Spec Will Cut External Drive Power Cords

Lucas123 writes "The Serial ATA International Organization just revealed that it is well along the way to finishing a specification that would remove separate power cords to external SATA drives or optical disk drives, allowing them to draw power from the host system. The resulting new cable, being called Power Over eSATA, will be compatible with the existing eSATA connector and support the current maximum interface transfer rate of 3Gb/s. The SATA organization expects the new cables to be released later this year to drive makers."

Comments

Cables

[ackthpt]

I wish they'd do something about this piss-poor connectors. I've had a number of them fail and had to junk them because they do not make a good solid connection, nothing prevents vibration from letting them slip.

Re:Cables

[Anonymous Coward]

Maybe you are not properly placing your vibrator?

Re:Cables

[pilgrim23]

GOSH! they might even invent... Firewire! Who'da think it.

Re:Cables

[jandrese]

I have to agree with you on this. Even the internal cables feel pretty loose to me and will come off with the slightest tug. Sometimes I find a nice cable and a compatible drive and they'll click together nice and solid, but for the most part they just don't have the holding power I would like. Even if they don't fall off, they're probably the flimsiest connector you're likely to use on an external connection of any sort on your computer these days. I have a few that have metal reinforcement on them, but they are an exception.

Re:Cables

[flaming error]

I agree. Any strain, and they come right out. So I bend the cable over the external drive, and duct tape it to the unit so it doesn't slip.

Re:Cables

[Jackie_Chan_Fan]

They also snap off and break! The piece of shit plastic slot on the hard drives, snaps off with ease.

The durability of Sata connectors suck.

Re:Cables

[Anonymous Coward]

I have to agree with you on this. Even the internal cables feel pretty loose to me and will come off with the slightest tug. Sometimes I find a nice cable and a compatible drive and they'll click together nice and solid, but for the most part they just don't have the holding power I would like. Even if they don't fall off, they're probably the flimsiest connector you're likely to use on an external connection of any sort on your computer these days. I have a few that have metal reinforcement on them, but they are an exception. I hate SATA connectors. A few months ago I replaced two ide hard drives with satas. I pulled out the drive bay, installed the new drives, connected the cables, and then slid the bay back in. When I did that I heard a snap. Pulled it out, and had two snapped sata connenctors. I've never had that problem with IDE. I snapped one off accidentally at work, too. At least NewEgg replaced them.

Re:Cables

[Jeff DeMaagd]

Firewire is more expensive and a non-native interface. FW800 is the closest to eSATA in performance, and even then, it's more expensive and slower.

Re:Cables

[Yo Grark]

Tell me about it.

Had 3 of them "snap" off of 2 separate motherboards after only a few connect/disconnects.

And before people tell me "it must be you", I had a technician call me and tell me he had to replace the mobo because he broke the remaining one off when he unplugged it to test a new HD.

Granted it was one of the first generation mobo's, but we're talking ASUS boards here, not asrock (and YES I know, god you guys are picky!)

Yo Grark

Re:Cables

[TheRaven64]

and a non-native interface Every single interface since IDE and SCSI have been non-native. That's what Integrated Drive Electronics means. The controller no longer sends native commands like 'move disk head' and 'read bytes' it sends abstract commands like 'read block 1406.' Your on disk controller translates these in to native commands. It is no harder to build a device that accepts FireWire commands directly and translates them in to native commands than it is to build one that understands SCSI or SATA. Most current FireWire drives go via SCSI or ATA because there is a much bigger demand for ATA drives than FireWire, but FireWire commands are almost identical to SCSI commands so such a drive could easily be built.

Re:Cables

[Supergibbs]

I just built a new system and the cables that came with my ABIT motherboard had clips.

Re:Cables

[pilgrim23]

the first Firewire drives built to the 1394 standard were firewire direct. Granted they cost a bucketful! FirewireATA, Firewire SCSI and the like interfaces soon came and cut the price since Seagate et al said a big NO to native Firewire in an industrial way.. Now, my point was: Firewire is easy to use, supplies power, can be daisy chained like SCSI (more then one drive on a port) comes in the 400 and 800 flavors, and I have used it for many years on both PC and Mac.
External SATA strikes me as similar to the REAL reason behind ISA-EISA-MCA-PCI or FPM-EDO-SDRAM-DDR-DDR2-DDR3 that is: Gee WHiz its faster! , and KA-CHING goes the cash register when a form factor requires buying new. I remember when SATA showed up and was OH SO MUCH FASTER then PATA or when Vista was better then XP.

Re:Cables

[afidel]

You just reminded me, I need to RMA that drive with Seagate. Or perhaps I should wait a couple years and abuse the 5 year warranty to get a bigger drive since I already bought a replacement and don't need another 160G drive right now =)

Re:Cables

[SkyDude]

I remember when......when Vista was better then XP.

Now I know I occasionally nod off during the day, but when exactly did THAT happen?

Re:Cables

[loki_tiwaz]

some cables come with retainer clips. the sockets on the drives and motherboards have bits to connect with them. also, if you are referring to e-sata, those are a similar kind of connector as hdmi and ipod docks, and they have retainers.

Re:Cables

[misleb]

Firewire is more expensive and a non-native interface.

SATA -> Firewire drive enclosures are pretty cheap. The difference in cost is insignificant.

FW800 is the closest to eSATA in performance, and even then, it's more expensive and slower

Yes, but Firewire is a much more flexible. Supports high end scanners, printers, networking, hard drives, optical drives, cameras, audio mixers, all kinds of stuff. Firewire is also very efficient. If I had a choice between a Firewire port and eSATA (even with power), I'd not hesitate to go with Firewire. No contest. Does eSATA even allow device chaining? That is a big advatange of Firewire right there.

The ONLY case I can think of where I might want 3G/s eSATA is for external RAID enclosures on a high performance workstation (server would use fibre channel or good ol' fashioned SCSI). Other than that, I don't see the point of eSATA.

Re:Cables

[Anonymous Coward]

but FireWire commands are almost identical to SCSI commands This both is and isn't true.

Firewire is actually completely different from SCSI. There is no resemblance between native FW protocol and native SCSI protocol.

However, SCSI has been branching out for a long time and has become separated into different layers. Many encapsulation schemes have been invented to transport SCSI commands and responses across non-SCSI busses. And that, as it turns out, is the normal way of talking to disks using Firewire: you use SBP-2, a SCSI encapsulation protocol for Firewire. The host computer generates SCSI commands, encapsulates them in Firewire packets per the SBP-2 spec, and sends them off to a FW disk controller, which generates standard SCSI responses and SBP-2 encapsulates them for transmission back to the host.

Since most Firewire disk enclosures use ATA or SATA disks, there's also a SCSI-to-ATA translation going on in the FW-to-ATA bridgeboard.

Summary: if people wanted to it would be possible to put a FW port right on the back of a disk and have it speak SCSI using SBP-2. Unfortunately (or fortunately depending on your POV) this idea fizzled in the mid-to-late 90s.

Re:Cables

[tdelaney]

I find that plastic SATA cable connectors often split apart if any kind of force is applied to them (e.g. bending the cable too near the connector to make it fit). Older connectors were more prone to this - newer ones seem better made.

