Serial ATA Coming

John Doe writes "Heatseekerz.net Has a new article dedicated to Serial ATA @ Cebit 2002. This technology will be here sooner then you think!" The article is a little thin, but I haven't heard a lot about what looks to be a very common standard in the not so distant future.

Comments

The Real Info...

[GeekLife.com]

Find specs and other technical info here.

Re:The Real Info...

[leibnizme]

Other links for further information:

Cnet

SATA and ISCSI

Intel Dev paper

Maxtor Whitepaper

Re:The Real Info...

[Anonymous Coward]

There is also this web sight. http://www.serialattachedscsi.com which discusses the future of SCSI and how it will share the same physical connectsions as sATA.

Re:The Real Info...

[tenman]

where is the guy with the funny mirror links when we need him?

Re:The Real Info...

[StillaCoward]

Reposted for those who can't see anonymous messages:

by Anonymous Coward on Wednesday March 20, @10:31AM (#3194064) There is also this web sight. SAS which discusses the future of SCSI and how it will share the same physical connectsions as sATA.

This site has some interesting info to read up on.

Odds?

[Fucky the troll]

Intel worked on the project. What're the odds that all the Pentium supporting motherboards get Serial ATA waaay before any AMD versions?

Re:Odds?

[larien]

My understanding is that the Serial ATA standard is open, and any lack of a Serial ATA AMD mobo is down to the AMD mobo manufacturers not being as fast as Intel.

Chances are, Intel will have a chipset out before AMD, if only because they have a larger R&D department.

Re:Odds?

[MtViewGuy]

Not to worry. :-)

I checked the web page of the Serial ATA working group and since AMD, Acer Labs (ALi chipset builder), SiS and VIA Technologies are contributors to the Serial ATA standard, expect Athlon-compatible motherboards to have Serial ATA connectors by at least late this summer. They have to anyway given that we'll be starting to see Serial ATA interface hard drives and optical drives about the same time.

I would not be surprised that both nVidia and ATI will have motherboard chipsets with Serial ATA support about the same time, too.

Re:Odds?

[Anonymous Coward]

"SiS and VIA Technologies"

Good God, man! It's like the motherboard special olympics!

Why don't you just crap in his hat instead of recommending dogs like that?!!! No wonder Athlon boards are as unstable as a drunk on a unicycle.

Re:Odds?

[Amizell]

If it's anything like DDR then the odds aren't good at all.

For those of us...

[PhysicsGenius]

...that don't obsessively read trade rags, I found a snippet explaining what this is:

Until now hardrives have been limited to a master and a slave on a single controller. The Serial ATA standard allows you to connect more than two in a daisy chain similar to SCSI.

Hope that clears things up, it did for me.

Re:For those of us...

[cisco_rob]

Also, until now, the fastest (theoretically) that a drive could muster was 133MB/sec. Serial ATA is supposed to have a (theoretical) transfer speed of up to 800MB/s.

Re:For those of us...

[Anonymous Coward]

I thought that Serial ATA 1 would not support daisy chain. As far as I know you need as many Serial ATA ports on the motherboard/controller as drives you want to use.

For more information go to here

Re:For those of us...

[Xaoswolf]

I thought that serial ATA 1 would be just like regular IDE, only 2 devices per controller. Then it would go up from there.

Re:For those of us...

[The Ego]

The sad part is that SATA is NOT daisy-chained !

Note: I am not complaining about the standard, but about the article.

Re:For those of us...

[Anonymous Coward]

Unfortunately, this will essentially be wrong.

sATA is point-to-point, and with-out placing an extra chip on each hard drive, and re-writing the ATA stacks of all OSes, this will not happen from a sATA point of view. Also, given that serial attached SCSI will use the same physical interconnect and not have this software stack issue, configurations like this are unlikely.

Re:For those of us...

[tenman]

Can I be the first to say "All anybody will ever need is 800MB/s"

You may begin to immortalize me now. :)

Re:For those of us...

[The Ego]

But why are people posting such misinformation ?

IIRC, first generation SATA is a 1500Mb/s physical link, with a 8b/10b encoding (not so sure, look at the specs), so it would provide 1200Mb/s of effective bandwith, 150MB.

From what I've heard, SATA II would double the signaling rate, among other niceties, and there are already plans for 3rd generation signaling rate.

Re:For those of us...

[JonWan]

Until now hardrives have been limited to a master and a slave on a single controller. The Serial ATA standard allows you to connect more than two in a daisy chain similar to SCSI.

I know this is a new "standard", but this is the way drives on my Commodore 64 were setup.They were not ATA, but I could have something like 5 drives as long as they had different IDs. Heck I even have a SCSI drive from CMD that attaches to the Commodre serial port.

Re:For those of us...

[Anonymous Coward]

At 1.2G bits/sec, the best topology is a point to point connection. (see page 24 on the specs 1.0)

No one in their right mind know anything signal integrity would do a daisy chain. But then again, I wouldn't expect the average Slashdot readers to uinderstand that.

Re:For those of us...

[Anonymous Coward]

The drives on your Commodore 64 were probably SCSI, which is why you could do that.

Re:For those of us...

[JonWan]

No the serial port on the C64 were bit bangers. Just load up the buffer and toggle the bits thru. The drives had their own 6502 and memory, they set waiting for their ID to appear on the bus and then answered back to start the read/write. They worked pretty well for what they were. The SCSI drive has an adapter with a 6502 and memroy just like the standard 1571 drive, but with a custom driver that translates the serial bits to the SCSI bus. Works very well and is about 10 times faster than the 1571 drive.

Re:For those of us...

[Ed Avis]

I don't get it, why not just commoditize SCSI and make it cheaper? Nice features like power and data in the same cable can be done that way too.

Alternatively, if Serial ATA is really good, does that mean SCSI will become obsolete?

Are there good technical reasons for having both? Or just marketing reasons to do with charging more for SCSI drives, and keeping ATA for the low-end?

Re:For those of us...

[Klox]

No, no, no!

This is so wrong. SATA is a point to point connection. A SATA system uses a star topology. There is no way to add more devices than there are connectors on your HBA.

This is a major difference that Serial Attached SCSI (SAS) has over SATA. SAS uses the same point to point hardware as SATA but defines expanders that allow you to replicate the point to point connection to multiple devices (or more expanders). Think of the expanders as switches. (If you think about it, it seems like SAS would have bandwidth problems with this expander technique, but it's more complicated than this and I can't get more specific at the moment.)

Re:For those of us...

[Ted Nitz]

Standard disclaimer applies, I'm not an analog engineer, nor am I an expert in Serial ATA, I just work for a company that does a lot of work in Serial ATA

Correct, Gen1 doesn't daisy chain. The HBA's we have here in the office are set to a two port configuration, but can be made as big as 8 ports.

Gen1 serial ATA is also a strictly in-box connection solution.

It does have plenty of benefits over ATA and SCSI as they stand now, hot-plug, blind-mateable connectors, really thin wires, the wires you see in all of the demo pictures are test wires, we'd always hoped to have thinner wires than those on the end product, but at this point those cables may become standard.

The connector is also tiney, we have a drive here that looks like a standard ATA drive on first inspection, but if you look closer, you notice the jumper block is only one row of pins, and that where the other row should be is a serial ATA connector.

The inital plan for Serial ATA involved features like daisy chain, and a few others, but as the problems started to arise in development of the spec (as they always do) it was decided to push most of them back to gen2.

Oh, I forgot my favorite feature, with Serial ATA, the legacy power connectors will eventually disappear, the new power connector is blind-mateable, just like the serial cable, and doesn't have most of the problems assosiated with "Molex" power plugs (ability to insert backwards with sufficient force, variation in size causing some connector/drive combos to be nearly impossible to connect dis-connect, etc).

Also, remember that when people talk about the 1.5Gbps speed, that each connector has the full bandwidth available to it, unlike U160 SCSI or ATA/133, and that it's also the Gen1 speed, Gen2 runs twice that, 3Gbps. Well, back to work for me.

-Ted

Re:For those of us...

[stripes]

I don't get it, why not just commoditize SCSI and make it cheaper? Nice features like power and data in the same cable can be done that way too.

Why then hard drive makers wouldn't be able to have both high-profit and low-profit items that are basically the same, would they? (FYI there has been a serial SCSI working group for a long while, they even produced working samples of controllers, I think maybe 5 years ago!)

Or less cynically (and less realistically) SCSI has a lot of gunk in it, starting over with a new spec gives them a chance to purge it all...oh, wait, starting with the ATA spec gives us all that ATA baggage, including SCSI over ATA...

Are there good technical reasons for having both? Or just marketing reasons to do with charging more for SCSI drives, and keeping ATA for the low-end?

Don't forget FireWire, which does at least all the big things serial ATA does, has been around a while, and is already in the market on many Sony PCs, and all Apples for the last few years. No, on the other hand, lets forget FireWire... (yeah, I know this is at least partly Apple's fault for it's patent stance, giving a reason for not using FireWire in some places and an excuse in many more).

Re:For those of us...

[Anonymous Coward]

> The sad part is that SATA is NOT daisy-chained !

SATA gen 1 is not chainable; gen 2+ are.

Re:For those of us...

[HeUnique]

You could actually have until 4 drives - count with me the device numbers - 8,9,10, and 11 - look at your dip switches in the 1581 (or jumpers in the 1541 drives) - you got only 2 :)

Re:For those of us...

[Anonymous Coward]

You just described FireWire (IEEE 1394), which is actually based on SCSI, but has all the features of SATA (except software compatibility w/ ATA) in addtion to power and up to 64 devices in a chain (well, 63 with the host controller which counts as one). The patent holders if traditional SCSI have kept its prices way up forever. FireWire is licensed by a pool who wanted it to become consumer oriented, so it is just 25 cents per a device which includes FireWire technology for the IP rights. Basically, FireWire would be better than SATA except for SATA's more direct backwards compatibility with ATA. But FireWire exists now and it rocks (ONLY ONE FRIGGIN' CONNECTOR!!).

Re:For those of us...

[connorbd]

That is one thing I've been curious about: why bother with sATA in the first place, when we have both USB 2.0 (okay, relatively new) and Firewire (not new, and faster than any ATA out there so far)?

/Brian

Re:For those of us...

[JonWan]

Yep... you're right 4 drives, It's been over 15 years. ;-)

I have a 1541, 2 1571s, and a 85M CMD HD. That makes 4 alright. Damn those things were fun to play with. The really scary part is that I gave $400 for that 85M CMD drive!

Re:For those of us...

[bbqBrain]

The drives had their own 6502 and memory

Heh, that brings up an interesting bit (no pun intended) I heard about C64 development. If my source is correct, programmers use to detect whether a drive was connected and use its memory (8k?) for applications.

At the time, I was just a little turd poking registers for screen color and typing programs in from _Family_Computing_. I started a cheesy text adventure game at one point but dropped it because it wasn't that interesting. But hey, how 'bout that 3-voice SID?

We were also lucky enough to have a VIC-20 with tape drive before that. I suppose that's because my dad is an EE and a little geeky in his own right. He never reached full geekhood, unfortunately. If he'd been born even 10 years later, I bet it would have been quite different.

Re:For those of us...

[jkujawa]

Firewire _is_ serial SCSI.

Re:For those of us...

[stripes]

Firewire _is_ serial SCSI.

The people running the FW working group seem to disagree, at the very least they are intent on not tracking all of the future SCSI logical changes, nor are the logical changes for FireWire being funneled back into "non serial" SCSI.

Other then that I agree, FireWire does seem really a whole like like SCSI with a different transport, hot plug, and support for bandwidth and timeslice reservation...

Re:For those of us...

[Ed Avis]

Why bother with ATA at all when you could just make 'hard cards' and plug a disk straight into the PCI bus? It seems kinda pointless to go to all that trouble of defining a complex protocol just for that ten centimetre cable between the motherboard and the disk.

With SCSI I can see some point to it, because you have potentially many disks over longer distances and you want to mix-and-match and have disks support the same features. But since ATA is for the low end anyway, why not make a really basic interface where any fancy stuff is done in software on the host machine? As long as you have DMA for transfers to/from the disk, and some way to move the disk head, that's it really. Today's machines are fast enough to do all the rest for themselves.

I realize this would create a whole nightmarish set of WinDisks... but I'm surprised the 'hard card' concept hasn't caught on again since they were first introduced on XT-type machines.

Re:For those of us...

[jafac]

No. This is INTEL'S fault, for hating SCSI and hating FireWire, and any technology that does not sell CPU chips.

ATA and USB require a CPU to intevene and manage transfers. (Do a tranfer of data from one SCSI disk to another, then one ATA disk and another, in the same machine, and the ATA transfer will eat up CPU cycles) SCSI and FireWire do not. Intel LOVES ATA and USB, because it sells more CPUs and drives up demand for faster CPUs.

Now, since Intel pretty much owns the hardware side of the PC industry, isn't it obvious why technologies intel does not favor would fall by the wayside?

Re:For those of us...

[smaug195]

Firewire isn't faster then any ATA spec, it can only transfer 400MEGABITS/Second while ata/133 can transfer 133MEGABYTES/Second :).

Re:For those of us...

[(outer-limits)]

Those drives always sounded like they were about to die, making incredible banging and chunking noises. Commodore always seemed to be able to develop something so it was just cheap enough to be unreliable.

Re:For those of us...

[Ed Avis]

I'm not sure you can blame Intel for trying to kill things like SCSI which reduce load on the host CPU. Because such cards require their own separate CPU, which is obviously good news for Intel! Intel chips like the i960, StrongARM, even the good ol' 80186 are all suitable for building intelligent disk controllers.

A better reason for Intel to dislike a technology is that it's controlled by patent holders who happen not to be Intel.

Re:For those of us...

[Deton8]

SATA is stictly point-to-point. You can't daisy chain multiple drives.

Re:For those of us...

[connorbd]

Hardcards? The only way that could work is if the disks were so thin they could be used surface-mount; there's a reason they stopped using them (structural integrity and vibration being the big ones).

/Brian

Re:For those of us...

[Ed Avis]

I was thinking of 2.5 inch disks; if they can be bashed around in a laptop they can surely be mounted on a PCI card in a stationary machine. The disks might have to be thinner, just reduce the number of platters (so the capacity is 40 gigabytes rather than 80 - so what). And the card would probably have to be a bit chunkier than an ordinary PCB. But it's surely doable.

Re:For those of us...

[connorbd]

My point being that you'd have to go even thinner than that to pull it off. IBM surely has the beginnings of the technology with the MicroDrives; all they need to do is find an effective way to scale up the area without increasing the thickness of the drive. And, like I said, make it into an SMD so it doesn't go flopping around. While they're at it, why don't they stick it in a drive bay to free up a slot :-)

/Brian

Re:For those of us...

[Ed Avis]

You could indeed stick it into a drive bay, if you can connect the PCI bus over the distance required. Some of the early PS/2 models had a so-called 'Direct Bus Attachment' drive which plugged straight into the MCA bus. There was a ribbon cable with a dirty great ferrite ring around it plugging straight into the riser (which also contained the ordinary MCA slots). Then no separate disk controller was necessary.

The same might not be possible for PCI because of the faster bus speed (33MHz or higher, rather than MCA's 10MHz), unless you have an even bigger ferrite or something.

Re:For those of us...

[connorbd]

The next step of course is to move the controller off the motherboard yada yada yada...

My point being that you're not gaining anything except another iteration through the loop that led to the invention of ATA and SCSI hard drives in the first place. They did it with graphics cards as well, moving graphics capability into the local bus with VESA. When VLB flamed out because they couldn't quite get it working with fast 486s and Pentia it was scrapped in favor of PCI. PCI wasn't *quite* fast enough so they moved it closer to the processor bus again with AGP. Figure on at least one generation of bypassing AGP for integrated graphics in the northbridge and feeding full-speed off the processor or memory bus before they come up with something else... hey, could happen...

/Brian

Re:For those of us...

[Ed Avis]

I agree, it is a bit of an endless cycle. My feeling is just that Serial ATA is (a) not quite good enough to compete with SCSI in the 'let's build an intelligent disk subsystem and take load of the host processor' market, and (b) too complex for the 'let's build something really cheap and dirty' market. The original ATA disk controller could be built for about 50 cents in an ISA bus machine. Could there be an equivalent for PCI?

