Seagate to Drop IDE Drives by Year End

ianare writes "Seagate plans to cease manufacturing IDE hard drives by the end of the year and will focus exclusively on SATA-based products. Seagate is the first major hard drive manufacturer to announce such plans, though others will likely follow suit. That's not to say support for the 21-year-old PATA standard is going to vanish overnight; similar to how ISA slots were available long after most of us had ditched our old ISA peripherals."

Comments

Good

[Espectr0]

At the least, this will drive the price of SATA drive down. Maybe it will be the same like RAM, where DDR2 is actually cheaper than the old DDR memory standard.

Re:Good

[CastrTroy]

All of us with DDR RAM are pretty bitter about that. I was pretty bitter about a year ago when I tried to buy SDRAM. That stuff is expensive. Still, I can hope that we can go back to the good old days (march 2001???) when SDRAM was $CDN 30 for 512 megs. That was when RAM was the cheapest it has ever been, at least considering how much you could do with 512 MB back then. Now that's that won't even get you the shiny desktop on windows vista.

Oh fuck.

[r00t]

What will I do when my drive dies again?

I happen to like my computer. Being fanless and well-built, it is quite reliable except for the damn hard drive.

What about osdev?

[Anonymous Coward]

I am currently writing a kernel that will depend on IDE (ATA, now called parallel ATA) for hard disk drive access. I will be using pio mode 0 (around since ATA "Advanced Technology Attachment" cam from the IBM AT) for the best compatibility with both old and new i386 compatible machines. What does this mean for kernel programmers doing small projects to learn? How much hard is Serial ATA to use from the kernel's perspective? Is it backwards compatible with the old PATA?

- The captcha is "faceted."

Re:What about osdev?

[larry bagina]

As a consumer, I'd rather get rid of the legacy shit (ATA, ps2 keyboards, bios, DOS/Windows :-). But for hardware hacking/os writing, a USB stack, firewire stack, etc are more work (and don't provide the immediate feedback like 100 lines of assembly to read the raw keystrokes).

You an still have fun with an ARM breadboard kit, though :-)

Re:What about osdev?

[OrangeTide]

Well Intel's SATA interface has become the standard and at least some vendors are duplicating the interface (while providing an extended interface of their own). Also the cards have ROM, so if you're writing something for 16-bit real mode you can just make normal bios calls to your favorite ide, scsi or sata controller.

It's sad that it's far easier to use an Ethernet card at the lowest level than it is to use a USB host controller.

If I were to write an OS from scratch I would probably only implement support for SATA for the start. I would have every block device go through a SCSI abstraction. It would only support 64-bit on x86 version of the OS. I figure by the time I finished the OS, that IDE, 32-bit, PS/2 and what not would be obsolete (it already sort of is). I'd probably not even support AGP and only pci-e style memory apertures.

To be honest the days of easy hardware hacking are over on the PC. I think for that kind of thrill you need to pick up a Nintendo DS, PSP, etc.

Right now I'm toying with the idea of making a cheap hacking "game system/home computer". Apparently Winbond makes an all-in-one chip for making those direct-to-tv toy game systems. 27MHz 65816(same cpu as SNES and Apple IIgs, it's like a 16-bit Commodore-64), sprite memory, generic I/O pins(nice for hooking up to an MMC/SD socket), and other goodies.

I probably won't get anywhere with the idea unless someone wants to chip in to invest in the idea so we could put enough systems together to sell on Thinkgeek or something.

It would be like the XGameStation or Hydra in terms of being a learning tool/hacker toy, but a fraction of the price and more like a real game system than some weird collection of off-the-shelf chips.

Re:What about osdev?

[Hal_Porter]

Really now? Ever heard of a thing called ACPI? If you have a laptop and have used the hibernation mode, you're executing code that is more or less in the BIOS.

That's true of APM - the OS actually made Bios calls and the Bios responded to events like pressing the suspend button directly. Since the Bios is real mode and non reentrant that was an issue. But it's not true of ACPI - the bios has methods in AML byte code but the OS is responsible for executing them via an interpreter. And the reason it uses byte code rather than native code is because it was designed to work on both x86 and Itanium. So EFI uses ACPI too for power management. Of course byte code in a virtual machine is hopefully a bit safer too.

And lets not forget that booting is still an important role in itself. Not only is there hardware initialisation, but there's the important role of loading the OS and/or boot loader. In fact, the reason that boot loaders exist (e.g NT boot loader, LILO, GRUB) is because the PC BIOS (interface) is so simple and unable to do anything more than load the first sector from a device and jump into it.

Which is an excellent place to stop. Trying to do more like ACPI or ARC firmware which it evolved from means you need to have filesystem drivers and network stacks in ROM. And magic system partitions which you need to start the machine and are mean a reinstall of everything if they get corrupted.