It's a pain, but nothing that the application if gaffer tape won't fix.

I haven't had any cables come out of either the motherboard or hard drive except of the connector has come apart, but they're not very well held in.

I havn't used eSATA, so I can't comment on cables for that.

Re:Cables

[dfghjk]

"Every single interface since IDE and SCSI have been non-native."

What a nonsense statement.

"That's what Integrated Drive Electronics means."

No, the "integrated" in "IDE" referred to the ST-506 task file being integrated into the drive. "Non-native" means there's translation hardware that converts one interface to another. There is no standard drive interface below IDE/SCSI/SATA/SAS today. Those ARE the native interfaces.

"The controller no longer sends native commands like 'move disk head' and 'read bytes' it sends abstract commands like 'read block 1406.'"

That definition of a drive interface is arbitrary, meaningless and irrelevant. Firewire drives are implemented typically using IDE drives and translator boards. They are therefore non-native.

"It is no harder to build a device that accepts FireWire commands directly and translates them in to native commands than it is to build one that understands SCSI or SATA."

That's beside the point. There are no such drives.

"Most current FireWire drives go via SCSI or ATA because there is a much bigger demand for ATA drives than FireWire, but FireWire commands are almost identical to SCSI commands so such a drive could easily be built."

Again, irrelevant. All firewire drives are non-native. SATA is a more efficient protocol that Firewire/SCSI anyway.

Re:Cables

[dashslotter]

It's not as trivial to change external interfaces as you may think. I've worked on hdd firmware, so I can say that the abstractions aren't as clean as you may hope. Performance is everything. Performance over ATA, that is. ATA specifies how to handle internal cache, for example, so you may need to do some ATA-specific tricks further down then you would like. Otherwise your latest drive may not make it onto that new Dell system. Keeping things clean and abstract so that you can just pop on another interface isn't always the highest priority.

Re:Cables

[Jackie_Chan_Fan]

The slot on my 500GB seagate SATA drive snapped. Instead of RMA'ing it, i stuck it in an external enclosure and fixed the connector with crazy glue :)

Atleast in the external ESATA case, it isnt likely to wiggle out or anything. Its in there pretty dam secure.

Re:Cables

[GrievousMistake]

Are you talking about SATA or eSATA? From the spec, the eSATA connectors are supposed to withstand 5000 insertions and removals. They are deeper than SATA connectors and have small indents for retention springs.

Should have been in the spec from day 1

[krog]

Seriously -- it's two more pins. Why wasn't the spec designed right in the first place?

Re:Should have been in the spec from day 1

[Anonymous Coward]

probably due to concerns over internal power supplies

Re:Should have been in the spec from day 1

[Daimanta]

Because it would require thinking and *gasp* some work.

Re:Should have been in the spec from day 1

[RingDev]

It could have been a political issue in the industry as well. If there was strong opposition to any specific power over SATA spec, it could have held up the spec. Where as, going live with the widely accepted standard, and gaining a foothold, the spec now has the power to determine what the manufacturers should do as opposed to the other way around.

-Rick

Re:Should have been in the spec from day 1

[Anonymous Coward]

Sometimes it's things that have always been done a certain way that are never noticed to need changing.

Re:Should have been in the spec from day 1

[jo42]

Why wasn't the spec designed right in the first place? Because most of the time, the people that write these specs, or design this stuff, don't seem to use any products in the real world...

Re:Should have been in the spec from day 1

[Anonymous Coward]

From a signal integrity point of view, you want the NOISY power connections to be away from the high speed SATA signal as much as you can.
It might not seem important right now, but it might come back and bite you on inferior cables or when they crank up the rates again.

From a easy of use point of view, it is certainly silly.

Re:Should have been in the spec from day 1

[WarlockD]

Maybe because of the fact that some drives use both 5v and 12v. Or that the 3.3v (does any drive use that yet?) was needed

To be honest, I don't even see if its possible for internal drives 3.5. Most of those drives use upwards of an amp off the 12v, and pushing 12watts down a little sata cable sounds like it would cause interference. Heck, it also means we have to add yet another 12v rail to the motherboard to support the power. It would be nice, however, for it on the eSATA connector. But thats the only time I would say it would be that useful.

Re:Should have been in the spec from day 1

[krog]

Of course proper shielding is necessary, but if you're in a position to design a cable which will support throughput of 3Gb/s, you are in a position to supply power in the same package. In practice, crosstalk from other data cables is a much greater problem than interference on the power supply rail. (Disclaimer: I'm an analog EE; I think about this crap.)

Re:Should have been in the spec from day 1

[Jeff DeMaagd]

It's not "just" two pins, definitely not the same size conductor. I think the drives need as much as one amp on 12V and 5V, and it looks like at least four conductors are used to do that.

Re:Should have been in the spec from day 1

[pclminion]

pushing 12watts down a little sata cable sounds like it would cause interference.

It's DC. How could that cause interference?

Re:Should have been in the spec from day 1

[truesaer]

That's an awfully simplistic view. A lot of devices use a lot of pins for power and ground, to ensure they can supply enough power. Otherwise if you use 2 you might need to use unusually large wires/pins, which can be awkward. Further, there are noise issues to consider. The power pins can't cause any interference on the data pins.

Re:Should have been in the spec from day 1

[Znork]

I can easily think of actual technical issues on this one.

See, historically, disks have had their power supported by the PSU directly. Now you want to replace IDE and put SATA connectors on the motherboard. That's fine. Then you want the SATA connectors to supply power enough to drive one disk? Ok... Then you want the SATA connectors to supply enough power to drive four disks? That's an 80W or more power bus over the motherboard; motherboard manufacturers had just about gotten over having to add new power connectors for CPU's and partly for PCIe, and now you want them to take the hit for disk power too, and presumably do some engineering for it too, (such as stabilizing the power so your computer doesn't crash as parts of the MB loses power when the disk spins up)?

Call it political or call it technical, but in this case it was probably not just simply adding the connectors that was the problem (you could have gotten around the issue by placing a PSU connector next to any SATA contact, of course, but imagine the bitching about that).

Re:Should have been in the spec from day 1

[Gat0r30y]

It's DC. How could that cause interference?

During spin up there are fairly significant current spikes, which could I suppose cause some interference.
Of course at that point there isn't any data transfer other than wait for ready from the SATA controller. So it could cause some issues. My question is will this still be hot swappable, which in my view is the main advantage to eSATA.

Re:Should have been in the spec from day 1

[magarity]

it also means we have to add yet another 12v rail to the motherboard to support the power
 
Not only that but any add-on cards for laptops will need a wall wart providing power in to them. Now talk about loose fitting; every USB2 and FW add on PCMCIA card I've had to use has had a miserable loose little socket for device power. I far prefer to use the wall wart that goes straight to the drive. I suspect I'm not the only one adding drives to laptops and this power over the data cable is not going to help there at all.

Re:Should have been in the spec from day 1

[kai6novice]

Because the computer industry wants to make more money by having 2 cables.

