Serial ATA Technology Explained

Mike Parsons writes "Explosive Labs has an interesting article on Serial ATA . Here is a quote: 'In the rapidly moving computer industry, there are rarely the kinds of revolutionary changes like what is about to take place in secondary storage segment. Soon the hard drives and configuration methods that have existed since the origins of the personal computer will change forever. The basic IDE technology has been around for nearly twenty years. When the lifetimes of other computer components like CPUs and video are measured in months, twenty years ago seems like prehistory.'"

Comments

Prehistory? Depends on context

[Hentai]

Think about this - how long has RS232 been defined? How long has the PC's parallel (i.e., LPT1) pin-out been defined? How long has the VGA pin-out been defined? How long has the PC keyboard pin-out and protocol been defined? A lot of things change pretty fast; a lot of things stay around forever. It all depends on whether upgrading them is worth the cost in the long run.

Re:Prehistory? Depends on context

[Thurn und Taxis]

For that matter, how long has the x86 architecture been defined? How many /. readers were even alive when the 8086 was released? I know I felt a few grey hairs pop out when a co-worker told me his first computer was a 286. Only superficial changes in computer architecture have happened during our lifetimes. The way we interact with computers is totally archaic, just like the way we interact with banks, cars, washing machines, and televisions is archaic. The world is dying for some clever person to come up with a way to make it just as easy to ask a machine to do something as it is to ask a person to do something.

Re:Prehistory? Depends on context

[sjbrown]

• RS232:
  Three modifications. Most recent in 1991
• Parallel port:
  Does anyone still use this? In the past 6 years, any printing I've done has either been off a network printer or at Kinko's
• VGA pin-out:
  Here comes DVI!
• keyboard pin-out:
  USB!

But yes, I acknowledge your point.

Re:Prehistory? Depends on context

[dutky]

With one exception, none of the "standards" you mention are either very old or very standard. The exception is, of course, RS-232, which was defined in 1969 by the Electronic Industries Association of Washington D.C. (my home town). All of the others are relative infants:

• The PC parallel port was defined, in a very loose manner, by IBM when the IBM-PC was introduced in the early 1980s. The original PC parallel port was unidirectional and good for little else than driving a printer. While a number of manufacturers offered bi-directional versions of the PC parallel port (including IBM, in the late 1980s, with the IBM-PS/2) no actual standard existed until the release of IEEE-1284, in the mid-1990s, which gives us the modern ECP/EPP option.
• The VGA pin-out dates from the introduction of the IBM-PS/2 in 1988/1989.
• The PC keyboard pin-out and protocol have only been around since the introduction of the IBM-PC, and have undergone at least two revisions, first for the IBM-AT keyboard, and second for the IBM-PS/2.

The reference to IDE having been around for 20 years is a pretty dubious one, unless you count IDE as just a variation on the vernerable IBM-PC expansion bus. (commonly known as the ISA bus) Still, 20 years ago, there was no such thing as IDE, even as a glimmer of a hope in anyone's eye.

Re:Prehistory? Depends on context

[Call Me Black Cloud]

You're correct. Consider the length of time the original male-female coupling system has been around...

Re:Prehistory? Depends on context

[chemmathguy]

I think that this "standard" will be around for a very long time...(andd I have no complaints about that =:p

Re:Prehistory? Depends on context

[Wesley Felter]

I don't have any of those ports on my computer; it's got USB and DVI instead.

Re:Prehistory? Depends on context

[Arandir]

My first computer was an 8086. And yes, I said 8086, and not 8088.

Re:Prehistory? Depends on context

[shepd]

>FYI, most modern bioses recognise USB keyboards and can be accessed from them.

Oh, I know that.

I also know its an option I disable on all my boards, and an option that isn't always the default.

This means that if you only have a USB keyboard and buy/receive a motherboard with USB disabled by default, you just bought a heap of trash.

Re:Prehistory? Depends on context

[mehfu]

Well, there are other standards available, which aren't as functual when it comes to the result, but enthusiasts say that the procedure is just as good, or better.

Re:Prehistory? Depends on context

[BWJones]

Well, the Macintosh boxes and laptops I use have none of those protocols. I use USB, Firewire, and ADC protocols. However, the desktop machines do use SCSI hard drive protocols.

There are reasons that older standards have been supplanted for bandwidth, convenience (hot plug and play is sweet), and useability and I would certainly argue that upgrading from all of these older protocols is certainly worth it. With USB I get decent bandwidth for tasks that require it (synching PDA, downloading digital camera images, hot plug and play of trackball and keyboard use). With Firewire I get hot plug and play 80 GB backups in under an hour and a half and very fast transfer of larger image datasets. Oh, and very fast iPod synching). With ADC I get an uncluttered single cable going from my flat panel to my computer that carries power, pure digital video signal and USB.

Re:Prehistory? Depends on context

[civilizedINTENSITY]

But did you ever see a 4004?

Re:Prehistory? Depends on context

[bobv-pillars-net]

Yeah, my Casio EC4004 calculator was based on a 4004.

Re:Prehistory? Depends on context

[sconeu]

The IDE I/O register mapping/interface dates to the ST-506 interface (early '80s), defined by Shugart Technologies.

The (16-bit) IDE physical interface is an extension of the AT bus (1984).

Re:Prehistory? Depends on context

[sconeu]

I was in college. I got my first 8086 manual (original Intel, onionskin) as a college text.

Re:Prehistory? Depends on context

[Oggust]

The PC keyboard interface has seen three revs, the original one was for the IBM PC RT and it had a lot of cool features that they removed for the XT. (Along with a lot of other stuff. Both the PC and the POWER-based RS/6000 architecture stems from it. It used a short-lived (but pretty cool) RISC CPU of a family called ROMP, and it ran AIX 2.XX)

It had a different physical connector too, most connectors on the RT was in the same style, kind of like a ribbon cable connector, but with a housing that prevented you from connecting stuff upside down.

The physical interface for RS-232 has changed quite a lot, too. Back in the olden days (terminal nets), 15-pin D-subs were very common, and you NEVER see those today. People would think it was a game port I guess. But you can sometimes see that there's one of those push-out D-sub blanks on the back side of a PC box that is too big for a 9-pin and to small for a 23-pin connector. That's what that's for.

Of course, the RT had the flat, small (8-pin IIRC) connectors for that too. It was pretty handy, you could get a 4-port serial card that didn't need octopus cables or external connector boxes or anything.

/August, got rid of my last 6150 a little over a year ago, but still has a few IBM 9332/400 disks.

Re:Prehistory? Depends on context

[slipgun]

Mine was a z80 - in fact the zx80

If you've still got a ZX80, I believe they're quite valuable these days... in the order of 250ukp.

Re:Prehistory? Depends on context

[hpa]

The PS/2 bidirectional serial port was all but 100% standard ever since the early 1990s. ECP/EPP came before IEEE-1284 (which was the standardization of these options); they were already ubiquitous when the first SuperIO chips came out

Furthermore, the electrical interface of the parallel port was defined by Centronics some time in the mid-1970s, and significantly predates the IBM PC -- IBM used it because it (and RS 232) exactly because it was *already* a standard!

Note also that on the original IBM PC all of these came on separate boards that you had to buy extra. Noone bought an "IBM Personal Computer Asynchronous Communication Adapter" (serial port) unless they had a modem to plug into, and noone bought a parallel port unless they had a printer.

Re:Prehistory? Depends on context

[shepd]

>Since buying a new keyboard costs $20-30 (U.S.), who cares?

Yeah, but why not just use a PS/2 keyboard to start with?

USB keyboards add nothing but trouble.

Re:Prehistory? Depends on context

[ka9dgx]

Let's see...
Commodore PET - 6502
Cosmac ELF - 1802
Heathkit Trainer - 6800
Radio Shack CoCo II - 6809 (OS-9 Rules!)
Custom built project - 8088 and glue logic
First computer repair - PDP11/34 with bad RAM
Got into it just after they phased out punch cards.

I have many friends who were there before CP/M, Xmodem, etc.

--Mike--

Re:Prehistory? Depends on context

[Wesley Felter]

We have a winner. Macs use USB for keyboards and mice.

Re:Prehistory? Depends on context

[Eil]

I agree to a certain extent. But follow me for a moment. Remember that just because something is old doesn't mean it's obsolete. The idea of using a machine or system to carry out lots and lots of trivial calculations in a short period of time goes back longer than most of us can trace our family tree. Yet today that idea is more important and ultimately more useful than it has ever been and things are only going to improve.

If we improve upon the technology and methods that exist today, there's simply nowhere to go but up. If we just scrap what we have and try to think up something better everytime a piece of technology seems old-school, we waste a lot of time sitting around and thinking.

Only superficial changes in computer architecture have happened during our lifetimes.

The first microprocessor had 2000 transistors. The Pentium 4 has 55 million. Modern processors also have features like optimizing cache logic, special-purpose instructions, and mechanisms to accelerate throughput. I'd call that more than a superficial change. Of course, the concepts of accumulators, registers, instructions, interrupts, stacks, and etc have been around for a long time, but like I mentioned above, that doesn't make them obsolete. If you took them out of the machine, you no longer have what anyone could call a microprocessor.

The way we interact with computers is totally archaic, just like the way we interact with banks, cars, washing machines, and televisions is archaic.

Er, they seem to work well enough for billions of people. My keyboard and mouse, for instance. I'm sure there are quicker ways to enter data and I'm sure there are more accurate ways. But not both. The keyboard and mouse are the best balance of speed and accuracy we have. Technology has not advanced enough to give us an increase in both, but I am confident that it will someday.

The world is dying for some clever person to come up with a way to make it just as easy to ask a machine to do something as it is to ask a person to do something.

I don't think I can speak for the rest of the world, but I personally agree. I hope to own, within my lifetime, a house with a computer that you can literally talk to and have it respond and perform actions based on your requests. It would be quite similar to ship-wide computers on Star Trek where you could tell it to adjust the room temperature, dim the lights, put on some Beethoven, give you the proper spelling for a word, list hibachi dealers in the area, record Enterprise at 9 o'clock on Friday, quiz my daughter on fractions, and compile a brief history of the Intel 4004 microprocessor from available sources.

Today's computers are capable of all of these functions (with the exception of the last one), just one key element is missing: artificial intelligence. Yes, computers can recognize speech, but the accuracy is horrible and they lack any way to figure out what you actually mean and neither can they recognize when you're talking to them or when you're talking to yourself or someone else.

However, it's not as simple as just waiting for some genius to come along and bestow AI upon the world. People used to believe that computers would never be able to spot patterns in a suitably complex set of data, but now we have a number of algorithms that do a sufficient job. It's going to be a constant evolution, but I'm confident that we'll get there. We probably won't ever see a "sentient" computer, (which is what most people think of when they think of AI) but someday we'll all be rambling off instructions and requests to our computers.

And then someone will make the comment on Slashdot that speech interfaces are archaic.

SCSI?

[GreyWolf3000]

Soon the hard drives and configuration methods that have existed since the origins of the personal computer will change forever.

All of IDE's shortcomings are fixed by SCSI (except for a small degree of added complexity). SCSI hardware is more expensive, and rarely does it come built-in to motherboards.

If more people used it, it would be a cheaper solution, and would fix all of IDE's problems without re-inventing the wheel--it's a solution that, right now, works.

15k rpm scsi drives get seek times in the low three range--that's three times faster than your average 5400 rpm ide hdd.

Re:SCSI?

[stratjakt]

> All of IDE's shortcomings are fixed by SCSI

You mean the oversized 40-conductor ribbon cables are solved by 68 conductor ones?

> If more people used it, it would be a cheaper solution

Heh, sure it would.

> 15k rpm scsi drives get seek times in the low three range--that's three times faster than your average 5400 rpm ide hdd.

And 10x more expensive.

You are the first SCSI fanboy I've ever seen.

Now bring on the cheap!

Re:SCSI?

[jandrese]

SCSI is largely expensive because it is sold to the server market. It would not be impossible to make a SCSI drive only a few dollars more than the comparable ATA drive, but why bother when you can sell it for twice the price of the comperable ATA drives. SCSI drives will stay expensive until people are no longer willing to pay the huge markup for them.

Re:SCSI?

[shepd]

>You mean the oversized 40-conductor ribbon cables are solved by 68 conductor ones?

Nope, they're solved by using high-density cable connectors, which IDE still hasn't figured out with 80-pin cables (instead IDE just cheats).

>And 10x more expensive.

Because... why?

Nobody is buying, that's why. Lower speed SCSI drives are still available, but are still expensive because IDE is stuck in everyone's heads as the only storage method for a PC.

>You are the first SCSI fanboy I've ever seen.

You haven't been here very long. Let me be the first to say welcome! :-)

Re:SCSI?

[Stigmata669]

"All of IDE's shortcomings are fixed by SCSI (except for a small degree of added complexity)."

Yeah, even if your statement were true. Price is a huge shortcoming in technology today, especialy when most people can't use the preformance hardware that they own.