Booting from the network or other unusual devices has always been a little difficult. OpenBoot and now EFI makes this stuff easy because it's based on an extensible framework instead of hacks and workarounds for the backward-compatible legacy from an ancient platform (the original IBM PC, over a quarter of a century ago).

You can boot off the network with a normal Bios. Or anything else - you just need an option Rom which implements int 19h. Or the Bios itself could support network booting. And just because you don't understand it, don't assume it's a mass of hacks and workarounds.

Re:What about osdev?

[Xiph1980]

By jonwil (467024)

The #1 reason I want something like EFI is to eliminate the world of proprietary bootloaders/selection mechanisms for good. Essentially the BIOS would be the one that displays the list of boot options.

Unfortunatly no vendor that supports EFI (including all Linux distros I have seen) gets it totally right (where any boot time configuration options are handled through EFI and not through another bootloader)
Well, EFI may not be the best way to get away from proprietary stuff. It seems that EFI explicitly vacilitates such behaviour by hardware manufacturers:

Interview with Ronald G. Minnich (Google cache)

What are your thoughts on the Extensible Firmware Interface (EFI)?

I have spoken with the EFI authors at length. They make no secret of the fact that a "core value" of EFI is the preservation of intellectual property related to chipset programming and internal architecture. To put it another way, EFI is dedicated to the preservation of "Hard" hardware (as defined above), and the provision of binary interfaces and subsystems to BIOS vendors and others.
It is not really possible to build a full open-source BIOS if EFI is involved. The Tiano system, which Intel claims is an open source BIOS, can not be used to build a BIOS unless it is attached to proprietary, binary-only BIOS code provided by a vendor.

Another important thing to realize about EFI is that it also contemplates enabling chipset features that will trap certain OS operations to an EFI-based control system running in System Management Mode. In other words, under EFI, there is no guarantee that the OS owns the platform.
Accesses to IDE I/O addresses, or certain memory addresses, can be trapped to EFI code and potentially examined and modified or aborted. Many see this as an effort to build a "DRM BIOS".
I am not sure what the real intent of this design is, but is is a real concern in secure environments (such as those found in governments, banks, and large search engine companies). A number of vendors and users have told me that they are not sure they can ship an EFI system they are willing to trust in a secure environment.

It's sour.

[Anonymous Coward]

The tech industry as a whole deprecates and wastes so much. It is a wasteful nightmare.

Re:It's sour.

[garett_spencley]

I'm glad it wasn't forced on the market within a year like the AGP to PCIe seemed to be. Sure, you could get AGP cards but the standard was relegated to second class treatment almost immediately.

Indeed. I was one of the poor unfortunate clods who went and upgraded his video card during the transition from AGP to PCI-e. I could have gotten a PCI-e version of my card but I only wanted to upgrade my video card, not my mother board etc. so I went AGP. I guess by now (about 2 years later) I got some good use out of it. But I'm the type of guy who likes to upgrade one component at a time as priority demands. Problem now is, in the last 2 - 3 years so many standards have changed so quickly. Much faster than I remember them changing (though that could just be due to aging). My current PC is pretty ok for my needs. But I'm starting to feel obsolete. It's single-core. 2Ghz. 1GB Ram. AGP card. IDE drives. When I upgrade I'm going to have to ditch this PC entirely and go BTX, dual or quad core, SATA, PCI-e etc. It will be an investment of a grand or two when I'm used to just investing a hundred or two here or there to upgrade what needs it.

I strongly believe that the main reason so many people are stuck with ancient old PCs from the mid - late 90's is price above anything else. Yes computer prices have come down dramatically. You can buy a PC for a couple hundred now. But a lot of people have WAY more important things to spend a couple hundred on. Like bills and food etc. And if their PCs fulfill their basic requirements then there's no reason to go brand spanking new. Right now we seem to be at a point where it's brand new or nothing. Simply because so many standards have been ditched for new ones in such a short period of time (ATX to BTX, 32-bit to 64-bit, single core to multi, IDE to SATA, DDR to DDR2 just off the top of my head).

Even if most of the standards have existed for some time, it's the manufacturers who, all of a sudden, decided to force the new ones all at once. That's how it feels from a budget conscious consumer.

Re:but the motherboards!

[karnal]

Serial ports are useful. Not so much in the home, but they're still useful.

Of course, a little USB-Serial dongle solved that issue for me when I had a thinkpad t42 at work a while ago...

When will old PCI die?

[Michael Woodhams]

My motherboard has great big old PCI slots, and tiny little 1xPCI-e slots which are just as capable. PCI-e has taken over for graphics cards, but I've never even seen a 1xPCI-e expansion card. (The motherboard manufacturers don't believe they'll be used either - they put them next to the 16x slot where double-width graphics cards will make them inaccessable.)

