardware that having to do this at layer 3 is an unnecessary overhead (although higher end routers will probably do it in hardware too anyway) when things such as TTL need to be recalculated at each step.

If the data is really important then higher level protocols will do their own checksums of the payload at either end, but it is largely down to the _lower_ level hardware error detection. Kind of the antithesis to the subject at hand.

Re:disable ECC?

[pclminion]

That's understating it. Bit errors occur on every sector read from the disk. Let me repeat that: every single attempt to read a sector results in errors. Without ECC, hard disks simply would not work, period. The data densities are too great.

Re:disable ECC?

[Anonymous Coward]

Stupid much? ZFS uses a "plain" checksum, which is something far inferior to ECC, how do you want to use one to replace the other?
Read every sector 100 times on average till you by chance get one without an error (not that this is probably optimistic).
Well, you could just start writing you data on paper, the performance will probably be better.

Re:disable ECC?

[dfghjk]

Sounds terrible. I don't want mandatory RAID to correct for errors that occur frequently, I don't want to burden the main system with work that has always been done effectively within hardware, and I want my error handling performed close to the metal. It isn't as though space is at premium these days.

The fundamental design of RAID assumes that drives are capable of knowing when they fail and that capability relies on embedded ECC. ZFS seems to think doing a poor imitation of that on the host CPU is a great idea but I don't.

Re:disable ECC?

[dfghjk]

The point is that the disk does this already and in a vastly better and more important way so why does ZFS bother and lose more space?

Re:disable ECC?

[TooMuchToDo]

Because I trust the disk's ECC over that provided by ZFS?

Re:disable ECC?

[beelsebob]

Because actually zfs can do this in a better way – that's why *it* closes write holes in such error checking schemes, while anything with no knowledge of the file system can't.

Re:disable ECC?

[rcw-home]

It doesn't seem like a great idea to me. There are a lot of different ECC algorithms and implementations. It seems to me that it would be better to let the hard drive manufacturer select one that closely matches the expected signal and noise characteristics of a particular disk drive rather than some generic algorithm in the filesystem.

First, I suspect that disk drive manufacturers are more keen on selecting an ECC implementation that matches their drive's marketing requirements.

Second, at work, I currently have a RAID1 whose backups have corrupt checksums nearly exactly half the time. As best I can tell, for whatever reason, there's a few sectors where different data got written to the two disks in the mirror. Without filesystem-level ECC, there is no guaranteed way to tell which disk has good data - if I tell the RAID controller to do a verify, I have a 50% chance of wiping out the good data (and if I don't do a verify, LSI tech support tells me to take a hike). End-to-end checksumming is the only answer here - it's not just happening to me.

Why are there sectors?

[AP31R0N]

i'm asking because i don't know, not to troll.

What is their purpose? Does the purpose still matter?

Re:Why are there sectors?

[JordanL]

A sector on a HDD is the minimum writeable space. Think of it as a lot in a subdevelopment. If each lot is 50,000 sq. ft. on a 20 acre plot, and you move to 60,000 sq. ft. lots instead, the plot is still 20 acres, but the development now has less lots on it.

In computers, larger sectors are often better for large files, while smaller sectors are better for smaller partitions and smaller files. If a sector is 4096 bytes, and you create a 1024 byte file, it still occupies 4096 bytes on the disk, as the HDD won't write anything else but that file to the sector. If you have files that are hundreds of megabytes though, you can access the file, with minimum wastage, by using fewer sectors, which reduces thrashing and similar issues.

The discrepancy between file sizes and sector sizes is what the difference is in Windows when you view a hard drive and it displays "size" and "size on disk". "Size" is the actual file size, while "size on disk" is the amount of space the file occupies on the hard drive.

Re:Why are there sectors?

[AP31R0N]

+1 Informative.

So why have the sectors at all? Bigger sectors seem to incur more waste, though with HDDs growing at the rate they do it might not matter. But if there was no sectoring, then a file could take as much space, and only as much space as it actually needs. The 1024 byte file could then take 1024 bytes.

i'm not saying sectors aren't needed, i just don't understand why they exist in the first place.

It is it some kind of addressing thing?

Re:Why are there sectors?

[JordanL]

It is indeed. Unless HDD makers were going to create firmware, and programmers made partition formats, which address each bit individually (which itself would require an enormous amount of space... much larger than the HDD in fact), you will always be unable to live without sectors. The subdivision idea is again relevant. Imagine if every part of the 20 acre plot had to be "addressable" down to the square inch.

Re:Why are there sectors?

[DarkOx]

You are confusing physical sector size with cluster size. May file systems are already addressing data in larger blocks. 4096 is very commonly used. They are generally multiples of 512 which is the physical sector size; so that its is easy to calculate the physical sector that needs to be changed when you know the logical.

Its quite possible to have a cluster size smaller than the sector size; the file system would need to be smart enough to determine what other clusters fall on that sector and write them all though.

Re:Why are there sectors?

[JordanL]

Sector does have several definitions, but from reading the article I'm fairly certain they are talking about sector size, and not clusters.

Re:Why are there sectors?

[MBCook]

Let's take a 1TB disk, which is becoming common. To address it all in 512B sectors, you need 31 address bits (since there are 2 billion sectors).

If you change to 4KB sectors, you now have 1/8 as many, so you only need to address ~270 million sectors, which takes 28 bits of address space.

The thing is, disks are given addresses of a certain size. If all address are 16 bits, and the sectors only have 512B in them, your disk can't be bigger than 32MB. By using 32 bits, you can go up to ~2GB. If you are using 4KB clusters, you could have 16GB.

In short, it saves address lines. That prevents having to change protocols, saves memory, saves hardware costs, etc.

The FAT filesystem article on Wikipedia discusses how this same basic problem happened to FAT.

Re:Why are there sectors?

[Thanshin]

So why have the sectors at all? [...]
The 1024 byte file could then take 1024 bytes.

That's not "not having sectors", that's having sectors 1 byte long.

Thus, apply the reasoning of "bigger sectors, faster treatment of bigger files, and vice-versa".

Re:Why are there sectors?

[AP31R0N]

Ah, so it's saying, "Turn left on Evergreen... it's on that block". And the monstrous estate is from Elm to Fern at State. As opposed to GEOCOORD 32'57"(bunchOfDigits) by 32'57"(more digits).

Got it.

With the 1024 byte example, could the address just be "from bit X to bit X+1023"? i guess that too would be too much. All those tiny .dlls and .inis would take more space to define than they actually take.

Thanks!

Re:Why are there sectors?

[Anonymous Coward]

The short answer is, essentially, addressing as you guessed. Sectors are essentially the addresses on the disk. Whereas computer memory (RAM) will have an address for each piece of information (e.g. a 'word' where 'word size' might be 8 bits or 32 bits or 64 bits), a disk will have an address for a sector or block. For RAM you could have 32 bit addressing which would give 2 ^32 bytes of memory (4 GiB) or 64 bit addressing could handled 2 ^ 64 bytes of memory. For disk sectors, you don't have an address for each byte, but for each sector. In theory, you could have access similar to memory with a 1 byte sector, but performance would suffer due to the overhead from addressing, access etc.

(This is kind of rough and can vary, but hopefully it gets the idea across. In a computer architecture class you would go into a ton of detail.)

You can read some here:
http://en.wikipedia.org/wiki/Logical_block_addressing
http://en.wikipedia.org/wiki/Hard_disk_drive_partitioning

Re:Why are there sectors?

[bdsesq]

Sectors used to be needed because the drives would lose sync. The sector header would help keep it in sync.

One thing that didn't get covered very well is "breakage". Breakage is the amount of space lost due to sector size. For each file you lose, on average, one half of a sector. This is because the last sector used by a file has somewhere between one and 512 bytes of data. For the new drives that is between one and 4096 bytes.

So if you have 1,000,000 files on a drive with 512 byte sectors you lose half of 512 bytes 1,000,000 times of 256,000,000 bytes. For a disk with a 4096 sector size you would lose half of 4096 bytes 1,000,000 times or 2,048,000,000 bytes.

This is not very important to the drive manufacturers because they are really interested in bragging rights for the biggest disk. And they know Joe Consumer does not understand any of this.

Re:Why are there sectors?

[Izmunuti]

Hard drives are random-access devices and sectors are the smallest atomic unit that a drive can normally physically read and write. It doesn't read or write half a sector. When emulating a write to a 512 byte logical block with 4096 byte physical blocks on the media, it has to read the whole 4K sector, modify it with the changed 512 bytes, and rewrite the entire 4K sector.

The concept of sectors could be hidden from the interface, theoretically. You could put the whole file system into the drive (OSD), for example, or allow the host to address bytes, hiding all the read/modify/writes. But, all the common hard drive interfaces (ATA/SCSI) use blocks/sectors.

Re:Why are there sectors?

[TheRaven64]

Yes, it's an addressing thing. The grandparent is confusing sectors with allocation units. A filesystem is perfectly at liberty to allocate sub-sectors to different files (some do). A 32-bit disk interface can address 2^32 sectors. If you have one-byte sectors then that means you're limited to 4GB disks. If you have 512 bytes sectors then you're limited to 2TB. If you want a disk bigger than 2TB then you can either make the interface wider or can make the sector size bigger. Making the address wider requires defining a new interface[1], although ATA currently supports 48-bit addresses, so this isn't really a problem for a while. It is convenient for filesystems, because they can continue to use 32-bit sector indexes for partitions larger than 2TB.