Re:Should have been in the spec from day 1

[araemo]

Why wasn't the spec designed right in the first place? That's my gripe with it..

SATA is a great standard, and eSATA is a great idea... but they decided, for secure connections(it seems that even the SATA-io thinks the internal plugs aren't secure enough for external drives), they'd use a different connector. So you have to buy a different kind of cable for eSATA than internal SATA anyways..

But instead of taking the opportunity to add power pins, and allow a full 3.5" drives' worth of power to be drawn from that port, they didn't add any power at all!

Now they're talking about doing PoE-esque negotiation and sending power over pins normally used for other things(Probably extra ground pins?) And the icing on the cake is they're still going to limit it to around 500mA, so it won't be able to power 3.5" disks, only the 2.5" ones that you can power off of USB or firewire just fine. What a waste of a spec.

Re:Should have been in the spec from day 1

[misleb]

Seriously -- it's two more pins. Why wasn't the spec designed right in the first place?

They probably just assumed that most people using portable storage (as opposed to say a fixed external RAID) would just go with Firewire, which can provide lots of power.

-matthew

Re:Should have been in the spec from day 1

[Anonymous Coward]

Why wasn't the spec designed right in the first place? This is the ATA committee we're talking about. ATA has never been anything but chewing-gum-and-baling-wire design from the start. SATA didn't change that, and eSATA wasn't even a spec until after the fact. (Initial eSATA hardware was not designed to any spec, it was just a manufacturer realizing 'hey! we can run a SATA cable between 2 boxes instead of only using it inside of a single box! and it'll mostly work so long as we don't violate the SATA signal integrity requirements TOO badly! Cool!')

Re:Should have been in the spec from day 1

[Agripa]

It's DC. How could that cause interference?

The current is not constant (unless they use shunt regulation at the drive) so there will be inductive coupling to nearby conductors. Since the current varies the voltage does change although at the low impedances normally associated with power circuits the change it will not be much. It will however still be enough to potentially capacitively couple to nearby conductors. Preventing both of these effects is one of the reasons for including decoupling capacitors and occasionally inductors in power supply wiring.

Re:Should have been in the spec from day 1

[zippthorne]

Why would it have to actually go through the motherboard? Why couldn't you just have the motherboard end be separated into a data chunk and a connector that mates with a free power supply dealy. Only the drive end benefits from the convenience of having integrated power.

allowing them to draw power from the host system?

[Anonymous Coward]

As opposed to what? The external nuclear reactors we are using now?

Obligatory remark...

[mass]

Imagine a beowulf cluster of these things ....

Re:Obligatory remark...

[krog]

No one obligated you to say something that threadbare and devoid of humor. No one. You did it on your own.

Re:Obligatory remark...

[colfer]

They could light up Cleveland!

Re:Obligatory remark...

[Anonymous Coward]

nothing can light up cleveland.

Re:allowing them to draw power from the host syste

[Anonymous Coward]

Time to polish up on your reading comprehension skills.

You do know what the word "external" means, right?

/anything that gets rid of some wall warts is good in my book

Re:allowing them to draw power from the host syste

[Anonymous Coward]

as opposed to a wall wart, anal wart

Re:allowing them to draw power from the host syste

[gEvil (beta)]

Clearly the AC duct taped a power strip to the side of his computer case, thereby making any extra power outlets he uses part of the host machine. Sheer brilliance I tells ya!

no excuse

[krog]

USB has supported bus power forever. There's a protocol (devices can use up to 100mA without asking, up to 500mA with host device permission) and it works. eSATA, a newer spec, did not learn from this??

Re:no excuse

[rucs_hack]

USB has supported bus power forever.

Yes, but only well for small devices. My iPod is supposed to be USB charged, but the trickle feed is useless for it. Apart from a joystick and keyboard I have, I avoid usb powered devices nowadays.

Re:no excuse

[RingDev]

...did not learn from this?? No idea, but a political argument theory holds more water than a "because they're dumb" theory IMO.

-Rick

Re:no excuse

[krog]

That's ok; USB isn't designed to power heavy loads. 2.5W (5V x 500mA) is enough to spin a hard drive, light up the Num Lock lamp, illuminate an optical mouse, etc. And no one is saying eSATA power should be as limited as USB. I'm just making the point that a protocol for bus power existed, it is a killer feature for an external HD connection scheme, and the designers of eSATA chose to ignore it. Lazy.

Re:no excuse

[Firethorn]

That's ok; USB isn't designed to power heavy loads. 2.5W (5V x 500mA) is enough to spin a hard drive

It's enough to spin up a laptop HD, but not enough for the cheaper and higher storage but larger3.5" desktop type drives.

Thus those drives need supplimental power, which is still annoying.

I'd have been happier with a limit around 12 watts, which is enough to power a 7200RPM HD, though you might need a capaciter to limit current draw during peaks.

12W@12V would be 1 Amp, so you'd only need a marginally thicker cable(or two), and you wouldn't be limited to trickle charging quite as many devices, or needing auxillery power sources for as many items.

You might even be able to operate an energy-efficient(if slow) laptop inkjet printer off of that.

Re:no excuse

[Anonymous Coward]

Never attribute to malice that which you can attribute to incompetence.

Re:no excuse

[cheater512]

I'd be surpised if this new standard used 12v. Chances are it uses 5v.
12v is considered pretty high to be part of a consumer device.

Re:no excuse

[ScrappyLaptop]

Never attribute to incompetence that which you can attribute to managment best practices.

Re:no excuse

[Znork]

12W@12V would be 1 Amp, so you'd only need a marginally thicker cable(or two)

Multiply it by 4 for sata and 4-8 for USB, and you would, however, have a noticably thicker motherboard (and/or separate PSU connectors and caps beside the USB and SATA connectors).

It's most likely not the cable that's the problem but the actual electronics that have to support the rated draw of the cable. Or worse, imagine having motherboards that dont support the rated draw and having users calling tech support with 'my computer crashed as I inserted my USB cupwarmer and the keyboard with LCD display and cooling fan at once!!!'.

You'd end up having to have a calculator to figure out what devices you could actually attach to your computer at any one time. Much as I'm loath to say it, I prefer the wallwarts over that.

Re:no excuse

[Chris Mattern]

My cell phone recharges via USB very well. That said, I have learned that on "power cord optional" USB hubs, the power cord really isn't optional.

Re:no excuse

[TooMuchToDo]

Couldn't you simply have control over the power? As you plug in each device, it "requests" power, and the USB controller would allow or deny the request, based on current draw. I know you say someone plugging everything in at once could break things, but if you only have two hands, you can only plug two things in at the same time (and they won't both be plugged in within milliseconds of each other, while the USB controlled would only need 50-100 milliseconds to decide if it should supply the power or not).

Power control is nothing new. It just has to be done intelligently.

Re:no excuse

[Znork]

Couldn't you simply have control over the power?

If I remember correctly, that's already done in USB; you dont have a guarantee of more than 100mA, but you can request it to step up to 500mA (so you can run a 4 port hub off one USB contact without extra power).