When will old PCI die? Perhaps very small format motherboards and laptops will eventually drive demand for 1xPCI-e cards?

For that matter - is there any reason for low-end PCI-e graphics cards to be 16x, rather than 8x or even 4x? (They'd still fit in a 16x slot.) I suppose there is no demand - any PCI-e motherboard has a 16x slot, and there isn't anything you'd want to put in it except a GPU. About the only use I can think of is if you wanted one computer to run many low-performance displays - e.g. 8 monitors off four GPUs, each using a 4x slot.

IDE graveyard

[Vskye]

This really kinda sucks. I have a computer that needs a few legacy items like IDE, Serial and a parallel port. Why? Well, serial port(s) for my ham radio stuff and a parallel port for my perfectly good HP 6L printer. (might be an unknown issue with the IDE side)
 
I also like to go back and play with a older OS sometimes which doesn't even see a SATA drive. Guess it's time to stock up on a few IDE drives.

Re:but the motherboards!

[Andrew_T366]

I don't know about you, but I'd MUCH rather have parallel, serial, PS/2, and IDE connectors--which are backwards-compatible with most everything and do what they are meant to do well--than a half-dozen more USB or FireWire ports that don't even correspond to any devices that I personally use.

USB keyboards require special drivers and offer no interface-speed advantages unless you type at superhuman speeds.

slashdot nerds not nerdy about terminology?

[Anonymous Coward]

May the collective Slashdot mind forgive my anonymous cowardliness if I am wrong, but it is my understanding that ALL modern hard drives are of the IDE variety, Integrated Drive Electronics. That just means that the electronics are soldered to the drive along with the platters rather than a card or something else, correct? I see multiple people referring to IDE drives and IDE channels but http://electronics.howstuffworks.com/ide.htm/ also agrees that IDE is not the true name for the interface standard; just that since almost all IDE drives are of the Parallel ATA variety, the two terms are used interchangeably. Should the rapid onset of Serial ATA (rightly) make us change our terminology?

What about USB enclosures?

[baeksu]

As far as I've seen, most USB enclosures have IDE harddisks inside them. The same is probably true for firewire as well. So there's still a lot of IDE harddisks on the market, and people do want bigger capacities as well.

Of course as a private company, Seagate are welcome to do as they please. There's still a few other manufacturers out there.

For desktop PCs, I think it would be silly to buy IDE-to-SATA converters. At least the ones in Korea cost close to 30 bucks. Most of the IDE harddisks people have are probably around 100-250 GB size, and you can already get that size SATA drives for less than 50 bucks. So the converter is not much of an investment really.

Compaq HD interface - the original IDE

[calidoscope]

Out of curiosity, did IDE have a standard interface when it started, or did everyone adopt the most popular one?

The original 'IDE' drives were made for Compaq by Control Data (whose disk drive division is now part of Seagate), so that could be thought of as the original standard. The intent was to have something that acted a bit like a standard MFM drive + controller to allow for a simple interface to the ISA bus. The original IDE port was on Compaq's multifunction I/O card that had the FD controller, parallel port, serial port and IDE port on one card. The original drives were 'dumb' with no information on drive geometry.

The P-ATA interface uses the same physical connector as the IDE interface, but incorporated much of the SCSI command set instead of the low level disk controller command set used on the original IDE drives.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:but the motherboards!

[Anonymous Coward]

The problem is, that most serial dongles don't support break signals correctly. Which is annoying when you want to persuade a Sun to the Boot PROM, or indicate to a datalogger that you'd like it to wake up, and start talking to you. Setting it to the lowest speed then sending 0xFFs for a bit usually works, but it's far from elegant.

Look into solid state (compactflash) replacements.

[Chonine]

There is a lot of perfectly usable hardware out there, which has the one reliability weakness with the hard drive. The latest IDE drives work great usually going back to some pretty old hardware, although you may be limited somewhat (depending on the MB and OS). The SATA drives break the compatibility, although you will probably be able to get SATA to IDE adapters for some time to come. Problem is, that will be for desktops only.

I have a small collection of some older Thinkpads. One thing that I have been using are notebook IDE (44-pin) to CompactFlash adapters. There are even some dual CF adapters available such as http://www.addonics.com/products/flash_memory_read er/ad44midecf.asp. Twenty-two bucks. Since it is IDE, the bus still has a master and a slave for it, and you can have two drives essentially in that one notebook HD slot. I think everyone is waiting for solid state drives to arrive on the scene (affordable ones), but most of those will probably be SATA. So this lets you get two 16GB CF cards into the single IDE slot on a laptop, and it runs silently. It is also cooler, weighs less, uses less power, faster access (not necessarily transfer), and they are much more reliable and rugged (the limited writes isn't as much of an issue now). It seems like a good way to patch up old hardware's Achilles' heal.