The real advantage of bigger sectors is that they reduce the command overhead. To write 4KB to the disk you just need to send one write command and the data, rather than eight. All modern operating systems cache data from disk in RAM and so will write it out or read it in as a group of pages. The smallest page size of any modern architecture is 4KB, so having 4KB sectors is a lot more convenient.

Re:Why are there sectors?

[Anonymous Coward]

Sectors allow the address space of the drive to be smaller by grouping bytes. This reduces the amount of space on the drive needed for keeping track of where the files are located. The downside, of course, is that in general, no more than one file can occupy a given sector (though a file usually occupies multiple sectors). Although it could be done to eliminate sectors by using a run-length addressing system within the filesystem, this would quickly break down if aggressive defragmentation was not performed on a highly active drive (though a filesystem could abstract sectoring on it's own, and virtual filesystems, such as ISO files do)

Re:Why are there sectors?

[blueg3]

The first part is true, but the second part isn't. The logical cluster size and its implications are all in software and dependent on the filesystem. It's entiirely possible to have filesystems that put multiple small files into a single logical cluster and, for that matter, into a single physical sector. This does mean things are a bit more complicated, though.

The real answer is simply that a sector is the size of a unit of data that can be read from or written to a hard drive. That constrains how the operating system accesses the disk, which has performance implications for different disk-use strategies.

Re:Why are there sectors?

[AP31R0N]

What does sync mean in this context?

So they are making the drives faster in the sense that there are fewer sectors, so it's easier to get to a city than to a particular block of a city. They are also keeping the address space small. And they are wasting space because most of the blocks of the city have huge yards and a tiny house.

Actually that first sentence should be a question. Does having bigger sectors make the drive appear to be faster?

Fascinating.

Re:Why are there sectors?

[donscarletti]

A sector used to be quite literally a sector of a disc in the mathematical sense, like a wedge shape that spins around. Now with LBA (labeling hard drive's blocks in series from zero rather than by their physical position) it is just like a block on your filesystem, but on the hardware instead, it is a blob of data that must be read or written as a whole. The rationale is that you are not likely to ever want to read or write one byte at a time, so there is no reason to make the hard disk handle requests for one byte. The difference between a "sector" and a block is that a block on a file system should not be smaller than a sector on the hard drive since an OS can pretend two, four, etc. sectors is a single block, it cannot cut a sector in half.

The upshot of this, is unlike memory which is addressable to the byte, hard discs can be much bigger compared to the address range since it only needs volume/blocksize addresses to locate the data, so even with a block size of 512, a 2 Terabyte (base2) volume may be sufficiently covered with a 32 bit address space, this makes everything a lot easier and more efficient.

Anyway, in answer to your question, sectors are still as useful as they ever were, just they might not actually be sectors anymore because of LBA. Maybe they are, I'm not sure, I've only written hard disc drivers, I've never built one of the things.

Re:Why are there sectors?

[Anonymous Coward]

...which proves the parent's point.

Your current FS uses 4096byte clusters so your 1024byte file ALREADY wastes 3072bytes of space, the difference is that now it's faster since it only reads one physical sector instead of issuing a consecutive read for 8*512byte sectors and throwing the last 6 sectors away.

Oh, and the other point you apparently missed is that you can pack multiple files into a sector if your filesystem supports 'flexible' clusters. ie. You can shove 4 1024byte files into one 4096byte physical sector if the FS is smart enough to remember that there are multiple files there and where each one starts and stops. No FS I'm aware of does this though since the overhead for storing that is too high and makes opening files more expensive, especially if you are using append mode.

Re:Why are there sectors?

[Anonymous Coward]

No they have not. A sector is the smallest unit you can write to on a disk. A cluster is related to file systems and always a multiple of the sector size (so when you update a cluster, you only update the minimum number of sectors). What you described was clusters, exactly like GP said.

The sector size is important because as all operating systems deal with large cluster sizes anyway, it is just more efficient to write a cluster as a sector anyway, and at the same time you get to store less redundancy (at the cost of losing more data in the case the ECC information is not enough).

In fact, your description of cluster sizes is only true for primitive examples (like FAT). In Apple's newest version of HFS+ (the only modern FS I know in some sort of detail), they store small files differently from larger files. Basically, very small files are stored in the directory entry (0-16 or so bytes), larger files (16-1024 bytes IIRC) are stored in a single database, and larger files are stored normally.

Re:Why are there sectors?

[Just Some Guy]

To write 4KB to the disk you just need to send one write command and the data, rather than eight.

So why can't the command be "write the following 17 bytes at 0x(64-bit address)"? I'm curious about the answer, too. I know how sectors are currently used, but what's the advantage of using them instead of directly addressing the bytes on the drive? The drive could still remap logical sectors analogous to how a virtual memory system maps logical addresses to physical addresses. Other than for that, why is the sector abstraction useful?

Re:Why are there sectors?

[Anonymous Coward]

Although I don't know exactly how HDDs work, as I recall, they do not act on bytes at all. They do reads and writes one (physical) sector at a time. The physical format most likely is some error-correcting code such that the 512 (4096) bytes are stored physically using enough space for maybe 600 (5000) bytes or so in a format which likely looks very little like the original data but is easy to convert (while correcting a small number of errors). Therefore, a write of 17 bytes would require the hard drive to read the entire sector, decode it, modify those 17 bytes, re-encode the sector, and re-write. Hard drives don't have an interface for that as filesystems can heavily optimize that operation (for example, when changing part of sector, you should always know what the rest of the sector contains as you probably read it earlier).

Re:Why are there sectors?

[will_die]

One thing the explanations are missing is that is this why if you bring up properties for a file on Windows XP, and others, you get two numbers. The size of the file and the size on disk.

Re:Why are there sectors?

[jameskojiro]

Small file - disk file?

Like in NTFS have a file that is structured liek a tar file that stores little files in it and in the process of defragmenting the drive it moves the small files that haven't been modified in say over a month over to this special file....

Re:Why are there sectors?

[BattleApple]

Not to be pedantic, but this is slashdot.. please rewrite your analogy to reference parking lots, size of parking spaces, and... cars.

Re:Why are there sectors?

[thuerrsch]

A sector on a HDD is the minimum writeable space. Think of it as a lot in a subdevelopment. If each lot is 50,000 sq. ft. on a 20 acre plot, and you move to 60,000 sq. ft. lots instead, the plot is still 20 acres, but the development now has less lots on it.

Now I'm even more confused. Could you give an exemple in metric units, for those of us not living in Liberia, Myanmar, or the US? Also, 50,000 square feet sounds awfully large for a hard disk to me. Perhaps you're referring to one of those contraptions they had back in 1960s, when you could fill several Libraries of Congress, (or their imperial equivalent, Royal Albert Halls), with a single hard disk.

Re:Why are there sectors?

[CityZen]

The correct technical term for "breakage" is "internal fragmentation".

Re:Why are there sectors?

[dotgain]

One thing that needs even more explanation is when you (in XP) bring up properties for a Folder.
You also get two numbers, the size of all the files, and the size on disk. Oh wait! It keeps changing! It keeps changing as it sums up the entire contents of the folder. How am I really supposed to know when it's finished?

Re:Why are there sectors?

[CityZen]

To understand sync, you need to understand that, at a certain level, the signal being read from the drive head is a stream of 1's & 0's. To make sense of this stream, you need to know the boundaries of things. At one level, you need to know where each byte begins and ends. At a higher level, you need to know where each sector begins and ends. That's what sync is all about. To "do" sync, you use a sequence of 1's & 0's that you'd never use for actual data. When you see that sequence, you know, "ah, this is the start of something". Once you see the start, then you'll know where you are as long as you keep track of things and certain rules are followed. For instance, one rule is that you can't have too many 1's or 0's consecutively, or else you'll lose your place, since you can't actually see the "start" of a 1 or 0, but you can only see the changes from a 1 to a 0 or vice-versa.

To further understand the need for sectors, you must also think about how you write data to the disk. First, the disk has to read the data off the disk while the platter is spinning in order to find the start of the right sector. Then it has to switch the head into write mode and write out the new data. It must stop writing data at the end of this sector in order to not overwrite the sync header for the next sector. Due to several sources of variations (platter spin speed, temperature, mechanical play, clock frequency, etc.), there needs to be a bit of "dead zone" (ie, inter-sector gap) between sectors to make sure that the sync headers are never overwritten.

By going to larger sectors, you have fewer of them, and therefore need fewer inter-sector gaps.

Computer people who've been around may remember that the Commodore Amiga could write 880KB on floppy disks when PCs could only write 720KB. This was because the Amiga didn't use multiple sectors per track, but rather always just read & wrote an entire track's worth of data for each access. Similar idea going on here.

Re:Why are there sectors?

[AP31R0N]

Ah, so it's a mechanical thing. We don't know what is a 1 or a 0 until we've seen both. We also need to know how wide each should be. Neat!

Thanks!

Re:Why are there sectors?

[the_enigma_1983]

This is going back a few years, but I do believe that will only work for single sector writes. Basically, if you're writing to a whole heap of sectors, there's no guarantee that they'll all be lined up one after another. You might be writing 8 sectors at locations 2, 15, 16, 46, 124, 3422, 3423. For each sector written, you also have to update the filesystem, so the filesystem knows that the next sector is at 15 (or maybe specify that the address of the second section is 15, depending on filesystem). I don't know if there are any improvements made to technology, however, that might streamline the process, especially if 8 sectors in a row are free for writing.