Re:For those of us...

[connorbd]

My real question is this: we are up to UltraATA 133 now, if you're willing (and/or crazy) enough to pay the price. This is too fast for all but the highest-performance flash memory devices; most hard drives don't need and can't use any more than ATA/133. We've had FireWire for a very long time now (five years, anyway), and it never got any serious play until a couple of years ago; we have USB 2.0 now, and those two standards will be leapfrogging each other for some time to come.

Why not just rewrite the BIOS code to create an abstraction layer between ATA and FireWire? (Granted I don't know much about BIOS writing...)

/Brian

too late, unless its way cheap

[TheGratefulNet]

there's already a high speed serial that can be used for ide drives. its called usb2 and also firewire.

I am using an external drive bay that takes FW in and converts (with a very small pcb) to 40pin ide (ata100). cost isn't much ($70) and the controller isn't either ($30).

I was able to copy an 80gig drive from native ide to a remote ide via firewire on the latest linux 2.4.18 kernel in about 3 hrs or less.

serial ide would probably JUST be ide. but serial usb2 and FW are more general purpose (video, etc).

I think serial ide is just too late in the market.

Re:too late, unless its way cheap

[Magila]

USB2 and FW are no where near fast enough to be a next gen HD interface, nor where they designed as such. SATA is many times faster than FW/USB2 (500 MB/s vs 50 MB/s for FW) and supports key features for HDs (daisy chaining etc). USB2 and FW are nice general purpose external interfaces which also do an ok job with HDs, but they were never meant to repace IDE and never will.

Re:too late, unless its way cheap

[Ami Ganguli]

I think cheap and fast is the whole idea. This is meant to replace current ATA interfaces, so that means cheap, mass produced, and designed for low-end systems (as opposed to big servers).

Re:too late, unless its way cheap

[Anonymous Coward]

sATA and SCSI would do an 80GB copy in under .5 hours.

In both FireWire and USB2, the bottleneck is still the physical connection a fraction of that of ATA, sATA, and SCSI. On sATA ~150MB/s and SCSI 160MB/s, you can handle two drives at full speed, and that is about it, possibly three.

Re:too late, unless its way cheap

[kaisyain]

Firewire can do daisy chaining. The speed difference also isn't huge. Serial ATA starts out at 150MB/s, while 1394b starts at 800Mb/s; SATA is planned up to 600MB/s while 1394b is 3.2Gb/s.

Since neither one is really what I would call widely available at the moment speed comparisons are kinda moot, kinda like asking if the Enterpise is faster than a Star Destroyer :-)

Re:too late, unless its way cheap

[Anonymous Coward]

USB 2.0 is only 480M bits/sec -> best case Ultra 66 performance if you ignore USB 2.0 overhead

Firewire is at 400M bits/sec will be going to 800M bits/sec

Serial ATA is 1.2G bits/sec -> Ultra 150. There are little overhead on the protocol for data transfer.

Which one would the smart person pick ?

Re:too late, unless its way cheap

[Thud457]

Still, I would rather have a standard for a fast way to connect any device to my computer rather than USB2/firewire + SATA. Why not just work on cranking up the speed on existing standards instead of reinventing the wheel (with a kludge based on ATA)?

I want 1500kb/s from my keyboard, dammit!

Re:too late, unless its way cheap

[Anonymous Coward]

I know you're not being serious, but that would make your keyboard pretty expensive.

Re:too late, unless its way cheap

[edmudama]

Have your read the official 1394 spec?

It's like 1400 pages. Holy transaction overhead.

Re:too late, unless its way cheap

[TheGratefulNet]

and why is the post marked troll?

it applies directly to the topic at hand. my point is that while serial-ide might be faster (in theory), it seems intended to be a host->drive controller, only. whereas both usb2 and firewire are more general purpose and are already shipping and functional.

and both FW and usb2 have more bandwidth on the channel (wire) than any drive can run at, external transfer rate wise.

'interesting' or 'informative' or even 'redundant' (if you must), but CLEARLY not a troll. sheesh.

Re:too late, unless its way cheap

[TheGratefulNet]

let me add more detail.

this was NOT a block-by-block copy. it was a recursive 'find . | cpio -pvd /mnt' copy between ext3fs (linux) and fat32. you have to read the file and then write it out in a very different filesystem. you cannot do this, no matter how fast your system is (by today's standards) in a half hour. no way.

Re:too late, unless its way cheap

[TheGratefulNet]

have you seen the controllers? some are single chip solutions (literally, just a single small chip).

even though the spec may be verbose, the silicon needed to support it isn't any bigger than a regular ide controller (roughly).

Re:too late, unless its way cheap

[stripes]

have you seen the controllers? some are single chip solutions (literally, just a single small chip).

even though the spec may be verbose, the silicon needed to support it isn't any bigger than a regular ide controller (roughly).

That is in part because a very large part of what the spec covers is stuff the hardware doesn't care about (contents of the "packets" vs. how to move them around). The FW chip on a PowerBook doesn't know how to send a "I'm not on the bus" packet out, or what to do with one when it comes in (other then what it does with all incoming packets - DMA it into memory and pull the interrupt line high so the OS can come by later and sniff at it).

The other reason it is huge is specs tend to be written huge, Internet RFCs are a marvel of brevity if you compare them to what most standards bodies produce (except the IPsec ones which are a lot closer to normal standards documents...).

Re:too late, unless its way cheap

[kolding]

Another advantage of Serial ATA is that, from a software point of view, it's supposed to look exactly like current ATA/IDE. You don't need to change your software stack at all.

Re:too late, unless its way cheap

[Chasing Amy]

I've seen Firewire HDs in action, and they are SLOW by comparison. My standard ATA-66 or ATA-100 5400RPM IDE drives are speed demons by comparison, let alone my 7200 RPM IDE drive. The fact is that the Firewire interface cannot compete, because of the time it takes to translate from Firewire to IDE since there are in fact no drives that natively speak 1394.

In practical terms this can result in *up to* about 500 KB (yes, big B) per second transfer speed difference in several parts of the drive's transfer curve. The most optimistic transfer speed chart I found on a quick search was the one at http://www.digit-life.com/articles/fireware/ which shows a close general match between the 1394 and IDE transfer graphs, but the Firewire graph still shows a lot of erratic activity which in real use would add up to very significant margins when transferring large amounts of data or incurring a lot of random accesses with a lot of small transfers of data. And this is the *most favorable* graph for Firewire I found; a few were less than stellar. And, bear in mind that these lags are very cumulative--transferring from one 1394-to-IDE drive to another might incur unnecessary penalties of up to 1 MBps in places. I routinely deal with multi-gigabyte transfers, and that is unacceptable since it can be avoided by using a native IDE/ATA transfer or, better yet, SCSI--and now, Serial ATA.

In addition, there's alays a danger when you translate from one protocol to another, particularly if it's needless in the first place. Haven't you known people who've b0rked data when transferring it to a Firewire HD? I have. Different enclosures=different chipsets=different drivers=different amounts of maturity, whereas the ATA standards are always--well--standardized and in common usage by the time people start buying their new drives and upgrade cards or new mobos. I've even known a friend who added a new Firewire drive to his B&W G3 Mac and managed to ruin some transferred data, and Macs are bred for Firewire. Problem was, his spiffy new VST (IIRC) external 1394 drive required 3rd-party drivers at the time, and OOPS! who knows what went wrong, but something did.

Could something go wrong with a standard ATA or Serial ATA transfer? Of course. I fell prey to IDE data corruption last year from the dreaded KT-133/SB Live!/BIOS bug. But the probability of data corruption is lessened because it's such a well-accepted standard by the time it gets into the wild, whereas Firewire drive enclosures are still the Wild West in comparison.

And the reason you can forget about native Firewire drives ever becoming mainstream is that manufacturers wouldn't want the expense--nor would consumers. It's cheaper in the long term to just invent Serial ATA than it would be to commoditize either SCSI or Firwire, with their more generalized and extensive and hence more expensive controllers for each drive. Firewire and SCSI devices have to be "smart," with a lot of decisions made in the device itself. ATA, and now Serial ATA, devices can be relatively "dumb", with most of the work being done at the main controller. The cost difference would probably add about $50 retail to the cost of each drive--Mac fanatics have been dealing with that because they overpay for everything, but PC folk wouldn't stand for it. We want big, fast, reliable--*and* inexpensive, and neither Firewire nor SCSI can really do all of that for hard drives.

Time has proven, though, that IDE/ATA/and soon (if all goes well) Serial-ATA can. Why pay %50 more per drive, when the cost of quality Maxtor/Seagate/etc. drives is getting so ridiculously low that that would add 50% to the cost of a low-end drive or 25% to the cost of a high-end drive, for no gain in HD performance? The only performance improvement would be a significant reduction in CPU usage, but that's not an issue on consumer-level and low-end workstation/server machines which are now well into the GHz+ range--and servers and high-end workstations would be running proven SCSI anyway.

Just my opinion, though. Firewire drives have their place--in Macs. ;-) Us real commodity computer users will almost always use cheaper ATA or more proven SCSI, depending on our needs.

Re:too late, unless its way cheap

[stripes]

Still, I would rather have a standard for a fast way to connect any device to my computer rather than USB2/firewire + SATA. Why not just work on cranking up the speed on existing standards instead of reinventing the wheel (with a kludge based on ATA)?

I share the desire for fewer interface types, but there are different tradeoffs to make with low speed (and frequently low cost required) devices like keyboards, mice and joysticks then with high speed devices (many network connections, hard drives, DVD players, displays)...and yet another set of isosocrnous devices (or whatever the term for latency intolerant is...like sound devices).

Making one standard that does it all will make it really complex (hi USB2!) and can make it costly to implement. I don't think USB2 will suffer in small devices because of how hubs have to work you can make the cheap devices cheap and the ones that have to be expensive can be expensive...except the hubs will be more costly because they are less like the simple bus USB1.x hubs could be and more like little rate converting ethernet switches.

Making two standards (say USB1.x and FireWire) have lots of political problems, but I think would have been simpler to deal with...unfortunetly that doesn't look like it will happen. Ah well, at least I can hope that USB2 and SATA don't suck more then FireWire (and I can wonder if USB2 will cut the legs out from under SATA since it might be "enough" to take the low end of the market leaving SATA in the same place ATA left SCSI...)

Re:too late, unless its way cheap

[Anonymous Coward]

USB 1.0!

Re:too late, unless its way cheap

[flatrock]

USB 2.0 just isn't fast enough. It will be good enough to connect a single drive, or an external CD-RW, but single IDE drives are already getting close to the total bandwidth of USB 2.0.

Firewire can be fast enough, depending on which version you're talking about. However, firewire is a more complicated interface with many more features than are needed by IDE drives. When Intel decided not to include Firewire in the BX chipset like was originally proposed, it lost a lot of steam. I don't know the implementation and licensing cost issues, but since it hasn't been integrated into a chipset yet, I guess they're significant enough that it's just not going to happen.

Re:too late, unless its way cheap

[Zathrus]

Probably because you showed an utter lack of knowledge regarding interface speeds, as well as different design goals for different interfaces.

Serial ATA doesn't fix all of the issues with ATA, but it fixes a lot of them. It also considerably ups the maximum speed on the device chain. It's designed for internal, high speed, drive-to-host communications.

USB is an entirely different beast. It's low speed (yes, it is), with a lot of overhead (since it supports pretty much anything connecting to it), and is designed purely for short-run external connections.

Could you internalize USB2? Sure. But you've tossed a boatload of bandwidth and added a boatload of additional data to the stream in order to handle external/anything devices. You're using a hammer instead of a screwdriver here.

And thank you for being foolish enough to say that interfaceXYZ has more bandwidth than any drive can run at. You know, once upon a time MFM had more bandwidth than the drives could handle. And we won't even talk about burst transmission - since drives do fill ATA100 to the brim. Oh, and we'll also ignore the joys of RAID, where even fibre channel is struggling to get enough bandwidth at times.

'interesting' or 'informative' or even 'redundant' (if you must), but CLEARLY not a troll. sheesh.

I'd personally suggest Flamebait.

Re:too late, unless its way cheap

[CanadaDave]

"there's already a high speed serial that can be used for ide drives. its called usb2 and also firewire."

usb2 and firewire are designed for external devices. SATA is better suited for internal motherboard connections. I am guessing that USB can never get as fast as you could with SATA simply because of the length of the conductors involved.

Re:too late, unless its way cheap

[ryanwright]

I was able to copy an 80gig drive from native ide to a remote ide via firewire on the latest linux 2.4.18 kernel in about 3 hrs or less.

Wow, a whopping 7.4 megs per second? I wouldn't exactly call that high speed...

Sooner than I think?

[Brento]

That's a matter of opinion. Remember, this requires new hard drives - something that doesn't exactly happen every day in big business. You're talking new hard drive duplicators, external hard drive enclosures, etc. This is like saying fibre channel hard drives are available today - well, sure they are, but they aren't getting big play in your typical home or business.

Re:Sooner than I think?

[jbrauer]

Currently there is a 1m cable length.

Although the signals look a lot like FC, it is only intended to be used internally, over short distances.

It also has all the same software layers as IDE, and is mainly aimed at the low-end and commodity markets.

The main benefits being hot-swap and the much smaller cables (only 2 conductors as opposed to 40/80). This will help cabling inside of cases and air-flow.

Re:Sooner than I think?

[TastesLikeChicken]

I think it will happen sooner rather than later. But largely because (VERY SOON) drives will zoom past the 150Gig limit imposed by IDE.

Re:Sooner than I think?

[Emil Brink]

Feel free to jump in the time machine any time, and speed forward to join the rest of us... Maxtor already did that, with ATA/133.

Re:Sooner than I think?

[ahaning]

Remember, this requires new hard drives - something that doesn't exactly happen every day in big business.

That is, unless you're using IBM drives :).

Re:Sooner than I think?

[mindriot]

Remember, this requires new hard drives [...]

No. At least according to the docs at serialata.org, you just need a simple adapter to connect old hard drives. I mean, it's still ATA, and of course downward-compatible. I'm really thinking about waiting for the standard to establish before buying a new PC. The star topology and the thin wires (no more 40-pin wide cables that clutter your PC case) make it worth it for me. Somewhere on the serialata.org website, there is a PDF presentation of the standard, with some pics on how the cables will look. I think it's about time for a standard like that.

Re:Sooner than I think?

[flatrock]

This is true, but it's because fibre channel controllers are much more expensive than and IDE controller, and for good reason. Fibre Channel drives themselves are available, and aren't priced significantly more than their SCSI counterparts.

Serial IDE drives will require a new interface logic and hardware. However, this has been comming for a long time. I doubt it's going to be a long delay before hard drive manufacturers can roll out drives with this interface.

Not ANOTHER standard

[qurob]

Firewire is so cool, they should just use it for hard drives also.

Integrate the controller on the motherboard if you have to.

Re:Not ANOTHER standard

[Anonymous Coward]

Why use a slower bus and a simplified SCSI command set, when ATA or SCSI is avilable?

Re:Not ANOTHER standard

[Anonymous Coward]

"Firewire is so cool, they should just use it for hard drives also.

Integrate the controller on the motherboard if you have to."

SATA doesn't have the Apple Computer tax on it that Firewire does. So once again Apple users are on slower preforming interfaces that cost more than their PC counter parts.

Lets hope for the sake of their users Apple can rise above it's greed and adopt SATA. It's not like the two techs can not exsist together.

Re:Not ANOTHER standard

[Anonymous Coward]

Hmm.
Firewire ~ 50 MBytes/sec
Ultra2 SCSI ~ 40 MBytes/sec
ATA/66 ~ 66.6 MBytes/sec
Yup, looks like it's another standard used for marketing purposes to help ensure a revenue stream.

Re:Not ANOTHER standard

[smagoun]

Apple already puts Firewire on the motherboard to allow connections to digital camera, iPods, and other goodies. Why would Apple want to spend MORE money by putting SATA on the mobo in addition to Firewire? Why not just use Firewire, which is already there + perfectly capable of working with fast hard drives?

Also, 400mb/sec Firewire is what, 3 years old now? There are 800mb/sec and 1600mb/sec versions just around the corner (800mb/sec might be here already).

