FireWire Spec to Boost Data Speeds to 3.2 Gbps

Stony Stevenson writes "A new set of data transfer specs may reach new Firewire speed records. The new transfer version is called S3200 and builds on the earlier specification approved by the IEEE.' The technology will be able to use existing FireWire 800 cables and connectors while delivering a major boost in performance. The new spec also will let users interconnect various home-networking appliances via coax cable, linking HDTVs with set-top boxes, TVs, and computers in various rooms around a home or office. The new release enables the transmission of FireWire data over distances of more than 100 meters. Home entertainment centers are likely to be an early application.'"

Comments

I think Apple....

[alta]

will be the earliest application. Remember when this was like e1394, or if you're sony i.Link. Those names never got any momentum, and they didn't push it. Heck, sony isn't too good at pushing standards anyway. Beta? Mini Disk? Memory Stick? Blueray has a chance.

Anyway, when apple calls it firewire2, then it'll get adopted.

Re:I think Apple....

[WinterSolstice]

My guess is they'd call it FireWire 3200, since they currently have "Firewire 400" and "Firewire 800". That's going off the spec here on the Apple page.

I haven't found too much use for my FW800 port... I just hope that I can find a sweet external drive for the FW3200 ;)

Re:I think Apple....

[Anonymous Coward]

Remember when this was like e1394, or if you're sony i.Link.

You're thinking of IEEE1394, which it still is. And Sony's iLink implementation was the unpowered 4-pin version of it.

Re:I think Apple....

[Metaphorically]

Yeah they've had some flops but they also had the 3.5" floppy, the CD (as part of a partnership), Hi-8, and S-PDIF (with Phillips). Personally I prefer open standards and Sony does have a history of trying to push more closed ones than open.

Re:I think Apple....

[kithrup]

I've got a drive with USB 2.0, FW400, FW800, and eSATA connectors. (It's a LaCie "quadra," if anyone is interested.)

In terms of performance, that's pretty much the order, there -- USB is slow, despite it's claims of 480mbits/sec. FW800 and eSATA get me about 85MBytes/sec reading from the raw device; writing appears to be a bit slower, but that was going through the filesystem, so I'll do another test on a drive I don't mind wiping out :).

As with SCSI, the big advantage the faster FW speeds have is that you can have multiple devices chained. So you can have six or seven disks on the chain, and not worry about running out of bandwidth. eSATA doesn't give me that advantage.

Re:I think Apple....

[WinterSolstice]

Oooooh - that drive is seriously lust-worthy :D

I see 3 of those 500GB models in that little rack replacing my current external FW-400 RAID in about 10 days...

Re:I think Apple....

[Colin Smith]

Beta? Mini Disk? Memory Stick? You know... I think you've found a pattern there!

Blueray has a chance. Then ... you blow it.

Re:I think Apple....

[TheRaven64]

I have two older LaCie disks on a FW800 chain. They are the triple-interface version with FW400, USB2 and two FW800 ports. With FW800, I can plug both disks into a laptop at the same time with a single cable (there's another FW800 cable strung between the two) and get 30MB/s copying from one disk to the other (like I said, these disks are a few years old now). The lack of FW800 on the old MacBook Pros was the reason I delayed upgrading my PowerBook for so long.

FireWire is also a good way of adding an extra network connection between a pair of Macs. If you're already using the wired and wireless adaptors you can string a FireWire cable between the two and use IP over FireWire. I tend to do this when I want to copy files between two machines that already have network settings for the other connections that I don't want to disrupt.

Re:I think Apple....

[FatMacDaddy]

I thought that Apple originally engineered the IEEE 1394 standard and that anyone was free to use the technology, but if you wanted to call it Firewire, you had to pay Apple some money. I was under the impression that other tech vendors are using the tech but no one wants to pay to use the recognizable brand name.

Re:I think Apple....

[iamacat]

Try TCP/IP over Firewire next time you need to transfer stuff between two computers. Way faster than ethernet through a hub and of course wireless. USB can not do the same thing easily since host and device do not use the same interface, hence you can not connect two computers with a cable without an adapter.

Re:I think Apple....

[ameline]

Chaining FW drives is great in theory; however, in practice, once you chain more than 2 drives on a single firewire chain (either FW400 or FW800), the performance of the entire chain degrades substantially -- I have measured this using a fairly new iMac, and GDrive-Q drives (using the oxford 924 chipset)

Re:I think Apple....

[DECS]

All the hating on Sony forgets that the company co-invented the CD format (heard of that?) with Philips, and the two later got behind DVD after their own video format based on the CD was hijacked by Toshiba. That puts Sony behind every successful consumer format of the last decade, not just the turkeys it has failed with.

Betamax has also been the basis for ED Beta, the prosumer format nearly every TV crew uses and has used over the last two decades. Trying to create a black and while picture by dialing up the contrast to ridiculous levels leaves you unable to see the big picture.

Further, it is silly to think that Blu-ray is going to lose out to HD-DVD given that it has one manufacturer (Toshiba), one studio (Paramount), and one tenth the installed base of BR. Will Vista turn the tide? Tee hee. Of course, I think there is more potential in content outside of HD in the near term, and don't see Sony contributing toward that potential or benefiting from it.

Why Low Def is the New HD
The video industry is heavily promoting HDTV as the biggest new thing since color. While it's uncontroversial that HDTV can deliver an exceptional picture for users of the latest large flat screen displays, sometimes a high pitched marketing message can drown out more interesting realities. In 2008, it appears that low definition video will actually have a bigger impact on consumers; Apple's strategies in video take that potential into consideration. Here's why Low Def is big and getting bigger--and why it's bigger than HD.

Re:I think Apple....

[Megane]

That is why I put a Firewire card in my G4 MDD tower, so that I can have each of my three external drives hooked up to its own dedicated FW400 port. (but of course the big drives are on the inside, since the MDD has four drive bays and two optical drive bays)

Of course that's a little more difficult to do with a laptop, but you're not likely to be leaving drives permanently attached like you would with a desktop.

Re:I think Apple....

[dgatwood]

Originally, probably. The trademark is now freely licensed, though.

http://www.1394ta.org/license/FireWire_License-Guides_v6.pdf

Re:I think Apple....

[Anonymous Coward]

eSATA doesn't give me that advantage.

You should look at SATA-PM (port multiplier). This can allow very efficient multiplexing of a single eSATA link, for example to build a multi-disk SATA external chassis and connect with one cable, or even just to add more disks to a system without enough internal SATA ports. Then you can use software RAID or anything like that on the OS. It's not exactly chaining, but it is sharing of the SATA bandwidth.

Note, you need a SATA controller that is PM-capable, as some older onboard SATA controllers won't allow this.

Re:I think Apple....

[tyrione]

Apple engineered and opened up the trademark by foregoing the $1 per interface charge people whined about. Apple holds the Chairman Seat:

http://www.1394ta.org/Contact/Board/

Chair:
Eric Anderson, Apple
1 Infinite Loop, MS 306-3FW
Cupertino, CA 95014
Phone (official business only): 1/408-974-1394
Customer information: http://www.apple.com/firewire

Re:I think Apple....

[magarity]

USB can not do the same thing easily since host and device do not use the same interface, hence you can not connect two computers with a cable without an adapter.
 
Not only that but also USB networking requires special software. Linux/WinXP/2K/98 all do FW networking out of the box. Vista does NOT, however, so keep that in mind if you need it.

Re:I think Apple....

[Khyber]

I have to disagree. I have a couple of computers setup to transmit over USB 2.0 thru NetBEUI to each other since I don't have firewire and only have a 10mbit NIC.

Also, when I worked at Solectron, we did HP image restores directly thru USB 2.0 from multiple servers scattered across the floor. USB has come a long way forwards.

Re:I think Apple....

[gb506]

I have to disagree. I have a couple of computers setup to transmit over USB 2.0 thru NetBEUI to each other since I don't have firewire and only have a 10mbit NIC.

Only a 10mb nic? Have you been frozen for the pat 15 years, or are you just cheap?

Re:I think Apple....

[OrangeTide]

What is there to disagree about?

USB2.0 ethernet is slower than Firewire networking.

USB2.0 cannot do it without an adapter.

Which of these two facts are you trying to disagree with? Yes it's certainly possible to connect computers together using USB with an adapter.

Re:I think Apple....

[OrangeTide]

"Further, it is silly to think that Blu-ray is going to lose out to HD-DVD given that it has one manufacturer (Toshiba), one studio (Paramount), and one tenth the installed base of BR. Will Vista turn the tide? Tee hee. Of course, I think there is more potential in content outside of HD in the near term, and don't see Sony contributing toward that potential or benefiting from it."

Other than your facts being incorrect, that is a completely logical argument. multiple manufactures and multiple studios (Universal, Paramount, and about 40 others) are continuing to make HD-DVD titles.

I don't count PS3's as part of the Blu-Ray install base. PS2 playing DVDs was not a significant market driver after the the first 6 months, I think PS3 will follow the same fate. But I will admit that is a SWAG on my part.