Re:Why are there sectors?

[CityZen]

The drive controller has a clock that it uses to control the width (or length, actually) of the bits that it writes. However, due to temperature variation (quartz oscillators are temperature sensitive) and other factors, the clock that is used when the bits are read may be slightly different than the clock that was used when the bits were written. A PLL (phase lock loop) can keep the clocks aligned, but it needs a certain minimum number of transitions per unit time to take place to keep them aligned.

In truth, reading & writing bits is actually more complicated than that, since drive manufacturers have pushed the technology past the point of simply reading distinct 1's & 0's from the platter. The signal is analog, and when you try to squeeze the bits closer & closer together, it no longer resembles just "high" & "low" levels (or positive and negative, given that it's sensing the alignment of a magnetic field). Instead, you get a signal that looks more like noise, and it requires some fancy signal processing to figure out from what is read what the original pattern written was. The technique is called "PRML" or partial response, maximum likelihood.

Re:Why are there sectors?

[owlstead]

You are confusing physical sector size with cluster size. May file systems are already addressing data in larger blocks. 4096 is very commonly used. They are generally multiples of 512 which is the physical sector size; so that its is easy to calculate the physical sector that needs to be changed when you know the logical.

Or more precisely, they are powers of two equal to or higher than 512. Maybe you can do things like having a cluster size of 1536 bytes, but it would not be a system that I have encountered in the wild.

Re:Why are there sectors?

[Twinbee]

Would there be a way to remove this sector/cluster business altogether, and operate on a bit by bit basis? Even theoretically?

Re:Why are there sectors?

[rollingcalf]

Theoretically, yes. Practically, no. Operating on a bit-by-bit level would be like selling rice by the individual grain, instead of in quantities like 8 oz, 1 pound, 2 pounds, etc.

The hardware can read an entire sector in an instant, just like you can pick up a bag of rice in one shot rather than gathering up individual grains. Then the operating system will piece out what part of the sector it needs, just like you will then scoop out the specific amount of rice you want.

What is going on here is like deciding to increase the minimum bag size from 8 oz to 4 pounds. That makes things more efficient by having to deal with fewer bags, but results in somewhat more waste if your needs don't align with the bag size. However, now that we're dealing with larger and larger quantities (i.e. larger disk drives and files), the waste becomes relatively trivial ... nobody cares about 3 lbs of rice being wasted out of a truckload.

Re:Why are there sectors?

[Just Some Guy]

OK, but that's still answering in terms of how sectors work and not why we're using them in the first place. :-)

use of current cultural context....

[Himring]

This may not be a smooth transition, because some OSes do not align partitions on 4K boundaries.

"One life ends; another begins"

Looks like 512

[midicase]

From the WD website:
http://www.wdc.com/en/products/products.asp?DriveID=763

Capacity 1 TB
User Sectors Per Drive 1,953,525,169

That would be 1 TB / 1,953,525,169 = 512. I tried to verify with the spec sheet but the model's pdf is password protected.

Re:Looks like 512

[butlerm]

Those are "logical" sectors, which can be different from the physical sector size. According to the Anandtech article the Western Digital hard drive model numbers that end with "EARS" use the larger, 4KB physical sector size, while presenting a 512 byte logical sector size to the operating system for compatibility reasons.

Please note, of course, that the logical sector size is a drive interface level concept distinct from the filesystem cluster or block size. Filesystem block sizes have generally been larger than the logical or physical sector size for quite some time.

Re:Looks like 512

[chill]

So, if you're using a compatible OS, will you be able to take advantage of the drive by having it present you with 4,096 byte logical sector sizes? Or is that all in the disk format?

Re:Looks like 512

[moonbender]

The drive will present 512-byte sectors to all operating systems (I assume the drive isn't even aware which OS is running). The 4k is strictly a behind-the-scenes thing, hidden from the interface by the drive's firmware. The OS, including a properly aligned XP, will still take advantage of the improvements of "Advanced Format" such as they are; slightly increased capacity (or slightly reduced price at the same capacity, whatever), and very slightly improved error detection and speed if I understand correctly.

Re:Isn't this just a firmware change?

[Anonymous Coward]

These new drives will most likely work with older OSs, it's just that when you go buy a 4k native drive, plug it into WinXP and format it at 512Bps, you'll end up with less capacity than you expected. Your new 1TB (4k native) may only give you 900GB at 512BpS.

By shipping drives in this configuration, and (finally, after 10 years!) getting OS suppliers to agree to it, hard drive manufacturers are going to be able to reduce their areal density design requirements, making it easier to ship drives.

You can probably format your existing hard drive at 4k if you want to, and if your OS supports it. You'll get anywhere from an 8 to 14% capacity increase in doing so. So, your 1TB drive (512 B native) would end up with perhaps 1.1TB at 4k sector size.

Give me some context

[KnownIssues]

How many Libraries of Congress can you store in that?

Re:Give me some context

[compro01]

Approximately 186 picoLOCs.

Windows XP end-of-life?

[Jim Efaw]

I don't know what "pretty much end-of-life Windows XP" you speak of. I'm replying to this from Windows XP Media Center Edition. 10-20% of the computers on display at Best Buy last week were netbooks and nettops with Windows XP. Most HP workstations have "Windows XP Professional 32-bit (available through downgrade rights from Genuine Windows® 7 Professional 32-bit)" and "Windows XP Professional 64-bit (available through downgrade rights from Genuine Windows® 7 Professional 64-bit)" as options as of today; until this week (last week of December 2009), if I remember, they didn't have any operating system options except "Vista® Business 32-bit with downgrade to Windows® XP Professional 32-bit custom installed" and "Genuine Windows Vista® Business 64-bit with downgrade to Windows® XP Professional 64-bit custom installed". Why? Because people who buy computers for a business environment will not buy Vista, at any price, for real production work — fair or not. I have clients who will not buy a computer unless it has Windows XP. Despite Microsoft again attempting to remove the previous OS from the supply chain by force despite overwhelming demand, just like they have before, XP is still being sold new on a very large portion of computers.

Re:Windows XP end-of-life?

[moonbender]

Well, it's in Extended Support which for one thing means MS doesn't give a rats ass whether or not XP works with the more efficient AF HDDs, since that's not a security related patch.

Re:Windows XP end-of-life?

[Jim Efaw]

Well, it's in Extended Support which for one thing means MS doesn't give a rats ass whether or not XP works with the more efficient AF HDDs, since that's not a security related patch.

Well, that's a fair assessment. Of course, that's a monopoly tactic — any business that dropped support for that widespread of a product in a legitimate competitive environment would find themselves with no customers for the newer product because customers would be trying to migrate out from under that vendor at all costs.

Re:Windows XP end-of-life?

[Anonymous Coward]

What other 10 year old OS is still supported and given new features?

Get real, XP needs to be retired. They shouldn't add more stuff.

Re:Windows XP end-of-life?

[petermgreen]

lukilly the HDD vendors do care and hence these drives will work fine with XP (as you would know if you'd actually read the fucking article) they just require a little more care at setup time if you want them to perform well.

Re:Windows XP end-of-life?

[twidarkling]

Or, in the real world, people realize that a product will eventually reach the end of where it can be tweaked usefully, and it becomes easier and cheaper to start on a fresh base. For some reason this never applies to computers though, people think it should last forever. Do people still insist that cars be remodels of the Model T? That houses be originally log cabins? No. People realize with damned near every other product that it changes, and you can't just keep tacking on to the old. Eventually you need to rip it down and start with something new.

Re:Windows XP end-of-life?

[scottv67]

What other 10 year old OS is still supported and given new features?

I can think of an OS that just celebrated its 32nd birthday a few months ago:

http://h71000.www7.hp.com/

Yes, it's still in use - I work with it every day.
ps. Get off my lawn.

Re:Windows XP end-of-life?

[TooMuchToDo]

Don't be surprised if someone comes out with an unsupported patch for XP that fixed the ADF problem. There's enough money riding on the issue with millions of installed XP boxes that someone is bound to do it.

Re:Windows XP end-of-life?

[moonbender]

I've read the article and half half a dozen other articles on the topic. I'm well aware that there's a utility to re-align the drives for XP. I though it wasn't necessary to point it out because basically every other comment to the article talks about it. Thanks for re-iterating it, though, foaming mouth and all. I'm just saying that MS doesn't really care either way, they want people to use 7 and won't support new features/different hardware setups (such as this) in XP.

Re:Windows XP end-of-life?

[thejynxed]

Well, given the length of a typical Microsoft support contract, the point is moot. Quite a few businesses sign up for the 5-10 year deals, instead of the yearlies. With XP in extended support, Microsoft will no doubt not even offer new support contracts (other than yearlies, only available for the next 4 years) or long-term contract renewals for the OS. This leaves you with Vista, Server 2008 or Windows7, with Windows 8 supposedly due out in 2012.

These customers will have to make some hard decisions within the next four years, and why XP needs to go sooner rather than later in the corporate environment, to take advantage of the new operating systems and the long-term support contracts that go along with them.

IE6 will finally be relegated to VMs, hopefully to vanish for good soon after.