Re:Not ANOTHER standard

[Anonymous Coward]

$1/port

Whoopity fuckin do

Re:Not ANOTHER standard

[psavo]

LOL
I'd like to see some drive that really puts out 66.6 MBytes/sec at ATA/66.
With scsi that's altogether another question. It beats shit out of ATA/66.
There's no point comparing theoretical outputs.

Re:Not ANOTHER standard

[Anonymous Coward]

Actually, it's $0.25/port.

Which makes it even *less* of an issue.

Re:Not ANOTHER standard

[thedbp]

I completely agree ... good FireWire PCI cards feature an internal FireWire port anyway. I've been waiting for native FW internal drives instead of the usual ATA-to-FW conversion external boxes. As of now, I've been using a short cable connected to the internal FW port with an open front bay to plug in an external and nestle it inside the box to save space. But I'd rather have a nice FW drive array internal ....

Re:Not ANOTHER standard

[subgeek]

and all of the devices you connect to that super fast firewire card are limited because they all share the same interupt and the same 33 MHz PCI bus.

Re:Not ANOTHER standard

[Anonymous Coward]

well then, heck, why not plug it straight to gigabit ethernet?

Re:Not ANOTHER standard

[phallen]

Firewire?

USB 2.0, baby! Chain as many as you want (well... almost) together and get BETTER THAN FIREWIRE SPEEDS!

Re:Not ANOTHER standard

[rmayes100]

Apple already has IDE or SCSI connectors on their motherboards as it is. SATA will probably replace those and not the Firewire. As other have posted Firewire is not a great solution for internal harddrives.

Re:Not ANOTHER standard

[qurob]

It's been explained before....Ethernet is for NETWORKS, Firewire is for drives, cameras, scanners....

Re:Not ANOTHER standard

[Bytenik]

Just because Ethernet is used PRIMARILY for networking doesn't mean that it can't be used for other purposes. It is just a data transfer medium.

Look at what Gibson Guitars is doing with it in their MaGIC technology.

Clearly it COULD be used as a hard drive interface. In fact, the whole Network Attached Storage (NAS) business basically uses it that way.

Re:Not ANOTHER standard

[Anonymous Coward]

Some time ago I saw a benchmark comparing ATA/100, USB2 and Firewire and you would probably think that the bandwidth provided by USB2 and Firewire will be enough to match ATA/100 or even ATA/66 ?

Well that wasn't the case and the drive attached to a ATA/100 port was minimum twice as performant as when attached to USB2 and Firewire.

USB2 and Firewire were pretty even, Firewire beating USB2 on a couple of benchmarks and USB2 beating Firewire on the other ones.

So yes, Serial ATA is very useful indeed and speed is not everything, being able to hot plug devices is a god send for cheap configurations.

Re:Not ANOTHER standard

[HeUnique]

Actually - the current answer is between you two - $0.50 ($0.25 X 2 since you have 2 ports on a standard motherboard)...

Re:Not ANOTHER standard

[Catbeller]

Not better than Firewire 2 devices, which will be out at about the same time as USB 2 devices. You have to compare apples:apples.

And Firewire supports up to 64 devices, daisy-chained, even on the day it was introduced: USB did not.

USB is still limited by being CPU-centric, while 1394 can move data without connection to a CPU.

Re:Not ANOTHER standard

[qurob]

A NAS device and a hard drive are very different.

HD's are pretty cheap, lets jack the pricing up by building gig ethernet into them all!

Re:Not ANOTHER standard

[jrwyant]

You have to be careful when comparing USB2 drives though, to other interfaces. Early USB2 drives use what's called a tailgate, which translates USB2 into ATA commands. This becomes the bottleneck, as there isn't much buffering provided, etc. in performing this translation. Native USB2 drives (those which have a discrete chip speaking USB2) will be faster.

Re:Not ANOTHER standard

[Anonymous Coward]

I have a Maxtor 53073H3 which is a 30.0GB hard drive. I get almost aroud 28.5 to 29.8 MB/s and I'm only on ATA 33. I know thats not 66MB/s, but if ATA scales the same way or nearly the same. Then I imagine 60MB/s should be attainable.
One thing that always cracks me up are the SCSI zealots on here like you. Don't get me wrong, I think SCSI is fairly cool. However I'm not going to get religious about a bus inteface. PIO-1 IDE is very slow batting in around 1MB/s. OTOH, I have a DEC Alpha AXP-PCI that runs on a 1.0GB SCSI-2 drive. I get about 3.2-3.5MB/s. That is 1/10 of what my IDE drive gets. Boy doesn't SCSI suck. IDE beats the shit out of SCSI.
Look at the new drives compared to the old drives for both buses. They have evolved.
Hell, Google uses IDE not SCSI in their farm.

Re:Not ANOTHER standard

[Anonymous Coward]

Not really. A NAS is just a bunch of drives with an OS to hold it together. Some are SCSI, some IDE - I must be missing the point here.

Re:Not ANOTHER standard

[Anonymous Coward]

Oh yeah, well USB 3 is going to be out soon thereafter, and support even more devices at even faster speeds! Take that!!

Re:Not ANOTHER standard

[Anonymous Coward]

Only the headline bit rate is higher. Any real transfer rates work out lower because the USB protocol is crap.

Already /.ed!

[Linuxthess]

Should have used Serial ATA!

Re:Already /.ed!

[Anonymous Coward]

My name is Tom, and my Dick isn't Harry. He's Bob.

Re:Already /.ed!

[distributed.karma]

Interesting point. Can we do networking over SATA? Now that would bang the shit out of my measly 10-baseT. Why do we need all these O(10^2) standards anyway? How about a single interface for everything, it can't be that hard.

Good reason cited for not...

[Numen]

[quote]
The reason for the late breakthrough of Serial ATA is that nobody wanted to, unnecessarily, spend time and money, developing a new interface.
[/quote]

Which seems like a good reason for not doing so.

I am in all likelihood missing something, but are we facing a bottleneck in ATA and ATAPI devices with their basic interfaces?... seems to be what we currently have has been able to improve and keep up.

Genuine question, not making rhetorical points.

It would seem to me, that unless Serial ATA offers clear and *necessary* benefits, it simply wont get support fro mthe chipset chaps.

I went looking for the benefits areticle on the cited page, but would appear they are struggling under their new load.

Re:Good reason cited for not...

[Anonymous Coward]

More than 2 drives on a chain is a massive benefit. No more fucking around with master/slave drive settings.

Re:Good reason cited for not...

[DaveRosserJr.]

ATA133 is getting close to the point where it makes sense in a good many server applications... just one setback... 2 drives per interface... ok how many ide interface cards can I put in a machine? If serial ATA allows me to expand that to say one interface and 15 drives then maybe I can get away with two controllers, four interfaces and 60 drives with software raid 10. I can think of a couple of places that would come in handy, provided it ramains cheaper than scsi.

Re:Good reason cited for not...

[NMerriam]

I would say the biggest benefit is the number of devices being increased. When the initial ATA specs came out, the notion of a desktop computer having more than 1 or 2 hard drives was insane.

Now it's not at all unusual to buy a new computer and find that every drive connector is taken because you have a DVD and CD-R on one channel, and two hard drives on the other.

Cabling issues will be a big benefit to OEMs, and end users who have fewer tech issues with adding drives, as it will be impossible to wire improperly and master/slave/cs jumpers will no longer exist.

If it was only speed improvements I tend to think nobody would care very much.

Re:Good reason cited for not...

[Anonymous Coward]

The number of devices is not being increased so much a the number of controllers is being increased. Each controller is still limitted to 2 drives, just like under ATA due to the software.

Re:Good reason cited for not...

[NMerriam]

I thought that was just for the "compatibility period" of the first generation serial ata spec.

Once everyone is using serial ata and software no longer expects a limit of 2 devices per controller I think it goes up to whatever limit (7,15?). Ahh, i'm too lazy at the moment to look it up :)

Re:Good reason cited for not...

[edmudama]

> Cabling issues will be a big benefit to OEMs,
> and end users who have fewer tech issues with
> adding drives, as it will be impossible to wire
> improperly and master/slave/cs jumpers will no
> longer exist.

Actually, that was solved with the cable-select jumper years ago. Almost all major OEMs buy their drives configured as such these days.

Re:Good reason cited for not...

[NMerriam]

Actually, that was solved with the cable-select jumper years ago.

I've had plenty of drives that didn't work well with CS, that's why eliminating the whole configuration issue is a big step. CS is a patch to the problem, but it doesn't eliminate it 100%

Mixing a sony CD-R with a Teac DVD-ROM both with CS and things like that sometimes cause weird issues. I had two drives where if one was first and the other second it would work on CS, but not reversed, unless I locked the master/slave correctly...

Re:Good reason cited for not...

[Anonymous Coward]

ATA100 is already here for low end (1-8 drive RAID) servers:

www.3ware.com

And no sane person would put more than 1 drive per channel in an IDE server.

SATA and 3ware are slowly but surely pushing SCSI further upmarket. Eventually (2-4 years) its going to run into FC, and get squeezed out entirely. Can't say I'll miss the ID jumper and termination hell.

Re:Good reason cited for not...

[DaveRosserJr.]

I must concur about no more than one drive per ide interface. I am hopeing that rule will not be pertinent for serial ATA.

FAQ from the SerialATA.org website

[blues5150]

Q1: What is Serial ATA and Why is it being developed?
A1: Serial ATA is an evolutionary replacement for the Parallel ATA physical storage
interface. Serial ATA is scalable and will allow future enhancements to the computing
platform.

Q2: Previous efforts to transition to a serial bus were not successful. Why do you
believe that Serial ATA will be successful?
A2: Serial ATA is a drop-in solution in that it is compatible with today's software, which
will run on the new architecture without modifacation. It will provide for systems which
are easier to design, with cables that are simple to route and install, smaller cable
connectors, improve silicon design, and lower voltages which alleviate current design
requirements in Parallel ATA.

Q3: Will there still be a parallel ATA bus when Serial ATA comes out?
A3: Serial ATA's adoption by the industry will follow a phased transition path. There
will be a point where both Parallel and Serial ATA capabilities are available.

Q4: You stated that PCs implementing Serial ATA will be in the marketplace in
2002. Why does it take so long to implement?
A4: The goal of the working group is to ensure the Serial ATA transition happens as
smoothly and quickly as possible. The Serial ATA specification is expected to be
complete in the fall of 2000 with adoption to happen in the following 12 to 18 months.

Q5: What are the end user benefit of Serial ATA?
A5: End users will benefit by being able to easily upgrade their storage devices.
Configuration of Serial ATA devices will be much simpler, with many of today's
requirements on jumper and settings no longer needed.

Q6: What is the cost to implement Serial ATA in a system?
A6: The cost of Serial ATA technology will be on par with today's Parallel ATA
technology.

Q7: Who are the members of the Serial ATA Working Group? Can new companies
join?
A7: The Serial ATA promoters group includes APT Technologies Inc, Dell Computer
Corporation, International Business Machines, Intel Corporation, Maxtor Corporation,
Quantum Corporation, and Seagate Technology. Information on joining the working
group is available at www.serialata.org and new members are welcome.

Q8: Hard disk data rates don't seem to be pushing the limits of current ATA66
technology. Why is Serial ATA being planned now?
A8: Serial ATA is an evolutionary replacement for the Parallel ATA physical storage
interface and will allow future enhancements to the computing platform. Specifically, the
thinner Serial ATA cable addresses OEM's concerns regarding airflow around the
Parallel ATA cable, and enables design of smaller PC chassis, as well as silicon vendors
concerns regarding 5 volt tolerance support in future designs.

Q9: Will Serial ATA be compatible with today's PCs?
A9: Serial ATA electronics and connectors will differ from Parallel ATA, however the
technology is software compatible and OS transparent. It is anticipated that there will be
adapters to facilitate forward- and backward-compatibility of hard disks on PC systems.

Q10: What is the impact of Serial ATA on OEMs?
A10: Industry benefits of Serial ATA include systems which are easier to design with
cables that are simple to route and install, smaller cable connectors with improved silicon
design, lower voltage which alleviates current design requirements in Parallel ATA and
compatibility with today's software which will run on the new architecture without
modification.

Q11: Beyond hard disks, will Serial ATA be used on floppy drives, optical drives,
DVDs, and ZIP drives?
A11: Serial ATA supports all ATA and ATAPI devices, including CDs, DVDs, tapes
devices, high capacity removeable devices, zip drives, and CDRW's.

Q12: What is the impact of Serial ATA on IEEE1394 (aka Firewire) and on USB2
in terms of PC system function?
A12: Serial ATA is planned to be the primary storage interface inside the PC system,
and is not planned as an external interface to PC storage or peripherals. USB2 and
IEEE1394 connections on the PC can be used where required as peripheral interfaces.

Q13: When does Microsoft plan to support Serial ATA in its OS's?
A13: Serial ATA is software compatible with Parallel ATA and requires no changes to
Microsoft operating systems, or any other OS as well.

Q14: What are the licensing requirements and costs of Serial ATA to companies
that want to use the technology?
A14: When the Serial ATA specification is complete, it will be made available at no
charge. The working group expects to complete the specification later this year.

Re:FAQ from the SerialATA.org website

[Tet]

Serial ATA is an evolutionary replacement for the Parallel ATA physical storage interface. Serial ATA is scalable and will allow future enhancements to the computing platform.

Questions not answered by the FAQ:

• Why the arbitrary distinction between internally and externally connected devices. Why target one and not the other? SCSI works fine for both, why not design SerialATA to do the same?
• Will I still be able to use a serial ATA device 10 years from now? I can (and do) use 10 year old SCSI devices. Will the SerialATA consortium guarantee backward compatibility, or is this yet another lock in to a perpetual upgrade cycle?

Re:FAQ from the SerialATA.org website

[TastesLikeChicken]

Why the arbitrary distinction between internally and externally connected devices. Why target one and not the other? SCSI works fine for both, why not design SerialATA to do the same?

Because drive manufacurers still want to be able to charge a premium for thier SCSI devices.

Will I still be able to use a serial ATA device 10 years from now? I can (and do) use 10 year old SCSI devices. Will the SerialATA consortium guarantee backward compatibility, or is this yet another lock in to a perpetual upgrade cycle?

I doubt we'll still be using harddrives in 10 years. I suspect SerialATA will kill SCSI and trying to get info off of old hard drives will be like trying to get information off of old 5 1/2" floppies.

Re:FAQ from the SerialATA.org website

[Tet]

I doubt we'll still be using harddrives in 10 years. I suspect SerialATA will kill SCSI and trying to get info off of old hard drives will be like trying to get information off of old 5 1/2" floppies.

Possible. But SerialATA isn't just aimed at hard drives. It's also designed to encompass tapes, CDs, DVDs and other high capacity storage devices (both fixed and removable). I think the chances of me still wanting to be able to access my CDs and DVDs in 10 years time is extremely high.

Re:FAQ from the SerialATA.org website

[p3d0]

Why the arbitrary distinction between internally and externally connected devices.

I don't think it's so arbitrary--one important difference is distance. Serial ATA goes to one meter, while (for instance) USB goes to about three meters. Less distance means you can get more throughput out of cheaper devices.

Re:FAQ from the SerialATA.org website

[Ted Nitz]

The distinction between internal and external drives is partally due to electrical problems with making external drives, at 1.5Gbps a 1m cable looks a whole lot like a transmission antenna....

Yes, SerialATA does have provisions for backwards compatability between the current Gen (whatever that may be) and earlier Gens, There are of course situations where backwards compatability cannot be maintained, but they're moderatly few.

-Ted

Re:FAQ from the SerialATA.org website

[Datafage]

Um, yeah, and you know what, in 10 years when we have SATA3 the CD/DVD specs will be exactly the same, and the SATA3 CD/DVD drives will be able to read your old discs. The only issue is with hard drives because the reader is integrated with the media.

Re:FAQ from the SerialATA.org website

[Anonymous Coward]

Then continue to use your old drives, there should still be your old PCI cards to use them. Also, I hear there will be adapters in the works to the old parallel drives to use the serial lines.

Re:FAQ from the SerialATA.org website

[Anonymous Coward]

1) A standard for internal devices can be much simpler, since the cable lengths are shorter, and the variety of devices is far less. Firewire has to be far more complex to handle things like no-host transfers and a huge variety of devices.