I think, generally, anyone who claims to know the winner of the Blu-Ray versus HD-DVD format war is completely delusional (undiagnosed megalomania?). There are likely plenty of CEOs, Investors, etc that are smarter than you with a lot more information. And they obviously haven't picked a clear winner yet. And I doubt stubbornness or pigheadedness could rule a multi billion dollar industry to the extent that a format war could persist for multiple years.

I think the only good deal right now is to get a hybrid player, and rent the videos from netflix or whatever so you're not burdened with a bunch of obsolete videos. Or stand by the side-lines and enjoy your upscaled DVDs.

Re:I think Apple....

[DECS]

You're right, HD-DVD has two large exclusive studios: Paramount and Universal (athough there are titles on both sides of the divide that jumped ship, from Spielberg's Paramount flicks to some BR titles that can be found on HD-DVD). Studios who make multiple format HD discs don't play a role in affecting the outcome. There are no makers building HD-DVD players apart from Toshiba and some efforts Toshiba has made to get Chinese companies to duplicate it.

FYI: megalomania is an obsession with power and control, not a secure assurance of a given outlook.

There really isn't a controversy however. The market has decided. That's why it took $100,000,000 to get Paramount to hold up its BR releases for two years. For things to balance out, but it would take a massively powerful shift of interest, and right now there isn't any mass of interest in HD discs in either format. They're projecting total sales of 1 million standalone players by the end of the year (not counting the 7 million PS3s sold) as the total installed base of HD players. That's absolutely nothing overall. It would be easier and more cost effective to roll out DVD-9 with HD content in H.264 at this point and beat both formats, if it were in anyone's financial interest to do so.

From that standpoint, the massive number of PS3s out there that will be sold, particularly in the next 6 months, do matter and will blow away any blip of HD-DVD, making it silly for studios to continue pressing HD-DVD movies. HD-DVD will become a vestigial addition to DVD just like DIVX and then go away. The players will become enhanced upscaling DVD players. This is so obvious that its painful to hear delusional stuff from HD-DVD buyers who insist that the format has legs because Microsoft promises cheap Chinese players. Microsoft promises a lot of things.

Microsoft said it would ship a video iPod platform and take over the market in 2004. I was late by half a year or so, and then Media2Go fizzled.

Microsoft planned to sell a million Zunes by June. That's really not very many, but it's still selling them at fire sale prices six months later. Apple sold 40 million in that time.

Microsoft is saying that Windows 7 in 2011-2012 (?) will improve upon the iPhone of 2007. Wow, I'd hope so.

Why Low Def is the New HD

Re:I think Apple....

[iamacat]

Ah the irony. You can no longer do high speed file transfers between two notebooks on the road. I am not aware of any battery-powered ethernet hubs, so I guess the only option is ad-hoc WiFi with inherent manual setup and security issues. Copying multi-gigabyte movie projects or DVD ISOs is going to be slow to the point of making two guys sitting with notebooks and doing nothing look comical or even drain the batteries before the transfer is complete. Other vendors generally enable their operating systems do more in each release.

Re:I think Apple....

[Anonymous Coward]

It's not the length of your Firewire cable that counts, but the bus width of your IEEE1394 Controller.

Re:I think Apple....

[dynamo]

Um, I think we still have ethernet cables these days.

Re:I think Apple....

[briancnorton]

I routinely move tens of gigs of data on and off of big drives, and I have actually timed it. If I'm transferring 100GB, the time difference between the four is in the 5-10 second range, and there isn't a consistent winner. The limiter is the HDD, not the connector.

Re:I think Apple....

[Ilgaz]

will be the earliest application. Remember when this was like e1394, or if you're sony i.Link. Those names never got any momentum, and they didn't push it. Heck, sony isn't too good at pushing standards anyway. Beta? Mini Disk? Memory Stick? Blueray has a chance.

Anyway, when apple calls it firewire2, then it'll get adopted. Lets hope their relations with Intel doesn't effect it. Intel isn't even on IEE1394 board. Firewire (400 and 800) is the reason why USB2 is still considered a joke in professional Video/Audio business for very legit reasons.

3200 standard is really needed in professional video business. Especially HD Field Cameras/Editing facilities need better bandwidth.

If Apple stays away from that standard, Sony should consider buying "Firewire" brand from Apple. i-link or IEE1394 really doesn't work. They should also make sure every Apple has option to adopt that standard, down to making PCI-(X) Cards for cheap prices.

Re:I think Apple....

[OrangeTide]

Microsoft has nothing to do with it. So I wouldn't worry about them.

I'm one of those folks who would rather watch this play out than decide early one which format will win. To be honest I wish a third format would jump in and just knock them both out. Neither of the formats are particularly appealing to the consumer I have interviewed.

ps - the megalomania comment is precise. it covers people who think, to a severe extreme, that they are more important (including knowledgeable) than they really are.

Re:I think Apple....

[Khyber]

Actually, the NIC is having issues, and only works in 10 mbit. But yes, I'm cheap, as I'm disabled and have no income. Oh, and I'm moving to California soon.

Re:I think Apple....

[DECS]

Microsoft has "nothing to do with" HD-DVD? Are you joking? The entire industry selected to align behind Blu-ray, with HP trying to pull Microsoft in to support it as well. Microsoft wanted to push its WinCE-based HDi authoring system.

Because Blu-ray had already selected a Java based system, Microsoft refused to support it, and single handily pushed Toshiba to continue to invest in HD-DVD, assuring it that HD-DVD support in Windows Vista and the Xbox 360 would turn the tide. The entire HD-DVD/Blu-ray divide is a Microsoft invention. That's not even controversial.*

HD-DVD has evolved into a misinformation effort rivaling "Get the Facts" and the Microsoft monopoly trial astroturf ploy combined. There are forums full of HD-DVD users all repeating the same talking points to the point where it sounds like the Zune. It's manufactured bullshit up the ying yang: the "Chinese will produce cheap players," and "Blu-ray has all sorts of hardware problems" and the "BR discs are expensive to manufacture", and Microsoft is "going to be all liberal and lenient about DRM enforcement" as opposed to Sony... it strains belief.

They count the 150,000 Xbox 360 HD-DVD sales to get a number that sounds significant, but exclude the several million PS3 units because they "are not standalone players," despite being the cheapest and therefore most popular BR players.

Supporting HD-DVD is a faith based initiative as George Bush might say. I am not a proponent of Blu-ray (I think HD is oversold, a position supported by sales of both players), but it's obviously in the lead by a wide margin. Dual format players sound great, but I find it unlikely based on history that that's going to happen. HD-DVD is only going to create VHS/Betamax uncertainty that prevents HD discs from taking off in 2008, offering the potential for more users to migrate toward downloads such as Apple TV and maybe Vudo.

Calling me names might get you excited, but using them improperly doesn't impress anyone. Megalomania is about power and control and domination. Asserting that what I'm pointing out is right in line with facts has nothing to do with forcing or subjecting anyone down. Insisting group-think obedience of Microsoft, other the other hand, is a bit excessive and domineering.

*Origins of the Blu-ray vs HD-DVD War
The arrival of DVD gave consumers far higher video quality in a new compact disc format with a variety of practical advantages over existing VHS tapes. There was no format war related to DVD because the two groups developing a new consumer video disc gave up their differences and worked together. Consumers didn't have to chose a format or worry about obsolescence. So why did the world return to a format war with HD?

Re:I think Apple....

[dirtyhippie]

Modern ethernet hardware is auto-sensing, so you don't need to worry about the crossover-ness of your cable. All gigabit (and many 100Mbps adapters) detect this and deal with it seemlessly.

Re:I think Apple....

[RzUpAnmsCwrds]

Oh, you again. For everyone who doesn't know what I'm talking about, go read this guy's page (roughlydrafted). You'll see how much of a crackpot Apple fanboy he is.

Unless, of course, you honestly believe that not having third-party apps on the iPhone is a good thing. In which case you are the crackpot Apple fanboy.

Re:I think Apple....

[RzUpAnmsCwrds]

There are no makers building HD-DVD players apart from Toshiba and some efforts Toshiba has made to get Chinese companies to duplicate it.

LG and Samsung have dual-format players. Thomson (RCA) is making players. There's the 360 HD-DVD drive. Onkyo has a high-end player. And, yes, now there are cheap Chinese made players too. Something that Blu-ray doesn't have.

There really isn't a controversy however. The market has decided.

Yes, the market has decided to stick with DVDs for the time being. DVDs are outselling both formats combined by a margin of 1000:1. Neither format has momentum here.

From that standpoint, the massive number of PS3s out there that will be sold, particularly in the next 6 months

I don't know if you're a Sony fanboy or just stupid. PS3 sales have hardly been "massive". Even if Sony moves 10 million units in the next 6 months, we're talking about a tiny fraction of the number of DVD players out there.

This is so obvious that its painful to hear delusional stuff from HD-DVD buyers who insist that the format has legs because Microsoft promises cheap Chinese players.

Now you manage to flame Microsoft in the same post. If only it made sense. Microsoft has never "promised" cheap Chinese players. People have looked at statements from Wal-Mart and other big-box retailers, and concluded that the market will be flooded with cheap HD-DVD players soon. Which it will.

DVD took off when DVD players dropped below $100. Remember Apex? Remember all of the other cheap no-name DVD players? They are what got the format into practically every home.