BTW, outside of the annoying UAC, I actually don't mind Vista 64-Bit too much. Then again, my hardware is not exactly standard quite yet, even though Core i7 and Core i5 is out (Lenovo K230, 6GB Ram, Core 2 Quad Q8200 @ 2.33GHz, 640GB SATAII, DVD-RW (Blu-Ray Optional), TPM-Equipped). The closest I found at Best Buy and Walmart (in my market), for instance were Dells and HPs that at the top-end offered the Q6600 with 4GB Ram and 500GB SATAII drives, Vista Home Premium 32-Bit.

Re:Windows XP end-of-life?

[mr_mischief]

I'm glad your 1982 Celica is still under warranty. Care to tell us how you conned Toyota into that?

WD's technical whitepaper

[cjjjer]

http://www.wdc.com/wdproducts/library/whitepapers/en/2579-771430.pdf

Re:Isn't this just a firmware change?

[Anonymous Coward]

It doesn't sound like the 512 bytes per sector is tightly bound to hardware.

The hardware is fundamentally designed to work optimally for whatever data unit, (sector, PDU, packet, whatever) is specified. There are buffers (physical memory, very fast, not general purpose) sized precisely for the packets. The time required to modulate the contents of the buffers across the bus(ses) is a function of the packet size and bus frequency; i.e. it does not vary and is therefore assumed throughout the system. This is real low level stuff we're talking about. The chips on the bottom of your disks are not general purpose devices that change their nature because you recompile something.

If your business depends on WinXP so much

Heh. I think it's the other way around. NTFS has been 4K aligned for a long time now; there are actually tools in the world to align legacy FAT file systems to 4K for conversion purposes. The phrase "some OSs" instead of "micro$oft" is a subtle clue that it's actually the non-microsoft systems that will have 4K unaligned partitions. You were expected to detect that.

Tail packing

[tepples]

Unless HDD makers were going to create firmware, and programmers made partition formats, which address each bit individually (which itself would require an enormous amount of space... much larger than the HDD in fact), you will always be unable to live without sectors. The subdivision idea is again relevant. Imagine if every part of the 20 acre plot had to be "addressable" down to the square inch.

It's called block suballocation: store a small file in its entirety in another file's slack space. And yes, it's a "killer" feature.

Re:Tail packing

[Anonymous Coward]

also known as "tail packing"

in the old neighborhood, tail packing would get you beat up.

Re:Tail packing

[Hatta]

So, I guess Hans Reiser has a lot of experience with tail packing then.

Paying for More Slack Space

[Doc Ruby]

Larger sectors means more empty space at the end of the last sector of a file. Lots of files means lots of wasted space. Modern OS'es, especially Windows, have many more, smaller files than in past versions, and the trend continues upwards.

So larger sectors means more space bought on a drive that isn't used. Which means more drives bought.

I can see how drive manufacturers would like that.

Re:Paying for More Slack Space

[ettlz]

I believe a filesystem with tail-packing support would overcome this.

Re:Paying for More Slack Space

[644bd346996]

Except that larger sectors also means more efficient ECC, so the same drive will present a higher OS-accessible capacity. Unless you plan to deal mostly in files that are less than 4K, you'll come out ahead.

Re:Paying for More Slack Space

[Volante3192]

Well, the trend today, especially with large drives, is to go for a cluster size of 4k anyway. Sure, there'll be a lot of system files under 4k, but there's going to be much more music and pictures over 4k that will likely take more space.

Re:Paying for More Slack Space

[PhrstBrn]

When they're able to make disks ~10% bigger just by switching to the larger sector size, I couldn't care less about this "wasted" space. You have a 1TB drive. Because of this change, they can fit 1.1TB of data onto the same size platter. So it would take ~28 million files worst-case scenario in this hypothetical situation before the 1.1TB drive would have less usable space. Of course, they like nice round numbers, so what it really means is they can make drives cheaper and faster. In the long run the consumer will win (there is enough competition where you shouldn't need to worry about price fixing). I like it.

Re:Paying for More Slack Space

[minerat]

NTFS's default cluster size for volumes up to 16TB is 4KB anyway. Anyone with a new computer (Vista+, anyone who wouldn't know what a file system is, have reason to choose another one and not know how to manually align it) will be using 4KB clusters.

Re:Paying for More Slack Space

[Anonymous Coward]

Maybe you're alluding to this, but several do, including btrfs.

Tail packing doesn't really "overcome" the problem, though. As another commenter noted, it is a restatement of the knapsack problem. Filesystems can either do it fast or do it optimally for space, not both; ie, it's a np-hard problem with a PTAS.

Re:Paying for More Slack Space

[Xeleema]

Hm....a few things in your comment caught my eye...

...disks ~10% bigger just by switching to the larger sector size...1TB drive...they can fit 1.1TB of data onto the same size platter...the consumer will win.

Anybody remember when WDC settled that class-action lawsuit for the whole "1GB = 1 Billion Bytes", causing people to pop out and buy a "80GB" disk, only to discover it was really a "72GB" disk? (Yes, yes, the whole "how many bytes in a Kilobyte" thing)

When I put on my Business Hat, this makes sense. They can keep marketing "1TB" disks, they're just "1.1TB" with 4k blocks, rather than "0.97TB".

Wonder what happens when I don my tin foil hat...

Re:Paying for More Slack Space

[PhrstBrn]

No, they'd still be 1.1TB disks because how many bytes a platter can store is completely independent to the sector size.

I can't help that people are too stupid to understand the difference between GB and GiB. Blame Windows for using the incorrect SI notation.

There is a standard. K means one thousand (1000). Not 1024 because it's a computer. Data transfer is measured in thousands of bytes as well (not 1024 of bytes), and they use KB/MB/GB.

The only thing inconsistent is how RAM manufactures report the numbers (they use MB and GB to represent numbers in powers of 2), but nobody complains because it's in their favor), and the notation Windows uses to report data (they should be using KiB/MiB/GiB, not KB/MB/GB).

Hard disks have *always* been measured in powers of 10. Nothing changed except people who don't care about this stuff and just want to go on AOL started using computers, and then noticed how everything didn't match up because they can't understand how binary notation works.

Re:Paying for More Slack Space

[Hatta]

I can't help that people are too stupid to understand the difference between GB and GiB

Yeah, those stupid OS coders. They're idiots. Every last one of them.

Re:Paying for More Slack Space

[PhrstBrn]

OS coders aren't the ones filing class action lawsuits.

Re:Paying for More Slack Space

[TheDreadedGMan]

Running Windows 7 64-Bit RC here... installed on a 750GB Samsung drive formatted using defaults via the installer.

Opened the C:\Windows folder, typed "size:<=4096" in the search box, found 23,347 files, selected all, right-clicked properties, I get:
Size: 43.0 MB (45,126,095 bytes)
Size on disk: 91.1 MB (95,563,776 bytes)

so they are slightly more then twice as big, these "more, smaller files"... personally 45MiB of a 698,000MiB is not bad wastage...

Re:Paying for More Slack Space

[mr_mischief]

I think the people worried about a few hundred config files wasting most 4K of space each (which their FS probably does unless they've already researched this issue and gone with one that tail packs) just don't want to admit that most of their files are 4G porn DVD images.

Okay. So, what if we have to image an old box?

[E-Sabbath]

Let's say, two situations.
A: Moving an XP box from an old 512 sector drive to a new, 4k sector drive. Image using, say, Acronis. Just have to run the CLI tool, after imaging?

B: Moving an XP box, using Acronis, from a 4k to a 4k. Would I have to run the tool?

Re:Okay. So, what if we have to image an old box?

[Mashiki]

Answer to A: Just run the tool
Answer to B: Run the tool to ensure that old sectors and tables align properly, more has to do with imaging software actually not properly recording sector counts.

Re:Paying for More Slack Space.

[butlerm]

Any change in sector size that doesn't affect the filesystem block size will not affect the number of KB required to store a file at all. Since virtually every filesystem already uses 4 KB block sizes by default a change to 4KB logical or physical sector sizes will not have an effect on storage requirements.

Flash's everyday life

[DrYak]

If the OS clusters aren't aligned to physical sectors, the hard drive's controller has to read-modify-write all the time.

Which is flash memory's every day life, btw. When reading, flash reads usually 1kb at a time, and might erases even a Mb at a time when writing. (hence firmware and TRIM optimising).

On the other hand, most OSes use much bigger clusters. NTFS and FAT32, for example, can use up to 64kb cluster. (Although, the ability to *create* huge FAT32 partitions is limited in recent Windowses. But who care ? FAT32 is mostly used on Flash and Flash is sold-preformated anyway)

So it doesn't really matter if sectors are 512byte (old HDD) 1kb (optical media) or 4kb (new HDD). They are all divisors of the various cluster sizes available.
The problem is that Windows XP and older, and most DOSes are hardwired to consider all HDD with 512byte sectors.

Re:Flash's everyday life

[dfghjk]

You like to talk a lot and say nothing interesting.

Re:Flash's everyday life

[Anonymous Coward]

You have nothing to say, and say it well.

Re: advantage of larger logical sector sizes

[butlerm]

For now, with SATA drives at least, the logical sector size exposed to the OS will remain 512 bytes. Several years down the road that will change, because it is simpler to have the logical and physical sector sizes be the same.