I strongly support the development of IDE standards. There are many situations when you need lots of hard drive space without bleeding edge preformance. Try to find a solution for doing a small (350GB) backup server to add to a tape backup system, you find 200GB IDE drives for the price of a 18GB SCSI drive. A thousand people will try and explain why the SCSI is a better deal because its more reliable and faster, but backing up 350 gb on IDE costs about $650, and on SCSI costs aboyt $4000. I don't believe the demand is the reason for the premium price for SCSI, but the hardware... It's just more expencive to make.

Re:SCSI?

[puppetman]

Lots of motherboards come with SCSI. Most Intel server boards do, as do boards from SuperMicro (P3TDE6-G, S370DL3, P3TDDR), Tyan (S2721, S2720), etc, etc. I've been looking into them for server upgrades (I just don't have time to track down a few dozen model numbers).

SCSI does have a faster spin time, and is in general much faster, but for small, random chunks of data, it can be slower (but it's a small penalty in a rare case). The width of the SCSI bus is much better (320 MB/s for SCSI 320), assuming you have alot of drives in big striped arrays.

Don't forget - most IDE drives (except for a few premium models) just had their warranty chopped from 3-years to 1-year. Not a reflection on declining quality of IDE drives, but rather the economics of the market place. SCSI still has decent warranties, and they last longer regardless.

SCSI is (much) better; could they be as cheap as IDE if everyone used them? Probably pretty close. But it's a chicken and an egg thing. Home-users don't buy them because they are expensive, and they are expensive because consumers won't buy them.

Ironically, SCSI stands for Small Computer Standard Interface, but SCSI is most frequently found in Servers (large, not small computers), in large RAID arrays. And more ironically, the SCSI drives usually used in RAID (Redundant Array of Inexpensive Disks) usually are not that inexpensive at 3x the cost of IDE.

Re:SCSI?

[Enry]

> You mean the oversized 40-conductor ribbon cables are solved by 68 conductor ones?

No, I'd guess the 80 pin ones that include power and configure the drive's ID, and allow you to just slap the drive in and let it go. IDE has NOTHING with that configurability.

> You are the first SCSI fanboy I've ever seen.

Now you've seen two.

Don't get me wrong - for home use, SCSI is overpowered. But if you're talking anything bigger than a desktop, make mine SCSI.

Re:SCSI?

[km790816]

Read the article.

The problem with SCSI is heat and noise, things that work fine in a server room, but are bad for the home user.

I had SCSI for a while in college. It was cool to show off, but having to turn up my TV to hear over the jet engine in my PC was annoying.

Re:SCSI?

[stratjakt]

Well, yeah, SCSI will always have a place in high end workstations and servers.

But my new desktop (as I've already posted) has an onboard SATA RAID controller.

Personally I cant wait to get some workstation-level performance at desktop-level prices.

And no more big ugly cables uglying up my side window. These little cables that came in the mobo box are going to look pretty good in there.

It's an improvement over PATA no matter how you slice it. It promises to be cheaper, faster, easier, hot-swappable, self-powered.

I dunno why people are complaining. Well, yes I do. This is slashdot, and any new technology is equated with a bid for world dominance from [CORPORATION].

Re:SCSI?

[megaduck]

Actually, even Macs don't use SCSI anymore. SCSI is strictly a "build-to-order" option on their PowerMac line. Even their Xserve server and storage products use multiple IDE channels instead of a SCSI bus.
</nitpicking>

You're right, though. SCSI only really survives in servers because it's just too darn expensive for everything else. Shame.

Re:SCSI?

[Sneftel]

Err... since when did a communications bus generatte any appreciable noise or sound? Nothing distinguishes the dynamics of hard drive operation between IDE and SCSI. SCSI hard drives tend to be faster, since people who buy them tend to want to attach them to servers, but if pushed into the home consumer arena you'd see SCSI drives with exactly the same noise/heat output as IDE drives. This is an example of market-driven, not technology-driven.

Re:SCSI?

[shepd]

>I would like to say that the price of SCSI drives is not 10x more because nobody wants a SCSI drive, but because they are simply more complicated to manufacture and interface.

That was true in 1980. It isn't now. I mean, it costs no more to build a RAID card (witness Promise Ultra Hack) than it does to build an IDE card, and that's massively more complicated than SCSI.

In fact, since all the smarts are moved onto the controller (which is dead simple to make today -- I bought a cheap one for $35 CDN two years ago -- it was cheaper than a cheap IDE card!) the drive itself is less complicated.

>Also, the preformance hit you take going from 7200rpm SCSI to 7200rpm IDE is not noticable at most times, but I suppose i am tolerant because i can wait more than 3 ms seek.

Agreed, but there's more to it than that. IDE requires new interfaces every drive (unless you want to take the horrible performance hit master/slave arrangements incurr). SCSI doesn't. IDE cables can only be 18". SCSI can be quite a bit longer. IDE only works for hard drives and CD-ROMS (and one or two other things). SCSI is for anything. There's more reasons than this to support SCSI over IDE (at the same price), but I think these three would be enough to sway users at the right price.

>SCSI is loud and hot and expencive, just like all preformance computing componants, and thus will never be a consumer standard.

Only because no manufacturer thinks there's a good market if they slap a SCSI controller on their current consumer hard drives. I think there is, and I'd be game to buy one, if they existed, for my next upgrade.

Re:SCSI?

[GreyWolf3000]

Yeah, even if your statement were true. Price is a huge shortcoming in technology today, especialy when most people can't use the preformance hardware that they own.

Here's why your point is moot: First, IDE drives are only cheaper because it's more widely used. Secondly, SCSI can scale to low end needs--you're not locked into having to be fast. What has happened is that SCSI has found its nitch in the server market. SCSI could downscale very well

I don't believe the demand is the reason for the premium price for SCSI, but the hardware... It's just more expencive to make.

That's only true because SCSI hard drive makers have to make fast server-oriented drives. If they had a desktop nitch, they could sell slower drives for similar prices--as it stands they couldn't compete.

SCSI can serve both low-end and high-end needs, if folks would give it a chance. The article's mentioning of SCSI being too loud is ridiculous--internally they all work the same. 15k rpm hard drives by nature make a lot of noise.

Re:SCSI?

[GreyWolf3000]

SCSI could scale down to fit desktop needs, but it wouldn't since IDE has such a high market share, it would start out more expensive. SCSI is expandable and scaleable (in both directions).

Re:SCSI?

[G-funk]

it's got nothing to do with how common it is, scsi is expensive because it raises the percieved value of the product, and it's common knowledge that this "elite"ness of more expensive alternatives to a product aid sales.

In short, it's expensive because they like it that way. There's no shortage of scsi drives, they're not particularly more expensive to produce.

Re:SCSI?

[PotatoHead]

Now you have seen three!

Scsi is fast and reliable. Notice that the price premium gives you a warranty these days where IDE does not. (Cheap has to come from somewhere, looks like it will be the IDE drives :)

SCSI devices last forever, or until they break, which is a long time. Move them from machine to machine, OS to OS as much as you want. They are almost all cross-platform and compatable. Just can't say that for IDE in general.

SCSI is really pretty easy. Where IDE can be a mess sometimes. Want to share a CD-ROM and DISK on same controller, want speed also? Not likely to happen unless you have a higher end IDE controller, but if you are going to go to this trouble, why not just get a SCSI one and be done with it?

Your 10X more expensive is packing a lot less punch these days and it's not true anymore. If you want the very high end SCSI devices you will pay a premium, but choose one notch down and the prices are fair. 3-5x maybe, not 10x but that does include nicely supported devices.

Most of the hard work has been done for SCSI, this new standard will cost all of us who depend on IDE. I would much rather keep my investment in SCSI hardware rolling rather than start paying again for another standard that is supposed to fix the broken IDE standard I have to live with now.

So, I may just do that. You see with SCSI, I have that choice, have always had it. Given a longer term useful life, it is fair to say that SCSI has saved me money! IDE? Not!

Re:SCSI?

[GooberToo]

Someone mod this guy up. He, "gets it!" SCSI is expensive because it's the premium, top-shelf product. Period. SCSI is expensive for the same reason that diamonds are. Both are highly desired but have no more value than the next drive (be it IDE or rock).

In other words, they are expensive because people want to keep them that way!

Re:SCSI?

[dan the person]

"Ironically, SCSI stands for Small Computer Standard Interface, but SCSI is most frequently found in Servers (large, not small computers), in large RAID arrays. And more ironically, the SCSI drives usually used in RAID (Redundant Array of Inexpensive Disks) usually are not that inexpensive at 3x the cost of IDE. "

You have to look at when these terms where invented.

The following definitions applied back then:
Small Computer = a computer that did not require an entire building to house it.
Inexpensive = cheaper than solid state.

Re:SCSI?

[fava]

I quote:
SCSI devices last forever, or until they break,

I can top that:
I will live forever, until I die.

What was your point again?

Re:SCSI?

[egarland]

> All of IDE's shortcomings are fixed by SCSI.

Heh. Here's a list of IDE's shortcomings SCSI makes worse:

• cable size
  
• interoperability issues caused by multiple drives per cable
  
• bandwidth per drive
  
• cost of the controllers
  
• cost of the cables
  
• cost of the drives
  
• Low reliability caused by multiple devices physically attached to the same cable that can bring parts of the bus or the whole bus down
  

The bandwidth per drive thing is one of the great things that SATA brings to the table. With a modern large SCSI setup it seems like you have a lot of bandwidth but on a per drive basis you really don't. 160MB/s divided by 12 drives = 13MB/s (1980's speed). To contrast that look at a 12 drive 3Ware SATA controller. That has a full 150 MBytes/Sec to each of the 12 drives.

To see the usefulness of this take the example of a 12 drive RAID 5 array doing a rebuild while the server is trying to read from the drives. The controller has at it's disposal 1800 MB/s worth of bandwidth that it can use. It can run those drives as fast as they can go keeping the write buffer full on the drive it's rebuilding and using the leftover bandwidth to service the server's requests. Modern ATA drives can read at up to 56 MB/s. With 12 drives you get a total of 672 MB/s throughput which is far more than even the new Ultra320 SCSI is capable of. With newer faster drives and 16 drive RAID controllers this problem gets even worse.

> If more people used it, it would be a cheaper solution

SCSI is quite widely used. There is a lot more SCSI out there than SATA and yet a motherboard with a SATA raid controller costs about the same as one without it whereas a motherboard with a SCSI raid controller on it costs about 3 times as much. SCSI is simply an expensive, complicated technology to implement.

>15k rpm scsi drives get seek times in the low three range--that's three times faster than your average 5400 rpm ide hdd.

The low seek times are a result of expensive server class drive technology, not of the interface. Seagate could just as easily drop a SATA interface on those 15K Cheetah drives and I suspect in the near future they will because:

All of SCSI's shortcomings are fixed by Serial ATA

Yea, I know, it's a cheap shot but really SATA is poised to replace SCSI in most of the markets SCSI still occupies. SCSI was mostly popular in server systems because of it's hot swapablility and plug and play operation (no jumpers to set on 80 pin sca drives.) These are advantages that Serial ATA shares. Motherboard integrated SATA RAID will take over for SCSI RAID in server class systems because of cost, size, power and bandwidth issues. 8 - 16 drive SATA RAID arrays will take over the low to mid-size storage array market. (If you can count 4.8 Terabytes as mid size.) Fiber channel will be left for SANs and large storage arrays. SCSI will be relegated to connecting external drive systems but I imagine fiber channel will eventually take most of that market.

People who like SCSI will probably like SATA even more. It will be faster, much cheaper, more reliable, more compatible, and easier to maintain and troubleshoot. True, you won't be able to run a printer or scanner off it but I doubt there will be a lot of people missing that particular piece of SCSI functionality.

Re:SCSI?

[Have Blue]

That's not counting the FC RAID add-on unit that should be released Real Soon Now(tm).

Re:SCSI?

[gad_zuki!]

> SCSI is expensive for the same reason that diamonds are.

Not at all. Diamonds are a controlled cartel and price-fixing is par for the course. SCSI in mass production anywhere near how much IDE stuff gets made will drop the price to the point where it would be affordable to sell it to the home market. Or so the theory goes.

If diamonds weren't price-controled they would be incredibly cheap. Read about it here: http://www.professionaljeweler.com/archives/news/2 000/020300story.html

Re:SCSI?

[PotatoHead]

You know there is *always* someone....

Well, if you take the rest of the sentence, which you left out, in context with the rest of what was written, it is not much of a stretch to understand that SCSI devices last longer in general than IDE ones.

This general idea expressed in the first paragraph nicely ties in with the part you left out; namely, ", which is a long time." Maybe that has just a little to do with why exactly you left it out, or could it be that you just knee jerked a smart ass response?

If you consider the entire first sentence, then you will come to realize that it sort of hangs at the end. You might wonder how long of a time after reading it? Well how long? Compared to what? Why so?

All three questions are answered in the first paragraph. Compared to what refers to IDE devices in a negative way. (Intended!)