Microsoft said it would ship a video iPod platform and take over the market in 2004. I was late by half a year or so, and then Media2Go fizzled.

This statement is poorly written ("I" was late?), and completely irrelevant to the HD format war. But that never stopped you DECS, did it?

Microsoft planned to sell a million Zunes by June. That's really not very many, but it's still selling them at fire sale prices six months later. Apple sold 40 million in that time.

Microsoft did sell a million Zunes. The fire sale prices were used to get rid of 30GB Zune inventory before the new Zunes launched. This is exactly what Nintendo did with the GameCube and GBA. It's what Apple does with the iPod. Look at the Apple store "Hot Deals" section.

Microsoft is saying that Windows 7 in 2011-2012 (?) will improve upon the iPhone of 2007. Wow, I'd hope so.

Windows 7 is scheduled for release in 2010, but they probably will miss that. How a desktop OS is going to compete with the iPhone I have no idea, which leads me to the conclusion that you pulled that information straight out of your ass.

Everyone, go read this guy's website. It doesn't take a rocket scientist to figure out that he's a crackpot Apple shill. His website has a Soviet propaganda poster modified to include the Windows logo (real tasteful there). It compares the Apple TV (which doesn't record TV) to the TiVo, and concludes that the Apple TV is better. It trashes the Zune using the installer bug that last was a problem 12 months ago. It claims that not having third-praty app compatibility is a good thing for the iPhone.

This guy is the WORST of the WORST. He makes other Apple fanboys look tame, which really takes a lot of effort. Even Windows fans like Paul Thurrott give Apple products their fair due (Thurrott owns a Mac and uses it regularly).

DECS will lie and misleed to prove his point, which is something that pisses me off greatly. It's people like you that make me think twice before I recommend Apple products (which, by the way, I do - I'm even giving an iPhones for Christmas).

Re:I think Apple....

[GregPK]

I dunno, when I was moving around Video Files. I noticed about double the speed with my Firewire 400 than I got with my USB 2.0 hookup. On the same hard drive too. Even in testing the Firewire still transfers faster than USB.

Re:I think Apple....

[OrangeTide]

I like Apple, and think they make some great products. So being an Apple fanboi doesn't seem like an insult to me. But yea, I will agree that the guy comes off like a crackpot. And so far none of his arguments have been convincing. I would love to know if HD-DVD or Blu-Ray is going to win the format war, I could probably move my financials around on my E*Trade account if I did have such information.

stupid tags

[farkus888]

what moron tagged a story about fire"wire" "wireless"?

Re:stupid tags

[doi]

I don't know, but I for one will welcome my new overpriced Monster Cable Gold Shielded Mega Quality Firewire 3200 Cable overlords!

Re:stupid tags

[Huntr]

I was sad it wasn't tagged "penis" like most of the stories were yesterday.

Re:stupid tags

[Anonymous Coward]

it's !wireless or for the programming challenged, 'not' wireless. but, yea, that can be applied to a great many topics. not sure how usefule the !x tag is for anything.

Re:stupid tags

[farkus888]

actually it was tagged wireless, then the wireless tag disappeared for a while, then the !wireless tag appeared. fortunately for me those with mod points were on when my joke was still relevant, no wasting of my "excellent" karma.

I would just like a single standard...

[explosivejared]

Maybe it's just me, but I don't see why USB and Firewire need to exist. Maybe I'm naive and don't see where there are ad hoc benefits to both. I would like to see a unified standard. I have both on my machine, so there is no compatibility annoyance. I don't see competition benefiting either one really.

Re:I would just like a single standard...

[hcdejong]

IIRC FireWire needed more complex (=more expensive) interfaces, so wasn't really suitable for mice etc.

Re:I would just like a single standard...

[speculatrix]

very crudely, firewire was specifically designed to be for high speed streaming data - like video - it's ideal for a hard drive moving large chunks of data, and would be no use for mice and keyboards. USB was originally designed for low bandwidth low latency peripherals; it allocates data bandwidth in inverse proportion to demand, so e.g. mass storage gets whatever's left after mice/keyboards and tablets have had their share.

Re:I would just like a single standard...

[zsouthboy]

USB is computer-centric - created by Intel. Somewhere in the chain, a computer needs to be connected.

Whereas 1394 is device-centric - designed for tranferring video, audio, data, etc. between "dumb" devices.

The PC world "snubbed" 1394 (not completely) because of the $.25(or something equally absurd) royalty to Apple; Intel pushed USB.

1394 is more expensive to implement in hardware (which should be obvious - it's for dumb devices!), while USB is cheap - the CPU does most of the work.

Re:I would just like a single standard...

[2ms]

The reason there are two is that Intel wanted to create a competitor to Firewire. So blame Intel.

Re:I would just like a single standard...

[2nd Post!]

Blame Intel then. Firewire was developed in 1990, released into PCs

Here are the technical specs for Firewire from Wiki:
FireWire can connect up to 63 peripherals in a tree topology (as opposed to Parallel SCSI's Electrical bus topology). It allows peer-to-peer device communication -- such as communication between a scanner and a printer -- to take place without using system memory or the CPU. FireWire also supports multiple hosts per bus. It is designed to support Plug-and-play and hot swapping. Its six-wire cable is more flexible than most Parallel SCSI cables and can supply up to 45 watts of power per port at up to 30 volts, allowing moderate-consumption devices to operate without a separate power supply. (As noted earlier, the Sony-branded i.LINK usually omits the power wiring of the cables and uses a 4-pin connector. Devices have to get their power by other means.)

FireWire devices implement the ISO/IEC 13213 "configuration ROM" model for device configuration and identification, to provide plug-and-play capability. All FireWire devices are identified by an IEEE EUI-64 unique identifier (an extension of the 48-bit Ethernet MAC address format) in addition to well-known codes indicating the type of device and the protocols it supports.

From the USB wiki:
USB was originally seen as a complement to FireWire (IEEE 1394), which was designed as a high-speed serial bus which could efficiently interconnect peripherals such as hard disks, audio interfaces, and video equipment. USB originally operated at a far lower data rate and used much simpler hardware, and was suitable for small peripherals such as keyboards and mice.

The most significant technical differences between FireWire and USB include the following:

        * USB networks use a tiered-star topology, while FireWire networks use a repeater-based topology.
        * USB uses a "speak-when-spoken-to" protocol; peripherals cannot communicate with the host unless the host specifically requests communication. A FireWire device can communicate with any other node at any time, subject to network conditions.
        * A USB network relies on a single host at the top of the tree to control the network. In a FireWire network, any capable node can control the network.

These and other differences reflect the differing design goals of the two buses: USB was designed for simplicity and low cost, while FireWire was designed for high performance, particularly in time-sensitive applications such as audio and video. Although similar in theoretical maximum transfer rate, in real-world use, especially for high-bandwidth use such as external hard-drives, FireWire 400 generally has a significantly higher throughput than USB 2.0 Hi-Speed.[13][14][15][16] The newer FireWire 800 standard is twice as fast as FireWire 400 and outperforms USB 2.0 Hi-Speed both theoretically and practically.[17]

There are technical reasons why USB 2.0 devices cannot efficiently utilize all the available bandwidth. USB communication is based on polling the devices; there is no pipelining of commands. After sending a command to a device, the USB host must wait for a reply to the command before a new command can be sent to the same device. The bandwidth of a USB bus is divided by all devices connected to the bus. The USB host cannot send commands to one device while waiting for reply from another device. Since all communication is initiated by a USB host, the host must periodically poll all those USB devices that can provide data at unexpected intervals, such as network cards and keyboards. This consumes unnecessary resources when the devices are idle. These issues are being addressed by the forthcoming USB 3.0 specification, although it is not clear whether USB 3.0 is going to match FireWire in bandwidth efficiency.

Re:I would just like a single standard...

[vux984]

USB is slow and cheap.
Firewire is fast and cost more.

So they both existed. One was good for mice and keyboards, one was good for digital video and external hard drives.

Then USB2 came out which is almost as fast as firewire, and the lines got blurry.

Firewire was still considerably better as a technology. It does a lot of its own processing while usb2 offloads a lot of processing to the host system... so firewire drives don't tie up the CPU the way a USB2 one does. Firewire supports more simultaneous devices, and seems to have fewer issues with power as well. It also doesn't have stupid rectangular connectors that users will try upside down 50% of the time.

Then Firewire 3200 was announced and santify was restored.

USB2 is slow and cheap.
"Firewire-3200" is fast and costs more.

Do we 'need' usb? no. We could get by on just firewire. But usb is cheaper and a penny saved is a penny earned.

Re:I would just like a single standard...

[timeOday]

But aren't today's firewire external hard drives only a little faster than USB2? I sure wish USB could simply be upgraded to these higher speeds (since it is admittedly undesirably slow for hard drives nowadays). But When I travel, I really like having a single cable for my GPS, MP3 Player, and camera.

Re:I would just like a single standard...

[Anonymous Coward]

"Maybe it's just me, but I don't see why USB and Firewire need to exist."

Well, currently, I'd rather have my external raid drives FW800 than USB. And now we have eSATA becoming popular. Yeah, it is a mixed bag alright.