Even so, devices like disks could theoretically store power in a battery or capacitors to satisfy their short-term spikes, but as we dont see that happening I suspect it's an actual power constraint, rather than a design mistake.

Re:no excuse

[TooMuchToDo]

Even so, devices like disks could theoretically store power in a battery or capacitors to satisfy their short-term spikes, but as we dont see that happening I suspect it's an actual power constraint, rather than a design mistake.

With the rate at which solid state drives are being researched and produced, I see the power draw of spinning drives as a moot point.

Re:no excuse

[afidel]

Yeah and enough transistors to support decent capacities and decent speed draw absolutely no power... Samsung's 32GB SSD uses .9W, only about half what an equivalent performance 100GB laptop drive uses.

Re:no excuse

[wgoodman]

Umm, couldn't someone plug in several devices at once if those devices are already plugged into a hub and the user just plugs the hub into their laptop? I know the topic is on PoeSATA, but you were talking about how USB handles things.

Re:no excuse

[darrylo]

While power-over-USB can appear to work in practice, you can (often?) violate the USB spec with USB-powered hard drives. Yes, 500mA may seem "enough", but even 2.5" laptop hard drives can draw 1A+ when spinning up. I've burned out some USB ports with cheap USB-powered 2.5" hard disk enclosures (bah, so much for overcurrent detection and protection). That sucks. ;-(

I hope the power-over-SATA specs are at least 1.5A.

Re:no excuse

[rubah]

The USB powered scanners are pretty awesome. I used to have a spot in my power strip dedicated to the scanner (it was under the desk and a pain to reach, so I didn't want to plug it in and out all the time), and now in a dorm, outlets are even more of a precious commodity, not mentioning the space. And since my computer is a MBP, I only have two USB ports, so I'm not going to waste one on a scanner full-time anyways (I could afford that at home with four), so it really works out well for me, except I have it stored on top of my shelf hiding some, um, incidental commodities.

Re:no excuse

[TooMuchToDo]

There is a scenario where many devices will be connected at the same time: boot time.

Many people leave external devices plugged in. Y'know, keyboards, mice, printers, drives, etc. Pretty much covers the board.

If the spec can control power (as USB can), then I expect it to be smart enough to handle a simple bootup.

Re:no excuse

[afedaken]

So you put down another pin header on the MOBO, and plug a pigtail with a molex connector into it. Connect this to your power supply. Not pretty, but it'd take care of that pesky amperage issue.

USB? Firewire?

[snib]

I guess I don't understand the value of eSATA. I don't see many eSATA drives, and I don't see many eSATA ports on computers or devices. Do we really need to add yet another port to laptops, in addition to the audio in/out, multimedia card, USB, Firewire, VGA, DVI, S-Video, Serial, Ethernet, Modem, etc etc? Wouldn't it make more sense to start eliminating ports and making everything work over USB, or Firewire, or some other spec?

As far as the article, it looks like a neat new development, but I know that you can get power over USB and Firewire. Maybe not enough for an external hard drive (I don't know), but IMHO it makes more sense to upgrade the power capabilities of universal technologies rather than promoting an exclusively hard drive-related format.

Re:USB? Firewire?

[jandrese]

I think the advantage is supposed to be cost and speed. eSATA is faster than USB and Firewire (I think, dunno about the latest Firewire) and requires absolutely _no_ on board logic to work. With this new spec an external eSATA case is literally a metal box with a hole in it, maybe a passthrough connector if they're feeling swanky. They don't even need the transformer anymore. That makes it cheaper than USB and especially firewire.

Re:USB? Firewire?

[Jackie_Chan_Fan]

try backing up your data to external 1TB drives through USB2. You'll soon see the importance of ESATA :)

SPEED.

USB is painful for disk transfers!

ESATA is the way.

Re:USB? Firewire?

[Solra Bizna]

FireWire is a fairly general-purpose specification, designed so that devices that require a fixed (and quite large) amount of bandwidth can be guaranteed it, and designed with device-to-device communication in mind. Its maximum bandwidth is 400Mbps (unless you count FW800, which I will as soon as I see a device that supports it).

SATA is a storage-device-oriented specification, designed pretty much so that drives can pump data over it as fast as they can read it, with a centralized paradigm and a much higher peak bandwidth at 1.5Gbps (or 3Gbps, but see the note about FW800 above).

Using USB for storage devices is perverted and wrong; it's synchronous, so your practical bandwidth is limited by the length of your cable and the response time of the nodes at either side. On the other hand, a design like that is pretty great for things like user input devices, which is one reason nobody ever talks about making FireWire mice.

So, in summary, SATA is more suitable for disks than FireWire, and USB is dog-slow. Any questions?

-:sigma.SB

Re:USB? Firewire?

[KingMotley]

They should make a USB-SATA so that we can have a univerally slow connection!

Re:USB? Firewire?

[TheRaven64]

unless you count FW800, which I will as soon as I see a device that supports it FireWire 800 has been around for ages. My old PowerBook (4 years old) and my 'new' MacBook Pro (1 year old) both support it and I have had two external LaCie hard drives for three years which have two FireWire 800 ports which allow drives to be chained together.

The spec has allowed 3200Mb/s over fibre for years but I've not seen any consumer products supporting it. The latest version of the spec (just approved) supports 3200Mb/s over the same cables and connectors as existing FireWire 800 systems.

Re:USB? Firewire?

[FreeBSDbigot]

unless you count FW800, which I will as soon as I see a device that supports it

Here are two: my coupla-years-old Powerbook G4, and the external drive that sits on my desk at home.

Re:USB? Firewire?

[waveformwafflehouse]

"unless you count FW800, which I will as soon as I see a device that supports it"

Two Firewire 800 devices that I use every day:

Lacie external drives http://www.lacie.com/us/products/product.htm?pid=10922

RME Fireface 800 http://www.rme-audio.de/en_products_fireface_800.php

While they don't use the full bandwidth individually, it's nice to be able to chain without worrying about audio/video dropouts.

So does coupling power with data restrict the potential to chain SATA devices in the future when the bandwidth out paces the drives?

Re:USB? Firewire?

[necro81]

FireWire is a fairly general-purpose specification, designed so that devices that require a fixed (and quite large) amount of bandwidth can be guaranteed it, and designed with device-to-device communication in mind. Its maximum bandwidth is 400Mbps (unless you count FW800, which I will as soon as I see a device that supports it).

One other thing that Firewire has going for it is its power spec, which (when using the full-sized connector) can provide up to 45 W. That compares to only 2.5 W for a (powered) USB port. There are 2.5" external hard drives that you can power through the USB port, but certainly not 3.5" external drives.

For reasons I can't fathom, however, I haven't ever seen an external 3.5" hard drive that can pull its power from Firewire, even though it would only require 12-15 W. Hell, my first iPod synced and recharged over Firewire, why not my external hard drive?

Re:USB? Firewire?

[geekoid]

Yeah:
If you are concerned about performance, why aren't you using SCSI? Compared to SCSI, all the are perverted and wrong.

I actually use USB2.0 for my external drive. It was cheap, and I only need it for back up of our pictures. So top speed wasn't a concern.

Re:USB? Firewire?