Why so is answered by a little critical thought surrounding the warranty part of that paragraph combined with the cheap part of it. Why would one cut the term of their warranty? Could product failure costs have anything to do with it? Maybe, but if that fails then cheap is there to back things up. Taking the cost out of things often means cutting a few corners. Well IDE drives are cheap right? Must have cut a few corners. Maybe those that affect lifespan?

Given this, how long is pretty easy. Longer than IDE devices in general.

There you go.

If that does not do it for you, then know it's late, and I'm done with this.

Re:SCSI?

[adolf]

There is a certain elegance to such things as SCA backplanes which ATA (in any incarnation) will never achieve.

Which is another point:

They botched SATA by not including power on the bus, having no standardized connector locations, and (at least so far) having no facility for connection of external devices.

The 1-device-per-wire rule of SATA is another detriment: Sure, the cables are fairly small, but can you imagine the rats nest that would be a 12-device SATA system?

SCSI daisychaining is an easy fix for that, and now-common LVD SCSI is quite able to support this number of devices with a single ribbon. And LVD, by its differential nature, is quite resistant to the electrical problems introduced by slice-and-jacket cable rounding techniques that are all the rage these days..

Oh. And you've got your math wrong, Son:

It doesn't matter if 3ware makes a controller with 12 150MBps ports. The 12-port Escalade 8500 you speak of has a 64-bit 33MHz PCI interface, topping out at no more than a paltry 264MBps to be shared by all connected devices.

If you're serious about throughput, try something like this: Two 320MBps channels on a 64-bit, 133MHz PCI bus. Good for real-world transfer rates in the realm of 640MBps, more than twice that of the 3ware product, while keeping a good portion of the PCI bus free the -other- 30 SCSI devices you've got plugged into it.

And none of this says anything about the benefit of SCSI for the home user:

Just bought a new DVD-R, but don't want to toss your old but dead-solid Plextor? SATA requires you to buy another port. SCSI just requires you to plug it in.

Re:SCSI?

[AlecC]

The cost difference between a Scsi interface and a Parallell ATA interface is negligible. At the worst it was only ever about $10, ten years ago, and now it is probably about $0.01. Manufacturers just use the Scsi interface as a way of separating "professional" grade drives from "consumer" grade drives. Scsi drives go up to 15,000 rpm, seek times of 4.5msec, five year warranty. IDE drives go to 7200rpm, 9 msec seek times, one year warranty. Scsi is not intended to be a consumer standard, it is intended to be a "hot rod".

On a single user computer, you will be unusual if you need the performance of a scsi disk. You may like the reliability (but there is always IDE based raid). For a central server, however, the scsi difference matters.

Re:SCSI?

[Spoing]

If more people used [SCSI devices], [SCSI] would be a cheaper solution, and would fix all of IDE's problems without re-inventing the wheel--it's a solution that, right now, works.

All my systems are SCSI-based, usually with one or two IDE devices added as adjuncts. While it is true that many of IDE's failings could be solved by using SCSI instead, the focus of each interface is entirely different. Both SCSI and IDE are valuable on thier own merits;

1. SCSI: A complex protocol to control propriatory and diverse devices transfered on standardized cables and handled by propriatory controllers. Hardware control of the device is handled by the protocol layer that is supported by a propriatory hardware and software layer plus the electronics on the device. These added layers are required because SCSI devices are so diverse and the controller-to-system interface is not standardized. SCSI devices can be any type of data storage unit (from simple hard drives to chained optical juke boxes), all types of scanners, network interfaces, and other standard and custom devices. The SCSI protocol allows extentions on a per-device basis. SCSI controlers require a chipset-specific software layer to link them to the rest of the system.

   IDE: A single physical and protocol interface located primarily on the drive itself. IDE was designed to be a cheap and simple extention of x86 chipsets for specific data storage devices; hard drives. CD, DVD, and enhancements to the IDE chipsets were added later. Because the drive electronics masks the true nature of the drive attached, and the system-side IDE support is generic, no chipset-specific drivers are required. (Yes, some IDE chipsets are garbage and require special support to prevent data loss. Most of these defects occured because the testing done was limited to "does it work with Windows 3.x/9x?" and things like DMA were largely ignored. Oops!)

All things considered SCSI would probably be cheaper now if it were used as the standard interface on x86 systems back before IDE, though it's impossible that it could underprice IDE just because of the propriatory nature of the hardware-heavy controller chipsets -- each flavor with it's own unique set of features. Keep in mind that Apple used SCSI as it's standard till reciently and switched to IDE mainly because of cost.

That said, I wish both IDE and SCSI devices a heart felt and fond "get lost". Newer device interfaces are cheap, fairly generic, fast, and allow arbitrary devices to be attached.

I don't expect a complete switch to happen soon, though fiber channel has replaced some of the functions that SCSI devices used to perform on the high end, and IDE is too cheap to pass up on the low end.

Re:SCSI?

[Zathrus]

IDE has NOTHING with that configurability.

Uh... yes it does. It's called SATA. See the article.

And, frankly, configuring IDE drives isn't the nightmare it used to be. As long as the jumpers are set right then the BIOS will usually auto detect and setup the drive properly. Only time I've had problems with that is jumper issues or having explicitly set the location to "disabled".

Don't get me wrong - for home use, SCSI is overpowered. But if you're talking anything bigger than a desktop, make mine SCSI.

That I can certainly agree with you on.

Re:SCSI?

[GooberToo]

If diamonds weren't price-controled they would be incredibly cheap.

And if you substitute SCSI drive for diamond, your original statement is also true! I'm fully aware of what the situation is with diamonds. Obviously. If I had not been, I wouldn't of made the reference. Because of the fact that it's controlled by cartels, that's the reason why diamonds sale for 1000x or more its actual value. Since SCSI is mearly a top shelf product, you don't see such extreme differences but the market mechanics are the same.

If SCSI weren't price-controled they would be incredibly cheap. See how the same is true? SCSI is expensive because they want it to be. Period. It's artificially inflated in place. This is VERY COMMON for ANY premium product! Argue all you want, this is how the real world works!

Re:SCSI?

[sfgoth]

15k rpm scsi drives get seek times in the low three range--that's three times faster than your average 5400 rpm ide hdd.

Mostly because they're spinning three times faster. It has nothing to do with IDE or SCSI.

-pmb

Only if...

[SirKron]

Serial ATA will only take off if it is not more expensive than parallel ATA. If we (meant users) want to spend more we would buy SCSI. What I want is a low cost way to stipe 4+ drives at home.

Re:Only if...

[rodgerd]

This gives it to you. The main problem with IDE-RAID is fucking around with cables and a standard that doesn't support hot-swap.

yeah, yeah, yeah enough with the sales pitch

[stratjakt]

Where are my drives?

I got a shiny new SATA RAID controller on my new motherboard, now when the hell am I gonna get a couple of 80 gig cheap, fast SATA drives to put into a striped set?

huh? huh?

Re:yeah, yeah, yeah enough with the sales pitch

[bjschrock]

Well... Seagate has announced their first SATA drives, the Cuda ATA V, but they haven't hit the shelves yet as far as I can tell. It is supposed to be available this fall... so they should be out soon. Also, the Cuda V is a true SATA drive, not just an IDE drive with a bridge slapped on it.

FIREWIRE?

[cybercomm]

So...why don't we use firewire? Isn't it faster than SATA? And the upcomming Firewire IEE 1394b should double the firewire speed to ~800mb/s. And let's not forget the fact that there are firewire HDD-s and other perhipeals on the market (though they are generaly external) or maybe, could this have anything to do with INTEL's desire to controll all components? I don't see the price as a limiting factor either.

Re:FIREWIRE?

[cheezedawg]

Um, no. Gen 1 SATA is 1.5 Gb/s. Firewire is not faster.

Re:FIREWIRE?

[Salamander]

Actually SATA is 1.2Gb/s, but that doesn't change your point.

Re:FIREWIRE?

[cheezedawg]

Actually, it is 1.5 Gb/s. There is a 8b/10b encoding at the link layer.

Re:FIREWIRE?

[Salamander]

Bits that the signalling throws away don't count.

Re:FIREWIRE?

[cheezedawg]

Well, I'm not the only one that calls it 1.5 Gb/s. The serial ATA website often refers to it as 1.5 Gb/s.

Also, signalling bits that are thrown away are often counted. Otherwise we would have 98 Mb/s ethernet instead of 100 :)

Re:FIREWIRE?

[Wesley Felter]

Well, I'm not the only one that calls it 1.5 Gb/s. The serial ATA website [serialata.org] often refers to it as 1.5 Gb/s.

That's what you get when you don't keep a tight leash on the marketing folks. InfiniBand is guilty of the same thing.

Also, signalling bits that are thrown away are often counted. Otherwise we would have 98 Mb/s ethernet instead of 100 :)

No, you've got it backwards. Fast Ethernet runs at 125MHz, so it would be 125Mbps if you count overhead and 100Mbps if you don't count the overhead. Guess which number people use.

This just looks expensive.

[jericho4.0]

The first generation doesn't look that impressive, as shown by the selling points given by the author. Thinner cables, up to a meter long?? Ok, sounds nice, but not worth paying extra for, for my needs. Serial ATA wil debut at 150 mb/s, not really an improvment at all.

The author then goes on to note that the 'roadmap' calls for the 2006 version to run at 600mb/s, which fits nicely with my roadmap to world domination in 2005. ...Ummmm, yeah, we'll see.

Although looking at the list of upcoming products and the manufactures making them, I don't doubt we'll all be useing this in a few years.

Re:This just looks expensive.

[pope nihil]

Actually they said SATA 3 (600 MB/sec) will debut in 2007.

I was a bit confused by this article. They talk as if this thing is the Second Coming of Christ, but then they talk about how desktop pcs are just going to keep taking baby steps. Also at the beginning of the article they say that serial seems to be a step back from parallel (ya think?) but it is faster and better and Oh! Look! An elephant!

Re:This just looks expensive.

[photon317]

The battle between serial and parallel communications is neverending. Show me a Serial WAN connection like a DS3, and I can say "Well, since you never send partial bytes, we could strap 8 of these side by side, send one byte at a time with the bits split up over the 8 DS3s in parallel frames, and we get an 8x speed improvement that's usable by a single connection and no additional latency".

Or show me a parallel bus like IDE, and I can say "Look, having all those data lines next to each other causes additional interference we have to account for, and they're bulky, cost more, overly complex, blah blah. If we just put a serial bitstream on a pair of wires, it would be so much simpler that for the same cost we can turn up the bitrate more than enough to make up for the lost parallelism."

It's all the same. Various communication technologies tend to rise and fall, serial replacing parallel replacing serial replacing parallel ad infinitum. In some cases (like PCI busses) parallel just makes a lot more sense, but in a lot of cases (network stuff, storage stuff especially) there's a tradeoff where both are better and worse than the other in different ways. You could just pick one and stick with it and do you incremental improvements, but the occasional switcheroo provides upgrade revenues and more user "wow" factor and buzzwords.

Re:This just looks expensive.

[Magila]

Eh, parallel is on the way out for good. You've obviously never heard of a little thing called clock skew. It's what makes your DS3 example impractical since the bits won't all arive at the same time, with the run lengths you're talking about with a DS3 the potential difference would be too large to effectivly compensate for.

Re:This just looks expensive.

[bjschrock]

It is an impressive jump ahead, and it's not going to be much more expensive. It's estimated that Serial ATA interfaces will be less than $10 more expensive than their parallel counterparts, and that will go down once it becomes more popular. Also, it's not 150 M-bits/sec, it's 150 M-Bytes/sec. Another big advantage is that Serial ATA is that the drives will be hot-swapable. The new power connectors have extended ground plugs so that the grounds contact before power... also drives will get to use a new voltage line (+3.3V) that they didn't get before.

Another interesting thing about the technology is that drives that are currently using the parallel SCSI interface will be moving to either SAS (Serial-Attached-SCSI) or Fibre Channel. SAS will use the SCSI protocol over the Serial ATA cables, so you can get rid of those nice giant ribbons.

Re:This just looks expensive.

[mosch]

Let's repeat that one more time for the Google archive. Magila has said that "parallel is on the way out for good."

As Will Rogers once said, "It's always risky to make predictions, especially about the future."

Re:This just looks expensive.

[Magila]

You want to come up with a real argument or just make stupid quips and quote people? Parallel interfaces are limited by well eshablished physical properies of electricity traveling through a conductor, thoes properties arn't going away any time soon and pretty much every known method of transmitting 1s and 0s has similar properties. Parallel interfaces came along because signal processors of the time were slow enough that these physical properties weren't a problem. Now that we can send signals of high enough frequency that things like clock skew do become a problem it makes sense that we abandon the parallel interface and never look back.

Re:This just looks expensive.

[bryan1945]

Interesting that no one else caught the one flaw in the serial/parallel debate- optics. Not quite price feasible now, but maybe 5 years or so(?), but run X number of single-mode fibers and you have X parallel bits with zero interference. It has its pros and cons, but nice performance!

Re:This just looks expensive.

[rodgerd]

Thinner cables are very impressive for those who like to stuff drives in boxes. In fact, thinner cables sell me on the standard - I don't care about the speed.