If you want to bitch about something, ask why the hell we have USB2 when firewire does the same job better. USB1, certainly has its place for mice, keyboards and other low bandwidth devices.

2) I still use a 202meg ATA HD from about 10 years ago on a modern controller. ATA is pretty good about backwards compatitibility. There will soon be adaptors to allow you to hook your parallel ATA devices up to your new SATA controller.

Re:FAQ from the SerialATA.org website

[flatrock]

Why the arbitrary distinction between internally and externally connected devices. Why target one and not the other? SCSI works fine for both, why not design SerialATA to do the same?

One meter cable length doesn't work real well for external devices, and for external devices you want to be able to hot plug the device like you can with Firewire or USB. Those interfaces are better designed for most external storage devices. It is a lot easier to design an interface that doesn't have to worry about drives disappearing and reappearing.

Will I still be able to use a serial ATA device 10 years from now? I can (and do) use 10 year old SCSI devices. Will the SerialATA consortium guarantee backward compatibility, or is this yet another lock in to a perpetual upgrade cycle?

I'm not sure how the people writing the standards can guarantee forward compatibility. You're asking them to say it will work with something that hasn't been invented yet. A more reasonable comparison is to ask if old ATA dirves will work with SerialATA, and they will with an adapter card.

Re:FAQ from the SerialATA.org website

[flatrock]

Ok, I got my facts strait and realized that Serial ATA is supposd to support hot plug devices. My next guess is that they're worried about FCC Class B and CE certification. The copper cables they are going to use probably won't pass the EMI limitations unless enclosed with in a computer case. I'm not sure why they couldn't propose a shielded cable for use outside the case, probably one of those product positioning, marketing decisions.

and now for something that's not slashdotted..

[cisco_rob]

and way more informative... want specs? go here

SerialATA doesn't seem very advanced

[spullara]

There are a number of issues that it seems that SerialATA doesn't address that it should:

1) Power to the device is still separated from the data connection.
2) Because it is backwards compatible with regular ATA it appears it will have the same limitations on the number of devices you can connect, i.e. 2 per channel.
3) It is unusable for external devices.

Why upgrade to a standard whose only advantage is a speed increase we don't need and smaller cables that can be done with parallel ATA ala "round" IDE cables? Seems like a huge investment that would be better made in FireWire 2.0 or something similar so that you can use the same interface internally and externally, with power provided, and have many devices on the same bus.

Re:SerialATA doesn't seem very advanced

[Junta]

Ok, let's see here:

For 1), I have heard that, indeed, power and data are together in a single connection, or at least it is tandardized to make hotswap feasible.

2) This is completely false, one of the main points of *serial* ATA was to increase the chain lengths to SCSI level capacities. The focus is on software compatibility, not transparent hardware compatibility. They say in the beginning, they expect motherboards that have Serial ATA to also have Parallel ATA on the same motherboard...

3) Why is it not usable for external devices? For one, they have extended cable length to three feet between points on the chain. No where near SCSI capability, but three feet from an interface card isn't bad. I suspect you could at the very least have SCSI solutions in ATA with this.

You are right that FireWire or USB2 might be worth a second look, but at the current rate, no one wants to bother scrapping everything they have based on ATA to pursue such a dream. I would much rather have Serial ATA than our current ATA. Of course, I have to wonder if the industry will even see this move as worth it. Even if from a software perspective it behaves similarly to ATA, I would think the hardware implementers have been holdig back. ATA is seen to meet the demands of home users, and SCSI supplies advanced features to businesses that need it. Hardware vendors have a vested interest in maintaining that dicotomy, since they can charge a huge premium for SCSI without problems coming up in the Desktop market...

Re:SerialATA doesn't seem very advanced

[Junta]

Open mouth, insert foot... My statement that *serial* ata definitely meant lengthening the chain is false, it seems that hardware compatibility is a significant issue, but it seems that at least initially chains are likely not to increase significant. 1) and 3) still stand though :)

Re:SerialATA doesn't seem very advanced

[saider]

...smaller cables that can be done with parallel ATA ala "round" IDE cables?

I was under the impression that bundling the cable created a huge amount of crosstalk between the various data lines. This could be solved with some shielding, but this would make the cable rather inflexible and difficult to work with. External SCSI cables are a good example of this. I have enough trouble working these things around outside of the box, let alone inside a confined space.

Firewire is good, but there are licensing issues which does not help produce a low cost solution. Remember ATA is for cheap drives. You'll still be able to buy firewire drives, they'll just be more expensive.

Re:SerialATA doesn't seem very advanced

[jbrauer]

From 1, you can't hot swap with a cable. Hot-swap requires a backplane. How you get power to the backplane is up to you. They have two connector styles. One standardizes the location of the old style power connectors, then they provide a single connector for power/signal, just like SCSI did, but they do not run power off of the adapter card and into the signal cable. If you are designing a backplane for drives, then you'd use the same connector, but that is different from the same cable.

2. This is incorrect, the signalling is shorter, and sATA does not want to compete with SCSI, in the future, SCSI and sATA will use the same interconnect and you choose based on how much money you want to spend. SCSI runs 25 meters point to point while sATA is 1m.

3. Nope, the distance is still 1m. No extra length. It is possible to change the spec a bit and give distance, but what would you run it to. Since it is point-to-point, and the software stacks can't handle more than two devices on a bus, what would you connect but one drive, or possibly two with a special chip?

SCSI is offering a compatible solution at http://www.serialattachedscsi.com

Re:SerialATA doesn't seem very advanced

[Anonymous Coward]

You don't want power on the same line...it creates noise!

Plus, the primary advantage of using SATA is that you can chain drives together...now then, if you have 5 high RPM drives spinning up at the same time all feeding off of the little shared line, that's not only a LOT of noise and a LOT of power draw (realatively) to have to shield against, you also run the risk of not being able to provide enough power to get them all to spin up. Makes for expensive and hard to use cables. Hardly ideal.

This very reason is why many drives (currently) allow for staggard spin ups as well flexibility in where each drive is obtaining it's power (seporate plug).

In an ideal (dual power supply or more) situation, half of your drives are each getting their power for a different power supply...

Re:SerialATA doesn't seem very advanced

[Junta]

to 3), I was under the impression that the "official" spec was 18" for traditional ATA, though 36" cables are available, they are not compliant with the SPEC. I still says that 36" (the sATA spec) is enough room from the adapter to make an external device possible, just not nearly as flexibile as SCSI external devices. In my experience, SCSI devices are typically placed frequently right next to the computer that uses it, well within 3 ft of the connector..

Re:SerialATA doesn't seem very advanced

[jbrauer]

Ah, it isn't the distance that is important, but the ability of the signalling layer to have external quality connectors, and EMI tolerances. While differential, you still have to worry about EMI creation. Further, 1m is the total distance, without repeaters and external termination, that is 1m from the chip to the disks chip, or close to it. Most SCSI connectors are run almost a foot across a card/mb, there goes one foot. Then you account for the fact that there is probably another 6" lost going to the device via connectors, which btw more than 2 of lower your distance to around 2 feet. Once you add up all the extra impediance changes and the board run lenghts, you really have between 1 foot an 18 inchs of external connect, assuming a cable which is rated for out of box connections and some form of connector specificaion.

Given that the physical layer of sATA is very similar to that of FC, there is no reason that the signals could not be boosted to do 15ms though. The big question is what would you hook up external, with out a hub of some kind, which the sATA 1.0 spec most definitely does not address, there is only the possability for one external device.

Re:SerialATA doesn't seem very advanced

[Krieger]

As much as I hate to say it, obviously you didn't read any of the background material. You are right about the hot swap and backplanes.

However the Maxtor presentation talks about using SATA as a replacement for SCSI, or at a minimum breaking into NAS and low end servers.

Cable length is up to 3M from the Intel presentation.

Also for the number of devices, since it's now PTP connections, it's relatively agnostic as to the number of devices since it no longer fits into the old model of channels. I still haven't found specific references for the number of devices, but the Maxtor presentation has a picture of a SATA drop-in PCI card with 6 SATA connectors.

Re:SerialATA doesn't seem very advanced

[gunix]

Well... Serial ATA 2 will be much nicer... I don't remember where I read about it... but it stated something about hotswapp... it sounded almost like SCSI+IDE=Serial ATA 2

Re:SerialATA doesn't seem very advanced

[edmudama]

> There are a number of issues that it seems that > SerialATA doesn't address that it should:

> 1) Power to the device is still separated from the data connection.

This is because the motors being currently used in hard drives pull a few amps at spinup time, and the wire guage used for signalling cannot possibly carry this much current. There is a thought that once the market is serial ATA native, the HD manufacturers will then standardize on a 5V, low current motor instead of the current 12V beasts, however, things like Microsoft's "on now" spec and other crap specify a minimum spinup time that force us to slam the motor to get up to speed.

Single connector vs two connectors has little bearing on whether you can hot swap. It is a function of how you isolate/protect your power circuitry from not having all the conductors touching at the same time. (e.g., in current ATA, what happens when of the +12,0,+5,-5, only the +12 and the -5 are connected because the pins are slightly out of tolerance?)

> 2) Because it is backwards compatible with
> regular ATA it appears it will have the same
> limitations on the number of devices you can
> connect, i.e. 2 per channel.

It is point to point. The notion of channels will disappear, and BIOSs in the future will simply allocate an 8-word I/O space address for the device, instead of todays "primary IDE" at 0x1F0 or whatever.

> 3) It is unusable for external devices

I don't believe that is correct, however, since it is point to point, a box of external drives (similar to a SCSI enclosure) would need a cable running to it for every drive in the enclosure.

There will be no daisy chaining or hub or star network of SATA devices.

Re:SerialATA doesn't seem very advanced

[Anonymous Coward]

It is pretty easy to go at much faster speed with a coax cable than a kludge like a round cable.

I would guess that power is separate because the power cable goes directly from the power supply to the drive not via the motherboard. Going to the motherboard requires routing the +12V which might required additional routing area on the board on a already congested 4 layer PCB.

Re:SerialATA doesn't seem very advanced

[stripes]

Also for the number of devices, since it's now PTP connections, it's relatively agnostic as to the number of devices since it no longer fits into the old model of channels. I still haven't found specific references for the number of devices, but the Maxtor presentation has a picture of a SATA drop-in PCI card with 6 SATA connectors.

But...how is a 6 drive SATA controller 100% software compatible with a older ATA spec that supported exactly two drives? Does the one SATA controller show up as 3 (or six) "normal" ATA controllers? Or is 100% really 100% except when you have more then two devices, send more then X bps, it is a day ending in Y, or the pope is waring a funny hat?

And isn't it sad that "requires no changes to Windows" has started being a major hardware feature?

Re:SerialATA doesn't seem very advanced

[Anonymous Coward]

Is Windows really the problem? Probably not on new systems. Major issue still seems to be DOS and the crappy PC BIOS. Got to be register-compatible with those PC ATs.

Re:SerialATA doesn't seem very advanced

[Krieger]

Note that it is software compatible, not hardware compatible. The two devices per channel was a hardware IDE thing. What is likely now is that they will show up as independant ATAPI devices. I am curious how the host card will register, but most likely you will get six independant ATAPI devices (assuming you use all six) and a single controller card. Again I urge you to read the various documentation as it does address most of this.

And yes it is sad. ;) Hell it's sad that they haven't found a way to phase out the BIOS yet.

Re:SerialATA doesn't seem very advanced

[MoneyT]

You are right that FireWire or USB2 might be worth a second look, but at the current rate, no one wants to bother scrapping everything they have based on ATA to pursue such a dream. I would much rather have Serial ATA than our current ATA. Of course, I have to wonder if the industry will even see this move as worth it. Even if from a software perspective it behaves similarly to ATA, I would think the hardware implementers have been holdig back. ATA is seen to meet the demands of home users, and SCSI supplies advanced features to businesses that need it. Hardware vendors have a vested interest in maintaining that dicotomy, since they can charge a huge premium for SCSI without problems coming up in the Desktop market...

Just wait until apple replaces all their ATA stuff with firewire, with no backward compatability built in (see iMacs and USB circa 1998). Then the PC manufacturers will begin including firewire faster than you can say "http:///..org"

Personaly, I'd rather see SCSI drives than a new version of ATA.

Re:SerialATA doesn't seem very advanced

[flatrock]

Just wait until apple replaces all their ATA stuff with firewire, with no backward compatability built in (see iMacs and USB circa 1998). Then the PC manufacturers will begin including firewire faster than you can say "http:///..org"

Since when have PC manufacturers followed what Apple has done, other than cosmeticly. There are also royalties to be paid on Firewire, but not on Serial ATA. Of course since Apple is one of the developers of Firewire, they're paying most of that royalty to themselves. If i remember correctly it was a pretty small royalty though, maybe a little under $1 a device, might even be less.

Personaly, I'd rather see SCSI drives than a new version of ATA.

If you want the benefits of SCSI, you're welcome to pay the extra that it costs. Most people won't notice any difference in their system from going from IDE to SCSI, why should they pay for SCSI.

Re:SerialATA doesn't seem very advanced

[MoneyT]

In terms of technology, espesialy when it comes to addopting new standards, PC's have looked very closely at what Apple has been doing. Notice the influx of USB and Firewire (IEEE 1394) devices recently? How about mini computers? Dual processors? Yes both PC and Mac have had DP systems for a while, but it wasn't until the G4 systems that DPs started to be marketed to the consumer and not just to the geek.

I'm not saying everything Apple does the PC manufacturers take (NuBus and SCSI never really caught on in the PC market, though Sun did quite a bit with SCSI) but Apple does have a decent amount of influence in the tech world, even if they don't have a lot of users

Re:SerialATA doesn't seem very advanced

[Anonymous Coward]

Since when exactly did IDE devices need -5 V ???

Re:SerialATA doesn't seem very advanced

[shoemakc]

Yeah, but a 5V motor would draw MORE current, wouldn't it?

This is because the motors being currently used in hard drives pull a few amps at spinup time, and the wire guage used for signalling cannot possibly carry this much current. There is a thought that once the market is serial ATA native, the HD manufacturers will then standardize on a 5V, low current motor instead of the current 12V beasts

Re:SerialATA doesn't seem very advanced

[flatrock]

USB was available on PCs before Macs. Apple waited a bit until it became a more accepted standard and then pushed it along by using it as the only option for the mouse and keyboard. The first all-in-one may have been the Apple Macintosh (not the iMac). PC vendors such as Compaq tried to compete in this market, but it never really took off for PC. Expandability seemed to be more important to PC customers. The iMac was definately a great success due to exellent marketing efforts by Apple. PC manufactures definitely tried to copy this with limited success at best.
I must be missing the push to sell Dual Processors systems to consumers rather than just Geeks. I agree that Dual Processor MACs are nice machines, we have one here at work running LINUX. I just really don't see them being marketed to home users, or dual processor PCs for that matter.
SCSI has always been a higher end technology. The advantages make more of a difference to high end users, and issues like termination and SCSI IDs confuse casual users.
I think the arrival of a larger number of Firewire devices is more attributable to it's use on digital camcorders which are getting much more popular, than it's use on MACs, but it is definately a technology that Apple developed. I'd actually have to say that both USB and Firewire have their roots in the good old Apple Desktop Bus, which though extremely slow, was way ahead of it's time compared to the PS/2 ports on a PC.
Apple has developed some interesting technology over the years. But that technology usually takes a long time to catch on in the PC community. Apple also seems to take it's time to adopt many things that are accepted as standard in the PC community, other than trying to cash in on some of Steve Jobs marketing sucesses.

it also...

[Anonymous Coward]

-replaces the long ribbon cable with a nice thin "serial" cable"
-replaces the seperate power connector and integrates it with the data cable
-standardizes the location of the data/power plugs
-allows for hotswap(partly because the location of the plugs are now standard)

600 mb/s??!?

[n3r0.m4dski11z]

Wow this looks really hype. at least 2wice as fast as U320 SCSI. but as with most new technology you really have to have price and market share. if the big mobo manufacturers (asus, abit, tyan) build this onboard their machines and the hdd manufacturers make new drives, this coudl easily squash ata 133 and scsi. in fact if it coudl support up to 15 devices scsi would have no benifits. there is so little info though its hard to tell. this could be really exciting stuff. i remember people sayign the same thing about firewire though and i have yet to see a total firewire system.