[fattmatt]

unless you count FW800, which I will as soon as I see a device that supports it FW800 is out there on the device side of things in the audio/video production world ... for example, meet the V3HD http://www.v3hd.com/overview.html

Re:USB? Firewire?

[Anonymous Coward]

It's probably more complex and expensive to optionally draw power from 1394 than to just rely on a separate power supply all the time. As you point out they need the separate power supply when running over USB, which is a must because a large majority of computers (e.g., every one I've ever owned) have no onboard 1394 ports.

Re:USB? Firewire?

[TeknoHog]

A "native" disk-oriented protocol is necessary for SMART, and probably for things like fancy hdparm settings as well.

Re:USB? Firewire?

[swordfishBob]

When SATA came out, it was doing less than 1/3rd what Firewire could. The same effort could have been put into developing "FW1600" etc, maintaining flexibility but gaining performance.
I don't see why SATA was necessary, but we have it now so it'll keep getting extended for a while, and it's about time this integrated power thing happened. Robust connectors would help.

Re:USB? Firewire?

[Anonymous Coward]

So, in summary, SATA is more suitable for disks than FireWire, and USB is dog-slow. Any questions?

Not really, but you did inspire a rant!

Every layer of my networking stack (except the very top) is also "general purpose", but somehow I manage to transfer all kinds of data over it. I'm sure you could make a faster "ethernet for gaming" or "IP for video", but it's so ridiculously more convenient to have one networking stack that you'd be stupid to do so.

USB is also pretty "general purpose". I remember when PC people used to try to justify different jacks for keyboards and mice, and printers, and tablets, and their UPS. Know what? It's so damned convenient to use USB for all of these devices that whatever theoretical advantages a "printer port" might have had are lost in the noise.

Even the specialized standards we have today aren't that great. Computer speeds are increasing by leaps and bounds, and I never buy top-of-the-line, anyway. I'd *gladly* give up 50% performance if I could have the same connector on every device I own.

I think a lot of people would. Maybe downloading photos from your digital camera would take a tiny bit longer, but my dad spends forever finding the right cable (USB but not a normal USB cable!), figuring out where to plug it in the camera, how to turn on the camera for the computer, and then when it's all hooked up, he hits "Import" and goes to get lunch anyway. Face it: for most people, humans are the limiting factor.

Re:USB? Firewire?

[PitaBred]

SATA's cheaper and easier to use than SCSI, but still nearly as fast in most cases. Many people who build computers just want to plug the drives in, they don't want to try to figure out which one is using which bus ID, and make sure they're all using the right ones and don't have two drives set to the same ID, etc.

Re:USB? Firewire?

[Agripa]

For reasons I can't fathom, however, I haven't ever seen an external 3.5" hard drive that can pull its power from Firewire, even though it would only require 12-15 W.

I would agree that the situation is a real shame however the two reasons for the lack of Firewire powered devices that occur to me are:

1. Not all Firewire ports have power and those that do will not power all devices so an alternative power adapter will be need to be made available anyway.

2. It is cheaper to use existing AC input power designs or adapters then to custom build the necessary regulated DC to DC converter to take advantage of Firewire.

I am actually a little surprised that Maxim or Linear Technology has not released an inexpensive dedicated Firewire DC to DC controller for external devices like they have for PoE or USB. I would find it fun and interesting to design such a circuit.

Re:USB? Firewire?

[cgenman]

eSATA is just SATA with bigger, less fragile connectors. As a standard, it's free to them to implement. eSata hard drives are just SATA hard drives plugged into the motherboard via a pair of pass-through connectors.

As a consumer, USB is much more convenient. But you're not going to be able to setup a RAID array of external USB drives, while you can do that with eSATA. And you can have an eSata port by just plugging some wires into an existing sata port on your mobo.

Re:USB? Firewire?

[Solra Bizna]

You make a valid point. However, your networking stack isn't synchronous; USB is. That's a lot more than a 50% performance drop, that's sometimes as much as a 99% performance drop. (in practice, usually around 75%.)

In any case, there is a place for USB hard disks, just not to the exclusion of IEEE 1394 and eSATA, which was the point I was responding to.

(I have yet to see an eSATA connector in person, but as other posters have already pointed out I'm routinely behind in hardware.)

-:sigma.SB

Re:USB? Firewire?

[Anonymous Coward]

The advantage is getting rid of an extra cable. Less cables means better airflow, less places to collect dust, and less stuff to get stuck in fans...

Now if only they could start adding power specs to various ports and such, there shouldn't be any excessive cabling clutter in the computer case and you'd have an opportunity for a lot better cooling airflow. Imagine if there were only one cable from the PSU to the motherboard, some sata cables, and the only other wires were to the case fans. (If they're smart, they'll add power connections to chip cooling fan clips so those wires go bye-bye too.) Also it'd be a lot faster to put the computer together.

Re:USB? Firewire?

[Shag]

Using USB for storage devices is perverted and wrong; it's synchronous, so your practical bandwidth is limited by the length of your cable and the response time of the nodes at either side. On the other hand, a design like that is pretty great for things like user input devices, which is one reason nobody ever talks about making FireWire mice. Somewhere out there, a hardcore gamer is bemoaning the slowness of his USB mouse, and wishing there were FireWire mice.

How about for internal drives as well?

[DoofusOfDeath]

Is there much of a reason that we couldn't have single power+data connectors for internal HDD / DVD drives as well? Things are better now that IDE cables are less common, but I'd still be happy for a cleaner interior of my cases.

Re:How about for internal drives as well?

[Anonymous Coward]

We do have them. It's just most people aren't willing to pay for it.

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

Re:How about for internal drives as well?

[hey]

Nice idea. I like it tidy inside my computer as well.

Re:How about for internal drives as well?

[Firethorn]

They take a lot of juice, potentially? Your standard wire going to the molex plugs in use are around 16 gauge, not exactly small. Mostly overkill, but traces are normally designed for fractions of an amp - think about video cards and their auxillery power ports today.

Re:How about for internal drives as well?

[Jeff DeMaagd]

SATA drives are already the equivalent to SCA, I don't think SCA was commonly done on a cable.

I can already slide a SATA drive into my computer and it plugs right into a back plane, without any cables. I also have an external hard drive enclosure whose drives automatically plug right into a backplane using the SATA power & data connector. eSATA is a little different.

Re:How about for internal drives as well?

[MBCook]

Not going to happen. There are certain considerations in external drives. Most won't take a ton of power, they'll be 7200 RPM or something like that. In a case, you see people with 10 and 15k drives that use much more power.

The biggest problem is that what we have works very well. It supplies a few different kinds of power (3.3/5/12v?) so they drive probably doesn't need to step that up or down. Using one power connector the external drive will have to step down the power from the max (12v?) to be able to power it's logic circuits. That wastes energy, space, and components. It's more expensive.

Then there is the circuit board part. Powered SATA means that you have to get the power to the connectors. The more power you want to run (like that needed for 10/15k drives, or CD or DVD writers) would be problematic. The more power you deliver, the larger the trace on the motherboard has to be so it won't burn out. So now you either have large traces running all over the place, or you have to put the ports next to the power supply connector. Or you make a new additional power connector going to that area of the motherboard.