Re:I would just like a single standard...

[0100010001010011]

USB devices are 'dumb' they just push bits back and forth. Firewire devices are smart and can talk to each other. Theoretically if someone put the right software on a camcorder, you could dump video directly to a Firewire hard drive.

Second, USB uses CPU. Firewire uses its own chip. (Which is why Apple removed the chip from the iPods, it wasn't going to fit with that and the video decoder in the same form factor).

USB is good because... there's no reason my mouse needs that much speed.

They're for completely different uses, problem is when someone tries to use one where it shouldn't be.

Re:I would just like a single standard...

[hcdejong]

In my limited tests, FW400 was about 30% faster than USB2 on the same computer/external harddrive combo. Plus I've the impression FW performs better when the load increases (more devices on the same bus).

Re:I would just like a single standard...

[kithrup]

One of the big advantages FW has over USB is that a device can request, and get, a specific bandwidth. So, for example, a video camera can say, "I need to have 19mbits/sec." And if it can, then it'll get that -- and nothing on the bus will be able to take it away until it releases it.

When doing real-time applications such as video, this is pretty important.

Re:I would just like a single standard...

[Mercano]

USB wound up being used for devices that were really more suited for firewire, such as external hard drives and CD burners, because it was cheaper to implement and there were (and still are) quite a few machines out there that have USB but not firewire.

Re:I would just like a single standard...

[Just Some Guy]

The bandwidth of a USB bus is divided by all devices connected to the bus.

Compare and contrast with Firewire?

Re:I would just like a single standard...

[MBCook]

I've seen the same thing on my MacBook Pro with my external drive. FireWire 400 is maybe 25% faster, FireWire 800 is 50% faster (same drive). The big thing is CPU utilization though. Maxing out the disk on FireWire 800 is no problem. Very little CPU usage (maybe under 10%, this is based on a little graph, I've never looked at hard numbers). Running it over USB has a very noticeable CPU impact.

FireWire is great at what it was made for. USB is very good at what it was designed for (mice, keyboards) and weak at things it was forced to do (hard drives).

It's all Intel's fault. They put USB on everything, but didn't put FireWire on anything until very recently, if they even have by now. So USB "won".

Re:I would just like a single standard...

[Anonymous Coward]

I've actually run MS Exchange mailboxes on an external Firewire 400 hard drive before. That was about 65GB of data for about 100 users. (I needed the extra space. It was only for a week while the new hard drives came in.)

Re:I would just like a single standard...

[2nd Post!]

Different devices can get different amounts of bandwidth on a Firewire tree.
Each device on a USB bus have to get equal bandwidth slices.

Re:I would just like a single standard...

[hyc]

The most obvious thing: USB was invented by Intel and designed to keep Intel microprocessors as an indispensible part of the system. That's why it uses that stupid "speak when spoken to" protocol. Intel did everything they could to proliferate USB in preference to Firewire, because they wanted to make sure the x86 remained king of the hill in system architecture. That's why USB is so cheap today. Nobody with equal marketing muscle was really pushing Firewire the same way, even though Firewire is better for 3rd party hardware manufacturers.

Re:I would just like a single standard...

[AaronW]

I consistently get noticeably better performance using Firewire 400 for external drives than USB with much less CPU overhead. In addition, at least on Linux, the stability has been much better. I found with USB that if I had multiple applications accessing an external USB drive at the same time the interface would hang, but no such problems with Firewire.

I haven't played around with the USB reliability with my latest Linux update since I just use Firewire instead.

Plus, I have no problem ganging Firewire drives together without the need for a hub.

After my first experience with Firewire vs USB I make sure any external drive I get supports Firewire, though I want to try playing around with eSata as well. I just need to pick up some cables one of these days to try it out.

Re:I would just like a single standard...

[InvalidError]

FW400 may be "only" 400Mbps but it is double-simplex 400Mbps - 800Mbps aggregate. USB2 may be "faster" at 480Mbps but it is half-duplex with 10% bandwidth reserved for host commands and more dead-time inserted between TX and RX packets while host and target devices' transceivers switch directions, wasting several microseconds each time.

The highest copy speed I have ever reached on FW400 is 32MB/s - limited by the ATA33-FW400 bridge chip (Oxford 911) while the highest speed I have ever seen with any combination of my USB2/480 external HDD boxes and PCs is 24MB/s with 18MB/s being more typical. Under the best circumstances, USB barely matches the SLOWEST speeds I take for granted on FW400. If I string two of my FireWire drives together and move data between the two boxes, I still get the same 24-30MB/s I am used to but if I try to do the same with USB2 and a hub, USB2 crawls at 8-10MB/s. In this scenario, FireWire is as much as 3X as fast as USB2.

With eSATA for external storage, USB3 (late 2008) for all sorts of high-bandwidth (4.8Gbps) devices and GbE for mainstream networking, FW3200 will become irrelevant soon enough: USB3 is supposed to be 4.8Gbps double-simplex fiber, blowing away FW3200 on raw speed and finally getting rid of USB1/2's half-duplex overhead.

Re:I would just like a single standard...

[99BottlesOfBeerInMyF]

USB is slow and cheap. Firewire is fast and cost more.

For an OEM to add a firewire port costs about $1.50 more than a USB port. Not exactly a huge difference.

Do we 'need' usb? no. We could get by on just firewire. But usb is cheaper and a penny saved is a penny earned.

There is nothing wrong with USB for what it was designed for. The problem is when it is applied to a task it just isn't very good at, like hard drives. Even if you have USB-2 for a hard drive, it is still pretty inferior. It uses up your CPU, you need software to manage it, it is slower, and you can't just plug it into your video camera and go, because it needs a computer in between. Intel's insistence on pushing USB-2 is just holding back progress, hopefully this will help solve that problem.

Re:I would just like a single standard...

[Anonymous Coward]

Firewire is also based on the SCSI standard. Its speeds are a sustained rate, whereas USB is a burst rate, peaking only briefly in very ideal situations. That's why in real-world applications, firewire 400 consistently outperforms USB 2.0, despite firewire being 80 mbit "slower."

Re:I would just like a single standard...

[TheSkyIsPurple]

>For an OEM to add a firewire port costs about $1.50 more than a USB port. Not exactly a huge difference.

Yeah, but when you spread that across a million machines, you're talking real money.

Re:I would just like a single standard...

[dangitman]

For an OEM to add a firewire port costs about $1.50 more than a USB port. Not exactly a huge difference.

To the OEM it is. They are cheap bastards. And of course, customers are cheap bastards, who tend to care more about price than quality. Keep adding "only $1.50 more" to other companents, and soon you don't have that $300 PC. Not that anybody should be buying a shitty $300 PC, but people like cheap junk, as I said.

Re:I would just like a single standard...

[Tetsujin]

Different devices can get different amounts of bandwidth on a Firewire tree.
Each device on a USB bus have to get equal bandwidth slices. That's not precisely true...

The bandwidth a USB device gets depends on two things: The speed of the device itself (naturally) and how frequently it's polled by the host computer. So one device can be polled more frequently than others (and HID devices can suggest how frequently they should be polled) - so the devices don't get "equal bandwidth slices" (at least, not equal slices of the total bandwidth when you average over time), one device can take more than 1/N of the total bandwidth of the bus if it needs it and if nothing else does...

('Course this doesn't compare to Firewire's actual [i]allocation[/i] of bus bandwidth...)

Re:I would just like a single standard...

[vux984]

For an OEM to add a firewire port costs about $1.50 more than a USB port. Not exactly a huge difference.

I said "a penny saved is a penny earned" for a reason.

Plus the average PC comes with what? 6, 8, 10 ports? Times every motherboard you sell... sure to you its a buck fifty on a $500 PC... but to the OEM its a whole LOT of $1 parts. And if they can shave even a few off it makes a big difference to the bottom line.

Re:I would just like a single standard...

[Nutria]

After my first experience with Firewire vs USB I make sure any external drive I get supports Firewire, though I want to try playing around with eSata as well. I just need to pick up some cables one of these days to try it out.

I echo your statements about FW vs. USB on Linux.

One thing to note: There aren't many external FW enclosures for SATA, and they're all on the expensive side.

Re:I would just like a single standard...

[Anonymous Coward]

I sure wish USB could simply be upgraded to these higher speeds

I wish lots of parts of the USB spec weren't completely braindead. As you may know, USB picks a speed by which data line (D+ or D-) you tie to 3.3V with a 1.5K resistor. Then when they wanted to add a third speed, they added another hack to allow that. It's kind of a mess.

As a rule, I don't trust specs who can't even get their units right. USB declares that devices must state their max power usage in milliamps. It's been that way since USB 1.0, and they've never fixed it.

USB is good at some things, and Firewire is good at some things, but all too often seems like the USB spec was written by high school kids. I've never built Firewire hardware, so I can't say whether it's any better, but I have trouble believing it could be any worse.

Re:I would just like a single standard...

[xlsior]

Then Firewire 3200 was announced and santify was restored.

USB2 is slow and cheap.
"Firewire-3200" is fast and costs more.