Re:This just looks expensive.

[jericho4.0]

What!?!?!?!?! Thats crazy. Oh, and btw, I have a nive Vax for sale, $20 + shipping. Obviously a great deal for the likes of you.

Re:This just looks expensive.

[Zeio]

I have to agree with alsta here and vehemently disagree with the moderation "flamebait."

Lets got over this supposed flamebait.
* SCSI has much better seek times than IDE disks
This is irrefutable. This is in industry rule. I refer people to the fastest hard drive I could think off the top of my head here: http://www.storage.ibm.com/hdd/desk/ds180gxp.htm Deskstar 180GXP - Average seek time - 8.5 ms
and to here: http://www.seagate.com/cda/products/discsales/mark eting/detail/0,1081,549,00.html 15K.3 Model Number: ST373453LC Seek time: 3.6 ms avg

Now, I don't know about you people but an increase in seek performance that is easily 2.4 times better that the best IDE for something that occurs that many times a second, well, I'm convinced that the drive is considerably faster. Think of any "worst case" scenarios, and this extra speed is likely never to go un-wasted on a computer.

I am also in accord with alsta's skepticism that SATA will be anything but a consumer grade technology. I am an ardent supporter of Firewire and SCSI, and it clear the SCSI standard has much to offer because implementations of SCSI over IP are planned to deprecate FC in favor of iSCSI over 10GE. To me this SATA is a day late and performance short. I think it is nice we can now clean up the crap connectors, but I will not be deluded into believing that SATA has a prayer in terms of beating SCSI.

I also have seen many reviews on IDE RAID and must concur that it is nearly impossible to IDE up to single drive SCSI performance without doing some hardcore RAID-ing. I cannot provide links to support this statement, but peruse through a site like Storage Review and you can see clearly where and why and how SCSI drives cremate their IDE counterparts.

I have IDE and would buy SATA for mass cheap storage at home, but it was a sad day when PC vendors stopped offerering SCSI upgrades from the factory on their cheesier machines and Apple stopped using them by default. This essentially was the advent of the "Geo" or "Kia" of the computer world. I hate hackneyed car analogies, but in this case, as I have seen in acerbic reviews of the said cars, "Why buy this cars where there are better used ones around?", directly applies. I can think of many "used" scsi drives and how badly they destroy consumer grade stuff with ease. This is both from L&F and measure performance metrics.

IDE does not support tagged commands, and it is still unclear if SATA's implementation will be as robust as SCSI. This alsta's statement is again correct with regards to terrible simultaneous access. Try running a busy Unix box with 100s of people copying files and doing this or that, compiling crap, etc. The disk needs to be serious about handling jumpy situations. Seek time and command tagging comes in handy here.

Unfortunately larger buffers can only mask problems. There are many situations where consistency is not guaranteed, and where the cache can easily be blown. I like buffers to be there to make an already fast situations move along more regularly, like a capacitor helps regulate power. I don't like an already bad situations being masked by case specific buffering optimizations.

I will ,as alsta indicates he will, continue to use SCSI as my most preferred storage bus. I cannot think of reasons not to, if I have an dumping ground for "crap" in my IDE large capacity hard drives. Also suspect is the standard being ratified over a year ago and nothing materializing yet. Also, I think the SMART works better in SCSI, I think sector sparing is essential and IDE and apparently SATA wont do that, and I like 5 year drive warranties not ONE like most IDE drives moved to now, and I liked support, and I like knowing that the company I bought the device from made an okay profit so I can get amenities like advanced replacement. It's a preferred choice, and if you can afford it, so it.

Just because you may not be able to afford it or don't think storage is not the way you should spend your hard earned money, don't knock SCSI. And certainly don't come to Slashdot and suppress other people's views on the subject.

Re:This just looks expensive.

[Ed Avis]

In the future couldn't there be two SATA connectors on a single device? Not running in lockstep like a parallel interface, but two separate connectors with data being sent along both. That sounds like an easy way to increase bandwidth without the syncronization problems of having parallel wires at the hardware level. But I don't know if the SATA standard supports this.

Re:This just looks expensive.

[Ed Avis]

SCSI has much better seek times than IDE disks

No, the interface makes no difference to the seek time. What you mean is that fast, high-end disks are typically sold with a SCSI interface, while slower models get IDE.

Re:This just looks expensive.

[photon317]

Of course you still have synchronization problems. Over an appreciable length, each fiber will be slightly different internally (and probably in path shape as well), which leads to the actual photons paths being significantly shorter or longer than each other.

Re:This just looks expensive.

[photon317]

Well, the DS3 example is of course crazy or people would be doing it. Over long wide area links serial tends to win, just like parallel wins on expansion card busses. However clock skew can be mitigated with buffering. However by eating the potential buffer time you increase the latency, which makes the tradeoff imperfect. YOu might gain 8x the bandwidth but double your latency or more. It would come down to actually measuring how things work out in a particular technology you're building to decide whether the amount of gained latency versus gained bandwidth paid off.

Re:This just looks expensive.

[Zathrus]

ATA wil debut at 150 mb/s, not really an improvment at all.

Does it matter? At all? No.

Frankly it could be 150 GB/s and it wouldn't matter in the least.

Go look at the manufacturer specs. Read the line that says "drive to host, sustained throughput". Note that no manufacturer claims more than 52 MB/s. Reality is closer to 48 MB/s for the fastest IDE drive. That's right! We're not even exceeding ATA/66 bandwidth yet. And still people are talking about 600 MB/s in a few years. Who cares? You can't reach that throughput anyway. Not to mention that the PCI bus is limited to 133 MB/s.

Ok, the bus speed does make a bit of difference. If the data you need is in cache then you can use the maximum theoretical bandwidth while reading from cache. So dumping a 8 MB cache via ATA-133 saves you about half a millisecond over ATA-66. You noticed that, right?

The advantages of SATA have nothing to do with the bus speed. The longer cable is useful in a select few tower cases. The hot swap will be nice for a small percentage of enthusiasts and idiot admins (I'm not a SCSI fan boy, but if you're running a server you really should be running SCSI). The small, thin cable is useful for everyone though -- the air traps created by ribbon cables are causing more and more problems as everything runs hotter. Most drives fail due to poor cooling. Want to bet that SATA drives have a significantly lower failure rate?

Re:This just looks expensive.

[mosch]

clock skew, a problem which we get better and better at handling, and which isn't particulary difficult to solve if the application allows for the addition of a small amount of latancy.

To use the DS3 example that somebody mentioned previously, it's entirely possible to create hardware on each side of the link that would send simultaneously over each link, and on the other side the bits would be reassembled, with some small reassembly delay.

Using this method you could greatly outstrip the bandwidth available from the current top of the line serial link, at the cost of a small amount of latency. And yes, I understand that 8 lines wouldn't get you 8 times the speed, because there'd need to be protocol overhead.

That being said, this is slashdot, so I tend to avoid real debate. Very few people have the capability to thoughtfully discuss technical topics, and those that do are often unreasonable, unable to recognize humor and generally unwilling to engage in anything interesting or useful. Thus, it's far easier to just use a quip and a quote to imply that your statement is clearly wrong than it is to explain why you're wrong.

Re:This just looks expensive.

[alsta]

Nonwithstanding, the comparison is the fastest disks from both camps. Thus it is valid.

Re:This just looks expensive.

[alsta]

I read the review and it speaks loads about how fast IDE is versus SCSI and how much cheaper it is and blah blah blah. But NOT ONE WORD about what made it faster or even how the test was performed to make the ruling.

Hence it is hearsay at best and a pointless comparison. Did you know that a Yugo is better than a Porsche? Because you get a lot farther on a tank of gas and it costs considerably less.

See? Now I made a comparison. All I have to do now is say that the Yugo is faster and puts the Porsche to shame. Now I made a ruling. What do I base this on? If I don't tell anybody how I came to that conclusion, it is a rather pointless thing to say, no?

Re:This just looks expensive.

[Wakko Warner]

Clearly, you're jealous.

- A.P.

Personally, I can't wait until it comes out...

[HoneyBunchesOfGoats]

...'cause that means prices will drop on hard disks that I can use.

even cheaper

[lseltzer]

The whole point of it is to be at least as cheap as parallel ATA, even cheaper. The connectors will be smaller and cheaper for example. It should also make system design more flexible since you won't have parallel ATA's infuriating cable length limits.

Re:even cheaper

[rodgerd]

Indeed - it'll be a blessing for system builders.

Give me Firewire!

[xixax]

So what does SerialATA offer that a Firewire connector on my hard-drive won't?

And haven't we discussed this before?

Xix.

Re:Give me Firewire!

[cheezedawg]

So what does SerialATA offer that a Firewire connector on my hard-drive won't?

150 MB/sec?

Re:Give me Firewire!

[Com2Kid]

• So what does SerialATA offer that a Firewire connector on my hard-drive won't?

Bingo, it is exactly what Serial ATA does not have over firewire that makes it so desirable.

Namely the $50 or so price premium. . . .

too bad that

[LuckyJ]

All data goes through the PCI bus...and it's bandwidth is only 133MB/sec theoretical. So, what does 400/800/anything else greater than about 100MB/sec in a media interface get you? Not much!

Ever read the actual throughput specs on a drive? The media throughput is not much more than 40MB/sec!!! Read the data sheets, people!

Add this all up and what do you get? Ripped off is about it!

Re:too bad that

[Enry]

Haven't seen 64bit PCI busses running at 66 or 133 Mhz, have you?

Re:too bad that

[Magila]

1. The "PCI problem" is going to get fixed next year when the next gen busses (Hypertansport, PCI-X, etc.) start appearing on motherboards.

2. That 40MB/s is sustained throughput, HDs can burst from cache much faster than that. This is going to become more of an issue as HD manufactures have finally started putting more cache on their drives.

Re:too bad that

[cheezedawg]

All data goes through the PCI bus

No it doesnt. Data goes through the PCI bus if the address is not claimed by something else along the way. That means that everything from the southbridge up is not limited by the PCI bus bandwidth. That means that integrated SATA controllers (not available until next year) are only limited by the bandwidth between the northbridge and southbridge.

Ever read the actual throughput specs on a drive?

Drive throughput has been steadily increasing, and it is predicted to pass up PATA within a few years, and that is not counting RAID striping or the 8 MB drive caches. Its always desirable for the bottleneck to be the drive rather than the controller.

Re:too bad that

[Magila]

4 of them in RAID 0?

This won't be an issue since SATA is strictly point-to-point, every drive gets it's own 150MB/s link.

Re:too bad that

[ltwally]

For starters, the PCI spec isn't limited to 133 MB/s. PCI 2.1 specs allow for 66 MHz 64-bit transfers, which equals 528 MB/s. PCI 2.2 specs allow for PCI-X mode, which adds 133 MHz 64-bit transfers: 1056 MB/s.

That being said, it is entirely possible to reach throughputs in excess of 133 MB/s using a PCI bus... though currently most desktop motherboards do not support anything faster than 133 MB/s. In time this will change as NICs, hard disks, and other gear requires it.

And your hard disk performance is barely par by today's standards. IDE drives are currently topping 50MB/s, while SCSI gear is hittin > 70MB/s. Though I am a SCSI man, i can see the future need for SATA. Right now it may be mainly a marketing ploy... But in a couple years it will be a necessity. Parallel cabling is nearing the end of the road.. all those wires in a cable allow for too much signal interference. Serial is the answer. Though it has less wires, the dramatic increase in signal strength allows for insane transfer rates.

Anyhoo.. personally I don't see any reason to go out and buy a new system just to have SATA. At the current it offers few advantages.. but in the not so distant future it will be a necessity for desktop systems. As for me, i plan on going Fibre-Channel SCSI :)

Re:too bad that

[Toraz Chryx]

" All data goes through the PCI bus...and it's bandwidth is only 133MB/sec theoretical."

Except of course on boards that have dedicated interlinks between the northbridge and southbridge for IDE/legacy/integrated sound etc.. (VIA KT266 onwards, SiS's recent chipsets, Intel i810 onwards, nForce etc.)

Re:too bad that

[rodgerd]

The media throughput is, yes. At the moment (I remember when my old 4 MB/s Seagate Barracuda were considered quite fast; that was only oin the mid-90s), which will doubtless change as densities and rotational speed increases.

But a large chunk of the performance in a moden drive isn't the platter to interface performance, it's the cache to interface. Adding RAM to a drive is a relatively inexpensive way to improve performance.

More Information

[leibnizme]

Since the site is slashdotted, here are further links about Serial ATA:

Cnet

SATA and ISCSI

Intel Dev Paper

Maxtor White Paper

Re:More Information

[Salamander]

I really love how the Maxtor paper compares SATA to parallel ATA, USB (?!) and Firewire...but not to SCSI or FC. I wonder why that is. Actually, no I don't. ;-)

Not important yet

[thetzar]

I can't wait to get my hands on some SATA devices. However, we're still stuck with PCI, here on the desktop end. WHEN will we finally start seeing the old original PCI spec phased out on the desktop end? Not until then will new technologies like SATA be able to shine. Bus bandwidth is everything these days.