In practice, though, as long as you are writing blocks aligned on 4KB boundaries to partitions aligned on 4KB boundaries (such that the blocks being written are aligned with the underlying physical sectors), it won't make much of a difference whether the logical sector size is 512 bytes or 4KB bytes.

With contemporary filesystems and databases that will almost always be the case, provided the partition alignment is correct for 4KB writes.

Re:Isn't this just a firmware change?

[Anonymous Coward]

You just repeated what TFA states.

Actually no.

[Mashiki]

Most of the drive manufactures are releasing tools to align the drives to 4k clusters so they can be used under XP. WDC already has theirs out here: WDC Adv Format Plus instructions on all of their new 1TB and higher drives on how to set them up properly. You do have to jumper them, then format them specially but the drives work fine with 4k clusters. I put one in my work machine on Saturday, works flawlessly.

*I only used WDC because that's the brand I picked up recently. I do know other companies have similar tools and jumper settings on their newer drives as well.

Re-Format old drives?

[Mojo66]

Are there tools to low-level format 512-byte drives into 4096-byte ones? I gather this would increse capacity by 13%.

Re:Re-Format old drives?

[butlerm]

This would require a firmware change, and for SATA drives, it is just not going to happen. High end SCSI drives maybe.

Re:Isn't this just a firmware change?

[butlerm]

NTFS has been 4K aligned for a long time now.

That doesn't do any good if the partition it is on starts with an LBA that is not a multiple of 8. Windows versions prior to Vista create the first partition starting at LBA 63, which is not 4KB aligned.

The people who will have performance problems will primarily be Windows XP users who purchase the newer style drives and do not realign the first partition accordingly. Some versions of "fdisk" on Linux have a similar deficiency, with an "cylinder" based user interface and odd size cylinders in the name of MSDOS compatibility. Not sure if that has been fixed yet.

Moot Point

[Archangel Michael]

The point is going to be Moot here really shortly. The whole idea of sectoring and block sizes and such is going to go the way of the DODO in a few short years as we move from Magnetic Media to Solid State.

And the moment we go to SSD drives, the whole game changes. The idea of physical drives spaces and such disappears, and we come a lot closer to what we now see in high end Drive Array storage, where everything is abstracted away from the OS anyways.

With SSDs we'll see new ways of upgrading / managing drive spaces, such that when the time comes to put in a bigger drive, you just drop it in, and all your data moves to the new drive tranperently in the background and when the process is completed, you just remove the old SSD, and add in another new drive(wash, rinse repeat).

We'll stop using terms like "format", "defragment", "drive", and even "volume", except to express things in terms for some of us older folks, who spent the last 30-35 years with those terms ingrained in our brain.

We're already starting to see the end of Magnetic Media. I would suspect that in 4 or 5 years, magnetic drives will be mostly gone.

Re:Moot Point

[TD-Linux]

Nevermind, of course, that flash memory is sector-based as well. I've never seen flash memory with a sector size below 512 bytes. To make matters worse, you can only erase Flash a block at a time - a block being a group of sectors that I've never seen smaller than 64KiB.

... I think I've been trolled.

P.S. Would you count FeRAM and MRAM as magnetic media?

Re:Moot Point

[Archangel Michael]

I do believe those are carry overs from Magnetic Media. There is no need for it to be that way (I think).

My point, was more or less, that we'll need to RETHINK how we define things. SSD will become more of an extension of the Operating Space we call "RAM". Much like we now have RAM, L2, L3, and L4 cache (and even maybe RAID Cache) are now.

I think, and this is just my opinion at this point, that we'll start to name memory by nearness to the Core(s), and SSD will join that space.

I'm not sure we need block level devices any longer. It will require re-thinking much of how we view things no doubt, but I think it is inevitable at this point.

Re:Moot Point

[DarkOx]

We might not need block devices when everyone has a 256-bit CPU in their laptop but until then your idea is completely impractical.

You have to address memory. Primary memory has to be addressed at the if not the byte level than certainly at whatever size the registers are in the CPU, be it dwords, qwords or whatever else. The CPU needs to be able to fetch and store single arguments. You could have some sort of address translation were each address from the CPU's perspective is a 4 a number of 4 byte offsets or something but that still won't get you more than a 2^63 bytes even on modern cpus. People want bigger drives than that.

So you are still going to have to address large(ish) ranges of bytes in secondary storage to have any reasonable capacity; without royal pain. I suppose if you really wanted to solve the problem a different way and bring persistent secondary storage into the "direct" address space you could go back to some kind of multiple addressing mode scheme but the software people will certainly hunt you down and kill you.

Re:Moot Point

[butlerm]

Actually, with flash drives everything related to the issue is much *worse*, due to the way flash works - the internal block sizes are huge (128KB). Because of this, contemporary flash drives are enormously complex, sort of like a filesystem with a 128 KB internal block size that sub allocates (and moves around, and coalesces) lots of little 512 byte "sectors" for presentation to the outside world.

Re:Moot Point

[Archangel Michael]

FYI, 64 bit = 16.8 million terabytes or 16 exabytes.

Source .. http://en.wikipedia.org/wiki/64-bit

How much data do you really have? (and no, your PORN collection doesn't count.

Re:Moot Point

[drsmithy]

The point is going to be Moot here really shortly. The whole idea of sectoring and block sizes and such is going to go the way of the DODO in a few short years as we move from Magnetic Media to Solid State.

I don't know what you consider "short" in this context, but current methods of addressing storage sure as hell aren't going anywhere for at least the next 15 years, and more likely 25.

We're already starting to see the end of Magnetic Media. I would suspect that in 4 or 5 years, magnetic drives will be mostly gone.

Not going to happen. 10-15 years, maybe. SSDs just aren't going to increase in $/GB fast enough for a shorter timeframe to be practical. A "cheap" 500GB SSD costs $1700ish today. A 2TB drive costs $150ish.

Another example: contemporary SANs are still primarily magnetic disk, and that certainly doesn't look to be changing any time soon (ie: within 18 months). Any SAN investment of significance will be used for *at least* 3-5 years after purchase.

Re:Moot Point

[drsmithy]

My point, was more or less, that we'll need to RETHINK how we define things. SSD will become more of an extension of the Operating Space we call "RAM".

No, it won't, it's too slow.

Re:Moot Point

[Archangel Michael]

It is faster than fastest HDD. And getting faster.

And RAM is not fast enough for Processors either, which is why we have L1, L2, L3 and L4 cache.

We're just increasing the speed of memory in stages. SSD are just another form of RAM, why not call it what it is? Why not plan to use it for what it is?

Re:Moot Point

[mr_mischief]

A chalk board and chalk is a form of RAM. Damned if it's going to be part of my main system memory, though. I don't think FRAPS would even bother giving me a reading.

remember HPFS, OS/2 ?

[Anonymous Coward]

This brings back memories of the HPFS disasters in Asia where they used 4k disks back in the day... and HPFS was set to use 512 sector sizes.....

Surprising nothing changes that much.

Re:Isn't this just a firmware change?

[Anonymous Coward]

if the business depends on xp so much, the IT dept actually failed at being capable of such things...

no, HDD sector size was 100 bytes long before that

[Anonymous Coward]

http://en.wikipedia.org/wiki/Early_IBM_disk_storage

The IBM 1301 HDD had a sector size of 100 bytes and I think at least one of the even older IBM HDDs had a sector size of 600 bytes.

"Time immemorial" goes back a wee bit further than the IBM PC/XT...

You don't seem to get this whole base 2 thing :-(

[Zero__Kelvin]

"If they want to use base10 the first thing they should do is respecify a byte to be 10bits."

Why would you want to change a byte to three bits? ;-)

All joking aside, changing a "byte" to be capable of storing a number from 0 through 1023 rather than 0 through 254 doesn't help matters one bit (OK, maybe not all joking aside)

Well it's about time that they caught up

[sloepoke51]

Back in the late 70's there was a new OS, CP/M 80 Version 2. Included in the BIOS support was a simple and very effctive sector blocking / Deblocking algorithm. The default support was 128 byte sectors on the common disk environment, 8 inch floppy disks. With the sector blocking and deblocking code, any size of physical sectors could be supported. On my Morrow Thinker Toy's double density controller, I was able to go from 256 byte sectors to 1024 byte sectors gaining a nice huck of space. In 1981 I worked for Micromation and had a chance to play with a 14 inch winchester hard drive, which had a huge 20 megabyte capacity. It had a "fixed" 512 byte sector size. After a little messing around with the drive, I found that the drive really could support larger physical sectors. I went from the 20 megabyte tot disk size to 1024 byte sectors and go another 6 megabytes for a total 26 MB out of a 20 MB drive just by enlarging the sector size.

ECC is NOT linearly proportional

[A nonymous Coward]

See the comment below. It takes 320 bytes for 8 512-byte ECCs, only 100 bytes for a single 4096-byte ECC.

Re:ECC is NOT linearly proportional

[pclminion]

That's because it's a different (more efficient) code. You could, in theory, use the same or similar code for 512-byte sectors. If you are changing the sector size, it's a good opportunity to switch codes.

Re:ECC is NOT linearly proportional

[Anpheus]

It's my understanding that to correct n bits of errors in m bytes requires a particular length of ECC, determined by algorithm choice and other things. What they've decided on doing is rather than protecting n bytes per 512 bytes, they'd rather protect q bytes per 4096 bytes. Even though the sector is eight times larger, the probability of eight simultaneous errors is far less likely and not worth protecting against.

tl;dr: it's complicated.