Except that USB 3.0 has already been announced as well:
http://arstechnica.com/news.ars/post/20070918-intel-announces-demonstrates-usb-3-0.html

The USB 3.0 target speed is 5 GBps, which would still make it one and a half times faster than the new Firewire 3200 standard... While retaining USB 2.0 backwards compatibility, which is a pretty big deal as well.

Re:I would just like a single standard...

[Obfuscant]

IEEE1394b (FireWire) has defined two modes of bus operation. Asynchronous and Isochronous.

Async is used for devices like disks where you want to read a block but it doesn't really matter if the block doesn't come back right away.

Isochronous provides a fixed amount of bus bandwidth at fixed times, which is important if you are streaming audio or video.

Iso is guaranteed up to 80% of the bandwidth. Asynch is guaranteed at least 20%.

As for this "new standard" the story is talking about, no, it isn't new. 1394b has had S3200 defined for a long time. VERY long time. Nobody has implemented it. We needed it not for the bandwidth, but for the long-cable capabilities, and we've been waiting ... Fortunately, there are a couple of 1394b PHY (physical interface) chips on the market, and people are making 1394a-connectored long-cable hubs.

The 1394 standard is really interesting, including the hardware handshakes etc and bus topology maps you can use to know what is connected where.

Re:I would just like a single standard...

[Televiper2000]

Firewire B does support 100Mb/s over 100M of CAT-5 cable. It's the exact same physical layer as Ethernet, you just need a cross-over cable since there's no hubs or routers in the chain. There's about 3 or 4 media converters available on the market and at least 2 IC's that interface directly with the B physical layers. I really wish more people would pick up on Firewire. They're currently looking for the ultimate synchronous Ethernet network (IEEE1588, Powerlink, etc) and as far as I'm concerned Firewire already exceeds anything they're capable of outside of being able to plug directly into a switch that the CEO is connected to (apparently CEOs get wet over being able to check sensors on an assembly line). It will be interesting to see if Firewire finally takes off the way it should in the next couple years.

Re:I would just like a single standard...

[eharvill]

Second, USB uses CPU. Firewire uses its own chip. (Which is why Apple removed the chip from the iPods, it wasn't going to fit with that and the video decoder in the same form factor). I've seen several posts about USB using CPU. With today's (this year's anyway - dual/quad/beyond) CPUs and going forward, does that even matter in the discussion? Maybe us sheeple have been trained fairly well over the years, but I'm guessing the extra few seconds, even minutes to transfer stuff from cameras, HDs, usb sticks, etc really doesn't matter at this point.

All that being said, networking over coax (which is wired all over the place in most homes) and transmission lengths of 100M really interests me. I wonder if I can use the same coax that delivers cable to my TV and splice it to PCs and set-top boxes all in one fell swoop. Would love to have an HTPC setup with a PC in the basement, not generating extra noise and heat in my living room.

Re:I would just like a single standard...

[dgatwood]

I wouldn't assume that USB3 will do much at all. If it doesn't require a more intelligent bus controller to offload the work, it will be torture on the CPU at those speeds, and if it does, a lot of manufacturers will reject it because it costs too much more to build cards for it, and a lot of users will reject it because of the substantial extra expense of the cards compared with other technologies. Similarly, for external devices, cost will probably play a more important role in driving adoption than any other factor, since neither USB3 nor FW3200 would be present in existing computers. To that end, FireWire has a major leg up on USB3. There's at least a possibility that it would require nothing more than a silicon change. Same connectors, same basic technology, same basic board layout, etc. USB3 will require new, upgraded connectors (optical) in order to reach those speeds, which presumably brings additional board layout requirements for extra traces, etc.

The reason you got so much better performance with FireWire is that you don't use any CPU power with FireWire. DMA from your hard drive goes straight into the host's memory without the CPU lifting a finger (except to ACK an interrupt and let the higher layers of the software stack know that data has been delivered). It's a similar story for audio and video. If your driver is written correctly, DCL programs running on the FireWire host controller do all the data copying for you without the CPU being involved except to receive a notification that data is ready. Compare that with USB audio which is an endless disaster because the CPU has to service it reliably and predictably, time stamp everything precisely, etc.

I don't see USB replacing FireWire for professional audio/video applications any time soon. It just isn't up to the task, and the additional parts cost of bringing it up to spec would eliminate the sole advantage USB has over FireWire: cost. For disks, USB 3 could replace FireWire if and only if they put true DMA support into the host silicon. Unfortunately, such a change would not only increase the host silicon cost, but also might increase the cost of the device silicon significantly as well, which would be a disaster for USB.

Note: I haven't read any specifications for either USB 3 or FW3200. I'm basing this off just a handful of articles on the subject and interjecting my opinion of the likely hardware costs. This is, of course, largely conjecture until such time as both technologies reach final silicon.

Re:I would just like a single standard...

[0100010001010011]

"Does it really matter"

Now imagine a TV station or the latest Movie producer. Those seconds do matter. Imagine a 50 GB HD movie. Few extra seconds add up, and if it didn't we'd all be content running our 300 mHz Pentiums because "what's an extra second, right?"

Firewire 800 already supports multiple forms of transmission, including coax, optical, and Cat 5E. Here's an article from 2005 about FW800 and coax: http://www.engadgethd.com/2005/09/14/coax-making-a-comeback/ It's just that most people don't know about it. So when the next technology is announced and people get all amazed at the "new" features, I laugh and say "no, that was there with the old revision".

See also: Target Disk Mode.

Re:I would just like a single standard...

[Hes Nikke]

Maybe it's just me, but I don't see why USB and Firewire need to exist. Maybe I'm naive and don't see where there are ad hoc benefits to both. I would like to see a unified standard. I have both on my machine, so there is no compatibility annoyance. I don't see competition benefiting either one really. FireWire = fast lane, USB = slow and middle lane. Do you drive a VW beetle in the fast lane? So why put a mouse on FireWire?

Putting a mouse on FireWire does the same thing as driving a beetle in the fast lane - it causes a backup in traffic. The presence of your mouse would slow down your hard drive.

Re:I would just like a single standard...

[eharvill]

Can you show me a 50GB HD movie? What would you be doing with it that requires USB or Firewire?? My real question is what is the CPU hit for USB vs the non-CPU hit for firewire. And would it really matter if I had a quad-core (or greater) CPU. And if it did, I don't think we are dealing with a typical desktop grade system (Mac/Windows/Linux/whatever) at that point.

I guess it's great that technology from two years ago still isn't mainstream. Laugh away...what's your point?

Re:I would just like a single standard...

[macshit]

With eSATA for external storage, USB3 (late 2008) for all sorts of high-bandwidth (4.8Gbps) devices and GbE for mainstream networking, FW3200 will become irrelevant soon enough: USB3 is supposed to be 4.8Gbps double-simplex fiber, blowing away FW3200 on raw speed and finally getting rid of USB1/2's half-duplex overhead.

So is USB3 also supposed to get rid of USB1/2's insane CPU overhead at high speeds?

USB2 seems like it's something that can maybe-kinda go fast in certain circumstances, but you certainly wouldn't want to use it for any routine tasks; it's more something to use as an emergency backup method, or for tasks you do very rarely.

[BTW, this is an honest question. Like many people, I love the idea of an "works for everything, does it all well" desktop connection standard. It would be great if USB3 could actually pull it off (within reason) and substantially address the problems of previous USB versions, but in my cynicism I'm afraid that it will merely claim to do it -- helping to kill off alternatives -- but fail to actually deliver (or rather "deliver" with tons of caveats).]

Re:I would just like a single standard...

[afidel]

USB3 is pointless, use eSATA instead. FW3200 will have a use as a transport for HD video between non-PC machines.

Re:I would just like a single standard...

[eh2o]

All valid points considering the situation of today, but on a future multi-core processor, the CPU expense may not be such an issue, e.g. with 16 cores, one can easily be spared just to deal with data bussing. This may be commonplace by the time USB3 matures.

Re:I would just like a single standard...

[eh2o]

Not unique to Firewire -- USB also has an isochronous transfer mode with guaranteed bandwidth and bounded latency, which is used by audio and video devices. If the device supports it, the host can also select which endpoint configuration to use depending on how much bandwidth is available.

Re:I would just like a single standard...

[jandrese]

The thing is, USB2 (and USB1 for that matter) were built on the same assumption: That by the time they were mature CPU speeds would be so fast that it would not matter if they had a lot of overhead. In practice that has left USB straggling behind Firewire because even though you have a fast CPU, if it is busy doing things then the transfer (and the stuff you're doing) is going to suffer. Also, although the CPUs get faster, the amount of data you want to send tends to grow as well and some of the gains on the CPU are negated.

Re:I would just like a single standard...

[jandrese]

I am curious how they're going to retain USB 2 and 1 backwards compatibility when they switch over to an optical cable.

Re:I would just like a single standard...

[xlsior]

combo cable -- optical, with fall-back to the normal metal pins.

Re:I would just like a single standard...

[heinousjay]

In practice that has left USB straggling behind Firewire

Although ridiculously more popular. Technology is not the only concern.

Re:I would just like a single standard...