Re:Not important yet

[Cyno01]

hopefully soon we'll be seeing AGP8 mobos w/ multiple agp slots and then finally agp cards, then SATA will get a foothold

RDRAM Redux

[grendelkhan]

This sounds remarkably like the plugs we got for Rambus RDRAM: serial interface is better than parallel, first gen won't see real performance gains, stick with us kids, this is gonna be really good.

I see a decided lack of Sun, IBM, AMD, or HP listed in the adopters, which leads me to believe that this is much like the above. Sorry guys, I'm not riding the first wave of any new tech on my salary. I'll sit on the sidelines for awhile and see how this pans out.

Re:RDRAM Redux

[Magila]

Excpect SerialATA is actualy cheaper to implement than ParallelATA and there's no latency issues like those that killed RDRAM, SATA is unquestionably faster than ATA100. SATA will be taking over in place of parallel, it's only a matter of time.

Re:RDRAM Redux

[MtViewGuy]

This is NOT RDRAM redux.

For one thing, both the Intel and AMD camps are heavily supporting Serial ATA, with the first motherboards that have Serial ATA connectors starting to appear on the market as I type this. Not only is the throughput going to be way higher than standard dual-conductor IDE cable connections for ATA-33/66/100/133, but because Serial ATA cable is a small rounded cable this means there are no wide flat ribbon cable connections; this makes for a much neater-looking system case interior and also means better system case cooling as a side benefit.

I expect the first CD-R/RW and DVD-R/RW drives with Serial ATA connectors to hit the market any time now.

Re:Parallel vs Serial?

[shepd]

>Forgive me if I sound a bit naive but wouldn't parallel be faster than serial?

Yes, but just like with memory, serial is cheap, parallel costs. Those extra wires just ain't free.

SCSI TROLLS: READ THIS

[Vengie]

For all of you waxing locquatious on the merits of scsi....please read the article, noticeably:

SCSI drives are notorious for their noise, heat and vibration levels. These low points are not acceptable at the consumer level. Noise and heat don't matter too much in huge server rooms, but they can quickly become a problem in desktops. Unfortunately, this means that the extremely low seek times and high spindle speeds that make SCSI drives so fast are not available in consumer market. Basically until cooler, quieter, smoother drives can be manufactured in high volume, desktop hard drives will continue to make baby steps in speed.

For the time being, IDE isn't going anywhere.
NOISE & HEAT will tend to outweigh (relatively) minor performance gains in consumer systems. (Enterprise hardware is another matter entirely)
sigh....we need to start using those annoying javascripts that make people read the article BEFORE posting.

Re:SCSI TROLLS: READ THIS

[Drogo Knotwise]

Not to mention the #1 stumbling block for most people: price. SCSI costs significantly more for comparatively little actual performance benefit. On a server, which is hard-drive intensive, this performance difference is big enough to make SCSI worthwhile. On the desktop, on the other hand, it costs more, allows smaller sizes, and apparently makes more noise and heat. No wonder IDE is here to stay...

Re:SCSI TROLLS: READ THIS

[mosch]

That has nothing to do with SCSI. The only reason it's even remotely related is because people have a tendancy to buy SCSI drives like this one, which is a 15K RPM drive with a 3.2ms seek, and the ability to sustain a 75MB/sec data throughput rate, whereas high-end ATA drives like this one spin at 7200 RPM, have a 9ms seek, and don't list their maximum sustained data throughput rate on the data sheet.

Is IDE appropriate for the desktop? absolutely.

Will retards continue using IDE in applications where SCSI is far more appropriate? definitely.

Does your post make any fucking sense at all? nope.

Re:SCSI TROLLS: READ THIS

[Ed Avis]

I'm surprised disks aren't sold with both interfaces, SCSI and IDE (perhaps using a small external dongle to make the plugs fit). It sounds like it would be cheaper to have a single production line for both.

command queueing

[Tester]

I think I read in somewhere that SATA could do command queueing. Does that mean that it will allow the driver to re-order them like SCSI drives? That was, I think, one of the main advantage of SCSI over IDE/ATA drives.. That they could re-order the commands and send answers in a different order to maximize performance.

Re:command queueing

[cheezedawg]

Yes- but AFAIK command queuing is not implemented in a lot of the 1st generation controllers because it can break backwards compatibility with PATA software. Most vendors went for an easy upgrade path instead. Look for command queuing in the next generation of controllers.

Re:command queueing

[bjschrock]

Most of the command queueing and multi-threaded operation in SCSI devices is implemented in the device. I think the protocol can handle re-ordering, but it's up to the device to figure out what order it wants it's data in... It is a bit interesting to send some write commands to a hard drive and then have it come back and ask you for the data in a different order that it decided was best.

whats new?

[nonane]

narrower flexible cables is all the SATA has to offer? its like switching from a two-headed screw driver standard to a philips one. the heads are different but the screw gose in the same way. give me a few features of SCSI in an affordable package and then your talking.

on the other hand: installing more ram/new-cpu wouldn't be such a pain .. 2 hard disks, a dvd and cd writer later your computer would have enough ribbons to host a toddlers birthday party. and to get rid of the freaking master/slave shite ... ahhh. damn it I want SATA!

Re:whats new?

[Cyno01]

round cables do help, when u've got 6 drives in a case (dvd, cd-r/rw, floppy, zip and 2 HDs, it can get pretty crowded, round cables do help airflow quite a bit, and every few degrees helps

Re:whats new?

[Cyno01]

round floppy cables

I want the connector!

[aronc]

Personally, I could give a rats butt about the speed. I don't want SATA so my drives go faster. I want it so I'm not having to spend twenty minutes doing finger gymnastics everytime I need to do _anything_ in my case.

Is it worth upgrading for? No, probably not. But id damn sure is worth waiting an extra few months for that next machine to save the hassel of those f'ing ribbon cables.

Re:I want the connector!

[HoneyBunchesOfGoats]

From the Cnet article: Parallel ATA ribbon cable is about two inches wide and can be only 18 inches long. The Serial ATA cables can be made up to three feet long, allowing for more elaborate routing, which would aid in creating cooler-running PCs. This just isn't true. For those who want narrower, longer cables, there are solutions available. I am currently using 36 inch long ATA/133 rounded cables, which I bought from Cable and Connector Technologies for $5 each. Yes, rounded cables to hurt the performace a *tiny* bit, but what drives run at 100% of their peak capable performance anyway? It is more than worth the extra length/reduced hassle that the rounded cables provide.

Re:I want the connector!

[aronc]

Can i say.... Waah? I've been working on these things for 9 years and it sounds to me like you're just whining.

I've been working on them for 12.. but anyway. It's hardly 'whining'. Those cables are a true pain and if I can get rid of them at no extra cost, why not? It's not like I'm blindly bitching about the current standard just to hear myself complain. Just voicing a positive (the best one, imho) of a new standard that's about to become available.

Re:I want the connector!

[aronc]

I am currently using 36 inch long ATA/133 rounded cables, which I bought from Cable and Connector Technologies [callcct.com] for $5 each. Yes, rounded cables to hurt the performace a *tiny* bit, but what drives run at 100% of their peak capable performance anyway? It is more than worth the extra length/reduced hassle that the rounded cables provide.

Having used rounded cables before (not currently, but before) and having messed with the SATA connectors on a product demo I can say that SATA beats them hands down, no contest. Did we really need a new standard for this? Nope. But they did it anyway. Would I pay extra for it? Maybe a few bucks.. say like the 5 or so the rounded cables would cost me.

ridiculous

[tps12]

Why are they wasting their time on this ancient technology? Serial is too slow to sync my damn Palm Pilot. I can't even imagine what it would be like to try and transfer 60GB of media files over it. These companies should just accept that USB is the way of the future (no extra power required either!) and get to work on something that stands a chance of selling.

This is a GOOD thing. FINALLY

[krray]

I've been waiting for Serial-ATA ever since reading about it. Faster speeds/bandwidth - which is actually finally needed in the IDE type world.

NO MORE RIBBON CABLE. My favorite Linux configuration is 1 whatever IDE drive for the OS, 1 IDE CDROM, and two (RAID-1) large IDE's for data and configurations. Quick and cheap for non-critical type functions/services. I rolled through a complete failure on the core OS drive, CD died -- while trying to roll up in size on the RAID-1 and hit *FOUR* defective WD drives...while never losing data _and_ configurations. IBM sits in there right now... :)

High end servers and workstations? Yeah, Serial-ATA is nice with the coming 40M/sec IDE type drives...but I'm also going to go after that 320M/sec SCSI technology too. Same IDE game, just a different connector basically.

NO MORE RIBBON CABLE.

Try stuffing four drives in a case. Not only is the IDE chain full, but cabling is a complete joke. Not anymore. Kind of like Firewire in the box, if you will. Except I think their screwing it up and keeping power separate where Firewire _can_ cary power to the devices.

So instead of tiny IDE connectors in the current Firewire and external type drives there will be tiny Serial-ATA hookups. So what. Now get inside a PC (and/or Mac) and do a little work.

With this and pricing for LARGE amounts of data ... I could see easily wanting a tall tower (remember those?) and building a rock and roll back end storage system for personal use. Quick and cheap ... and now VERY EASY to do. Personal RAID-50 500G personal array anybody?

I could record so many hours of anything I wanted and never worry about losing it ... even with el'crap-ola IDE no-warranty technology.

Of course when I have a few extra thousand lying around (not likely any time soon with the current economy outlook) I'd love to try SCSI-320.

Now, IDE is rolling into ~40M/sec. Firewire *has* been ready for those speeds for a while. At least USB2 can keep up for a bit as well. Even faster drives is a must though. Firewire-2 is just around the corner (either 800 or 1.6Gbit's).

It's sad that your typical/standard Mac type network (1Gbit) is faster than the typical drive being hosted. Your typical Windows network at 100Mbit is pretty muched caqpped by the current typical drives top performance at 10M/sec.

Serial-ATA, oh yeah. One card (1Gbit) in the Linux box and I could saturate their bandwidth. Why not?

Microwho?

Re:This is a GOOD thing. FINALLY

[NineNine]

This post is an *excellent* example of what watching too much super-fast cutting TV (like MTV) can do to you.

Huh?

[WD]

What on earth does the type of interface have to do with heat or noise?

SATA Linux Support

[yamcha666]

I noticed this little remark:

... although it wont officially be supported until Microsoft's Longhorn OS, Serial ATA devices will work with all versions of Windows.

And this made me wonder... how long will it take until Linux (and the *BSDs) support this new standard? Will it happen after Longhorn's release? Or has it already been done?

Re:SATA Linux Support

[jensend]

Serial ATA support is in Linux 2.5.35 and up, as noted here.

Bandwidth

[vlad_petric]

Serial is the wave of the future, and it will give ATA hard drives the bandwidth and features to compete with SCSI.

I don't get it ... I quite agree that, as a serial bus, it'll be clocked a lot faster than IDE ... but a simple back-of-the-envelope calculation tells us that it has to be at least 8 times as fast as the current devices (it'd have to be 533 MHz to be on par with ATA-66)

It looks like a technology whose main purpose is to make things incompatible, and thus require people upgrade more stuff. And anyway, it's not the speed of the bus the limiting factor (for the vast majority of users), but the mechanics of the harddrive (SCSI hardrives are faster than IDEs because they almost always are top-of-the-line products with higher rotational speeds - anybody saw a 15000 RPM IDE ?)

The Raven

Re:Bandwidth

[TheGratefulNet]

scsi drives share the same rotational stuff as ide cousins. just translate different host language to the disk itself.

I'm a big believer in scsi, don't get me wrong. but its only the controller that's different (the one on the drive itself).

there's no technical reason for why ide can't be 15krpm. its pure marketing.

and besides, thinner cables that can run longer are MUCH nicer than fat parallel ones (like scsi and current ata).

Re:Bandwidth

[rweir]

a simple back-of-the-envelope calculation tells us that it has to be at least 8 times as fast as the current devices

You needed an envelope to calculate 8x=8x? Geez, schools these days. Back in my day...:)

Re:i dont see a huge rush for people to upgrade

[Anonymous Coward]

Psst.

Don't look now, but you're momma dont drive the hi-tech industry.

I do.

Re:More "standards"

[Wesley Felter]

Serial ATA connectors are keyed and designed for hot-swap.

IDE TROLLS: READ THIS

[darkwiz]

SCSI drives are notorious for their noise, heat and vibration levels. These low points are not acceptable at the consumer level. Noise and heat don't matter too much in huge server rooms, but they can quickly become a problem in desktops.

For the time being, IDE isn't going anywhere.
NOISE & HEAT will tend to outweigh (relatively) minor performance gains in consumer systems. (Enterprise hardware is another matter entirely)
sigh....we need to start using those annoying javascripts that make people read the article BEFORE posting.

Absolutely. But this has nothing to do with SCSI, it has to do with the high spindle speeds at the bleeding edge. The card on the underside of the drive is not making that ear shattering racket. They even acknowledge that in your quote.