Re:ECC is NOT linearly proportional

[Anonymous Coward]

It's different and more efficient *because* it's for a longer data size. It's the same type of code, and the efficiency depends on the length of the data. You cannot "use the same code" for 512 sectors because it's more efficient precisely due to the sector size increase.

Re:ECC is NOT linearly proportional

[loose electron]

The subject of ECC's and the unique aspects associated with magnetic storage can fill multiple books. (start googling run length limited codes, reed-solomon, convolutional and cyclic codes, and your eyes will glaze over PDQ.)Most of whats been posted here is either wrong or pretty oversimplified (mostly just wrong) - The quick and dirty on whats unique about ECC's in disk drives is that the errors tend to be bursts rather than individualized - so getting a long string of bad data at one place is the norm, rather than the exception. (simple way of thinking about it - think of a scratch, and thats a supersimplified analogy)

Bigger sectors make sense - the 512KB standard has been around since before I designed drives (got to go back 30 years, argh!) but in the grand scheme of things it doesnt matter a whole heck of a lot. Access time is alway going to be dominated by rotational latency and there is no way around that, short of redundancy locations of data (at a heavy cost on storage density) and total storage capacity needs vs. capability are generally dominated by the areal density improvements. (Who cares about 10% of overhead for formatting, sector size, ECC placement etc. when the total storage doubles every 6 months anyhow?)

Oh, and the death of the HDD due to SSD has been greatly exaggerated, IMHO, because the need for total storage keeps going up, and the cost for $$ per GByte needs to keep that competitive. The SSD drive will be sweet for things like laptops/netbooks, thats for sure, but nobody wants to pay the serious money for terabytes of movies on a SSD, when a HDD will get it done at less cost.

Oh - and if anyone wants to say the price of SSD storage will come down and surpass that of HDD storage? Well, maybe, but the problem there is that Moore's law is running up against the limits of Physics right now. THe state of the art transistor has a gate oxide thickness of 4 atoms, and a channel length of 13 atoms. Good luck with doing fractions of atoms, there might just be a few problems with that.... :)

Re:ECC is NOT linearly proportional

[mswhippingboy]

Oh - and if anyone wants to say the price of SSD storage will come down and surpass that of HDD storage? Well, maybe, but the problem there is that Moore's law is running up against the limits of Physics right now. THe state of the art transistor has a gate oxide thickness of 4 atoms, and a channel length of 13 atoms. Good luck with doing fractions of atoms, there might just be a few problems with that.... :)

The atomic size limitations apply to HDDs too. I fail to understand your implication that more data can be packed into a mechanical, whirling set of magnetic plates than can be packed into a solid state circuit (even if we don't get below a couple atoms thickness)? So what if the physical dimensions of an SSD device "might" eventually need to grow to to be able to match the capacity of larger HDDs. SSDs have a lot of room to grow before they hit up against the (relatively gargantuan) physical dimensions of even a laptop HDD.

From TOFA ( O == other)

[MiniMike]

From TechReport:

you'll have to download a free WD Align application to align partitions correctly with Windows XP.

Looks like we can continue to use our archaic XP (stifled yay). For now, at least.

Re:From TOFA ( O == other)

[hairyfeet]

But what about dual boots? I keep an XP partition alongside my Windows 7 HP X64 for those older programs that don't like Windows 7, and have several customers that do the same. Does this mean we that dual boot are gonna have to research our asses off not to get one of these "improved" drives or suffer?

Re:From TOFA ( O == other)

[petermgreen]

But what about dual boots?
either create your partitions with a tool that aligns them on 4KB boundries or use western digitals partition realignment tool.

Re:From TOFA ( O == other)

[rts008]

*disclaimer-all of my home network consists of *nix machines*
I'm guessing an added step won't be too painful.
Instead of just partioning and formatting to install the respective OSs, you'll have to apply the HDD mfg supplied conversion tool to the intended XP partition before formatting it.
I could be wrong, but it seems a sensible scenario to expect. :-)

Re:From TOFA ( O == other)

[dysan27]

It all depends on what you use to partition the drive. As long as the partitions are 4k aligned it's fine, so if you use 4k aware partitioning software (ie Win 7) all should be fine. You'll only hit the performance issue if you use XP to set up the partitions. And even then you can use the WD align tool that they provide to align the partitions.

Re:From TOFA ( O == other)

[mr_mischief]

Have you tried XP Mode for Windows 7? I've not had any problems, and your XP just has to deal with the image file. No direct access to the drive means the drive's characteristics are moot.

If you're dual booting because you've had problems, please let us know what to watch out for. If you just haven't tried XP Mode, you might want to try it if you have the virtualization support necessary.

Re:From TOFA ( O == other)

[hairyfeet]

I'm using Win7 HP x64, so no XP mode, not that it would matter as a good 90% of the problems I've run into are with games which of course suck in VMs. Some of them that went tits up are the earlier MoH games, KB:The Legend (To be fair it don't like XP x64 either), some of the bargain basement games I've been picking up at Gamestop like Blacksite:Area 51(fun but short) and TP:Fall of Liberty. TP was right, damned thing is just an X360 game, can't even choose items with the mouse!

Funny part is many of my business customers have decided to go HP as well. They looked at the MSFT chart showing features and didn't see anything they would pay the higher cost of Pro for. They are SOHO/SMBs without AD servers, and they decided they would rather just replace programs than deal with a VM. MSFT really should have kept the $50 for Home, $100 for Pro pricing, as many (including myself) were seriously thinking about going Pro after the holidays, but after the price hike forget it. Home Premium works fine, and it isn't worth the extra cash just for XP Mode.

Question: Does anybody know an easy way to get the data out of Quickbooks and into something less expensive or free, like GNUCash? That one program is the only reason some of my customers are dual booting. At $400 a pop I can't say as I blame them, but surely there has to be a FOSS solution that will let them make receipts and invoices like Quickbooks without losing their data? Most are running 05 or 07 and it just shits itself and dies in HP x64. probably the craptastic DRM they employ, but sadly many of the SOHO/SMB users have lots of data tied up in the thing and it would be a royal PITA to replicate. Any ideas?

WRONG 2 TB limit due to 4G sectors

[Anonymous Coward]

32 bits is enough for anyone, unless you need more than 2^32 things to track.

Going to 4K sectors takes that to 16 TB, assuming the industry needs to stick wiht 32 bit. I am foggy about the finer details.

Re:WRONG 2 TB limit due to 4G sectors

[yuhong]

Yea, that is important, because the MBR partition scheme is limited to 32-bit LBA, meaning it is limited to 2 TB with 512-byte sectors. To fix this you can either go to GPT or as above increase the logical sector size to 4 KB, which one do you think is better?

Flash doesn't have a spinning platter

[tepples]

If the OS clusters aren't aligned to physical sectors, the hard drive's controller has to read-modify-write all the time.

Which is flash memory's every day life, btw.

But then flash doesn't have to wait for a full platter rotation after the read before it can modify and write.

On the other hand, most OSes use much bigger clusters. NTFS and FAT32, for example, can use up to 64kb cluster.

But if the start of the partition isn't aligned to the start of a hardware sector, cue the RMW and waits for rotation. For example, in a 512-byte sector environment, it's perfectly valid for a partition (and thus all the clusters) to start on an odd sector number.

Re:Flash doesn't have a spinning platter

[FuckingNickName]

But if the start of the partition isn't aligned to the start of a hardware sector, cue...

...chastisement for not setting up your drive using the vendor-provided software.

The main problem is solved, as always, by a layer of indirection. Your theoretical worst-case scenario of an extra rotation for every write is going to be offset by controller buffering, operating system caching, etc. Don't forget, when I am changing a couple of bytes in a file, I have to read at least a logical sector, which means the controller has to read (and may cache) at least a physical sector.

For modern high-performance environments where the driver optimises based on the physical sector size, geometry, and so on, there is going to be no assumption of a fixed sector size. More likely is the trade-off by queueing commands which are coalesced for most efficient read/write by the controller, where you let the firmware worry about such details and accept that premature optimisation on your part may reduce efficiency.

Re:You don't seem to get this whole base 2 thing :

[Magic5Ball]

But I like 255. It keeps my white page backgrounds white.

Characters were not always 8-bit

[klubar]

I believe that some of the early CDC machines (a company that is no longer around) had a 6-bit character. The Digital Equipment Company (DEC, alos a company that is no longer around) PDP-1, maybe the PDP-20, and some others also had a 6-bit character. The PDP's had 36-bit words, packing 6 characters into a word. And of course, the IBM machines (a company that is still around) used EBCDIC rather than ASCII (but did use an 8-bits per character). Some of the earlier (and even the 370's) IBM machines used BCD (binary coded decimal) for arithmetic (packing a number from 0 to 9 in 4 bits, with some sign and unassigned bits left over).

Also, back in the IBM JCL days, when allocating disk space for a file you could specify the number of cylinders (or tracks) that you wanted, the block size and the packing factor.

Re:Characters were not always 8-bit

[Anonymous Coward]

What *have* you been reading?

Many machines utilised a 6-bit byte (the term 'byte' even originally referred to 6 bits), not just CDC, and they stored data in BCD - a character set which could handle digits and upper case alphabetics.