[Anonymous Coward]

Howdy,

I agree a lot what you said and which is correct, but using the word "dumb" is quite pejorative and limits the scope how to understand how the mentioned technology works. You can connect devices with any level of 'intelligence' to IEEE-1394. A camcorder to a camcorder, a camcorder to a harddrive to a camcorder etc. daisychaining the devices.

IEEE-1394 (Firewire, iLink) is a bus where attached 'devices' get automatically unique address, register and announce services they provide for others perusal.

Thus any two devices on the same bus can talk and use each others services without any interconnecting device(s) like hub, switch or mediating service between. The bus, protocol used and capabilities provided by non general computer ( != dumb ) device or general computer like devices can be actually very intelligent.

A common use today with a camcorder is to get rid of those DC -tapes and use attached Firewire drive with built in Firewire DC-tapemode feature which it annouces and provides to devices connected to the bus. Notice, that that drive works with any camcorder with IEEE-1394 connectivity and which is able to cope with external DC-tape unit, even old Sony camcorders sold 5-7 years ago work nicely with that kind of firewire diskdrive and _without any additional hardware or software upgrade_.

I was quite supprised when I bougth late -99 a Sony PC100E camcorder and a colleague of mine had just bought a Ricoh camcorder and after a trip to local sights (Deerah, Riyadh if I remember correctly) we were able to simply interconnect our camcorders with a very thin and simple 4-pinn firewire cable that costed just few bucks (riyals actually) and copy easlily each others footage to another camcorder. No computer required, full-speed and no loss of quality ofcourse.

The most amazing thing then was that once you had the new blank tape on the other camcorder and were ready to record. You naturally set it to record and noticed that it actually atuomatically commanded the other camcorder in the other end of the wire to play. The system worked as one system. It did not matter which of the units you did commands the other end did exactly the correct thing with play/pause/stop/record etc. With different manufacturers devices! Wow, that was quite amazing, at least then it was.

If you call that kind of devices "dump" I dunno what most of the devices I have worked (& hacked) from -80 would be called.

I encourage you to test interconnecting different firewire devics yourselves. Find out how "dumb" you still find it all.

HTH,

ac

Re:I would just like a single standard...

[MikeyVB]

It's all Intel's fault. They put USB on everything, but didn't put FireWire on anything until very recently, if they even have by now. So USB "won".

If you really wanted to blame someone, you can at least partially blame Steve Jobs. Wikipedia mentions that one of the reason Firewire wasn't as popular as USB was the high licensing and manufacturing costs compared to USB. (One of the IP owners is Apple). This gentleman mentions that just when Apple was going to give away licenses for a fairly cheap price (1997 or something), Steve came back to Apple and decided that maybe they should charge more. This made everyone unhappy (including Intel) and the rest is history.

I suppose nobody will read this as the topic is pretty old now, but oh well.

Re:I would just like a single standard...

[0100010001010011]

Cold Mountain was cut and edited exclusively in Final Cut Pro on a G5 I have some 3-4 minute movies in DV format that are around a gig or two. just because YOU don't use it doesn't mean that the technology doesn't exist. You know BetaMax actually caught on and was used over VHS for Television production?

I can't find specific numbers right now. However I do remember some one saying that FW was more stable under linux because if multiple devices tried fighting for the USB port someone lost. Second. Say you have 3-4 hard drives. Because of USB topography it's going to limit your max speed to any one hard drive. USB's "peak" speed is also 480 MBits while FW is 'only' 400, but because of the setup FW can sustain that 400 MBit much more of the time than USB can at 480.

Re:I would just like a single standard...

[Anonymous Coward]

Somewhere in the chain, a computer needs to be connected.

More specifically, exactly 1 computer needs to be connected, and the entire chain is dependent on it; if the host fails the other devices cannot communicate with each other.

Re:I would just like a single standard...

[petermgreen]

Firewire was meant to be complementry to USB. USB for relatively slow stuff firewire for fast stuff and stuff that needed to communicate directly with each other without the help of a host PC. Firewire also carried more power allowing bigger devices to be bus powered.

Then intel released USB 2 which had a higher headline speed than firewire and was cheaper but ended up slower in practice. It also was a simple master-slave system unlike firewire.

Then firewire got an upgrade to double it's speed.

I would compare it to IDE vs SCSI, USB like IDE is cheap and basic. Firewire like SCSI is faster and more flexible but at a price.

Re:I would just like a single standard...

[Anonymous Coward]

Yes, your links to Wikipedia and one guy's blog where he states it is his opinion and that "this is the version of the story I like" really make a good case. Pleas stop perpetuating rumors. It is not informative. It is crap.

Re:I would just like a single standard...

[Anonymous Coward]

Shut up fanboy. A quick Google search would reveal that yes, the licensing fees quoted in the Wikipedia article are the correct numbers.

Re:I would just like a single standard...

[toddestan]

Do we 'need' usb? no. We could get by on just firewire.

Do you really want a Firewire mouse?

Re:I would just like a single standard...

[InvalidError]

So is USB3 also supposed to get rid of USB1/2's insane CPU overhead at high speeds?

How much CPU time USB needs is a matter of which hardware interface the USB host has.
- UHCI is the dumbest controller that only does SERDES and other continuous real-time stuff in hardware - only supposed to be used on USB1.x
- OHCI does more data/protocol handling in hardware - the intended HCI for early USB2 hosts
- EHCI does most handling in hardware

My desktop PC (3GHz P4-HT with i865G chipset) has UHCI USB2 controllers and I get ~20% CPU usage while writing to an USB2 disk (200GB Maxtor 6B200M0 in a NexStar3 eSATA/USB box) at 20MB/s. On the other hand, my laptop (1.8GHz A64 with nF3 chipset) has OHCI USB2 controllers and I can write at 24MB/s to the same external USB2 box with ~10% CPU load. With EHCI, the CPU load should be even lower.

USB is only a protocol, how much CPU overhead it will have only depends on how much of the protocol ASIC designers are willing or deem reasonable to implement in hardware. This is the same thing for TCP offload engines: extreme gamers want ultra-low latencies and are willing to pay extra for hardware-assisted network traffic processing on their network cards.

Re:I would just like a single standard...

[dgatwood]

No, CPU overhead will always be a problem. Most computers these days are laptops. Powering up a general-purpose CPU core to do the work in software, no matter how efficiently written the software, will always require a lot more power than custom-designed gate silicon that does data shuffling using a specifically crafted language designed for the purpose.

Using a CPU to do this work is like using Photoshop to capture handwritten notes. Yes, it is capable of doing the job, but it's massive overkill, and as a result, can never do it as efficiently or as well as a tool designed to do the job (e.g. perhaps a custom tool that recognizes the difference between the writing end and the eraser end of a stylus). Even if you added that feature, you're still running a massive application that sucks down lots of CPU power and RAM to do things that are largely unnecessary for the task.

Re:I would just like a single standard...

[Anonymous Coward]

I can get 32MB/sec or so out of my USB2 disks all day with under 10% CPU utilization. Firewire should be lower, but in reality, with good USB drivers the USB CPU utilization is not high. From what I've seen Windows and OSX both have USB drivers that are less than they could be though.

          That said, Firewire should have higher performance and less CPU usage (I haven't tried it.)

Why still no optical link?

[hcdejong]

Instead of trying to cram RF signals at ever higher frequencies down a coax or twisted pair cable with all the problems that entails, why don't we switch to optical cabling?

(I know, it's because that's not profitable enough for the manufacturers. They wouldn't be able to sell us new cables every 5 years)

Re:Why still no optical link?

[ChronoReverse]

Actually, USB 3.0 will have optical connects (while retaining the metal connectors for USB 2.0 compatibility (the socket, not the cable).

Re:Why still no optical link?

[TheRaven64]

The 1394 spec has included an option for fibre for a while. There used to be a nice graph on the Apple site showing the different lengths you could use (I think you got something like 1000 metres with optical FireWire). As far as I know, no one has implemented the fibre version in consumer devices.

Re:Why still no optical link?

[SoupIsGoodFood_42]

That's already in the full FireWire 800 spec, at 3.2 Gbps, too. There's a port on my iMac right now.

It seems that this boost is a way of running the optical speeds over copper, but I could be wrong -- the article isn't very technical.

Re:Why still no optical link?

[ceoyoyo]

Optical cable is fragile. I don't think it would like getting tossed in my backpack and dragged on a plane ride.

Re:Why still no optical link?

[afidel]

If you're worried about your optical cables then get armored ones like these from monoprice.

Re:Why still no optical link?

[Anonymous Coward]

Exactly.

Beyond general durability and everyday robustnes fibers tend to get stiff while undisturbed some time (few years for example). Once you need to touch them you better have replacement cables available if those in use happen to break when you have to reroute or otherwise move them. I've been working with fibers (datanetworking since -85, mostly with large scale installations) and problems with 'fibre tails' (a drop fibres from patch panel to device) or device to device fibers tend to break actually quite often. I would compare such fiber failure occurences to amount equivalent aproximately the same as common disk drive failures.

So, you can't ignore the issue, instead you need to address the issue.