SCSI is better than ATA. Even SATA. ATA has been trying to catch up by stealing some of the best parts of SCSI (like TCQ). But it just isn't quite as good yet. Quite frankly, I agree with the majority of SCSI zealots: if the damn PC makers would embrace SCSI, then the cost of SCSI would come down to near parity from the volume of sales.

Now, is SCSI better for your average Joe? Maybe not significantly. Neither is 7200 vs 5400, 2MB vs 8MB buffers, or 8.9 vs 9.1 ms access times.

However, if they could use one cable to connect 15 devices in their tower, they'd be alot happier than having the 8 cables they'd need to do it with current IDE tech (let alone IDE's relative inability to be used externally).

This is the one to watch

[Gerry Gleason]

It looks like it is designed correctly from a physical standpoint. Hot swap should and blind mating should be part of the program. I looked at it in some detail a while back. They even planned for progressive mating of the ground first. All the big drive manufacturers seem to be behind it.

The only thing I haven't seen is any noise about chip sets that support in on the system side. As soon as these are available, you'll see MBs and systems. SCSI will probably stay important for larger faster arrays, but scaling bandwidth seems to look pretty good for this as well.

As soon as mainstream MBs are there, these will quickly become the commodity drives for all the manufacturers, and they will phase out Parallel ATA stuff.

Obligatory crazy conspiracy theory

[spun]

They are implementing this new standard rather than utilizing firewire so They can create a whole new generation of "trusted devices" and make us all buy them.

Re:More "standards"

[cscx]

The red stripe points towards the power supply connector, or marked pin 1 on the mobo. That's SOOO difficult. 8-)

Firewire Drive != Pure Firewire

[nuxx]

Note that almost all the Firewire hard drives that you see on the market are ATA hard drives with FirewireATA hardware in the enclosures. As far as I'm aware, the only disks that you can readily get your hands on will have interfaces of IDE/ATA, SATA, SCSI (of various connectors), and FC-AL. That's why you can't use Firewire inside a PC. Using SATA makes far more sense, especially for migrating to a new standard, as it's most likely easier to make a SATAATA adapter since the protocol is very similar.

Logic?

[wyrmBait]

> (less wires = less logic).

If you've ever looked at the circuit diagram for a simple multiplexer, you'd take that back. Really, how complex the is logic depends on what you want to do with the data off of those wires. To really simplify the idea: if you use serial, you've got to have a muxer/demuxer on the end of that line, and if you use parallel you need to have a clock to syncronise communications on those lines.

In any case, the difference in circuit complexity due to parallel vs. serial, even if one required less than the other, would be a few orders of magnitude less than the complexity of the circuits required to manage the disk itself.

SATA == Future

[nuxx]

Note that Sun, IBM, AMD, nor HP are disk manufacturers. (Well, IBM might still be, my memory is being bad tonight, but I digress...) Some AMD and Intel motherboards are already coming with SATA RAID interfaces. Intel is right behind the technology, as they are a chipset manufacturer. AMD isn't. VIA, a chipset manufacturer is, along with a ton of other manufacturers who are core to desktop/workstation storage. Just because the big power houses that you name aren't on board doesn't mean anything. Most of these places leave their disk interfaces up to someone else. And those companies *are* adopters.

Re:SATA == Future

[nuxx]

Doh. Sorry. My bad. When I think of AMD boards I generally think of VIA (or SiS, ick) stuff. I guess I'm forgetting the wonderful AMD PCnet32 chip that's powered so many different workstations and servers without a hitch. I shouldn't forget all their other semiconductors, either...

Re:Parallel vs Serial?

[CMiYC]

Forgive me if I sound a bit naive but wouldn't parallel be faster than serial?

On paper, parellel can be made to be faster than serial. However, in the practical world it is very difficult to make a high bandwidth parallel bus. It is even more difficult to run that bus any considerable amount of length. By using a serial bus with embedded clock you only need 2 signal wires. If those signal wires are a differential pair (perferrable Low-Voltage) then you can run them a considerable length at an extremely fast rate. If you have a parallel bus, you don't have the option of embedding the clock. If you are running with out an embedded clock you must send a clock syhncronous with the data. Now you have to deal with skew issues between each individual channel as well as all channels relative to the clock. Not to mention other aspects such as crosstalk between the channels. If all you have to worry about is two signals (which if are differential can be considered 1) then many of those issues ago away.

There's a lot of physical behind why its extremely difficult to run long lengths of parallel lines. Yes a parallel bus is faster, but it is almost impossible to implement a reliable parellel bus running at 1.5Gb/s, through cables, and with connectors. Take a look at a bus like hypertransport. It can be up to 16bit wide and run 1.6Gb/s. However, it is a point to point protocol, is run over a control impendance, and is run over very short distances.

I hope that helps.

Am I missing something here???

[Lobsang]

I think I am. Really.

The article seems immensely biased and lacking in technical detail. It also raises some "dubious" points IMHO. Let's see:

- P-ATA cables cannot be longer than 40cm. S-ATA cables can be up to 1m long:

Granted, those cables are annoying. But really, how many times have you felt the need for a cable much longer than 40cm? People with full-sized cases may benefit, but then the author says that the current trend is "small footprint machines". So, why do I need a cable that is bigger than my server?

Also, if you dislike flat cables, buy "rounded" P-ATA cables (available today, just google for it).

- P-ATA connectors are big!

Yes, they are! But you'll require at least twice as many S-ATA connectors, as only one device is supported... In the end, the real state on the mobo is going to be similar.

- One device per controller is an "Advantage".

C'mon... This guy must be joking. I couldn't believe my eyes when I read it! One device per controller is an *advantage*???? Why??? I wish I could add more devices (like SCSI and Firewire) to my curreny P-ATA technology. And then he says ONE is good for me? Don't think so...

- High transfer rates are useful for multi-disk RAIDS.

What kind of RAID? RAID 5 is slow in writes due to the computational power needed to calculate the XOR. Adding bandwidth won't help. And I can't see why or how only RAIDs will benefit from higher throughput.

- Speed:

Granted. It may be faster than P-ATA. But what about established technologies like SCSI and Firewire? I *think* (not sure) Firewire can go much faster than S-ATA in its initial version.

I'm disappointed...

Re:Am I missing something here???

[Kevin Stevens]

Longer, thinner cables are a huge advantage. As processors start catching up to space heaters in the amount of heat they throw off, airflow is ever more critical in a case. Ribbon cables can really screw up the flow in a case. In the case of meter long cables, they could even be ran along the side of the case. And do you honestly like ribbon cables? serial ata will have a keyed connector... no more guessing as to which way it should plug in. Similarly, one device per header should reduce aggravation between master and slave drives. Have you seen the size of the new headers? They are smaller than half the size of the ribbon connectors, and also, since there are fewer data lines, there is alot less space needed on the mobo for connections. And yeah the speed will be faster. Serial ATA was never intended to be a major milestone in computing like the article hypes it up to be. However, it IS an improvement, and a step in the right direction. I personally can not wait until I build a system without those dreaded annoying ribbon cables. Its nice, its better, it IS backward compatible (they will sell adapters to convert the two connections). You are not losing anything here, but you are gaining some nice advantages.

Why not firewire?

[splorf]

Unlike some here, I'm not fond of SCSI (obnoxious cabling and termination issues even worse than ATA). But Firewire seems to have every advantage claimed for SATA and then some. Why not just put firewire on motherboards and in disk drives? Then we can finally ditch ATA in all its incarnations.

Be a SCSI fanboy too

[Lobsang]

SCSI is full of annoyances. Price, incompatibility between controllers, etc.

However, if you're using linux, try this on both IDE and SCSI:

time dd if=/dev/your_disk_device_here of=/dev/null bs=1k count=100000

Then compare the CPU used for IDE and SCSI.

You might too become a SCSI fanboy. :)

Re:Be a SCSI fanboy too

[Lobsang]

SCSI is definitely somewhat faster at some things, but having to spend 3x - 5x as money to get there, and have less space to boot, is means it simply ain't worth it for 99.9% of people.

Yes, yes! We agree 100% on that. I don't have SCSI on my home desktop. But I wouldn't have a Data Warehouse server running on IDE. To each tool, its use...

A never-ending game of leapfrog

[Salamander]

Of course, no discussion of Serial ATA would be complete without mentioning the answer from the SCSI camp - Serial Attached SCSI. SAS will use the same connector as SATA, but will support longer cable lengths, multiple initiators (if you don't know what an initiator is you don't even belong in this discussion), full SCSI semantics instead of lame-o ATA semantics, etc. Even so, the SAS folks are still ceding the high end to Fibre Channel and talking about three coexisting technologies for the low-end/midrange/enterprise market segments. Sorry, kiddies, but SATA is still low-end.

If there's one mistake you should try not to make more than once in this business, it's that competitors have been standing still since their previous generation. Announcing something brand new and having it be less than half a generation ahead of the competitor's last version is a failure.

argueing with everyone

[squarefish]

ok, SCSI is cool and fast- however it is not marketed to general pulic. *** It's a higher standard who's taget is the server market where it can draw a more technical user base willing to pay the higher prices. I *** IDE is very limited, but also easier to use and way cheaper. *** Serial ATA is an itelligent combination of both that targets the way even new users would like to equip their machines to do more and be more flexible. It will be marketed to the general public and at a much cheaper price because of the collaboration of so many different companies that will mutally benefit from it's popularity. *** In the end it will be better for everyone (the companies and the users) *** It will also help drive a hurting industry if it ever actually becomes available!

Re:argueing with everyone

[GreyWolf3000]

SCSI is not an interface limited to the server by any physical means--merely marketing means. SCSI drives are built almost identically to IDE hard drives. They are usually faster, but you could make a SCSI hard drive of any speed/noise level you damn well pleased. SCSI is a superior interface to SATA. Period.

Re:argueing with everyone

[AlecC]

> SCSI drives are built almost identically to IDE hard drives.

Not according to the presentations I have had from the manufacturers. And, puttint their money where their mouth is, compare warranties (5 yrs vs 1 year), seek times (4 msc vs 9). They put better bearings in - one mfr claimes that bearings in their scsi drives cost him twice what the bearings in teh IDE drive cost.

Re:argueing with everyone

[Zathrus]

Compare apples to apples.

Take a 7200 RPM IDE drive. Take a 7200 RPM SCSI drive (ok, these are rather hard to find now). Rip off the electronics boards.

Now try to find a difference. You won't find one.

The physical drives are identical, at least as long as they have the same performance criteria.

Seek times? Bearing costs? You're comparing a 7200 RPM IDE drive to a 15,000 RPM SCSI drive. Amazing how the seek time is lower and the bearing costs are higher. It's simple math.

Re:what the fuck?

[Cyno01]

i wish i could find a USB 5.25" drive, i dunno why, but that'd be cool

Re:I want a

[Hadlock]

i'll second that. until then, windows XP has TCP/IP drivers for firewire, which is 400mbps, and just around the corner, 800mbps. firewire can also go longer distances than SATA

SATA propagates all the crap of PATA

[mnemotronic]

Serial ATA has kept all the truely horrible, obscure, and performance limiting features of parallel ATA (which was originally designed for 40 Mbyte drives), and shoves it down a thin cable. A course in ATA protocols will leave anyone who is computer literate going : You have got to be kidding!! What were they smoking????? The Serial spec has to maintain all kinds of bizarre, obsolete behavior just so that a SATA drive is backwards compatible with old software (i.e. that WfWG 3.11 system you've got tucked away in the closet). Throughout the history of ATA, when drive capacities climb towards the addressable limit of the spec, the protocol is kludged by a team of drunken baboons to extend it for another generation. The SATA committee munched an opportunity to drill a much-needed stake through the heart of ATA and give us a new interface for the next 20 years. Instead we get a change in connectors for a protocol with it's feet firmly encased in the concrete of MS-DOS 2.

Firewire (1394) was killed by Apple's licensing fees and Intel's sudden backstabbing policy change on building it into south-bridge, along with their NIH attitude. There existed working 1394 Device Bay drives over 6 years ago, with OS support from m-soft. 1394 was an attempt to keep the good parts of SCSI protocol, while leaving out as much of the useless stuff as possible (MODE SELECT).

Fibre-channel is still Real Pricey, for the same reason that SCSI is -- "just because". Or, as the hardware vendors say "harrumph, well, it's all about volume".

Re:SATA propagates all the crap of PATA

[Zathrus]

Throughout the history of ATA, when drive capacities climb towards the addressable limit of the spec, the protocol is kludged by a team of drunken baboons to extend it for another generation

Yes. We should run out of space in the latest incarnation in roughly 50 or 60 years.

Unless, of course, you're expecting to implement a single drive with more than 144,115,188,075,855,872 bytes (that's 128 petabytes or 131072 terabytes) anytime soon.

Yes, previous extentions have been poor. Maxtor got it right for ATAPI-6 which has been adopted by the industry. 48-bit addressing of 512 byte sectors.

Re:SATA propagates all the crap of PATA

[ez76]

Maxtor got it right for ATAPI-6 which has been adopted by the industry

Disingenuous.

Re:what the fuck?