The PDP's did *not* have 36-bit words: the PDP-8, for example, was a 12-bit machine. The DEC-10 and DEC-20 (perhaps occasionally called PDPs) did have 36 bit words, as did many other mainframes. The Honeywell 6000/66/DPS 8 for example could pack 6 BCD characters into a word, or 4 ASCII characters right aligned in 9-bit bytes.

I also think you are probably referring to IBM DOS JCL rather than OS.

Re:Characters were not always 8-bit

[slicerwizard]

Yes, 370s had base 10 math opcodes for business-type apps, along with the normal binary add/sub/shift/etc. opcodes - just like x86 processors do. So what's your point?

Re:Characters were not always 8-bit

[VGPowerlord]

As strange as it may seem, there actually is precedence for 8-bit bytes... in Currency of all places.

The Spanish Milled Dollar was often split into eight pieces to make change. Hence the term "Pieces of Eight."

However, for whatever reason, those were often termed bits in the US. So, there were 8 bits to the Spanish Milled Dollar.

And two bits made up a quarter-dollar, hence the "Shave and a hair cut... two bits!" routine.

Re:Characters were not always 8-bit

[david_thornley]

I believe that some of the early CDC machines (a company that is no longer around) had a 6-bit character.

They had 60-bit words, divided into 6-bit characters, and the I/O computers had 12-bit words. There was a system for representing lowercase characters that involved mixing 6- and 12-bit characters. I never got the hang of it myself.

Also, back in the IBM JCL days, when allocating disk space for a file you could specify the number of cylinders (or tracks) that you wanted, the block size and the packing factor.

Do you realize how much time and effort I've put into blocking all traces of IBM JCL from my memory? And now you've got to stir those memories up again....

Re:Characters were not always 8-bit

[Babylon Rocker]

It was even worse than that on the PDP-10s.... The operating system used both SIXBIT characters and 7-bit ASCII characters using variable-length bit-field instructions (36 bit words => 5 chars and 1 bit left over).

Re:Characters were not always 8-bit

[Bigjeff5]

Just a wild guess, but I think his point was that "Characters were not always 8-bit".

Re:Characters were not always 8-bit

[unitron]

...I think his point was that "Characters were not always 8-bit".

Quite true. I've known any number of two-bit characters.

Re:Characters were not always 8-bit

[jonadab]

Yes, at one time long ago, there were other "byte" sizes. But that's reaching pretty far back into the history books. By the time IBM sold their first x86-based "Personal Computer", the entire computing world had fully standardized on the eight-bit byte. DEC, for instance, after dipping their toes in the pool with the PDP-11, finally came fully over to eight-bit bytes with the Vax and thereafter never looked back -- and they were one of the last holdouts.

Re:Characters were not always 8-bit

[sincewhen]

Also, back in the IBM JCL days, when allocating disk space for a file you could specify the number of cylinders (or tracks) that you wanted, the block size and the packing factor.

Hey, I resemble that remark! Some of us are still coding JCL every day. But you no longer need to specify the block size and packing factor, SMS does most of the grunt work for you.

Re:Characters were not always 8-bit

[spitzak]

Those machines did not address units smaller than the 36 bit word.

I do not know of any machine that could address units small enough that anybody would think to put one character in each other than 8 bit addressing. There may have been 12-bit addressable units on PDP-8 style machines but there was a tendency to pack 2 characters into each of them.

misalignment costs 3X

[Anonymous Coward]

Back in the day I had an underperforming RAID-5 array from Data General. When their 3rd-tier engineering support explained to us how to properly align disk partitions to the stripe size, we _tripled_ our disk performance. This required, of course, a complete dump/restore cycle to tape, incurring significant downtime. Clueless XP users on 4096 byte drives are in for a world of hurt.

A correction regarding file systems and HDDs

[jonaskoelker]

If a sector is 4096 bytes, and you create a 1024 byte file, it still occupies 4096 bytes on the disk, as the HDD won't write anything else but that file to the sector.

Wrong. Probably on more than one level.

I'm not an expert on disks and their interfaces, but I know a little about file systems. I know fershur though that disks don't have any notion of files.

I assume that disks have the following interface: you can point it to a place p in RAM and ask it to write the next n bytes in n-byte-disk-block number k, where n is a constant specified by the disk; i.e. write(p, k). You can symmetrically read(p, k): copy n bytes from disk block number k to RAM, starting where p points at. You can also ask the disk for the block size, get_n().

An easy way to lay out files in n-byte blocks is that each block contains data from zero or one files---the implication being that to get the data that makes up a file f, you store a list of blocks containing the data in f, then access each block in sequence. The other implication being what my parent said: a 1024-byte file will take up a 4096-byte block for itself. (The assumption here is that the data for the file starts at the first byte of the 4096-byte block, for each block in the list, and runs to the end of the block, maybe except for the last block in the block list.)

It's also possible to store files more densely, but not requiring the file to start at the first byte of the 4096-byte block. That means you need to store an offset into the block as part of each element of the block list*.

So for two 2048-byte files to share block k, one would be stored at "(k, 0)" and the other at "(k, 2048)". Similarly, four 1024-byte files could be stored at (k, 0), (k, 1024), (k, 2048) and (k, 3072).

You may be familiar with the programming trick of stealing bits off an int pointer; this is similar to what goes on here---by making your pointer-to-somewhere-on-disk type more coarse-grained, it can be stored in less bits. Or reversely, by using more bits, you can point to more stuff (or more detailed stuff).

* Yes, there may be smarter ways. Yes, you can also store the file "tails" (pieces that go in the last, possibly-unfilled, block) together, multiple tails (or short files) in one block. It's faster to write a new block if you don't have to preserve old contents, so that's a strike against packing small files. It's more space efficient, but finding a good allocation of tails is essentially approximating the knapsack problem (which is NP-hard), so finding the optimal packing is infeasible. Et cetera. In short: There are trade-offs.

Now, please someone correct my model of disk interfaces :)

Time to Eliminate this problem

[FlyingGuy]

With the exception of recovery utilities no OS needs to know anything about the disk geometry.

Drives should simply be on a port and you either read or write a stream to that port.

The only commands sent to the drive would be to read, write, create and delete.

DMA would be handled by the drive, hell there really isn't a need for a controller.

The BIOS should be the only thing that knows anything about the drives and that would be limited to booting up.

Linux uses a "filename" to access everything from a disk, so this becomes a simple matter of the "drive subsystem" sending a command to the disk drive to create a "library" called "/" and one called "/boot" etc. etc. and even this is only a file with an index. The drives manufactures could then do all of the low level work to handle the cluster and sector stuff ( hell even on the fly if it notices that the file sizes are getting smaller or larger. it could re-adjust things to take advantage of that fact and keep the drive completely optimized at all times.) and the OS would be none the wiser simply because it does not need to be.

Re:Time to Eliminate this problem

[butlerm]

Making the drive handle things at the file level is the equivalent of turning it into a NAS device where the system software would generally be inaccessible, unmodifiable, and un-upgradeable, unfortunately. It would still be an interesting engineering challenge, of course.

Re:Time to Eliminate this problem

[the_enigma_1983]

From the sounds of it, you're recommending that we move all file system operations over to the HDD (all of the managing where files are, how big they are, what directories exist etc)?

Re:Time to Eliminate this problem

[FlyingGuy]

Yes, it would very closely resemble NAS, but there would be no TCP stack involved.

The thing is we are starting to really push drive technology. Solid state drives, drives with bigger sectors, is the structure of the drive layout better suited for one thing then another, or could some very interesting optimizations be had by removing this stuff from the OS and putting it in firmware and having that hardware / firmware combo be an ASIC or a Programable Gate Array?

A LOT of things are done faster and more robustly in Hardware and there are other tasks that really require a more completely software based approach and I think this is one of those things that could really make large gains if realized in hardware.

If firmware was involved it could be upgradeable much like flashing the bios.

Re:Time to Eliminate this problem

[FlyingGuy]

Yeah pretty much.

Pretty much like NAS but without the "N". With all of the new drive technologies comming out, we really havent advanced much in the way we use and access them. We are still having to deal with sectors / clusters / heads and all of that stuff.

I believe there is much to be gained in performance by realizing all of that in hardware that we have a very simple and straight forward conversation with the device instead of having to deal with all the drive geometry and all that other very complicated stuff.

Imagine if you will any of the popular file systems done in hardware. You would eliminate a lot of overhead, get it out of the kernel and deal with it with a far simpler set of instructions. An ASIC would do one thing and one thing only, deal with data requests and maintain the most efficient use of the available storage. We would still fopen() fclose() etc. but those would simply talk to the hardware port that is the storage device, then let the storage device decide how best to store the data.

Operating system compatibility?

[tepples]

chastisement for not setting up your drive using the vendor-provided software.

For which operating systems is this vendor-provided software made available?

Re:Operating system compatibility?

[FuckingNickName]

At a glance, at least Windows XP, Vista, 7... I assume that every enterprise with enough money to be in the kind of business to worry about sector size has at least one non-OEM licence of Windows lying about.

Otherwise, you could just boot into Linux and make sure to create your partition on appropriate boundaries using your preferred partitioning tool.

Remember, we're worrying about one-time setup here.