With the fibers always remember. Plan ahead routing and supporting of fibers so that there are no hard corners, they are well supported and don't need to be disturbed till you reengineer the site again. If that doesn't work, make sure you always have couple of spare cables with correct connectors and type (MM/SM). It's good idea to have those spares anyway, even if no immediate need to reroute fibres exists, but when the value or revenue lost is greater than the cost of spare fibers to users of the system.

The same would apply to any user at home. Fiber, core at least of it is class, which as amorphous solid substance and not able stay unchanged over the time in common conditions we tend to use it . Don't just toss those fibers cables behind or leave them hanging wildly if and when the system has any value to someone you care about.

And the optical transmitters/receivers still cost more than equivalent copper solutions. That is excactly the reason why FC (FibreChannel), GE (GigabitEthernet), 10GE etc. are still short distances over the copper. Copper works there well, is durable and still cheaper too.

At the times when I read postings where some 'fanboy' (not you certainly, but someone there above) screams over having the fiber now and immediately really does not understand much about the issue, sadly.

And why is that so? I think media did very bad job during the time ATM was new and spreading. Media misinterpreted the benefits of fiber, spreaded it wide and now we got these folks who demand fiber just because it's fiber -- nothing else is good enough apriori.

Fiber's got great virtues which all I'm not going to delve here, but it's also have its downs which seriously limit its usefulness as a sole connectivity solution.

ac.

Re:Why still no optical link?

[ceoyoyo]

I didn't find any armored fibre optic cables. I bet they're expensive though, at least compared to the copper cables we use now. Probably heavy too. And definitely bulky, since they can't be bent sharply. All good reasons why you wouldn't want one attached to your mouse, say.

Re:Why still no optical link?

[ceoyoyo]

Good advice for using fibre in a fairly permanent setting. Someone pointed out that Firewire does have a spec for fibre. USB and Firewire can't switch to fibre completely, or even in the mainstream though, because glass fibre is never going to be appropriate for the kind of things USB and Firewire are used most for: portable devices and ones that are plugged in and unplugged or moved frequently.

Re:Why still no optical link?

[afidel]

Damn, it showed up in preview, here's another attempt. They aren't exactly expensive at $3.36 ($2.69 in quantity) for a 1.5ft cable. Hell even a 50' cable is only $12 =) Of course they'd be a hell of a lot more expensive at BB or CC but that's what you get for instant gratification and expensive realestate.

Re:Why still no optical link?

[ceoyoyo]

Ah, I was looking at network cables, not audio.

Still doesn't look like something I want hanging off my notebook mouse though.

compared to sata 3Gbps

[speculatrix]

this seems clearly aimed at providing a long reach version of sata at 3Gb/s (Gigabits not GigaBytes per second). Incidentally, too many people call the higher speed sata "SataII" which is somewhat incorrect - sataII means a whole slew of features over and above the first version say Sata IO Org. Note that 3Gbps means 3 x 10^9, not 3 x 1024^3.

Re:compared to sata 3Gbps

[TheRaven64]

Unlike the high speed with SATA, the high speed with FireWire is actually useful. I can connect up to 63 devices on a FireWire chain (actually a tree, but few devices have more than two ports). The fastest hard drives you can buy at the moment (15KRPM SAS monsters) get around 125MB/s sustained transfers. Your average consumer drive is lucky to get 60MB/s (480Mb/s). While SATA has some nice future-proofing, the extra speed of FireWire 800 can be used now to plug in two drives and get decent speed with both (I do this with a chain of two LaCie disks that plug into my laptop). FW 3200 will let you get decent performance out of quite a few more drives; you could easily run a small RAID-5 array over it, connected to a laptop, and get good performance.

Oh, at a higher level, the protocol used for disk communication with FireWire is a lot closer to SCSI than SATA. I'm not sure how it compares to SAS (I've not looked at the SAS spec) but I wouldn't be surprised if they are similar.

Re:compared to sata 3Gbps

[imroy]

Still wrong on the "SATA II" front. The organisation that is now called "SATA IO" used to be called "SATA II" (I can't find what the acronym stood for). When they released the specs for SATA 3Gb/s, people saw "SATA II" and thought the two I's were roman numerals and concluded that this was SATA version 2. Unfortunately, this misunderstanding really caught on. It's been several years now and I still see hard drives and other gear (motherboards, PCI cards, enclosures, etc) being marked as "SATA II" or "SATA 2" compatible. *sigh*

Re:compared to sata 3Gbps

[afidel]

The speed of SATA isn't wasted in theory, SATA allows for a tree topology up to 3 levels deep and up to 128 devices. This has never been a problem in the past because the physical specs didn't allow for SATA to be the host connection so since you were bridging to a separate physical transport you would hide the details of the drives from the host. With SATA 1.2 you now have eSATA which is usable as a host connection. I expect to see cheap storage shelves that are nothing but a SATA->eSATA hub/mux and power distribution box. This would allow you to use one HBA for a few external enclosures just like with SCSI.

Too Little Too Late

[ChronoReverse]

USB 3.0 is slated for 2008. Optical connect, fully USB 2.0 compatible, 5GBPS. Barring a major mistake, Firewire is, again, DOA.

Re:Too Little Too Late

[Alcimedes]

Is USB still going to require that your CPU do all the heavy lifting while doing data transfers? If so then I doubt 1394 is dead.

Re:Too Little Too Late

[ChronoReverse]

That wasn't the real issue with USB 2.0 anyways. The real issue is that USB 2.0 was suited for burst data transfers (where you got the theoretical speed), but simply can't do streaming at the maximum speed due to the enormous overhead of the protocol.


With that said, CPU utilization will indeed be a HUGE concern since USB 3.0 is so fast. The relatively minor CPU overhead of USB 2.0 will give way to CPU stalling overhead unless USB 3.0 addresses it. There's not enough information to make a statement about this, so we'll have to wait for more information to be released.

Re:Too Little Too Late

[Anonymous Coward]

Tell that to people who work in pro Audio/Video, USB is a complete joke for serious applications!

Re:Too Little Too Late

[Anonymous Coward]

Well FireWire 400 is in theory slower than USB 2.0 (480) ... but in practice FireWire will kick USB's nuts. It's much faster in transfer and also it won't have sex with your CPU while transferring at full speed ;-)

Re:Too Little Too Late

[anagama]

I don't think firewire has been "DOA", at least from a useablity perspective. It is true that USB is more ubiquitous.

I have two shells for some laptop drives. The older Firewire shell, with an ATA drive, is very reliable, feels as fast as the internal drive, and requires only one cord to run. My newer one is a USB sata shell -- two wires, the drive seems to get bogged down during large transfers, and it can be finicky about mounting. Sadly, when I went looking for a shell to put that sata drive in, I couldn't find a firewire sata shell. USB is great for small devices -- thumb drives, mice, webcams and such -- but after using both firewire and usb for harddrives, in my mind, firewire is a winner and USB is relegated to the "if I have to use it I will" category.

Re:Too Little Too Late

[Joe The Dragon]

look hear
http://eshop.macsales.com/shop/firewire/

Re:Too Little Too Late

[Revotron]

Optical connect? 5GBps? (Or was that 5Gbps?)

By that logic, analog audio would have died in the 80's and everything would have TOSLink/EIAJ Optical. Hell, up until two years ago I never had optical audio on a computer.

Seemingly "obsolete" and "slower" technologies have their uses, and in some cases can be better than ultra-bleeding-edge.

Oh, and if 480Mbps USB transfers can bog down a CPU, I'd pay to see what 5Gbps+ would do.

Re:Too Little Too Late

[ChronoReverse]

You missed the fully compatible with USB 2.0 part. USB 3.0 still has the copper component since, as you've mentioned, it's still useful and important. So not only can you continue to use existing USB parts, new parts can still just use the copper part. And it indeed is 5.0 Gigabits per second, not bytes.

Re:Too Little Too Late

[ArbitraryConstant]

I share your skepticism, but it seems likely the new spec will address this. We'll have to wait to see how successful it is.

Re:Too Little Too Late

[SoupIsGoodFood_42]

FireWire 800/3200 already arrived quite a while ago, and given that 3.2 isn't as much as 5, I'm sure an updated spec will arrive in time for the people who actually need such high speed connections. Or perhaps you are one of those people who don't understand the many technical advantages of FireWire, and just compare the theoretical max speed of the two specs?

Re:Too Little Too Late

[eharvill]

With that said, CPU utilization will indeed be a HUGE concern since USB 3.0 is so fast. The relatively minor CPU overhead of USB 2.0 will give way to CPU stalling overhead unless USB 3.0 addresses it. There's not enough information to make a statement about this, so we'll have to wait for more information to be released. Sort of the same question I posed to an earlier post -

Can anyone define this CPU overhead? Will I really care with my snazzy new quad-core CPU? How about in a year with 8-cores or more? I'm already seeing complete Dell quad-core systems for less than $600. I think cheaper price/faster speed/more cpu overhead will win all the time, especially since CPU overhead (sure, use a whole core, I have 3 left) going forward will be almost negligible.

Re:Too Little Too Late

[anagama]

Yeah -- after I posted I went looking and saw quite a few. A year and half ago though, I had rough time -- looks like it's time to ebay that USB shell and replace it though.