Or you could make a connector that goes in the middle of the cable to supply power on route, instead of from the motherboard. But that would only mess up your case more.

The current specification works very well for internal stuff, I would be amazed if they changed it to provide this kind of thing for even modest internal drives.

Re:How about for internal drives as well?

[poot_rootbeer]

Is there much of a reason that we couldn't have single power+data connectors for internal HDD / DVD drives as well?

Nostalgia? The big ol' 4-pin Molex power connectors are practically the only thing inside a PC case that are still the same as they were when IBM first introduced them to the desktop twenty-seven years ago. If we get rid of those, we'll be severing the last remaining connection to the machine's origins.

Re:How about for internal drives as well?

[MMC Monster]

Take a look at modular PSUs. More expensive, but definitely better. You basically plug in whatever cables you want into the PSU.

Re:How about for internal drives as well?

[Guysmiley777]

(3.3/5/12v?)

The standard 4-pin molex connector has one 12v (yellow), one 5v (red) and two ground pins. There is no 3.3v pin.

Re:How about for internal drives as well?

[TheThiefMaster]

But a Sata power plug does have a 3.3V pin. A lot of sata hard-disks work without it, but this is either because they were designed for molex power (and may even have a molex socket), or have been designed to step 5V down to 3.3V if they need it and it isn't available.

Re:How about for internal drives as well?

[MrNemesis]

Think of every 12V line going into every hard drive in your machine. Now think of every 12V line having to be routed through the motherboard.

It essentially won't happen because it'll make motherboards much more complicated (read: expensive). That said, power-over-SATA shoudl have been in the e-SATA spec from the beginning, glad I didn't hop on the bandwagon earlier.

Re:How about for internal drives as well?

[RollingThunder]

How would the external drives use any less power than internal? I don't think anyone's expecting these cables to be as thin as internal SATA cables; nor do they expect internal "power over SATA" to not require a beefier unified cable. They just want one simple cable to avoid the clutter.

Re:How about for internal drives as well?

[British]

If we get rid of those, we'll be severing the last remaining connection to the machine's origins.

Good riddance. Trying to mitigate ribbon IDE cables + daisy-chained power cables was never fun. I just wonder why it took 27 years to do it. Really, I will have no gloating of the PCs of tomorrow having no inherent design flaws with PCs of yesteryear. The less of a rat's nest of cables in a PC, the happier I will be. No more worries of cables getting in the way of fan blades, etc.

Re:How about for internal drives as well?

[Anonymous Coward]

Don't fret, there's always the PC speaker!

Re:How about for internal drives as well?

[Anonymous Coward]

It's still just avoiding the real problem, which is that internal SATA hard disks shouldn't *need* cables at all.

SATA is designed with the connectors in the same place, so if the case is designed for them, you just slide them in. I recently bought an external enclosure, and it did this right: just slide the disk in, and push a lever. Apple, of course, is smart enough to do this with their tower, as well.

I guess the real question is: where are all the PC cases that are actually designed for SATA? You could even make a case that worked for any motherboard, using mostly-fixed cables from the case to the edge of the motherboard, so the non-standardization of motherboards is no excuse.

Re:How about for internal drives as well?

[maglor_83]

Actually they already have special SATA power cables. They have a similar connector to the data cable. So all thats left is fans...

Re:How about for internal drives as well?

[Firethorn]

I was thinking of the DVD part - and why you'd still want a seperate feed for powering internal drives.

For external drives, sure they use extra power, but you could incorporate a molex plug to keep the power off the motherboard. Or at least next to the powered external SATA plugs.

Externally unifying the cable matters a bit more than internally, at least in my mind.

Combining the power and data cable on the MB would/will complicate matters in motherboard design.

Re:How about for internal drives as well?

[Kjella]

Big connector on motherboard near PSU, wires glued to the underside leading to whereever the SATA ports are on the board and basicly punch through? I agree that if you make them part of the motherboard layers it'll make it more expensive, but I really don't see any reason to do that for a pass-through.

Re:How about for internal drives as well?

[cgenman]

If we get rid of those, we'll be severing the last remaining connection to the machine's origins.

They're also one of the only things that continues to work well. They're impossible to put in upside-down. They're beefy and long-lasting. They hold together well. They're easy to modify to fit special needs. They're dirt-cheap to make.

You could argue that smaller Molex connectors would be nice, but you might be able to reduce the size by 50% in exchange for a lot of retooling and standard confusion. But you can't argue that Molex is underperforming or is otherwise becoming a bottleneck.

We've also got the same door size in the backs of the cases, but we don't have a compelling reason to make those smaller or larger. They just work.

I'm all for power on one unified cable internally and externally. But the hearty Molex connector deserves our respect.

Re:How about for internal drives as well?

[owlstead]

Nope, those small piezo speakers on the motherboard don't look a bit like the big magneticky ones in the original.

Re:How about for internal drives as well?

[afidel]

They are only better if you prefer neatness and flexibility over efficiency. Modular PSU's are always ~10-15% less efficient than the same design in a nonmodular form.

At least its not Power Over SATA

[517714]

POS would be a less than desirable acronym.

Re:At least its not Power Over SATA

[theRiallatar]

You say POSATO, I say POSATA.

Re:At least its not Power Over SATA

[MiniMike]

Maybe they could call it PATA to avoid confusion...

Re:At least its not Power Over SATA

[Anonymous Coward]

eh... let's call the whole thing off?

lets hope it doesnt draw much power

[Anonymous Coward]

i have had USB drives that needed both USB connectors in just to power it up, some PC's are ok with 1 but can you take that chance on a site visit ? power from the host is a good idea as long as the host has enough power to give

Sweet

[nog_lorp]

Am I the only one who is just really happy about this?

Re:Sweet

[Anonymous Coward]

Yes.

Re:Sweet

[nog_lorp]

Ouch :( I didn't think it would be redundant in the face of all the other comments being negative on the topic :/

bah

[Anonymous Coward]

wake me up when we have sata over power

40% more pins!? ARE YOU CRAZY?

[hxnwix]

Why wasn't it in original parallel ATA? 2 more ATA-66 pins would be a mere 2.5% pin count increase, whereas here it is a 40% increase!

If only they'd thought it out...

[ricebowl]

Power Over eSata?

Look at those uppercases, the only acronym/abbreviation they can go for is either POeS (not too great, but better than) POS...

I can only hope it's a meta-commentary, the designers' own reaction to another port and yet-another-acronym...

Wha-d-ya-mean "power cable?"

[DanQuixote]

One of my tech support calls was about 1980, my friend's mom had a computer, and she bought a printer, which she tried to hook up herself, but it wasn't working.

I went over there and quickly spied the problem... the data cable was connected, but there was no power cable hooked up.

She quite innocently and logically asked, "why do I need a separate power cable?"

People don't really give a damn that the power system and the data system are two separate systems. It really is completely reasonable for them to expect a single cable to power as well as communicate.