Re:600 mb/s??!?

[Anonymous Coward]

SCSI still has many advantages over SATA. SCSI can communicate with multiple drives at once, the article doesn't mention if SATA fixes the current generation of ATA's ability to communicate with only 1 drive at a time. SCSI is a tried and tested, reliable standard, SATA could not aquire that in any less than 4 years, IMHO. Adaptech released a white paper explaing the core diffeences between USB, FW/IEEE1394, ATA, and SCSI. None of these technologies can realisticly compete with each other based on latency issues,multiple drive communication issues, etc.

CPRM on ATA?

[ryanvm]

For those who are still wondering about CPRM on the Serial ATA spec, these documents may be of some use.

Re:CPRM on ATA?

[Anonymous Coward]

CPRM was going to be put on the vanilla ATA spec.

oh joy, yet another "standard"

[jspectre]

to compete with the firewire (and upcoming gigawire) "standard" and the usb 1.x and 2.x "standards". oh and we have ata/133/100/66 "standards". scsi 1/2/3/4/5/ultra/wide/thin/mega-super-fun "standards" too.

how come none of my "standard" devices talk to each other very well?

well three cheers for the latest "standard". by the time it's on everyone's hardware it will be superceeded a dozen times or so.

Re:oh joy, yet another "standard"

[Conspiracy Theorist]

As I've always said: "Standards are good; everyone should have one."

Re:oh joy, yet another "standard"

[Anonymous Coward]

Hey, where can I get one of those new fangled scsi mega-super fun drives? :-)

Re:oh joy, yet another "standard"

[Wdomburg]

>to compete with the firewire (and upcoming
>gigawire) "standard" and the usb 1.x and 2.x
>"standards". oh and we have ata/133/100/66
>"standards". scsi
>1/2/3/4/5/ultra/wide/thin/mega-super-fun
>"standards" too.
>
>how come none of my "standard" devices talk to
>each other very well?

How do SCSI devices not talk to one another well? I've run SCSI-1 devices on Ultra2 controllers and vice versa without a single glich. Likewise, I've run ATA/33 devices on ATA/100 controllers and vice versa without a single glich.

I have little experience with either USB or Firewire, but I've never heard of problems with backwords compatibility on either of them.

Matt

Re:oh joy, yet another "standard"

[Anonymous Coward]

The great thing about standards is that there are so many to choose from.

Re:oh joy, yet another "standard"

[Anonymous Coward]

So would you rather have each company have 20 versions of 50 different protocols?

VIA motherboards only talk to IBM hard drives?
SiS only works with mice from Logitech?
If you're using an Intel chipset, you'd better have an NEC CD-ROM drive!

Standards are good.

Re:oh joy, yet another "standard"

[jspectre]

well consider yourself lucky. over the past 20 or so years i've had tons of "scsi voodoo" where devices wouldn't show for no apparent reason. chains that needed no, one or two terminators on them for strange reasons. two "identical" setups that didn't work the same.

both usb and firewire were supposed to be issue free yet there are many people who have strange problems with both. i know if i plug my harmon kardon usb speakers into my usb hub i lose my intellimouse and zip drive. some times. some times it works just fine. i can't plug my firewire hard drive into my cdrw and then into my powerbook, my cdrw has to be at the end of the chain. no idea why.

for what it's worth, there are no native firewire hard drives out there, they all are ide (or scsi) with bridge boards. doubt we'll ever see a native usb 2.0 drive. looks like the standards were very well accepted. :-( kudos for sony for at least starting a trend putting fire^H^H^H^Hi.link ports on their cameras

Re:oh joy, yet another "standard"

[brain159]

Addition:

If you're using an athlon board based around the VIA KT133 chipset and you're a happy windows xp user (don't laugh - we do exist), don't expect to use a common webcam like the Logitech Quickcam Express (among many others), because it won't work.

(yes, I know VIA chipsets blow. At least, I do now... *G*)

Re:oh joy, yet another "standard"

[Wdomburg]

Maybe you're just unlucky :) I worked support at a large company, mostly on Unix systems. So I've walked people through SCSI configurations on HP/9000s, Sparcs, Vaxen, Alphas, PeeCees, Macs running HP/UX, SunOS, Solaris, Digital Unix, OSF/1, Openserver, Unixware, SCO 3.2v4.2, Xenix, Linux, FreeBSD, Windows, MacOS... I think the only really weird problem I saw beyond driver issues was one machine which would only work unterminated.

I any case, its not perfect, but considering the SCSI standard deals with a mish-mosh of signaling, ranging from 5MHz to 80MHz, single ended or differential, 8 or 16 bit, single or double pumped, synchronous or asynchronous operation... Not to mention various extentions to the protocal including packetized scsi, domain validation, etc, etc...

Honestly, I'm amazed that they've managed to push the speed up well over 100x what the original 8-bit narrow async standard pushed, and you can still slap on an adapter and have a damn good chance of it working.

Matt

Seems /.ed... Here's the article:

[mnordstr]

(no pictures! :)

Serial ATA, A New Standard
Serial Advanced Technology A ttachment is an evolutionary high-speed serial link replacement for the parallel ATA attachment of internal storage devices. It connects hard disks, DVDs, CD-R/Ws, zip drives and all other ATA and ATAPI devices to the motherboard in desktop and mobile PCs, servers and network storage.

The new point-to-point device communicates trough a 4 layer interface:

4) Application Layer 3) Transport Layer 2) Link Layer 1) Physical Layer

If you want more information, you can download the Serial ATA 1.0 specifications at http://www.serialata.org

The reason for the late breakthrough of Serial ATA is that nobody wanted to, unnecessarily, spend time and money, developing a new interface. Because of the higher performance demands, several companies had to cooperate developing a much better, more intelligent ATA: Serial ATA will allow these future enhancements to the computing platform.
The Serial ATA working group that developed the Serial ATA specification was led by APT Technologies Inc, Dell Computer Corporation, International Business Machines (IBM), Intel Corporation, Maxtor Corporation, and Seagate Technology.
Seagate and Maxtor told us the new drives would be shipping in autumn, but the real breakthrough will take place when chipset manufacturers (Intel/Via) have integrated Serial ATA on their Southbridge.

Benefits
Let's take a look at some end user benefits of Serial ATA:

No software depency, it's 100% compatible with today's software and OS transparent.

Easier configuration of the storage devices (jumpers are no longer needed).

Supports lower cost device architectures.

Much better cabling and connectors: the thin and flexible cables result in better airflow trough the pc housing and enables design of smaller PC/motherboard chassis. Therefore, they are simple to route and install, and can be up to 1m long.

Last but not least, a higher bandwidth: the transfer rate exceeds all current ATA standards. Generation 1, 2 and 3 S-ATA supports respectively up to 150MB/s (1.2Gbits/sec), 300MB/s (2.4Gbits/sec), 600MB/s (4.8Gbits/sec). As Serial ATA works asynchronous, there are no isochronous requirements.

The Prototypes
Seagate's SATA prototype:

Maxtor's prototype:

When we take a closer look at the connectors, we see there's a slight difference: Maxtor used the 35B1 configuration, starting right: the Serial ATA connector, with both power and signal segments, legacy jumper and power connector. Seagate used the 35B4 configuration which has no legacy power connector.

The Serial ATA signal segment counts seven pins: three ground pins a transmitter signal pair and a receiver signal pair. The Serial ATA power segment counts 15 pins, containing three different voltages: 3.3V, 5V and 12V.

The prototypes momentary shown, need a PCI-to-SATA host controller or a SATA-to-Parallel ATA bridge chip.

Maxtor used the first one:

The Future
By the end of 2002, there should be SATA-sytems on the market. As you might have noticed, the parallel ATA is finally dead, though it could take up to four years to eliminata all parallel ATA devices. Nevertheless, I'm going to wait buying a new system, because most new technology suffers childhood disease.

Re:Seems /.ed... Here's the article:

[El_Nofx]

Layer 4 is the Transport Layer, not the Application layer.

1. Physical
2. DataLink
3. Network
4. Transport
5. Session
6. Presentation
7. Application

I just remember that from my Networking Classes in college, All People Seem To Need Data Processing.

Re:Seems /.ed... Here's the article:

[Anonymous Coward]

That's for tcp/ip networking, tcp/ip networking doesn't even need all those layers

Re:Seems /.ed... Here's the article:

[Kanasta]

Whoa, 2 concerns:

1) It's not even out yet and already there are at least 2 connectors? 35B1/35B4

Reminds me of 5-6yrs ago. DVD standard being finalised, and we had DVD-RAM and DVD-RW. Fast forward to now, and we have _3_ writable 'standards' with no end in sight.

Will this happen to S ATA?

2) P ATA is dead, S ATA will have teething problems no doubt.

Looks like I won't be able to buy a new HD (or any ATA drive) for many, many years. :)

how many devices ?

[overlord]

I think one problem is the number of devices in
a single PC. If this serial works as USB (but in this case for disk and renovable media) would be
a great success.

OverLord

The FAQ

[TheFlu]

Taken from the Serial ATA website:

Q1:What is Serial ATA and Why is it being developed?
A1: Serial ATA is an evolutionary replacement for the Parallel ATA physical storage interface. Serial ATA is scalable and will allow future enhancements to the computingplatform.

Q2: Previous efforts to transition to a serial bus were not successful. Why do you believe that Serial ATA will be successful?
A2: Serial ATA is a drop-in solution in that it is compatible with today's software, which will run on the new architecture without modifacation. It will provide for systems which are easier to design, with cables that are simple to route and install, smaller cable connectors, improve silicon design, and lower voltages which alleviate current design requirements in Parallel ATA.

Q3: Will there still be a parallel ATA bus when Serial ATA comes out?
A3: Serial ATA's adoption by the industry will follow a phased transition path. Therewill be a point where both Parallel and Serial ATA capabilities are available.

Q4: You stated that PCs implementing Serial ATA will be in the marketplace in2002. Why does it take so long to implement?
A4: The goal of the working group is to ensure the Serial ATA transition happens as smoothly and quickly as possible. The Serial ATA specification is expected to becomplete in the fall of 2000 with adoption to happen in the following 12 to 18 months.

Q5: What are the end user benefit of Serial ATA?
A5: End users will benefit by being able to easily upgrade their storage devices. Configuration of Serial ATA devices will be much simpler, with many of today's requirements on jumper and settings no longer needed.

Q6: What is the cost to implement Serial ATA in a system?
A6: The cost of Serial ATA technology will be on par with today's Parallel ATAtechnology.

Q7: Who are the members of the Serial ATA Working Group? Can new companies join?
A7: The Serial ATA promoters group includes APT Technologies Inc, Dell Computer Corporation, International Business Machines, Intel Corporation, Maxtor Corporation, Quantum Corporation, and Seagate Technology. Information on joining the working group is available at www.serialata.org and new members are welcome.

Q8: Hard disk data rates don't seem to be pushing the limits of current ATA66technology. Why is Serial ATA being planned now?
A8: Serial ATA is an evolutionary replacement for the Parallel ATA physical storageinterface and will allow future enhancements to the computing platform. Specifically, thethinner Serial ATA cable addresses OEM's concerns regarding airflow around the Parallel ATA cable, and enables design of smaller PC chassis, as well as silicon vendors concerns regarding 5 volt tolerance support in future designs.

Q9: Will Serial ATA be compatible with today'sPCs?
A9:Serial ATA electronics and connectors will differ from Parallel ATA, however the technology is software compatible and OS transparent. It is anticipated that there will be adapters to facilitate forward- and backward-compatibility of hard disks on PC systems.

Q10: What is the impact of Serial ATA on OEMs?
A10: Industry benefits of Serial ATA include systems which are easier to design withcables that are simple to route and install, smaller cable connectors with improved silicondesign, lower voltage which alleviates current design requirements in Parallel ATA and compatibility with today's software which will run on the new architecture withoutmodification.

Q11: Beyond hard disks, will Serial ATA be used on floppy drives, optical drives,DVDs, and ZIP drives?
A11: Serial ATA supports all ATA and ATAPI devices, including CDs, DVDs, tapesdevices, high capacity removeable devices, zip drives, and CDRW's.

Q12: What is the impact of Serial ATA on IEEE1394 (aka Firewire) and on USB2in terms of PC system function?
A12: Serial ATA is planned to be the primary storage interface inside the PC system, and is not planned as an external interface to PC storage or peripherals. USB2 and IEEE1394 connections on the PC can be used where required as peripheral interfaces.

Q13: When does Microsoft plan to support Serial ATA in its OS's?
A13: Serial ATA is software compatible with Parallel ATA and requires no changes toMicrosoft operating systems, or any other OS as well.

Q14: What are the licensing requirements and costs of Serial ATA to companiesthat want to use the technology?
A14: When the Serial ATA specification is complete, it will be made available at nocharge. The working group expects to complete the specification later this year.

Re:Broken

[Anonymous Coward]

Ah, the joys of MySQL shittiness.

Do we really need this thing?

[cholokoy]

For the average home user, the present size of hard disks available for his/her choosing is way too much for the needs. Who wants this thing - mostly the geeky types who want to show off their stuff to friends or maybe those that really want to squeeze performace from their systems and I suspect not too many.

How many people would want to make raid systems or install several hundred GB of storage when a 40 or 60 GB HD will do?

Heck i'm still using pentium 100 systems for a home firewall and its more than enough. How many really run their systems to the ground with games, databases, or simulation? Not much so this is the only market that will be looking at this new tech.

For servers though this might be compeling but if scsi keeps up, it will have a hard time breaking their superiority.

Re:Do we really need this thing?

[Anonymous Coward]

Yes, for crying out loud, we do need these advances. Sometimes advances turn out to be good things. Imagine that!

If people like you were in charge we'd all be stuck in the stone age of computing, banging on stone tablets (hard drives) with chisels.

Wait a sec... you were trolling! Ha-ha!

Re:Do we really need this thing?

[AlexDeGruven]

I think this would be a pretty good thing for gamers. Not even the ones that run their systems into the ground. The faster you can access the data on the CD/HDD/DVD, the better the game runs. There's nothing worse than waiting the extra time for a cut scene to load. For me, it totally throws off the flow of the game.

For servers though this might be compeling but if scsi keeps up, it will have a hard time breaking their superiority.

This type of technology would be a great boon for the low-mid end server market. Especially if the cost is the same/comparable to existing low-end server hardware.

Re:Do we really need this thing?

[Anonymous Coward]

I have 240G on my PC and looking for more. The reason is the proposed CDR tax, SSSAC etc. I want to be able to ride out the confusion for the next little while.

At what I would be paying for any mass store media in a year in Canada, RAID seems to make the most sense.

Look's like she's cooked :-�

[vlag]

Warning: Can't create a new thread (errno 11). If you are not out of available memory, you can consult the manual for a possible OS-dependent bug in /home/sites/site84/web/class_db.php on line 7

Warning: MySQL Connection Failed: Can't create a new thread (errno 11). If you are not out of available memory, you can consult the manual for a possible OS-dependent bug in /home/sites/site84/web/class_db.php on line 7

ERROR: Unable to connect to database.

Damn my prejudice

[Anonymous Coward]

I guess the C64 & 1541 scarred me for life, because whenever I hear about hooking up drives via any sort of serial setup, I instinctively think ,"That must be awefully slow." Yes, I know it doesn't make sense.

Re:Damn my prejudice

[Anonymous Coward]

The C64 serial bus was a very good standard. Its speed was the result of a minimalist hardware interface which consisted of open collector TTL drivers and software bit twiddling to feed the line. It could have been way faster without much effort at all (say 10 Mbits/s). The main limitation was that the processors that were connected to the bus were running at about 1 MHz. They could not handle anything much faster.

There is a good analogy similar to the ATA/Serial-ATA discussion at hand. The C64 serial bus was the serial version of the parallel IEEE-488 instrument bus (as also found on Hewlett-Packard test equipment). The earlier Commodore Pet actually used the IEEE-488 parallel bus.

What strikes me as very silly is how we truly reinvent the wheel every few years. Both the SCSI and IEEE-488 bus are very sophisiticated and well thought out advanced architectures. And of course there is the NIST standard IBM System/360 data channel which predated even them. The ATA interface started out simple. When it ran into limitations it was forced to answer the same questions as earlier bus architectures. So now this whole complex ATA spec has evolved, essentially duplicating what previous generations of bus designs have already accomplished. Sheesh.