[dan the person]

IDE also becomes outdated very quickly.

Try plugging your 7200RPM 120Gb IDE drive into a 386 era IDE controller and see what sort of performance you get. You'll probably only be able to access 8Gb of its capacity also.

IDE hasn't "just been there" it has been constantly evolving.

Real benefits ?

[IanBevan]

This article is very pro the S-ATA standard. However later on in the article it states that S-ATA hard disks will be of the same speed and size as current ATA. The article says that the only way to get real performance increases in disks is to make them faster - so then we get to the same problems as SCSI, namely noise and heat.
As far as I can see, a couple of pages of this article are denoted to the new smaller cable size and connector footprint - who really cares ? I run an overclocked PC and as such use readily-available rounded IDE cables to afford better air flow. What other tangible advantages is S-ATA going to offer ?

Actually, no.

[Rui del-Negro]

First, this is IDE (just as ATA-66 or ATA-133 are IDE).

Second, the reason why Betamax died (well, didn't actually die, but didn't take off, either) was Sony kept it a proprietary format, while JVC let pretty much everyone make VHS products.

Serial ATA is one of the most unrevolutionary evolutions ever made. Basically it just changes the cables. The drives stay pretty much the same, the controllers stay pretty much the same, the drivers can stay exactly the same. Instead of wide, flat cables and two disks per channel you now get thin round cables and just one disk per channel (but since the connectors are so much smaller, you can have many on the same board). It's a good thing.

There are basically three reasons for having multiple standards. The first is a purely commercial one. Brand A invents the A-link and patents it, and brand B decides to create B-link so they don't have to pay a fee to Brand A. The second is evolution. Sometimes, a standard needs to be replaced or updated to cope with new demands (ex., ATA-33 becomes ATA-66). The third is that some standards are specifically suited to some situations (ex., SCSI lets you connect a lot of drives, and has support for other kinds of peripherals, but IDE is cheaper to make, and enough for most people).

RMN
~~~

Re:Actually, no.

[Rui del-Negro]

The main difference from, say, (p)ATA-133 to SATA is in the transport (ie, the cables and the electrical signals). The (logical) protocol itself remains pretty much the same, and therefore so do the controllers and the drives.

The proof that this is an evolution, and not a revolution is that most SATA solutions in the market right now (including some high-end RAID cards, etc.) use exactly the same chips as the (p)ATA models, and simply add a converter and a couple of SATA connectors.

Even in terms of transfer speed, the change is minimal (from 133 to 150), and pretty mcuh irrelevant (no current IDE drive comes even close to 100 MB/s, let alone 133).

RMN
~~~

Re:Give me Firewire!-here we go again!

[Wdomburg]

# Data transfer rates of up to 400 Mb/sec with 1.2 Gb/sec speeds in development

400Mb/sec vs 150MB/sec

Pay attention to case, it does matter. As for what's in development, call me when its actually available.

Matt

Solid State

[GT_Alias]

This article seems to be largely extolling the virtues of a smaller hard-drive cable (yes, I know there's other stuff in there, but you can't deny that they make a really big deal out of those cables).

I don't think Serial ATA is the monstrous revolution they'd like for it to be. I think solid state storage will represent that...when we no longer have to rely on precision mechanical components that become royally fscked with the introduction of 1 speck of dust...THAT will be a revolution.

S-ATA is the Ghetto FC

[Duncan3]

It's real simple, ribbon cables SUCK, they cost more to make then serial so PC makers hate them.

So, here's how it is...

Fibre Channel - 2Mb/s(10Mb coming very soon), 126 drives, 10+ mile range, better then SCSI.

S-ATA - 1.2Mb/s(2.4Mb in 2004), 18" range?, IDE protocols for all your write-only data needs.

S-ATA is the Ghetto FibreChannel, just like IDE is crappy SCSI, expect similar suckiness and low quality to go with the low price and cheaper cables (to make, to buy they will cost more I'm sure).

But again, this is all about the creaper cables, since lets face it 95%+ of the machines out there only have one drive anyway.

Re:S-ATA is the Ghetto FC

[Duncan3]

yea yea, s/M/G/g, it's late.

Re:S-ATA is the Ghetto FC

[Vortran]

And a ghetto pc SHOULD have only one drive. That's another reason why it's a GHETTO pc.

At 320 MB/sec with LVD SCSI, I _still_ see my biggest system bottleneck in hard disk data throughput. It really irks me that I can spend $1,000 for a decent system (mobo, cpu, memory) and not even be able to come close to alleviating my hard disk bottleneck for another $1,000.

These days putting a hard drive in a new machine feels like putting a lawn mower engine in a Lamborghini.

I've looked at solid state drives. If I took a mortgage on my entire net worth and liquefied all my assets, I still could not afford solid state drives at $80,000 a piece. Especially on a network that runs over 250 GB of storage. :(

Vortran out

Re:S-ATA is the Ghetto FC

[bored]

Besides the fact that most FC drives are FC-SCSI hu? lol.. :>

Very poor article

[waltc]

Aside from the fact that this article was very poorly written, it's difficult to understand what the author is so excited about. Between phrases like "dawn of a new era" and so on, the author makes a terrible, rotten case for why anybody should be excited about this at all. Basically, all I got out of it that was of seemingly immediate importance was the fact that the author seems to think thinner, longer drive cabling is revolutionary and heralds a new day in personal computing.

The article was extremely misleading in that it said next to nothing about the kind of drive technology to be used in SATA drives. Is this because the author doesn't know anything about these drives, or is it because the author knows there's nothing new whatever in the current SATA technology as far as the actual drives go except for the interface and cabling?

From what little I've read the first SATA drives are standard parallel IDE drives with serial interfaces. Is this approach supposed to make them run faster, or something? *chuckle* (facetious question)

Without some interesting new drive technology to make the interface change worthwhile, what's the point, here? *IS* there some point aside from thinner/longer cables???

A couple of days ago I saw a 200GB WD SE with 8 megs of cache. I'm already enjoying the benefits of RAID 0 + 1--the current IDE subsystems are *already* much faster than the drives they host. What's the problem? Cables? These days you can buy well-made rounded IDE cables (that are not simple ribbon cables folded, spindled, and mutilated.)

Maybe I'll become impressed when I can read more about the drive technology planned here and how it differs (if it does) from current IDE drive technology, but right now I'm not impressed at all. It's articles like these that definitely give SATA a "bad name"--if it in fact deserves something better.

*cringes*

[Akardam]

Please, for the love of everything that's geek... DON'T use "drop" and "hard drive" in the same sentance.

IDE Technology -- What really needs to be fixed.

[Proudrooster]

You can stick a parallel to serial adapter on an IDE drive, reduce the cable size, but it's still a crappy IDE drive.

If IDE hardware developers read slashdot, here is a list of IDE problems I'd like to see fixed.

1. You can't HOTSWAP an IDE drive without risking blowing your drive, crontroller, or upsetting the powersupply.
2. You can't WARMSWAP an IDE drive, without risking blowing your drive, controller, or upsetting your powersupply.
3. IDE still only supports 2, yes 2 drivers per controller, which makes it impossible to do hardware RAID-5. That leaves us with software RAID-5 as our only option.
4. IDE cables can only stretch so far, so even if you could somehow manage to get 8 IDE controllers into a box, for a total of 16 drives, there would still be cable length issues. I think 1 m is max. We need differential IDE :)
5. IDE drives are just now able to verify data integrity, but thats good since we can start using IDE drives in servers that don't need 100% uptime.
6. ATA/100 Round IDE cables are already available. In fact I just ordered some that have a UV reflective coating for my next case mod which features a black light. Airflow isn't a big issue, in fact Compaq has been slicing up IDE cables for a long time now to increase airflow.
7. The SUSTAINED TRANSFER WRITE RATE of IDE drives is still not fast enough to store uncompressed NTSC video at 60 frames per second, or store high bandwidth Satellite streams.
8a. Size increase (GB's) are not keeping pace with read/write access speeds and simply adding cache RAM and tweaking seek algorithms isn't going to remedy this problem.
8b. As, internal volatile write caches grow larger, the risk of uncommitted writes being lost in a power outage or crash increases.

If serial ATA would let me connect 4 drives per controller, I might start getting excited. If I could start "hot swapping" IDE drives, I would get really excited.

However, going from "flat to round" and "parallel to serial" is about as exciting as Windows XP compared to Windows 2000. It does the same thing, only slightly different. Actually in the case of Windows XP, thats not true, since Windows XP is missing device drivers for older Digital Cameras, Scanners, Modems, Video Cards etc ... but assuming you could still use all your exiting hardware it would be about that exciting.

S-ATA exists for one reason...

[Francis Avila]

...and one reason only:

To get rid of those damn ribbon cables.

Don't believe the marketing hype. SATA isn't about faster speeds, or more advanced features, or any of that crap. S-ATA is about cables.

IDE is crippleware. At some point in the past there was probably a need for a simpler, less expensive counterpart to scsi for desktop systems, but frankly that need is gone. The price distinction between IDE and SCSI has long been totally artificial. Drive manufacturers make a drive, and then slap on whatever control board they need, IDE or SCSI. Makes no difference to them, except that they get to mark up the SCSI version. Pure marketing: they need to stratisfy their technology so the enterprise guys don't feel like they're sullying their hands with the same tech as those Walmart PC-consumer lusers.

Frankly I wish SCSI had those neat little connectors (and they soon will, with Serial attached SCSI), and I hate ribbon cables as much as the next guy, but I'm not going to be fooled into thinking this is any real improvement over IDE.

But even as little as this is, it's long overdue. Those ribbon cables are the enemy of all that is good and just and true in the world.

Remember folks, SATA is only one letter away from SATAN. Q.E.D. Evil.

Re:S-ATA exists for one reason...

[MikeBabcock]

On which note, serial scsi would be a great invention :)

Oh wait, that's firewire ... well, maybe we should put some push on the development of 800, 1600 and 3200Mbps speeds on firewire busses then; no SCSI IDs and no IDE channels either.

Serial vs Parallel

[bertok]

I've noticed that the argument for serial over parallel is usually something along the lines of: At higher frequencies, it becomes increasingly difficult to synchronize the signal travelling down multiple parallel conductors, but this is not a problem with serial protocols, so despite having less wires, serial interfaces can be faster thanks to higher signal frequencies.

This is all fine and good, but why not just treat the wires in a parallel cable as individual serial wires? Sure, if you increase the signal frequency, it becomes next to impossible to guarantee that all the signals arrive at exactly the same time, but I don't see the need for bit-level synchronization. If each wire has its own protocol, its own synchronization, and its own buffers, then as long as there is synchronization at the packet level, there should be no need to worry about synchronizing at the bit level. This would allow both high frequencies, and lots of wires.

Re:Serial vs Parallel

[amorsen]

This is all fine and good, but why not just treat the wires in a parallel cable as individual serial wires? Sure, if you increase the signal frequency, it becomes next to impossible to guarantee that all the signals arrive at exactly the same time, but I don't see the need for bit-level synchronization. If each wire has its own protocol, its own synchronization, and its own buffers, then as long as there is synchronization at the packet level, there should be no need to worry about synchronizing at the bit level. This would allow both high frequencies, and lots of wires.

Congratulations, you have just invented the RDRAM bus. The disadvantage is price. 150MB/s from one set of wires is more than enough for current drives, and serial speeds increase all the time, so it should not be necessary for many years for hard drives.

buyer beware

[frovingslosh]

from the article: and although it wont officially be supported until Microsoft's Longhorn OS

This seems to say something that I've never seen admitted about serial ATA: that it has DRM built in! If you want to buy hard drives that get to decide what you can and can't store on them, go ahead, but I'm not going to buy into any DRM technology. Extra speed and a smaller cable will not tempt me into doing it; I'll stock up on the last of the regular ATA drives as the serial ATA's replace them.

Re:buyer beware

[frovingslosh]

Longhorn doesn't work unless all parts of the system employ DRM. That Longhorn is already stated as the oficial point of support for serial ATA, while the drives already work under current Windows, tells me that there is a Longhorn related feature waiting for us.

Why put it into a drive? Part of the cradle to grave DRM philosphy. To keep you from storing something "they" don't want you to. To control how you can use something that you do store (it's only accessable by some DRM approved applications). To limit how long you can retain some files. To limit the number of times you can view a file and to keep you from making a copy. Basically to decide what you do with your computer.

You need to do a little research, DRM built into a drive has been talked about for a while, but it looks like it's in the serial ATA drives, just not a marketing feature they want to advertise to the people who actually end up with them!

Actually, no.

[Rui del-Negro]

Not quite. SATA controllers have one device per channel, with no master or slave.

You are probably thinking of the situation where the motherboard has a (p)ATA controller and a converter is used to connect it to SATA cables. In this situation, one SATA channel is assigned to the (p)ATA master and another to the (p)ATA slave. But from a SATA point of view, the two channels are completely independent, and only support one device each.