Linux is F'd

[Anonymous Coward]

All of the current Linux installers still use the legacy "CHS" to partition disks (as does gparted). It can be ignored if done manually with parted or sfdisk.

The 255*63 alignment may mis-align (depending on the disk) the data sectors of EXT* and XFS with regard to the disk blocks, forcing the disk to read-modify-write the data into the on-disk 4K sectors. This destroys performance.

In the ATA disk information, there is data to tell the user how big the physical disk blocks are, and what the alignment of logical blocks is within the physical blocks. Installers can then set up the partitions to minimize mis-alignment. In either case, the "cylinders" values are meaningless except to very old BIOS, which can still work OK, as long as the MBR table is correctly set up.

One further complication is that not all disk manufacturers are going to follow the ATA spec'. They will report disks with 4096 physical sectors as having 512-byte physical sectors, so there is no way to align them using the ATA information. The safest course of action, then, is that for disks reporting 512-byte physical sectors, align the partitions to some multiple-of-4K boundary, like 1 Mbyte (which also leaves room for a GPT). This will not guarantee that the partitions are aligned with the disk blocks, however.

Maybe some reviewers can start checking performance with various partition alignments and let us know which disk drives aren't following the spec'.

please stop spreading FUD

[Anonymous Coward]

What a bunch of misinformed drivel. That article is missing a couple of things:

firstly) The issue affects all Windows versions based on a 5.x kernel. That means Windows 2000, XP, 2003 server and Windows Home Server.

1) These drives are NOT strictly-4k-sector. The platters may be organized in 4k sectors, but the drive only talks to the OS in terms of 512 byte-sectors. And since we're discussing old Windows versions: NTFS has defaulted to using 4k (logical) sectors since its introduction, so there is NO performance penalty when using NTFS on these drives. You shouldn't be using FAT32 anyway.

2) The issue can be worked around by creating partitions with a tool that understands 4k sectors, or by re-aligning the partitions after creation/installation. If you only use a drive in those systems (i.e. no repartitioning), the drive will work as it should. Even if you create partitions that are unaligned, the drive will still work - you will only lose some performance.

3) The one genuine problem raised in the linked article comes when you want to use these drives in closed-firmware devices. In this case you still have two options: either you use the WD-provided jumper setting, or you pre-create the partitions before you insert the drive.

I fail to see what the fuss is all about.

correction

[petermgreen]

added text in bold

From the operating systems point of view these look no different from any other drive. it's just that they will perform badly if the partitions are misaligned so I don't see how it will cause drives formatted on one system to fail to work on another.

Re:correction

[raynet]

You could align the partitions to 4K manually via CLI tool in XP and it should work without performance penalty, atleast it works so with external RAIDs.

Re:You don't seem to get this whole base 2 thing :

[Zero__Kelvin]

"But I like 255. It keeps my white page backgrounds white."

But imagine how much more white they will be with 10 bit bytes! 1023 - 255 = 768. That's more than 3 times the wholesome white g00dness!

Doesn't Affect Virtualization

[Xeleema]

So, I'm seeing all of this hubris about "Windows XP going down because 4k disks are upon us!!!" which really makes me want to point out two things;

1) If you're using Virtualization with file-based storage (as opposed to disk-based storage), you're in the clear.

So we can all run a VM of WinXP for Development/Compatibility purposes if we *really* want to. The folks requiring 3D-acceleration will take a bit of a hit, but hey, if you haven't beaten that game after 10 years, cheat and get it over with already.

2) Are we seriously debating the *merits* of Windows XP?

The only reason it's any good is because of 9 years of patches and bugfixes. Re-Read that last part; It took them NINE years to get it going good.

Re:Doesn't Affect Virtualization

[RoboRay]

To be fair, it's been pretty decent for a couple of years now, so it "only" took them SEVEN years to get it going good.

Re:Paying for More Slack Space.

[sshir]

The problem is with alignment of partitions. With old partitioning (like the one XP uses) everything is off by one 512 byte sector. So it will be very messy when your clusters not align to 4k boundaries.

Linux is (not) F'd

[Anonymous Coward]

/me pulls out my bag of troll-feed and doles out a handful....

Although I agree this is really bad for the large-scale enterprises with 5,000+ Windows XP boxen to support and for grandma wanting to play Final Fantasy VII, Linux isn't F'd.

The deal with Microsoft is that their going to refuse to tweak a fix for WinXP, citing monetary concerns.

Linux isn't "F'd", because 2 weeks after someone hits this brick wall, patches will start to rain down and land in our hot little hands. There might even be a minor point release bump to the Kernel just for this sort of fix.

You sir, are and idiot.

[Anonymous Coward]

That's very much like saying you had a compression algorithm which could store 512B typical data in 320B, but could store 4kB of the same type of data in a mere 100B.

ECC sizes don't need to grow linearly with data sizes, this is true, but they cannot SHRINK. You can get smaller per byte of data, but you cannot get just plain smaller.

Microsofts Take Away

[Anonymous Coward]

But Windows is the only operating system anyone would even need. And Microsoft Money apparently is the only money HDD Manufacturers will ever need.

The move to 4k benefits SSDs as well.

[Platinumrat]

If you read anand's articles on SSDs this jump to 4k block sizes would naturally fit in with how SSDs keep track of used pages. Thus would boost performance in the long run, if this was adopted as the new block size.

Re: Performance in "compatibility mode"

[butlerm]

The "compatibility mode" jumper shifts the alignment of logical block addresses to the underlying physical sectors so that instead of the physical sectors starting at LBAs that are a multiple of 8, they instead start at LBAs that are a multiple of 8, minus 1. That way the first partition, which traditionally starts at LBA 63, will be physical sector aligned. No gratuitous read modify write, no performance penalty, at least for the first partition.

Of course you don't want to switch the jumper on a formatted drive, because the position of all the data will shift by 512 bytes, making it look like garbage to any ordinary filesystem.

Virtual Machine (vmware, virtual box, etc)

[Anonymous Coward]

What impact will this have on Virtual Machines with respect to their Guest OS?

Re:Isn't this just a firmware change?

[Anonymous Coward]

It is not, at least for this case. The article clearly states that while the partitioning will be done at the firmware level, the firwmare capability itself will NOT be rolled back, as it largely depends on the platters themselves.

Re:You don't seem to get this whole base 2 thing :

[geminidomino]

But I like 255. It keeps my white page backgrounds white.

But just think, by going to 1023, it will keep them BRIGHT, not just white! That's 3x the extra space for fabric softener, too!

also

[shentino]

Not just for the sake of ECC space savings

4Ki also happens to be the page granularity for most modern processors, so you wind up having to write in units of 4Ki anyway whenever pages get dirty.

Re:Isn't this just a firmware change?

[Rennt]

If your business applications aren't platform agnostic by 2010 your IT dept have not been doing their job.

4096B is the new 512B

[Duggeek]

Those are "logical" sectors, which can be different from the physical sector size. [...] ...logical sector size is a drive interface level concept distinct from the filesystem cluster or block size. Filesystem block sizes have generally been larger than the logical or physical sector size for quite some time.

Thank you, butlerm for some much needed sanity in this thread.

The notion of cylinder/head/sector in the literal sense has been completely deprecated, but the scheme was preserved as an ad-hoc standard... which has since run into five other "hard limits" in BIOS and Int0x13 addressing... but that's another story.

Around the time of the first BIOS addressable-space limitation of 528MB, most hard disks were already being mapped with ZBR and translated through firmware. (see Zoned Bit Recording) The standard of 512b/sector has simply been a case of tradition and best practice. It was "just they way they made them."

A new logical int0x13 hook driver is all that's needed to interpret C/H/S coordinates with a 4KB base instead of a 512B base, and M$/Apple/Linus (et al) can likely cook that up in their sleep. No applications—short of low-level virus scanning, low-level disk utilities and software RAID, to name a few—would be affected by a different "physical" sector size. Most apps are in "virtual mode" and treat files as objects, which is handled in turn by the core OS.

Most filesystems use 4K granularity as it is, all that needs to happen is to equate cluster with sector. (as a most simplified "patch") I'm sure they'll come up with a new scheme to keep it scaled up... such as 4 sectors per cluster. (becoming 16K cluster/block size)

The greatest headache will be the game of catch-up by the OS-dev and utilities arena, who now have to completely review their preconceptions about HDD storage. Think regression-testing-cubed.

While we're at it, let's correct TFP by saying that filesystems will always align partitions to the beginning of the next logical cylinder. (whether it's a multiple of 4KB or not) The most likely problem to arise would be with the utilities mis-reading the "Advanced Format" disks, (showing them at only 1/8 of their actual size) calculating capacity based on total-sectors*512b, instead of the correct formula of total-sectors*4096b.

If you think about it, it's really a practical step. Think of where HDD sizes were ten years ago; 40GB was gi-normous and 100GB was just a pipe-dream. Ten years before that? A 100MB disk was "spacious" ...just a 1GB disk wouldn't be available for another five years. In all that time, the sector size hasn't changed at all... just half a kilobyte. It's a sign of the times, people.

Think of it this way... now instead of the nerdy term "sectors," you can introduce a newer term like "quads" or "kilo-quads" and see if it catches on.

/. at its best!

[AP31R0N]

The replies to my question here exemplify /. at its best. Informative answers with no snark. Thanks to everyone who replied. i learned quite a bit.

More Space = More Porn?

[ttyX]

Whoopie!