Imagine...

[Anonymous Coward]

...a Beowulf cluster of one of these next to Natalie Portman and Alan Thicke smothered in hot grits!!!

.it .doesn't .work .anymore .because .kl erk .is .dead .from .a .semen .overdose .it .doesn 't .work .anymore .because .klerk .is .dead .from . a .semen .overdose .it .doesn't .work .anymore .be cause .klerk .is .dead .from .a .semen .overdose . it .doesn't .work .anymore .because .klerk .is .de ad .from .a .semen .overdose .PAGE .WIDENING .SHAL L .RETURN .LIKE .A .MIGHTY .PHOENIX .and .shall .c onquer .you .smelly .GNU .hippies

FireWire is already on the ball.

[elocutio]

Other than the touted 600Mb data rate versus 400Mb for FireWire, I don't see any real advantages to this new standard. FireWire (aka IEEE 1394 is already planned for use in internal devices, and it looks like a competition footrace may emerge between SATA and FireWire. Because of Apple's noted efforts in founding the FireWire project, It will be interesting to see what side Microsoft will support the strongest.

Re:FireWire is already on the ball.

[hattig]

Well, with 1394B coming RSN with support for 1600Mbps (200MB/s) I think that this is more than enough for most home systems in terms of hard disk interface speed for the next couple of years at least. Future enhancements to Firewire will, of course, push that speed up to 3200MBps, and possibly further.

Firewire could support up to 63 internal hard drives, although they would each have to share the same bus. SerialATA allows for chaining of drives, but doesn't mention how many can be chained. I imagine that motherboards will have 2 SerialATA channels late this year, and 4 at some point next year - allowing SATA RAID 5 (3+1 drives) at a combined throughput of 600MBps...

But SerialATA has one major advantage. To the OS, is looks like a standard ATA interface - no driver changes are required at all. This isn't the case with Firewire currently. Hence it will succeed in my opinion (as well as being on every motherboard come 2003).

Re:FireWire is already on the ball.

[Anonymous Coward]

Sorry, these numbers are off. That is 6000Mb or 600MB(bytes not bits) vs. 400Mb(bits) or 50MBs for FireWire.

Re:FireWire is already on the ball.

[Anonymous Coward]

Um...actually the data rate specified for Serial ATA was going to 600MB/sec, with the initial spec. calling for 1200Mb/sec MB=Megabyte...Mb=Megabit

Re:FireWire is already on the ball.

[Wdomburg]

>Other than the touted 600Mb data rate versus 400Mb
>for FireWire [apple.com],

It's 600MB, not 600Mb, and that's in a future revision. The initial devices will be 150MB vs 400Mb on firewire, or about three times faster than firewire.

Matt

Re:FireWire is already on the ball.

[Anonymous Coward]

Hey! Firewire is 400megabytes/sec, cause steve said so!!

IT's a good thing

[ruiner13]

What, now that Intel says firewire is dead, we need this. I mean firewire isn't on any devices, and isn't a proven technology. And this standard can work without a computer host, like firewire. And this supplies power on the data cable. Oh, wait, intel is dreaming again.

Why is it faster?

[DeltaOne]

Could someone explain me why a serial interface is faster than a parallel one?

I thought that sending bits in parellel would be faster since you can send a bunch of them in one pack, instead of getting them one by one.

Of course, if you process the bits at each end of the wire one bit at a time, there is no gain, but then, shouldn't you work on the controller so that it works fully in parallel too?

Re:Why is it faster?

[blackwings]

The reason that there is a movement from parallel standards towards serial standards is not because serial transfer is faster, in most cases it is not, a parallel solution will in most cases always be faster (unless the time it takes to de-paralize the data is to high), the problem with paralell transfer at hig speed is isolation.

When you transfer data at high speed with cheep cables like the normal IDE once, the signals in the different cables tend to polute each other.

Three different solution exists to this problem:

1. We could build better cables (like the ones you usualy have with Ultra2/Ultra160/320 scsi interface).

2. We could send the data with cheap cables but with a better error correction. (like is done with ATA66/100/133)

3. We could develop a serial interface.(Like Serial ATA)

Re:Why is it faster?

[Anonymous Coward]

Intuitively, its not.
n wires carrying data gives n times more info than one wire. Trouble is, synchronization - skew of the data with wires not being exactly the same length / quality, etc.. So when you ramp up the rate, you're more prone to skewing, etc. Serial doesn't have that problem. Look at the length restrictions on ribbon cables for examples - after so long, the skew hits you. And all those extra wires aren't data - theres control signals in there, too.

Re:Why is it faster?

[Pussy Is Money]

With a parallel bus you can only send signals so fast before you hit clock skew problems: at high frequencies it becomes difficult (read: expensive) to ensure that all the wires remain in sync. A serial bus allow much higher frequencies (at the same cost), plus it saves a lot of space (more cost savings).

Re:Why is it faster?

[Anonymous Coward]

In simple term:

When you are running them at a fast speed, then you have to start looking at the skew (i.e. difference in arrival time) of the data bits etc... to sync them up. This skew budget can really eat into your data window.

Re:Why is it faster?

[DeltaOne]

Thanks for the reply!

It's been helpful.

Serial is a bit soggy...

[Anonymous Coward]

Now what is so special about serial ATA? I don't see anyone bragging about their serial mouse...

Re:Look's like she's cooked :-�

[Anonymous Coward]

At least it said what's wrong. If that would have been a Windows server, it wouldn't physically exist anymore!

Serial ATA could REALLY cut into SCSI sales

[MtViewGuy]

I think if you're willing to live with the four-device limitation of Serial ATA, the new standard will have a number of great benefits.

First, Serial ATA will offer a major leap up in terms of transfer speeds. The initial speeds will be equivalent of ATA-150, but we will eventually see the equivalent of ATA-300 and ATA-600 speeds, which far surpasses even the current Ultra-Wide SCSI 160 standard. I'm sure companies like Promise are working on RAID controllers for Serial ATA that will allow an even bigger leap up in hard disk performance.

Second, because Serial ATA has finally dispensed with using a ribbon cable, this means way less clutter inside the computer and could also mean system cases can be designed for more efficient air cooling.

Third, Serial ATA--unlike SCSI--doesn't require you to load device drivers out of the wazoo to support devices on the bus. The only driver you'll probably need is the driver for the motherboard chipset that incorporates Serial ATA support.

I mean, let's face it--SCSI is still pretty expensive due to the cost of host adapters, cabling and SCSI peripherals. SCSI is only really useful where lots of SCSI device access (for example, hard disks) is required, primarily in server environments. For workstation and home computer use, Serial ATA offer most of the high-speed disk access of SCSI, but at a much lower cost.

In short, SCSI will become increasingly a niche product. And SCSI may eventually get competition from the high end of mass storage interfaces as the cost of Fibre Channel devices drops.

Re:Serial ATA could REALLY cut into SCSI sales

[paulbd]

Third, Serial ATA--unlike SCSI--doesn't require you to load device drivers out of the wazoo to support devices on the bus. The only driver you'll probably need is the driver for the motherboard chipset that incorporates Serial ATA support. this is an OS design issue. you don't have to do this with Linux. there is a single SCSI driver, based on the identity of your SCSI controller. All other SCSI devices attached to the bus are accessed using this driver. this has never really been true under Windows or MacOS, but it has nothing to do with SCSI itself, just the rather silly way developers of and for those platforms have gone about creating the driver architecture.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

It is important to remember that products cost as much as the consumer is willing to pay. SCSI costs more because it still has a performance advantage and people are willing to pay it.

ATA is cheap because a lot of it is sold, and it isn't as high performance.

FC is still very expinsive because only a few larger organizations want it, and it is difficult to administrate, but it serves a purpose.

sATA will probably steal a bit of share from SCSI in the low end server/workstation market where only 1-2 drives are used. But these servers would still benefit from SCSI, the value add though is not enough to justify the higher costs. But since SCSI past U-640 will be compatible with sATA using Serial Attached SCSI, the cost of SCSI may come down comparably.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

Under Linux, each SCSI device still has it's own device driver, not a single device driver. Under Linux, there is a single device driver for disks just like there is under windows.

Linux isn't supperior in this, if anything, given the initrd method and the way in which ordering of driver loading is specific and the bugs in RH 7.2s installer, the MS method of signing drives and not changing their names depending upon the loading of SCSI drivers gives MS a bit of an edge in this storage model.

Re:Serial ATA could REALLY cut into SCSI sales

[morbid]

Well, yes it could cut into SCSI sales, but high-end workstations and servers are using Fiber Channel nowadays, so U160 SCSI has already been superceded. I think this serial ATA stuff is just a marketing exercise - to the low end currently occupied by IDE/ATA.

Re:Serial ATA could REALLY cut into SCSI sales

[MtViewGuy]

I agree with at least your first paragraph.

Because today's bus-mastering SCSI host adapters off-load storage device access to minimize CPU cycle usage, this already means very fast access to any storage device connected to the SCSI bus. That's why SCSI is very popular for servers, where fast access to the hard drive is a must.

However, for standalone computer users, SCSI is way overkill except in a very few cases (for example, multimedia authoring). The majority of computer users don't really stand to benefit from bus-mastered SCSI, alas....

Re:Serial ATA could REALLY cut into SCSI sales

[MtViewGuy]

If you read my original message, the two biggest benefits of Serial ATA is faster disk access and dispensing with that finicky ribbon cable.

Just the switch to thin cables will definitely help the system case cooling issue.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

ATA is bus mastered, thats not the problem ... the problem is the lack of tagged queing causing lots of interrupts (or rather lack of support for it, its in the spec but noone is using it). SCSI drives also tend to have lower latency, but that is more an artificial differentiation by the manufacturers ... has little to do with the actual interface.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

Tagged queues are used by almost every drive and adapter manufacturer, and are the real advantage of SCSI. This is what allows SCSI to be multi-threaded, as opposed to ATA, which is single threaded at the hardware level.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

I disagree that high-end workstations are using FC. FC is not a good protocol for running your local disks to, almost all servers and high-end workstations still use SCSI for their disks. FC is only of benefit for clusters of people wishing to share drives. And if you think SCSI was expansive, try FC. FC is also much more difficult to adminitrate. U-160 still out sells FC by a greater than 100-1 ratio.

Re:Serial ATA could REALLY cut into SCSI sales

[pmz]

...if you're willing to live with the four-device limitation of Serial ATA...

Many of us simply can't live with the four-device limitation. SCSI and Fiber Channel shine when scalability is needed. There are many applications for multi-controller multi-drive RAID devices that ATA simply isn't cut out for. It is also nice to just be able to add another device when needed--SCSI is very convenient.

...eventually see the equivalent of ATA-300 and ATA-600 speeds, which far surpasses even the current Ultra-Wide SCSI 160

This is comparing something that doesn't exist to something that does. Also, Ultra320 SCSI is just around the corner.

Third, Serial ATA--unlike SCSI--doesn't require you to load device drivers out of the wazoo to support devices on the bus.

This is untrue. One SCSI driver allows me to connect any SCSI device: hard disks, ZIP drives, scanners, etc. The only additional drivers are those needed for non-SCSI devices, such as the parallel port or a modem.

...SCSI is still pretty expensive...

Not in the long-term. Good system administrators are more expensive than SCSI controllers, and the time and frustration saved more than pays for the SCSI controllers.

And in the home, SCSI really never had a foot-hold, so Serial ATA changes nothing.

In short, ATA never really competed with SCSI and never will. As long as ATA is crippled to be useful only in personal computers, it will never appear in big computers, multi-user computers, or high-performance workstations. These are not niche markets, either.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

Yes, thats what I said ... the problem with ATA is the lack of tagged queing support.

Its not inherent to the ATA protocol though, as I also said. Its in the ATA specs, just not in the drives and drivers.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

It's really too bad but I realize why you had to post this as AC. It's definately a +5 informative, but it will be modded to -1 (anti-Linux) in a shot. :( That's why the moderation system on /. is a complete failure and should be scrapped.

Re:Serial ATA could REALLY cut into SCSI sales

[brer_rabbit]

I mean, let's face it--SCSI is still pretty expensive due to the cost of host adapters, cabling and SCSI peripherals.

The only reason SCSI is really expensive is because Joe Consumer hasn't adopted it. If the industry went with SCSI then SCSI vendors could lower their prices. Of course, that won't happen until vendors lower their prices, a catch 22 really.

Also, I really doubt serial ATA would bite into the server market. SCSI has such a robust history (once you get it terminated correctly!), admins like to go with what they know works. Plus scsi supports tagged command queueing, which has yet to be implemented to any reasonable degree on the ATA side.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

What fucked SCSI up was the greed of the controller manufacturers. Adaptec comes to mind. SCSI was not designed to be expensive, but it turned out that way. Low end SCSI controllers were no faster than PIO ATA. And good SCSI controller were priced out of this world. The old Adaptec 1542 was over $200. It was more expensive than most motherboards with processor. And what was a 1542? A Z80 (earlier an 8085), and a ROM, and some IO drivers. Less than $5 dollars worth of silicon.

Of course even low-end SCSI is more flexible than than ATA. And thanks to LSI/Symbios/NCR, high performance SCSI controllers are no longer a high-priced Adaptec monopoly. But the damage has been done. Even Apple has abandoned SCSI. OK, I still use SCSI. After you have been spoiled by SCSI, it is hard to switch to anything else.

Re:Serial ATA could REALLY cut into SCSI sales

[stripes]

there is a single SCSI driver, based on the identity of your SCSI controller. All other SCSI devices attached to the bus are accessed using this driver. this has never really been true under Windows or MacOS

Er, that is exactly how MacOS X does it, and I think how Mac OS prior to OSX also did it.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

Another good example is Apple -- They embraced SCSI, pushed it out to 10% or so of the market, and then they sat on their asses for 10 years. Ever seen one of those G3s with built-in SCSI? It's the same 5mbs dogshit that they were shipping on the Mac IIx, with the same retard 25-pin connector.

Re:Serial ATA could REALLY cut into SCSI sales

[Anonymous Coward]

Yeah, I know. Apple used the same old NatSemi 5380 SCSI chips used in Sun3 workstations. Apple never upgraded. Oh, and typical of Apple, that 25 pin connector. What fucking cheapskates. That connector was not part of the SCSI standard. It screwed up the imepdance match something horrid. A high density SCSI-II connector would have set Apple back maybe 75 cents extra in cost above that bogus 25 pin job.

Re:Look's like she's cooked :-�

[battjt]

Interesting. I'm reading /. waiting on our DBA's to fix our SQL server.

No YOU have to remember!!!

[tenman]

"something that doesn't exactly happen every day in big business"

Big business isn't the point here. The point of ATA is to be a cheap consumer product. And, if it becomes a standard, it will only take a metter of weeks for Gateway, Dell, HP, IBM, Micron, Etc, to start shipping the new drives. Then the change starts to take over the world.

Do you think manufacturers are waiting for a gun to fire so that they can start making duplicators? Design plans for this type of equipment started as soon as the standard started to take shape. I expect to be able to buy a SATA machine from stores two or three months after the standard is approved.

I see where you are coming from, but the wheels of the electronics business wait on nothing.

Three paragraphs per page???

[Mr. Neutron]

I understand the need to maximize ad revenue, but this Heatseekerz.net is absurd! Perhaps someone can post a link about Serial ATA where we can read more than six words between page loads.

As far as the name "Serial ATA," it's a smart move. It will create the impression in people's mind that it's an "extention" or "enhancement" of standard ATA, without necessarily being backwards compatible at all. But, hey, once it gains market share, and the SATA drives start filling the shelves at Best Buy, it won't really matter.

Re:Three paragraphs per page???

[HiThere]

And the devices that use it can be called
Serial ATA Nodes.

I do wonder what would be involved in cable extender devices (relays? repeaters? hubs? whatever...).
Built-in is often nice, but especially if it's hot swappable, external drives are often quite useful. E.g., if they're cheap enough you could plug in a disk for a back-up, then remove it and put it on the shelf until it was time for it to go off-site.

Now I'll grant that one can do that with drives built into the chasis, but unless they start puttin more slots in a tower that's not a practical answer. (Currently we use an external tape drive, but as disk sizes increase it takes longer and longer to do the backup. And as disk prices drop, the price of the inconvenience of using tapes becomes less justifiable. So an external disk drive seems headed towards the right answer.)
.