It is probably a good thing to look into for the 3.5" desktop drives too. As CF cards continue to grow and fall in price, I expect in a few years all my modern SATA equipment will be using SSDs, and my older PATA equipment will have large cheap dual compactflash cards. Some of the hardware is so slow that all I really need is a 1GB CF card to store a minimal Linux distribution on it anyways.

Re:How nice for you.

[Anonymous Coward]

I second that.

Generally, I get sad with ever widening gap between users and technology. USB is "new" serial/parallel port but I cannot use and control it like I could do with legacy ports. I can't fiddle with it without buying expensive, underperforming "USB-to-whatever" bridge chips that obscures what is really going on on the wires. Like back then when PCI superseded ISA, "new, better" replacement is vendor locked, if you want to use it, make your devices that run over it, you need to buy yourself an ID from a regulating body.

Each new hardware "improvement" is more and more anti-hacker, more exclusive club, more "keep out!".

We are often discussing free software issues here on /. , but DRM, TrustedComputing and TiVoisation are in fact problems with closed, obscure hardware. Back in days of dawn of GNU, all you needed to control your electronic estate, beside software source, was to get or deduct the schematics of hardware. IMHO, that is why RMS concentrated on, at the time, only part that was obscure. Things changed immensely. We have almost all of our software needs covered with software we can control but we are still at mercy of hardware manufacturers.

We need to get back to basics, perhaps as far as pre - IBM PC era and reinvent our computers, making all the right, logical and natural decisions this time, the way they should have been from the start - simple, robust, flexible, extensible, transparent.

Re:PS2 keyboards

[walt-sjc]

Went down to the local OfficeMax the other day... No SATA optical drives at all. Ditto for Staples. The industry needs a big kick in the nuts to dump old legacy shit. Seagate dumping IDE is a kick in the nuts to OfficeMax and other retailers to wake the fuck up, and start carrying modern accessories. Even buying a DVI cable is a painful process - you are lucky if you have ONE to choose from (there are 8325 flavors of the frickin pinout, with monitors and cards keyed so only ONE cable type works...)

If you have a legacy IDE system, you can always get IDE to sata converters. Ditto for PS/2 to USB.

Really old legacy PC's just are not worth the trouble. If you have need for a low-end firewall box (always the stated use for an ancient box) you are better off with an embedded device running openWRT or something similar. A big old Pentium 133 that can't boot off a CD just needs to be retired already.

I'm just blown away that nearly every modern motherboard still has IDE, parallel, serial, and PS/2 ports. Hard to find ones that don't. I don't want the interrupts wasted! I don't want the board real estate wasted! I want more USB and ESATA ports on the back panel instead... Heck, if you feel you REALLY need the ports on the motherboard, put them on a header that I can extend to a few jacks on a PCI slot bracket, but I would prefer that they not be there at all.

Re:PS2 keyboards

[Nullav]

Next to go should be PCI slots.

I have a feeling you'll be waiting a while for that. Not all of us like to buy new sound/tuner cards when we build a new machine. (Although, I suppose everything fails eventually.) It'd probably take around five years to wean everyone off PCI.

I really don't see so many people want to keep PS/2 ports around. I can pick up a PS/2-USB adapter for $0.30.

Re:Serial ports

[TheRaven64]

for some reason they still haven't standardised a usb to serial protocol, dammit Interestingly, there is a standard for serial over Bluetooth which is very well supported. I haven't, however, seen any Bluetooth serial adaptors for under $100. Something I could plug into an RS-232 port and have it route the serial signal to bluetooth on my laptop would be very useful, especially if it could be powered from the RS-232 port.

Re:PS2 keyboards

[raddan]

Yes, but you know-- there's actually no reason not to allow you to plug a mouse into a keyboard's PS/2 port and vice-versa, except that it allows motherboard manufacturers to cut some costs on the second controller for the mouse. That's why the color-coding was introduced-- so that people wouldn't try plugging one into the other. Before AC'97, I had several computers (including my beloved ThinkPad 365CD) that didn't care which one you plugged it into, because the controller was the same on both ports. It's basically just a fancy serial port.

Re:Does it really matter?

[Kjella]

I've never had a problem with the SATA connectors, even though I hear of several people that have managed to break them. And I work mostly with Shuttle XPCs which aren't exactly spacy. I've had them come undone a few times but that's better than excessive force being applied to the connector, if you ask me. The only time I can recall having a near-fatal accident was in a mixed SATA-PATA environment, because the "yank" when you loosen a molex connector causes all kinds of hell with all the other cables. In a clean SATA environment, connect/disconnect the cables with no use of force and thus no damage or accidentally disconnecting anything else. Perhaps not idiot-proof but if you're not an idiot, a lot better to work with IMO.