These folks shouldn't pat themselves on the back for a "new feature", they should try harder next time to close a bug out in something much less than 30 years!

This is a basic usability requirement that people persistantly ignore despite the rat's nests of cables running around all their gear. This is certainly one of the biggest reasons for the popularity of USB!

Re:Wha-d-ya-mean "power cable?"

[harrkev]

She quite innocently and logically asked, "why do I need a separate power cable?"

Does she also plug her cable feed (or sattelite receiver) into her television and expect it to work without plugging the TV into 120V?

Re:Wha-d-ya-mean "power cable?"

[Khaed]

Maybe she's Amish and doesn't have a TV you insensitive cl- oh wait, printer, right.

Re:Wha-d-ya-mean "power cable?"

[AKAImBatman]

It really is completely reasonable for them to expect a single cable to power as well as communicate.

Should have got her a Coleco Adam. Nothing like a computer that's tied its printer at the hip! ;-)

Re:Wha-d-ya-mean "power cable?"

[DanQuixote]

Clearly you are a good engineer, and as every good engineer knows, it's all about trade-offs. If Tesla had his way, there would indeed NOT be a separate power cord for the TV.

Overall historically, we've made pretty good decisions about how to handle power. However, in the last 10 years I have been very disappointed with consumer electronics. Powering a device is a major requirement for anything we design, yet batteries still suck, wall-warts continue to proliferate, mp3 players don't charge via a standard USB port, and I STILL have to plug every last item into it's own special power cord, despite the inconvenience.

<rant>Why is power still an after-thought during product or specification design???!!!</rant>

Re:Wha-d-ya-mean "power cable?"

[geekoid]

Did she complain when she needed to plug in her VCR separate from the TV?

Re:Wha-d-ya-mean "power cable?"

[darkwing_bmf]

However, I don't expect my monitor to have a single connection to the PC any time soon.

We PC users may not expect good design, but it would still be nice to have. I want a less cluttered computing area and I've read more than one article stating that consumers in general want less cluttered computer areas.

Re:Wha-d-ya-mean "power cable?"

[noidentity]

I went over there and quickly spied the problem... the data cable was connected, but there was no power cable hooked up.

She quite innocently and logically asked, "why do I need a separate power cable?"

For the same reason your house needs a separate hookup for phone and electricity. Or alternately, for the same reason you have a toaster for small things and a full-sized oven for big things. And I'd like to keep these things separate, rather than try to cram them all into one thing. Though with hard drive power, there's little downside to having power delivered over the data cable.

Re:Wha-d-ya-mean "power cable?"

[DanQuixote]

Any idea, taken to the extreme, will always break. This is another thing good engineers learn quickly.

"house needs a separate hookup for phone and electricity"

Evidently you're not familiar with the Broadband over Power Line (BPL) specification. I know it gives the ham operators problems, but there are definitely people out there who want data and power over the same line.

Everyone who buys phone, cable TV and internet service in a package would agree with me... given today's technology, we should no longer have to live with such a tangle of cables.

Re:Wha-d-ya-mean "power cable?"

[grumbel]

### However, I don't expect my monitor to have a single connection to the PC any time soon.

NeXT monitor doesn't need a seperate power connector, comes all in one cable.

Re:Wha-d-ya-mean "power cable?"

[grumbel]

### For the same reason your house needs a separate hookup for phone and electricity.

Which reason would that be? A good one or just "because thats how it always was"?

Just for the record, my phone works without connecting it to a seperate power supply, power comes right from the phone cable.

Re:Wha-d-ya-mean "power cable?"

[Frenchy_2001]

"why do I need a separate power cable?"
because most computer systems are not designed to provide the few amps of current that a laser printer can need?
Because most of those interface are designed for low power peripherals and have specs mirroring that (USB for example can feed 0.5A into its own cable), but more powerful peripherals get plugged into it. So, to work, they need more power and get an external adapter.

The *REAL* problem comes from people unable (or, much more likely unwilling) to follow directions. Most recent piece of equipment, be them electronics or computer peripherals include a quick plug guide, but those people will NOT read it.
My wife is one of them. She will NOT follow any instruction, but will then complain loudly that it does not work. Packages get ripped open and manuals discarded to the side while she tries to figure out the way to use it, gets frustrated after 5 minutes and complain it wont work. She will not open a manual, no matter what amount of frustration she will endure.

Is that new spec good and interesting? Sure, but then dont complain that you get crashes on your computer, because that new magic peripheral just pushed your power supply over its limit (most brand computer has very little overhead in its power supply).

Re:Wha-d-ya-mean "power cable?"

[PitaBred]

In every single major disaster that has happened lately in the US, ham operators have been essential in communicating with the outside world when traditional lines of communications are down. Denying them the ability to practice is a bad idea, IHMO. It'll just make things like Katrina impact people that much worse.

Re:40% more pins!? ARE YOU CRAZY?

[rrkap]

It did get included eventually. 2.5" drives use 44pin IDE cables that carry signal and power.

Re:40% more pins!? ARE YOU CRAZY?

[harrkev]

Well, the Parallel ATA interface typically uses small wires (probably around 24-28 gauge). Do you really think that you can power a whole hard drive over a pair of #24 wires? Nope, not even close. You would either need to use different wires, which would require a custom connector, or you would have to use MULTIPLE wires (probably about 12-20 should be enough. That, however, makes the connector wider, the cable wider, and increases cost for everybody.

Thick wires are cheap, and molex connectors are cheap.

Re:allowing them to draw power from the host syste

[sexconker]

This is actually a damned good idea - why haven't I seen cases with power strips???

Re:allowing them to draw power from the host syste

[miscz]

Have you ever used external SATA drive?

Existing power situation with SATA is hilarious

[dwalsh]

Parallel ATA (A.K.A IDE): Big Parallel data cable with a shitload of pins. 4 pin power cable.

Serial ATA: Serial data cable with just 7 pins. Power cable has twice as many pins!

Did they just move the lines to the power cable? :-)

Re:Existing power situation with SATA is hilarious

[Gat0r30y]

7 pins, but just 4 wires going in (unless you got a 3.3V line too which is not necessary on the bulk of SATA drives). The 7 pins make it hot swapable, you can just yank power to a drive, and it doesn't hose the whole thing. I do failure analysis on hard drives all day (its my job), and for me it is about the best part of the SATA spec (since I don't have to reboot machines just to throw a different drive in for testing).
On another note, I'd guess this is also why it took so long to come out with an eSATA power spec that would work, since occasionally it matters whether you pull the SATA cable or the power cable first. Pulling power first is generally the way to go, as sometimes if you pull the SATA cable first the machine will get pissed, and you gotta reboot as your SATA controller will just sit there waiting to hear back from the drive that no longer has a connection.

Re:Existing power situation with SATA is hilarious

[nerdguy0]

You realize they did this because those little pins that contact inside the connect can only handle a couple 100 mA of current before they start to melt through the plastic, right? They had to tipple up each power and ground rail just to the high current requirements of current spindle based HDs. However, with the new SSDs coming out, it may no longer be the case for much longer.