Its coming.

[CMiYC]

I work for a T&M company and deal with our computer solutions. Customer interest is building in our Serial-ATA analysis tools. Developers find it challenging, but appears to be the way they want to go. I wouldn't hold my breath for a desktop system any soon then 6 months. If it does come out before 6 months from now, don't expect it to be very stable.

The Benefits of Serial ATA

[rtos]

So you want to know what Serial ATA is all about but you don't want to read the article? Well, of all places Dell has a decent page about Serial ATA. It takes a bit of the "this is the best thing since sliced bread" stance, but there is some good info in there nevertheless. It is this info that I will now blockquote:

Benefits of Serial ATA
Serial ATA offers a number of benefits over Parallel ATA, including:

• Reductions in voltage and pin count
• Smaller, easier-to-route cables; elimination of the cable-length limitation
• Improved data robustness
• Backward compatibility

Voltage Reduction
Serial ATA's low-voltage requirement (500 millivolts [mV] peak-to-peak) will effectively alleviate the increasingly difficult-to-accommodate 5-volt signaling requirement that hampers the current Parallel ATA interface.

Cabling
The Serial ATA architecture replaces the wide Parallel ATA ribbon cable with a thin, flexible cable that can be up to 1 meter in length. The serial cable is smaller and easier to route inside the chassis (see Figure 2). The small-diameter cable can help improve air flow inside the PC system chassis and will facilitate future designs of smaller PC systems.

Improved Data Robustness
Serial ATA will offer more thorough error checking and error correcting capabilities than are currently available with Parallel ATA. The end-to-end integrity of transferred commands and data can be guaranteed across the serial bus.

Backward Compatibility
Serial ATA will provide backward compatibility for legacy Parallel ATA and ATAPI devices.

More information can be found at the Serial ATA FAQ (again, rather 'pro' biased).

Performance - doesn't anyone care?

[Vortran]

A few posts have hinted at this already, but one major problem I see with serial ATA is performance. I also have to ask (along with some others) why? We already have FireWire and USB for those who believe serial I/O is the holy grail (and in some instances it is a great answer). For storage devices, serial I/O technologies are to be avoided IMO.

For REAL performance, give me SCSI (or fiber channel) any day. Here's why: The serial I/O technologies for the PC, as well as the abomination called IDE, utilize CPU interrupts and the system bus to move data. The devices cannot "talk" to each other without utilizing CPU power.

For me, this is a major problem. I installed an IDE drive last weekend (because they're cheap) and while I was copying files (some 15 GB worth), my CPU usage was near 90%. All other processes were slowed down, so much so that I couldn't compile, render, play games or even edit code or browse http comfortably while this was happening.

On the other hand, when I copy files or burn CDs from one SCSI device to another, my CPU goes unmolested while these fine (Ultra320 LVD) devices talk (scream, rather) to eachother at lightning speeds.

Why bother creating new serial architectures that are no better than what we already have? Why not work at getting a SCSI drive down to the same price as an IDE drive.. or ( GASP! ) change the sorry PC architecture to have more than 16 IRQs! This would make me slightly less averse to burning up my scarce IRQs on IDE controllers.

To summarize - you lose system performance when using CPU intensive I/O. So avoid it if you like to use your computer and your storage devices at the same time... and if you can afford it.

Vortran out

Re:Performance - doesn't anyone care?

[Elbereth]

If we used what we had instead of reinventing the wheel, then all those engineers and marketing types would be out of a job.

If anyone is curious about getting into the world of SCSI, here's what I suggest:

1. Get a cheap Ultra Wide SCSI host adapter or motherboard with integrated Ultra Wide
   
2. Buy a refurbished Plextor SCSI CDROM
   
3. Buy a refurbished Plextor CD-RW
   
4. Get a cheap Ultra Wide, Ultra2 Wide, or Ultra 160 hard drive (preferably not refurbished)
   

You can get refurbished Plextor optical drives at many online stores. I use www.hypermicro.com the most. They have great products and service. I've bought almost all of my SCSI peripherals from them.

I bought an Asus P2L97-DS motherboard (dual processor, slot 1, PII 266-333 and Celeron 266 to 533, 0 - 512MB PC66 to PC133 RAM, 4 DIMM slots, integrated Ultra Wide SCSI, 4PCI/3ISA, 2x AGP) on ebay for about $50. This is a great intro board to both SMP and SCSI. All you need is two cheap Celeron processors and you're all set.

If you want to use refurbished hard drives, I suggest searching ebay for someone who's got lots of positive feedback and a non-DOA warranty. You can get 9GB and 18GB Ultra 160 hard drives incredibly cheap, due to the cut-throat competition.

Tekram makes inexpensive, powerful SCSI host adapters. They use the LSI Logic chipset, which is extremely well supported under Linux. Adaptec SCSI host adapters are lot more expensive, but you can find them much easier at retail stores.

If you wanted to, you could get a dual processor Asus motherboard, Plexter SCSI CDROM, Plextor SCSI CDRW, and a 4.55GB Ultra Wide SCSI hard drive for something like $250. Then, grab two Celeron or Pentium II processors off ebay (about $25-$50 each), an Enlight 7237 case ($50-$75, depending on the power supply), and a Matrox G200 or other cheap, well supported AGP card. In the end, I doubt it would even come close to $500. And now you have your very own SMP SCSI workstation.

Re:Performance - doesn't anyone care?

[Anonymous Coward]

SCSI devices cant talk to eachother without utilizing CPU power either, one of the differences is indeed that data has to go through memory but the main difference is the lack of tagged queing support for ATA (its in the spec but noone supports it). You could even make an ATA controller with a local memory pool so the data could be copied locally instead of through system memory if you wanted to, luckily the CPU does not have to touch the data itself even with ATA.

Both SCSI and ATA have to be interrupted when the drives have to be told which blocks of data have to go where, the SCSI drive has no concept of a filesystem after all, but SCSI can que a whole lot of copy's while ATA has to be interrupted after each individual one has completed.

Re:Performance - doesn't anyone care?

[stripes]

SCSI devices cant talk to eachother without utilizing CPU power either, one of the differences is indeed that data has to go through memory

SCSI devices can talk to each other, not just to a "SCSI CPU Device". The very very seldom do. When someone added the experimental splice system call to Ultrix (or was it DIGITAL Unix at the time? or maybe NetBSD/Alpha) they did support SCSI disk to tape and disk to disk copies. As far as I know that was the first and last time it was done in any "common" OS.

Both SCSI and ATA have to be interrupted when the drives have to be told which blocks of data have to go where, the SCSI drive has no concept of a filesystem after all

That is true, what the splice code did was on the read side look at the block numbers in the inode, and the first and second (and third if that FFS used it) level indirect block tables. On the write side to a tape device it was trivial, for another disk device it had to use the normal FFS code to find space for new blocks, which caused a normal read of the free block map, and normal writes to the block map and the block (indirect) tables for the inode (and the cache for the block to be written had to be invalidated). However in both cases the data writes went from device to device without going through main memory.

The speed up from skipping the CPU/memory step was modest. The speed up from being able to do the reads and writes somewhat out of order was less modest. The speed up or slowdown from not getting the recently written blocks in the block cache was frequently larger then the other effects.

As it turned out the larger win for splice was in dealing with network I/O and skipping the user-to-user context switches, which may be why neither Linux nor the BSD's implement their splice-like calls (sendfile I think) in a way that skips the CPU, they just avoid user-to-user context switches.

If the experiment was re-done it might be different, the memory wall keeps getting larger, and modern CPUs throw away a lot more pending work then the old Alpha did.

All this stuff was written up in one of the mid-90's Usenix reports if you want to read it in more detail then I remember.

Re:Performance - doesn't anyone care?

[jafac]

I had understood that FireWire did not suffer from this problem (CPU utilization)- that FireWire was derived from SCSI, and in fact, you can even transfer data between two FireWire devices NOT connected to a computer (provided you had a file-system interface of some kind).

In any case, the ATA/SCSI thing is true, ATA does use CPU, and SCSI doesn't (as much). And this is the reason why Intel promotes ATA and not SCSI, and why computer manufacturers include ATA controllers by default, and not SCSI. Because SCSI does not drive demand for more/faster Intel CPUs.

Adaptec's Serial ATA controller

[ArcticChicken]

Adaptec has a press release concerning their new Serial ATA ASIC/controller here. I'm sure many other manufacturers have similar news as well.

This is one new standard I'm willing to accept. In fact, I'm a bit surprised by the number of people here scoffing at Serial ATA. With performance of some parallel ATA drives matching mainstream SCSI drives for months now, with capacities closing in, and with SCSI manufacturers continuing to slowly drop production of SCSI optical drives, I think the end of SCSI is near. I never thought I'd say that, but I really think it is.

So to all you people saying that this just introduces a new standard to a "mess", I think you're wrong. This will end the division between desktop storage and mid-level server storage. Firewire and USB will stick around - but only as the external storage interface options they should be.

Serial ATA Replacing SCSI in Workstations?

[nuxx]

One thing I'd like to know about Serial ATA, and can't seem to find, is weather or not it eliminates one of the biggest issues with IDE. That is the fact that only one device per channel (the master or the slave) can be 'talking' at a time. This is the reason why IDE RAID is generally only faster if your two drives are on different channels, why you don't want to master/slave a CD-ROM and CD-Writer, etc. If this limitation remains, Serial ATA will still only be truely usable in home machines. If this limitation remains, I'll still keep on running my perihperals (CD-RW, CD-ROM, DAT) on SCSI and hard drives via Fibre Channel.

-Steve

I would've posted this earlier....

[teamhasnoi]

The reason for the late breakthrough of Serial ATA is that nobody wanted to, unnecessarily, spend time and money, developing a new interface. but I had to pause my internal narrative three times.

I think, perhaps, the overuse of, commas, let to the late breakthrough, of Serial ATA.

Re:I would've posted this earlier....

[Anonymous Coward]

Note to self: Make sure "This is a joke" appears after tongue-in-cheek comments. Perhaps a winky dude ;) or this guy :P will help.

Re:I would've posted this earlier....

[Anonymous Coward]

Perhaps, William Shatner, was reading, the article, and it was being, dictated.

temporary mirror due to ./ing

[millette]

Here's a temporary mirror due to ./ing:
http://mirrors.waglo.com/slashdotted/www.heatseeke rz.net/
I'm still building it...

Re:temporary mirror due to ./ing

[microsost]

due to ./ing ay... pretty good site that dotslash.org

Re:temporary mirror due to ./ing

[millette]

Look who is talking ;)

Software Upgrades

[chiph]

All the disk-utility vendors have gotta love this. Right now their software probably assumes only two devices per ATA/EIDE channel. But with SATA, you can daisy-chain additional drives, which means a software upgrade is required for all your disk-utilities and backup software to be able to address the additional devices.

Must make a note to put in a "buy" order with the broker for Symantec/Norton, Powerquest, Iomega, Veritas, and Computer Associates.

Chip H.

Serial SCSI replacement?

[jonr]

Wasn't IBM developing serial standard decade ago? Whatever happened to that? (I think it was called SSA or Fibre-Channel)

Re:Serial SCSI replacement?

[Anonymous Coward]

Wasn't IBM developing serial standard decade ago? Whatever happened to that? (I think it was called SSA or Fibre-Channel)

SSA and Fibre Channel are two seperate technologies. You can find some (biased) information about them here.

The way I heard it is, historically SSA and Fibre Channel competed somewhat, and since IBM was pushing SSA and got off to a late start on Fibre Channel development, they were actually trying to kill Fibre Channel for awhile.

Though I have no idea what's up with SSA these days, Fibre Channel is mostly alive and well. It is intended to be a higher-performance replacement for SCSI. It is also higher cost, even though the components are supposed to cost less to manufacture, and I think this is hurting it. I think companies inflate the cost of their FC products because of the economies of scale *and* simply because it performs better than SCSI. Well, and also the FC protocol (in all its glory) is wayyyy complex. The designers went for total flexibility (multiple topologies, multiple classes of service.. heck you can run TCP/IP over it) and it shows -- most vendors only support a limited subset of the possible functionality.

One other interesting thing about Fibre Channel is that gigabit ethernet borrowed its physical interface (up through the 8B/10B encoding method).

Still has 137GB Limitation

[KieranElby]

One of the current limitations of the ATA standard is that maximum drive size is 137.4GB. While we're not quite there yet, it seems like this could become a problem at least by mid-2003.
I'm surprised that the opportunity was missed to address this with the introduction of Serial ATA.
For the curious, the limit comes about since only 28bits are used for the sector number in the ATA protocol. (2^28 * 512 bytes = 137.4Gb).
This is straying dangerously off topic now, but its quite amusing to look at the history of arbitrary hard disk size limits: (from The Storage Review)
PC/XT Parameter (10.4 MiB / 10.9 MB) Barrier
FAT12 Partition Size (16 MiB / 16.7 MB) Barrier
DOS 3 (32 MiB / 33.6 MB) Barrier
The 1,024 Cylinder (504 MiB / 528 MB) Barrier
The 4,096 Cylinder (1.97 GiB / 2.11 GB) Barrier
The FAT16 Partition Size (2.00 GiB / 2.15 GB) Barrier
The 6,322 Cylinder (3.04 GiB / 3.26 GB) Barrier
The Phoenix BIOS 4.03 / 4.04 Bug (3.05 GiB / 3.28 GB) Barrier
The 8,192 Cylinder (3.94 GiB / 4.22 GB) Barrier
The 240 Head Int 13 Interface (7.38 GiB / 7.93 GB) Barrier
The Int 13 Interface (7.88 GiB / 8.46 GB) Barrier
The Windows 95 Limit (29.8 GiB / 32.0 GB) Barrier
The 65,536 Cylinder (31.5 GiB / 33.8 GB) Barrier
The ATA Interface Limit (128 GiB / 137 GB) Barrier
And only four of them are due to Microsoft...

Re:Still has 137GB Limitation

[Anonymous Coward]

You'd figure they'd have learned by now.

I guess it must be good for the upgrade market.

Re:Still has 137GB Limitation

[edmudama]

The 48-bit command set is part of the ATA-6 specification that you can read at www.t13.org. Serial ATA will support this command set.

Most vendors don't need to support 48-bits yet because they don't have drives that are big enough. Many manufacturers do not make 4-platter IDE drives anymore, and with the current technology of 40GB/platter, the 3-platter disks are only 120GB.

When the next generation comes in at 60 or 80 GB/platter, they'll support 48-bit commands as needed.

Re:Still has 137GB Limitation

[Anonymous Coward]

They *did* learn by now. You'd think you'd learn by now that anytime you say anything on /. your're most likely to be proven wrong. ATA is up to 48 bit addressing giving enough room for multi terrabyte drives.

Why is serial ATA faster?

[kirkb]

Pardon my ignorance, but can somebody explain why serial ATA is faster than the current (parallel) ATA?

On PC's, parallel ports are significantly faster than serial ports because they transmit 8 bits at a time instead of serial's one bit at a time. Wouldn't the same thing hold true for parallel vs serial ATA?

Please explain.

Re:Why is serial ATA faster?

[jyak]

Parallel ATA cannot scale to support several more speed doublings, and is nearing its
performance capacity. By contrast, Serial ATA's roadmap starts at 1.5 gigabits per second
(equivalent to a data rate of 150 MB/s) and migrates to 3.0 gigabits per second (300 MB/s), then
to 6.0 gigabits per second (600 MB/s). This roadmap supports up to 10 years of storage
evolution, based on historical trends.

Re:Why is serial ATA faster?

[edmudama]

> Pardon my ignorance, but can somebody explain
> why serial ATA is faster than the current
> (parallel) ATA?
>
> On PC's, parallel ports are significantly
> faster than serial ports because they transmit
> 8 bits at a time instead of serial's one bit at
> a time. Wouldn't the same thing hold true for
> parallel vs serial ATA?
>
> Please explain.

In a perfect world, parallel would always be faster than serial. However, what happens is that due to outside factors (shape of the cable, EM interference at the time, etc) when you send those 8 bits down a parallel port, they don't all arrive at the destination at exactly the same time. The faster you send them, the more likely they are to not arrive when you send them since your tolerances get lower. This is referred to as signal skew.