RMN
~~~

Comment removed

[account_deleted]

Comment removed based on user account deletion

Comment removed

[account_deleted]

Comment removed based on user account deletion

More info

[Daniel Rutter]

I wrote a piece on this subject a while ago.

It, um, reads less like a press release than does the Explosive Labs piece :-).

Argh.

[Wakko Warner]

No they aren't.

They're expensive because they're better.

Are they somewhat overpriced? Why, yes, they are. But that does not diminish the fact that SCSI kicks IDE's ass all over the place.

Plz do not attempt to refute this. Thx.

- A.P.

Re:Argh.

[Nogami_Saeko]

Why should I care about SCSI? I don't run a server, and I don't care if my applications and data take a couple extra seconds to load?

For the price of the equivilent SCSI system, I can get at the very least 5x the storage on an IDE system. That's plenty of storage AND redundancy.

Don't get me wrong, I used to run all SCSI in my Amiga, and for a while I had SCSI in my PC, but SCSI just can't compete with monster cheap IDE drives.

$799 - Seagate Cheetah 73.4GB SCSI 10,000rpm
$575 - Seagate Cheetah 36.7GB SCSI 15,000rpm

$279 - Western Digital 120.0GB IDE

(cdn OEM prices)

Point.

Re:Argh.

[G-funk]

I never said they're not better.... But they're maybe worth twice the price. They're sold at 5X the price for a 10k drive, and abou 8X for a 15k drive. It's an artificial pricing. They've been making SCSI drives for at least 15 years, I think they're as common as they're going to get, and it's not like they're rare. Most servers around the world run on scsi (not as much now as it used to be, but still most).

Re:Argh.

[runderwo]

but SCSI just can't compete with monster cheap IDE drives.

Wrong. The manufacturers have chosen to segregate the market and to place SCSI drives at the high end. This is not a limitation of "SCSI", it is a marketing choice.

Re:Argh.

[Znork]

True, not as much now as it used to be, we're switching to FibreChannel attached storage for servers instead... ...which is even more expensive...

Re:Argh.

[Sabalon]

And what does that fiber channel plug into? A box with a fiber channel interface to the external world and a RAID array of SCSI disks inside the box.

Re:Argh.

[GooberToo]

They're expensive because they're better.

Back on planet earth, no one said otherwise. This is why they are a top shelf product. Having said that, you can make a better product and have it cost only a single digit percent in difference. If I had to spend 9% more to make a SCSI drive and can sale it for 5-20 times the price of an IDE drive, I'm fairly sure I would. Then again, that's fairly obvious...isn't it. ;)

In the example about diamonds that I gave, diamonds really are better than the typical pebble out my front door...but they are not 1000x better (price wise). A $1000.00 diamond should be able to be purchased for a couple of bucks (if that). They can't be because it's a premium product. The reasons it's 1000x more expensive rather than mearly 5-20x is because of the cartels.

Plz do not attempt to refute this. Thx.

Please do not attempt to refute how the real world works. Thanks. ;)

A compromise between SCSI and IDE

[iankerickson]

ACARD makes a series of SCSI-IDE bridge cards which connect to the SCSI chain on one side and an ATA hard disk on the other. They have several models, mostly depending on what type of SCSI cabling you have, costing from $50 to $80. They support large ATA disks, the cost of which plus the $70 for the bridge is still cheaper than most SCSI drives. If you don't need the warranty and physical traits of SCSI hard disks, but you want to be able to hook up 6 drives to your PC with only 1 IRQ and IO address or add 60-80 gigs of space for under $200, this might do the trick. They also come in handy for old workstation-era machines, like PowerMacs, SparcStations, or VAXes. The bridge doesn't require any drivers or software to work, since it just tunnels ATAPI and makes the IDE drive look like just another SCSI disk in the chain.

http://www.acard.com/eng/product/scside.html

Microland sells them in the US:
http://www.microlandusa.com/microland/

Some downsides:
- The hard disk has to be formatted while cabled to the SCSI-IDE brige. You can't move a drive from a regular IDE controller to the SCSI-IDE bridge without getting geometry errors.
- The interface is ATAPI only, so not all commands for the device may work. FE, firmware updaters and vendor utilities designed for the hard disk probably won't work the bridge.
- The utility to update the bridge's firmware is only for DOS/Windows.

There will probably be LVD-SATA bridges too in the future, if SATA truly catches on.

Various peripheral evolutions...

[phorm]

Parallel port: Does anyone still use this
I work at a school... they still have a DOT-Matrix or two in use (both as network or local printers. The cool thing is that these buggers still work almost as good as I remember them being 10 years ago (which is to say noisy as all heck, but still functional). We also had one of these for printing out (blah) COBOL code in college.
Parallel printing has evolved though. At some point we got EPP in conjunction to ECP.

I'd also like to recommend the usage of the PS2-style keyboard connector as a friendly successor to the old AT-style standard.

And in an addition for the parent
How long has the VGA pin-out been defined

Really old printers can actually have physically different pinsets. The old ones also didn't seem to have as many pins (though the places for them were there). Guess they thought ahead when designing the pin layout for monitors?

Re:Various peripheral evolutions...

[Clover_Kicker]

>I work at a school... they still have a DOT-Matrix
>or two in use

Dot matrix printers are great for schools. They're so slow/loud/annoying that people only print what they need, instead of using the school as their personal print shop.

A lot of places got "sticker shock" when they first introduced laser printers to student labs... Labs that had never bothered with printer quotas suddenly needed them.

Re:IDE Technology -- What really needs to be fixed

[amorsen]

If you had read the article...

1. You can't HOTSWAP an IDE drive without risking blowing your drive, crontroller, or upsetting the powersupply.

With SATA you can.

2. You can't WARMSWAP an IDE drive, without risking blowing your drive, controller, or upsetting your powersupply.

With SATA you can.

3. IDE still only supports 2, yes 2 drivers per controller, which makes it impossible to do hardware RAID-5. That leaves us with software RAID-5 as our only option.

Who cares when you can get hardware RAID-controllers with 12 ports on one card? What is the great advantage of having the cable be the single point of failure for your whole RAID, like SCSI does?

4. IDE cables can only stretch so far, so even if you could somehow manage to get 8 IDE controllers into a box, for a total of 16 drives, there would still be cable length issues. I think 1 m is max. We need differential IDE :)

Ok, 1m can be a problem for some people. However most people do not have cases larger than 1m.

5. IDE drives are just now able to verify data integrity, but thats good since we can start using IDE drives in servers that don't need 100% uptime.

Err, why is it a problem when it is already fixed as you say?

6. ATA/100 Round IDE cables are already available. In fact I just ordered some that have a UV reflective coating for my next case mod which features a black light. Airflow isn't a big issue, in fact Compaq has been slicing up IDE cables for a long time now to increase airflow.

Round IDE-cables are expensive to produce and still large and inflexible. SATA solves it.

7. The SUSTAINED TRANSFER WRITE RATE of IDE drives is still not fast enough to store uncompressed NTSC video at 60 frames per second, or store high bandwidth Satellite streams.

So get the hardware RAID-controller and start streaming away. Oh wait, hardware RAID for SATA doesn't exist. 3ware is a figment of my imagination.

8a. Size increase (GB's) are not keeping pace with read/write access speeds and simply adding cache RAM and tweaking seek algorithms isn't going to remedy this problem.

You can't blame the interface for that. 150MB/s per drive for 12 drives on one card is way more than any SCSI solution supports -- and way more than current drives need.

8b. As, internal volatile write caches grow larger, the risk of uncommitted writes being lost in a power outage or crash increases.

So turn off the write cache. ATA supports Transaction Command Queueing although not all drives support it yet. By the time SATA drives become available, TCQ should be common.

Re:i dont see a huge rush for people to upgrade

[rodgerd]

s/hardware manufacturer/OEM/

Dell will be drooling. Cheaper cables that are easier to install will be a big win for them.

it's not all the same

[g4dget]

It's not all the same. You have a limited budget: are you going to put it into additional wires or additional electronics? Since the cost of high speed electronics has dropped through the floor while wires aren't getting cheaper, serial is becoming increasingly attractive. And that's even not taking into account all its other advantages.

SCSI is hard to configure

[g4dget]

SCSI configuration is not exactly user friendly: you need to work out the SCSI IDs and worry about termination. There are also a bunch of different SCSI versions you need to worry about. If you get things wrong, nothing may tell you about it, the drives may just fail in subtle ways.

IDE is much simpler: with cable select, you can just plug in anything anywhere and it works. Serial ATA will preserve that simplicity and improve on it.

Then WHEN

[ackthpt]

I've been following these and there's a number of manufacturers planning to make them available, Maxtor, Seagate, Western Digital, Fujitsu, et al, but dates have been pushed back. Seagate was to be shipping ST380023AS and ST3120023AS drives in late October, now I'm seeing late November or even December. Maxtor has stated they will ship in December, others I haven't found out about. There will be a SATA group presence at COMDEX. Here's a source of information, but it tends to be general and dated, aside from having some technical docs online, too.

Average Joe? Puhleeze.

[gosand]

Now, is SCSI better for your average Joe? Maybe not significantly. Neither is 7200 vs 5400, 2MB vs 8MB buffers, or 8.9 vs 9.1 ms access times. However, if they could use one cable to connect 15 devices in their tower, they'd be alot happier than having the 8 cables they'd need to do it with current IDE tech (let alone IDE's relative inability to be used externally).

The average Joe will simply ship the whole computer back to Dell. Ask average Joe if he has an IDE drive or a SCSI drive in his computer, and see what kind of reaction you get. You'll either get Mr. Self-taught computer Joe, who will tell you he has a 30 Gee Bee and a 256 Em Bee, or someone who doesn't know and doesn't care.

People can say SCSI is better (or equivalent) to SATA, but the bottom line is it doesn't matter. The market, or more accurately the people driving the market, will decide. In the end, we are going to have technology that keeps getting better and better. Maybe SCSI missed the boat because of price. I know I wouldn't buy a SCSI drive for home, they are just too damn expensive. Servers - sure. But SATA is a way of sparking interest. If SCSI drives were cheaper, people would buy them, and if there was more interest in them, they would be cheaper. So SATA comes along, will garner interest because it is new, and it will probably take off. I don't think SCSI will go away any time soon, because of legacy support, but if it does, then are we really missing out on anything if SATA can pick up the slack?

Lamenting wide spread SCSI adoption on the PC

[tbuskey]

The best thing Apple did was put SCSI in the mac plus. Suddenly you could add hard drives, cd roms, scanners, tape drives, ethernet adapters, serial port adapters.

I used to put my Syquest and DAT tape drive on a mac, PC, and Sun. I still use the DAT drive I bought in 1993. How many 10 year old peripherals do you use?

If PCs had gone to SCSI instead of IDE (remember MFM?) we'd all have cheap SCSI drives we could use on any platform. There'd be more innovation in SCSI.

Instead, I get IDE in my Ultra 5, Macintosh, and PC. Only 2 devices unless you want speed issues. It can only really be internal devices.

And I still need SCSI for my tape drive or other external media.

Who needs SATA when we've got Firewire?

[Valdrax]

Firewire solves all the major shortcomings of both SCSI and SATA. It provides fast transfer rates, intelligent, cheap controllers that offload CPU utilization, and has no problems with multiple devices daisy-chained off of one port. It's cheap and fast, only costing about $7 to add a Firewire port to a motherboard.

This addresses all of the SCSI issues you listed.

However, your SATA data paints a rosier picture of things then reality allows. Your 1.8 GBps controller is somewhat useless seeing as it's a PCI card, which won't be able to come close to using all that bandwidth. I'll be more impressed when ports are finally built into the motherboard. Also the lack of integrated power connector means that routing cable around a case is still not as simple as if Intel had continued to suggest internal Firewire for future PCs before getting huffy at Apple and starting its USB 2.0 movement. Furthermore the one port per one device limitation is onerous. While future versions are expected to provide daisy-chaining of drives, current revisions do not. This means you need one port per device, and I guarantee you no current devices will be using half the bandwidth of their connection. It's a waste. Furthermore, SATA still eats CPU just like all other forms of IDE because it lacks intelligent controllers.

I say forget SATA, what we need is internal Firewire. That was the nirvana of case wiring that we passed up a few years ago. I know that the SFF system I'm building will be quite cramped due to ATA/100 cables and power cables. If we'd adopted Firewire internally a few years ago like Intel originally suggested, my air flow problems would be solved already.

Re:Sticker Shock (OT)

[Clover_Kicker]

"sticker shock" isn't really a technical term, it refers to getting a nasty surprise about the cost of something, in this case huge quantities of toner and paper.

I wasn't able to find a precise definition, or the origin of the phrase. I've usually heard it in the context of buyng a car, i.e. once you've added in freight+tax+rust treatment etc. etc. etc.

Inexpensive Disks?

[Ionizor]

I'm afraid you have your acro wrong.

RAID is Redundant Array of Independent Disks.

Re:Give me Firewire!-here we go again!

[Salamander]

Yeah, but paying attention to case only works if the people providing numbers do so as well, and so many people screw up MB vs. Mb that you can't count on it. It's one of those things where it's better just to leave no room whatsoever for error by spelling it out.