Re:Existing power situation with SATA is hilarious

[dwalsh]

You realize they did this because those little pins that contact inside the connect can only handle a couple 100 mA of current before they start to melt through the plastic, right?

... yes I suspected as much. Kind of suggests they should not have used such small connectors? A molex or new connector with 2-4 big pins (molex are naff) would have done the trick.

Something I have never understood

[brunes69]

"The spec has allowed 3200Mb/s over fibre for years but I've not seen any consumer products supporting it. The latest version of the spec (just approved) supports 3200Mb/s over the same cables and connectors as existing FireWire 800 systems."

I have never understood this about "specs". If the cable already had the physical bandwidth to transmit 3200Mb/s from day 1, then why did the orginal "spec", which is just a document after all, specify that higher number? I have the same question for USB 1.1 vs. 2.0 - why wasn't it just 480 Mbps from the get go? Was it because of chip processing limitations at the time or something? Because that argument doesn't hold water for me; it would be perfectly valid to write the spec for the THEORETICAL FUTURE when such speeds are possible and just wait for hardware to catch up with the spec.

I know it can't be as simple as it seems but I just have never understood this.

Re:Something I have never understood

[rrkap]

Because that argument doesn't hold water for me; it would be perfectly valid to write the spec for the THEORETICAL FUTURE when such speeds are possible and just wait for hardware to catch up with the spec.

When you see a device described as USB, USB 2.0, FireWire 400, SCSI-160 or whatever, you should be able to make assumptions about how quickly the bus will perform. If your standard is for a performance level that you can't achieve then knowing a device conforms to a specific spec says nothing about how it performs. Also, you don't know what the future will hold. I don't think that most people back in the days of 10 base T would have believed that unshielded twisted pair copper was capable of carrying 100 times as much data as it was carrying then. In that case, it turned out that lots of existing cable was perfectly suitable for far greater performance than would have been the theoretical spec 15 years ago.

It's worse!

[swb]

Powered External SATA.

PEeSATA anyone?

Re:40% more pins!? ARE YOU CRAZY?

[magarity]

Do you really think that you can power a whole hard drive over a pair of #24 wires?
 
My first thought was a standard skinny 80 pin flat cable with a couple of fat wires for power running down one side making it look all lopsided. what would be the retail prices on something like that, lol...

First make a working Expresscard eSATA

[Anonymous Coward]

Good, now maybe someone can make a working implementation of eSATA on an expresscard.

Apparently Silicon Image, who makes the SII3132 chipset and driver used in ALL expresscards supporting OS X, is now claiming that the failure of their drivers after updating to OS X 10.5 (they work with 10.4 & Windows) when using a Western Digital MyBook Premium ES or a Seagate Freeagent Pro external drive is a problem with the drive and not their drivers. They point to some issue with the Oxford chipset used in the enclosures. Oddly, the drives work with the card and with Windows and OS X 10.4.

Where does that leave everyone who owned a working card and drive, and upgraded to Leopard? It leaves us with the door slammed in our faces. Fortunately, this problem has now been covered by my bloggers and others are becoming aware of just how shitty the Silicon Image Chipset is. If you search for them on wikipedia, you'll find that this isn't the first time they've released a buggy chipset and not fixed it. Another of their SATA chipsets would corrupt drives under heavy loads.

So lets see if these tech companies can first implement the existing spec properly, and then worry about introducing a new standard.

Bad acronym

[dustmite]

Yikes, did some funny South Africans tag this article? 'Poes' is a 'lekker' South Africanism meaning, more or less, 'pussy' ...

12V unusual?

[Firethorn]

Well, 5V is pretty odd for consumer devices as well - you have the actual choice of 4.5 or 6 using normal cells.

Still, it's pretty close to 4 rechargable NiMH cells (4.8v).

12v is considered pretty high to be part of a consumer device.

What sort of consumer device? There's lots of consumer devices that run at 110V, for example. ;)

Yes, 12V is high because you need 8 cells in series to get that much voltage - most devices just don't need that much wattage today. I'd mostly recommend 12V for USB simply to be able to ship a 'decent' amount of power over a thin wire.

Re:12V unusual?

[cheater512]

Why are you talking about batteries?

5v is standard for TTL (transistor-transistor logic) digital circuits. 3.3v for more complex chips and 1.8v for low power stuff.
Good luck getting batteries to produce any of those voltages.
You will find all three of them plus 12v in your computer however.

Re:12V unusual?

[Firethorn]

However, you talked about consumer devices - which are normally powered off of batteries.

The internal voltage can be handled by various converters today rather easily.

Re:12V unusual?

[cheater512]

Yeah but they use DC-DC converters to get the required voltages.

Re:12V unusual?

[Firethorn]

That's what I said, I simply dropped the DC-DC part.

Comment removed

[account_deleted]

Comment removed based on user account deletion

I have switched back to PATA drives only

[Anonymous Coward]

The main reason is that they don't get re-assigned to different devices. With PATA, you know which physical drive will get /dev/hdc every time.

I understand that some of this can be corrected by configuring udev rules.

However, if you are booting off of a linux software RAID1 disk, it is impossible to write an MBR to handle all types of drive failure. If the two MBRs are set to load the kernel from their respective drives, and one fails such that it disappears to the system altogether, if the other one now becomes the first one's device name, the machine can't boot. One the other hand, if the first (normally booted) one fails such that it is still detected but not a member of the RAID, the machine will boot by reading the kernel off of the bad drive; even if that doesn't cause problems, when you swap the bad drive out, the machine will no longer boot.

I have also had problems with using USB and Firewire disks for scripted backups. They were mounted automatically at night, backed up, and then unmounted, and someone physically rotated the backup set during the day. After a few months the device number would change, and then the backups would stop working; in some cases, the wrong disk was mounted and had a backup dumped on it filling it up. This seniario is fixable by the correct udev rules configuration, of course. But, udev should not have to be a kludgy patch on top of a broken-by-design way to hook up things to a computer.

I am boycotting all USB, Firewire, and SATA devices until something better comes along.

Connect Everything Over Ethernet

[dogganos]

Why SATA and eSATA and IDE and USB, FireWire and DVI and and and??...
If everything just used a simple, yet as it has been proven efficient protocol like Ethernet, then our lives would be much easier.
Oh, and Ethernet also has Power Over Ethernet for the hungry devices. I wait for the day when I will plug an RJ-45 jack into my hard drive (which will not be a hard-drive, but an SSD).

My $0.02

My prediction: solid state drives

[Schraegstrichpunkt]

My prediction is that this will be more useful for solid-state "disks" than it will be for mechanical hard disks.

K.I.S.S.

[hurfy]

How about we just make a external connector that we plug a molex connector into the back end of? Could be molex or another shape or whatever. Laptops could biuld one in and desktop could have an extremely simple one that goes in a slot cover with the external connector(s) on outside and a molex on the inside. Could build-in something to step down voltage to have several choices available at connector even. Then a power cord going to that could replace the wall wart powering SATA,USB,speakers or virtually anything by grabbing power off the correct pins. Still 2 cords but at least they would going the same direction now ;)

Comment removed

[account_deleted]

Comment removed based on user account deletion