Serial ATA borrows a technique from LVD SCSI devices which is low-voltage differential signaling. They send the pulses down 2 lines polarity reversed. By using 2 wires instead of 9 (8 data bits + a clock) or in UDMA land 20 wires (16 data lines, IOR, IOW, DMARQ, DREQ) the chances of them being significantly different than one another is less, because they're closer to following a consistent path through space. This allows them to toggle the lines MUCH faster in LVD applications than parallel applications, which gives us much higher data rates.

Another thing is that the IDE bus still uses TTL signal levels (5V/0V), meaning that it takes a HUGE amount of power to wiggle all those 40 conductors up and down to get some data across. The little chipsets on motherboards these days have trouble supplying enough internal power to do that, so LVD will help them make less complicated circuits in the chipset to talk serial ATA.

eric

It'll be interesting to see

[Scratch-O-Matic]

how they design the new, thinner cable so it just barely reaches, then falls short when you have to flip it 180 degrees to get pin 1 in the right place. If it can't do that, I don't want it.

Re:It'll be interesting to see

[p3d0]

Doesn't it suck when you tell a joke, and someone mods you "informative"?

Bundling the cable

[hey!]

Some uninformed speculations from somebody who used to know something about electronics but is now hopelessly out of date.

I suppose there will be fewer lines than parallel ATA because serial ATA is, well, serial. Also, instead of the standard flat ATA cable with its parallel wires just made for crosstalk and interference, they could use a balanced twisted pair for greater noise immunity. Because the the wires in the pair are twisted and the signal is taken as the difference between the two wires, any stray EM tends to affect both wires equally and thus cancel itself out. Also radiation from the pair is reduced because balanced current running in opposite directions tends to cancel out its own raditions.

Serial ATA won't replace SCSI

[Demon-Xanth]

Why? Because Serial ATA is still ATA and SCSI is still SCSI. All the inherant benefits of SCSI are still limited to SCSI and all of the inherant drawbacks of ATA is still in ATA. Serial ATA doesn't really change the protocol, just the interconnects.

And Ultra320 is on the horizon already. And by the time Serial ATA reaches 600MB/sec I'd expect SCSI to have continued to grow.

Just because you can put two cows side by side doesn't make them an ox team.

Re:Serial ATA won't replace SCSI

[Anonymous Coward]

Or, you can just buy a 3ware (www.3ware.com) card and get performance with cheap IDE drives that would require a multichannel hardware RAID SCSI card to match.

Low CPU usage too, since the 3ware card looks to the host computer like a SCSI card.

And you can hot-swap the drives with an appropriate drive cage.

Its even more reliable too, since a single cable problem will take exactly one device out of service, as opposed to the whole chain for SCSI.

SCSI (the wire) has been hanging on by moving further and further upmarket, but its about to run into FC, at which point it will have no place else to go. SCSI (the protocol) will live on in ATAPI, FW and FC.

Serial ATA and MS

[Sri Lumpa]

In another related new, a MS spokesman said Serial ATA is so good that it shouldn't be restricted only to interfaces inside the PC and that MS was planning to improve on it to form a network-centric (a la firewire(TM)) Serial ATA specification, for now dubbed SATAN. The beast is scheduled to be unleashed soon upon the world.

A linux evangelist said inresponse to Microsofts project "Repent now and see the light before it is too late for your soul".

Number of Devices

[Krieger]

This is one thing I'm concerned about. I have only been able to find the following passage about the number of devices you can use.

From the Intel pdf from http://developer.intel.com/update/departments/init ech/it03012.pdf.

"By Contrast Serial ATA is a point to point interface where each device is directly connected to the host via a dedicated link. Each device, therefore, has the entire interface bandwidth dedicated to it, and there is no interaction between devices. This means that software can be streamlined, eliminating the overhead associated with coordinating accesses between the master and slave device sharing the same cable."

So this makes me wonder what the typical number of Serial ATA devices per mobo will be?

Yuo = ignoramus

[bwoodring]

Both drives on an ATA channel share that bandwith. A lot of newer drives push out > 33 MB/s second. Which makes firewire an inferior solution for drives.

Hell, the new WD1200JB has a peak transfer rate of 48.8 MB/s, which would saturate firewire all by itself. Serial ATA will debut at 150 MB/s which is much more realistic.

Plus, while $0.25 sounds like a small amount of money, if you multiply it times the number of drives sold by companies like WD and Seagate, it adds up, why would they want to pay that money, when they can get a superior solution for free?

Re:Yuo = ignoramus

[Anonymous Coward]

'superior solution'? sorry, this is a solution that stinks of Intel's classic 'i didn't make it so i don't want it' line. firewire, and its internal counterpart, are wonderful. you say the drive would saturate firewire, but firewire is a scalable standard, and the 800 megabit (100MB/s) spec is *ready to ship today*. Serial ATA, like every Intel technology, is faster, but is 'just around the corner.'

why wait? why introduce another standard, that requires another controller and an entirely new line of drives, when an existing standard is there, is ready, is fast and is useful for more than just hard drives?

Re:Yuo = ignoramus

[Kymermosst]

Simple. Because it requires no change in software on the host PC

Is that too easy or what?

No new drivers to install, it looks to the computer just like any other standard ATA interface.

Turn DMA on, dumbass.

[bwoodring]

see subject

SATA design

[bwoodring]

With serial ATA, the design is point-to-point, with only one device per channel. With fewer pins, each channel will be cheaper to implement, so you will have more channels per controller.

Re:SATA design

[nuxx]

Hrm, thanks for the info... That might not be so bad then. Do you have any pointers to a more realistic spec, one that details length of cable, etc? serialata.com doesn't seem to have anything, and the linked site is dead (of course).

-Steve

Re:SATA design

[Catbeller]

length of cable 1 meter, just like the present ATA.

too expensive

[Hugonz]

Firewire is too expesive to be implemented widely at .tw ad .hk manufacturers....the technology has heavy patent royalties (about 1 USD) per chip.

misc stuff I've found on Serial ATA

[noodley]

lovely pic of a SATA RAID prototype I can't find the link, but I remember seeing a SATA hub, where it would plug into the board and supply 4-5 connectors for devices. IIRC, this would/could also be the way the SATA smaller raid controllers would be hooked up

Re:misc stuff I've found on Serial ATA

[noodley]

http://www.anandtech.com/cpu/showdoc.html?i=1590&p =2

clarification - from the spec

[Klox]

There is a lot of misinformation being thrown around, so I thought I'd quote the spec:

2.1 Goals and objectives
Setial ATA is defined with the following goals and requirements listed in no particular order:
* Primarily inside-the-box storage connection (no outside the box)
* Completely SW transparent w/ ATA (easy transition)
* Low pin count for both host and devices (2 pairs)
* Favorable (low) voltages
* Supports lower cost device architectures
* Higher performance than equivalent ATA (data rate, queuing, overlap) w/ scalability to higher
* Much better cabling/connectors (thin, flexible)
* Includes efficient power delivery
* No software dependency. Relatively easy transition (price, IHV NRE and capital inventory risk, wide variety of devices at intro, etc.)
* Power management and power consunption suitable for mobile use
* Allows roadmap spanning ~10 years
* Cable length comparable to ATA (<1 m)
* Transfer rate exceeding best ATA (~150 MB/s) with scalability to higher rates
* Light protocol allowing overhead latencies to be minimized
* Asynchronous only (no isochronous requirements)
* No Peer-peer transfer support (to/from host only)
* Provides support for 1st party DMA access to host
* Cost competitive with equivalent parallel ATA solution at introduction (host + device + cable)
* Storage device centric (no cameras/scanners/printers)
* Easy installation/configuration (plug/play, no jumpers, no external terminators)
* Single host (no multi-initiators or host/host networking)

Re:Get SCSI, dumbass.

[Anonymous Coward]

IDE with UDMA on still uses alot of CPU resources while SCSI uses almost none because the information doesn't have to leave the SCSI chain like IDE

Reliability

[leandrod]

I remain faithful to SCSI because of IDE's (ATA's) problems of reliability, not only because SCSI drives are usually better tested but also because ATA's specification is defective. Is sATA supposed to fix this? Or I will continue to have to shell out the buck to get SCSI reliability and speed?

Re:Get SCSI, dumbass.

[Anonymous Coward]

That hardly matters, you arent going to get near to saturating even a 32 bits PCI bus copying from HD to HD ... and the memory bandwith it takes is only a very small percentage. The main problem is the huge volume of interrupts.

It's called DMA, dork

[Anonymous Coward]

And IDE drives have supported it for about 5 years.

The latest IDE HDDs have yet to...

[Mode Frozen]

...reach the data rate of 50MB/s. The new drives from the likes of seagate, maxtor, western digital BURST in the range of 48-49.5 MB/s and as such have yet to even take full advantage of ATA66. What are these HDD manufacturers developing that will take advantage of thei new standard? I hope that 10000 RPM HDDs will bring better performance.

Ethernet-disks??

[uncl_bob]

I am probably a moron, but _why_ cant we use a separate ethernet inside? Use gigabit-ethernet and you will have 100 meg/s to play with. 10 gigabit ethernet is not that far away so it would also be pretty forward compatible.

serial ide

[perversion00]

now maybe the answer is right in front of our face and were not thinking right now there finally coming out with a revision on firewire now wouldnt it be nice compact and fast to creat and internal firewire chain instead of sticking with and old technology ...

The is no royalty fee on 1394

[Catbeller]

NO, it does not have $1/PC patent royalties per chip. Apple waived that years ago.

So there is not a price problem caused by Apple.

It's expensive because it's expensive. Because Intel invented USB, and Apple invented 1394, Intel has doggedly refused, even up to the present day, to support the standard on its own mobos. Intel FUD took care of the rest. So it has taken years to reduce the price of the chipsets -- but not because of the licensing fees. It's a matter of unit cost. Since Intel was actively hostile to the (superior) tech, it retarded the acceptance of 1394 and kept production costs artificially high. Chicken and egg...

But thanks to Apple, and common sense, people realized that Firewire was simply superior to USB in every way but price. Just compare an iPod to a standard USB MP3 player - transfer speed enormously faster, and Firewire also charges the iPod's batteries during the process.

So the prices came down despite determined opposition -- the market actually worked, sort of.

Firewired external CD-RW drives and hard drives work fine, and speedily. Putting the drive into the PC itself seems obvious.

The fly in this soup: 1394 developers seem determined to insert copy control into the cable/controller hardware. If there will be a choice between mediocre USB with no DRM, and 1394 with DRM, I'd throw the 1394/DRM equipment into the garbage, even if it were free!

The best standard is irrelavant

[Neil Watson]

It doesn't matter how fast or how many devices "could" work with any standard.

The only thing that determines its success is who holds the patents and whether or not anyone else want's to pay the roylaties.

Distiguishing between latency and throughput

[neirboj]

Given the physical dimensions of the platters, the RPM of the motor, and the density of the bits you can come up with the rate at which bits pass under the sensor. This figure is the continuous throughput of the drive. On several of the high-end drives I've looked at lately this number is in the neighborhood of 40MB/s.

Because of various tricks - like reading more data in one go than is necessary to fill the current request - the drive can, from time to time, exceed the limitation described above by sending data out on the bus from it's cache (which is a heckuvalot faster than the platters). The cache has a finite capacity, however, so this technique is only useful for smallish periods of time and for smallish blocks of data, or when very particular demands are made on the the drive.

In my little geeky heart, I can't help but get excited when I discover that storage bus speeds have yet again been doubled. The excitement doesn't last long though because the hard drive is still the slowest part of the computer, and until the RPM doubles, or I can afford to hook up a big stripe set (a situation that makes the "I" in "RAID" seem rediculous), that extra bandwidth won't mean a thing.

Corrections and flames are welcome, as are large cash donations.

Regards,
John

Just to clarify

[MoneyT]

When I said mini computer, I should have said all-in-one or compact computers

Yes, SerialATA will take over and it will be nice

[egarland]

Anyone who has dealt with one of the 8 port RAID controllers like a 3Ware 7850 knows that it is time for SerialATA. In order to move from an 8 drive to a 12 or 16 drive controller the connector needs to get smaller, the pin count needs to go way down, and the cables need to get longer.

Why SerialATA and not USB 2.0 or Firewire? Two big reasons: cost and design.

Firewire and USB are not designed for what SCSI and SerialATA are. If you pull a failed drive off a firewire chain and plug a new one in it's place the new drive will have a different ID than the old one did. Firewire and USB are loose protocols that are designed for plugging in optional components not for critical drives and drive arrays. With Firewire and USB the ID goes with the device not with the connector. With SCA SCSI and SerialATA it is the other way around.

As far as cost, those bridge chips between USB/Firewire and IDE are very expensive. IDE is the cheapest drive technology by far and from what the industry is saying SerialATA will be as cheap or cheaper. The chips on the motherboards and drives will have less pins. The connectors will take less real estate on the motherboard, the cables will have fewer wires. Because it's new technology it will probably carry a price premium for a little while but in the end, this stuff will be really cheap.

I am a big fan of IDE RAID. It's price/performance is absolutely amazing compared to every other RAID option. The problem is you can never get more than 8 drives attached to a controller. 8 Drives is nice but a standard 3 channel LVD/SE scsi controller can handle 45. Even a 1 channel can take 15 drives. SerialATA will open new doors for larger IDE RAID arrays, with smaller connectors that allow more drives per controller and longer cable lengths to allow more drives per system. A standard hot plug connector will also come in handy when building compact hot swap enclosures. I expect to see SerialATA RAID controllers supporting up to 16 drives and the ability to put 4 in one machine for a massive storage server (10+TB). This will be a great help to IDE RAID.

In short, I think SerialATA is the only hard drive interface technology that has a bright future. Fiber channel is just too expensive and SCSI, while nice, is losing it's advantages over IDE. DMA mode is now standard on all modern IDE chipsets nearly eliminating the performance differences between SCSI and IDE. Now with SerialATA the cabling on IDE is much more graceful than before. While SCSI is nice for it's ability to daisy chain, that's really a bad idea for production systems. Point to point is much more reliable and much easier to troubleshoot.

Fast, cheap, flexible, huge industry support. Sounds like winning technology to me.

Little white lies

[billcopc]

It's just like what I tell my s.o. every night "Don't worry, I'm coming in a minute".. of course that minute usually happens only after a few hours of merciless coding and tweaking, by which time she is long asleep.

Just the same with Serial ATA. They've been spouting a whole bunch of buzz about it, but we have yet to see any concrete applications of the proposed standard. It's kinda like the hype that preceded USB, which was already obsolete by the time it hit the market. Methinks the same is going to happen to Serial-ATA. Why don't we just stick to SCSI instead ?

Re:Little white lies

[Nerant]

It's just like what I tell my s.o. every night "Don't worry, I'm coming in a minute".. of course that minute usually happens only after a few hours of merciless coding and tweaking, by which time she is long asleep.

I hope the rest of you aren't thinking what I'm thinking. ;)

Re:First!

[(outer-limits)]

That is why the US had such a blood thirsty civil war, and Europe has been at war for the past 3,000 years.

Just when parallel cables are going serial....

[(outer-limits)]

The power cables are going to take over the inside of the box. I hate cables.

Re:Cnet needs to check their facts.

[east_bay_pete]

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.

If Parallel ATA cables can only be 18 inches long, then I guess those 36" cables I have in my computer right now must not exist.

It's too bad, because for things that don't exist, they work great.

WRONG

[guacamole]

These are the corrected peak speeds:

UltraSCSI (wide): 40MB/second
Utra2 SCSI: 80MB/second
Ultra160: 160MB/second
Ultra320: 320MB/second

Firewire: 40MB/second

As you canse, firewire compares well only to the old'good UltraSCSI standard. 40MB/second is still pretty fast for most users though ..

Re:First!

[Anonymous Coward]

The US Civil War would not have occured if not for the Negro, and the timeframe you mention for Europe corresponds with the arrival of the first merchant Jews.

Re:First!

[(outer-limits)]

Now that is a novel interpretation of history, blame the negros for the civil war. As i recall, they were forcibly brought to the us for the purposes of slavery. And the arrival of the first merchant jews plunges europe instantly into war, where there before then been utter peace. I don't recall the roman conquest of europe as being a jewish act.