Re:Too Little Too Late

[mrL1nX]

Yes, exactly! You want professional audio interfaces? Firewire is the only way to go. USB just has too much latency for serious recording purposes and half-duplex just doesn't cut it for software monitoring at sub-5ms latency in most cases.

Yeah -- so what?

[ThousandStars]

If a protocol is released in a forest, and no one is there to hear it, does it make a sound?

I've read a variety of posts about the problems with FireWire (see here and here from what I found on Google), and the big problem is that FireWire didn't become a de facto standard seven or eight years ago when it was really needed. These days, it seems like few computers other than Macs ship with FireWire standard, and I've never seen a laptop in the wild outside of Macs with a six-pin FireWire 400 port, let alone 800.

I've heard this is chiefly due to Apple's initial intransigence regarding licensing; they demanded $1 per computer to use the "FireWire" name, making other device makers really angry. Considering how slim hardware margins are, no one was going to go for it. FireWire 400 is still technically superior to USB 2.0 in many ways, even today, but it's never reached the market penetration it needs, and now USB 2.0 is "good enough" for most purposes.

I use a Mac and so do many family members, and I've long counseled them to get only FireWire drives for backups. When Leopard came out, some were shopping for drives, and I found that I could not find FW400/USB 2 drives for as little as plain USB 2.0 drives. In other words, the FireWire premium for HDs appears to be at least $30. Not a good sign for market penetration.

Now FW 3200 is being discussed when FW 800 already seems dead on arrival in consumer land, and only supported to the limited extent it is by Apple. Not making it backwards compatible with FW400 was an idiotic decision that ensured whatever chance it had in the market was gone. In the meantime, eSATA and the like have come along and perhaps obviated the need for many FireWire applications altogether.

Re:Yeah -- so what?

[kithrup]

FW800 is backwards compatible with FW400. The connector, however, is not.

This is a good thing, because the FW800 connector will be the same one used for FW3200 (or whatever it ends up being called :)).

You can, however, get an adapter pretty easily and cheaply. This will allow your FW800 device to plug into your FW400 chain (and run at 400Mbits/sec)... or your FW400 device to plug into your FW800 chain (and, while it will run at 400Mbits/sec, your FW800 devices will still be able to run at 800Mbits/sec!).

You can also run FW800 over Cat5 cable, and there are small, plastic, optical cables available for it. (The auto industry uses FW quite a bit -- or did, anyway, I'm not sure if they're still using it. The real-time capabilities were very important to htem.)

Re:Yeah -- so what?

[Anonymous Coward]

Firewire didn't become the de-facto standard? Which rocker did you fall off of? *IT IS* (and has been for a while) currently the golden standard, in the areas for which it was designed! It was not ever designed to do the stuff that USB1 was designed to do: mice, keyboards, other human interface devices like tablets, and accessories like webcams, desktop printers, and low end scanners.

Firewire was never a competitor in that market.

Re:Yeah -- so what?

[ThousandStars]

What you say is true -- and it doesn't matter.

Connectors are all fine well, but the time and energy necessary to buy, hang on to, and connect them makes such connectors a niche market. The operative word is convenience, and the more technical version of it is Metcalfe's law. Everyone has USB, therefore everyone makes USB devices, and therefore more people get USB. Apple killed (or at least harmed) the virtuous cycle that might've made FW a standard in 1998 - 2000. Now only a very small number of machines (relative to the size of the market) have FW800, and that number is unlikely to rise because of the chicken and egg problem.

Re:Yeah -- so what?

[CompMD]

Every Dell Precision workstation I have purchased over the last six years has had Firewire implemented on it, along with front and rear ports. I also have two Dell laptops that came with Firewire. If by "few computers" you mean "computers built by the largest manufacturer of computers in the world" then sure, "few computers" other than Macs have Firewire.

You should probably shop around for drive enclosures. $30 premium for Firewire? You can buy encrypting Firewire/USB2.0 enclosures for less than $50.

No, the eSATA stuff has come along and is useful for NEW equipment, but the millions of machines that corporate users have all include Firewire, and they don't all include eSATA, and thus there is definitely a strong market for Firewire equipment that still exists, and will continue to exist for several years. Capability-wise, there is far more that the 1394 spec can do than SATA; they aren't true competitors.

Computers aren't just for geeks and home users. When a company like Boeing decides to stop buying computers with Firewire, then maybe you can say Firewire is dead.

Re:Yeah -- so what?

[Anonymous Coward]

I have a 6 year old Dell workstation tower with multiple firewire ports built into the base machine. Linux handles them just fine and supports the disks connected to them. So yes, it looks like it made a sound....

Re:Yeah -- so what?

[2nd Post!]

It's a good thing that Macs (in the US) are about 9% of the market and grow at about 30% a year; that means this time next year they will be about 12% of the market. I don't think Firewire is going away, any more than SCSI went away despite the success of ATA.

Re:Yeah -- so what?

[99BottlesOfBeerInMyF]

These days, it seems like few computers other than Macs ship with FireWire standard, and I've never seen a laptop in the wild outside of Macs with a six-pin FireWire 400 port, let alone 800.

Toshiba laptops come with Firewire ports as well. It is, however, a rarity... which is sad. The main reason OS X is my base OS running my VMs, instead of in a VM under Linux is Apple's "boot in firewire" mode along with the "upgrade from previous computer" option for installs. When I get new hardware I enjoy the one-click feature to have all my old stuff on my new machine and it all configured for me. The easiest way to do this with Linux and Windows is to run them in VMs and let OS X handle it. Unless Firewire becomes more common, this option will never be cloned by Windows(and probably not by Linux) since the feature doesn't work for USB-2 (as it requires an active CPU).

Re:Yeah -- so what?

[Lost Engineer]

Sony laptops also ship with the "mini" Firewire ports, which is annoying as shit b/c your iPod discharges while you're forcing songs onto it.

Re:Yeah -- so what?

[Jeff DeMaagd]

Apple did reduce the fee, their fee is now only part of the total $0.25 per unit royalty.

I've seen the mini firewire port on many notebooks. I think the problem with puting the six pin port in it is that it would allow the device to consume power and that's not necessarily desirable when everything has to be so small and allowing more power to power a high-draw FW device can be a problem.

I think eSATA is the replacement for Firewire in many of its previous niches other than video decks and camcorders. It helps that it really doesn't require a controller chip on both ends of the cable, and it should be exactly as fast as it would be if it were an internal drive. But eSATA is for only one drive per port, port multipliers are still an expensive and rare exception.

Re:Yeah -- so what?

[Anonymous Coward]

I don't even know where to begin.

iPods are consumer devices. They do not need FireWire. The 2005 article you linked to was a little bit off in its rumor/prediction of MacBook laptops eschewing FireWire. It is two years later and every Macintosh computer still ships with FireWire standard. Every Mac Pro computer has FireWire 800 in addition to regular FireWire 400. That 2002 article you linked to? Well, it is 5 years later and it has nothing relevant to say. Maybe you should Google harder?

Your EETimes articles is even less relevant. From 1999, it is 8 years late to the commentary table. Rumors from unnamed sources do not make good articles.

I hope you are not conseling your friends and family on buying external hard drives, considering your inability to find relevant sources to back up your faulty assertions. I can fully see why you can only find FireWire 400 enclosures that cost $30 more: you suck at shopping!

I guess your final paragraph is the coup-de-grace where you display your ignorance for all the world. No, wonder-boy, FireWIre 800 is not-not backwards compatible. In other words, FireWire 800 is fully backwards compatible with FireWire 400, at least in every consumer implementation on the market. All it takes is a passive adapter cable, much like when using a 4-pin (often called i.Link) FireWire connect with a standard 6-pin FireWire 400 connector.

I am quite happy using FireWire 800 on my 24-inch iMac. But, then again, I am a real Mac user. I do not play on Slashdot in an attempt to misinform people.

Re:Yeah -- so what?

[remahl]

USB certainly has FW beat on adapters: http://en.wikipedia.org/wiki/Image:USB_types_2.jpg

Re:Yeah -- so what?

[Nazlfrag]

Well, if you go by one set of figures, Apple has an 8.1% share in the US with a whopping 37% increase over last year, while a more conservative estimate puts it at 6.3% share and 15.9% growth. The difference in these figures comes from the different predictions, one at 1127k sales and the other 1338k sales, the difference is 211,000 sales which is a fairly decent chunk. It's a little too soon to confirm the 37% prediction AFAIK, but it is likely to be closer to that than the 15% one. Whether it will sustain that growth another year is a tough call.

Either way, it's a very healthy market for Apple in the US but unfortunately they only hold around 2-3% of the world market. As such, the world market share is what will decide the fate of firewire, and it's not looking too promising. Until Dell, HP, Gateway etc. come on board it doesn't stand much of a chance. Even Toshiba including a port as standard would have a greater world impact than Apple.

Re:Yeah -- so what?

[ThousandStars]

Apple reduced the fee too late, and if it's anything above $0 it's too high. Go read Joel's software on the chicken and egg to discover the problems with both.

Re:Yeah -- so what?

[Anonymous Coward]

"In other words, the FireWire premium for HDs appears to be at least $30."

Buy the enclosure separately and the difference isn't that much....last I checked.

Re:Yeah -- so what?

[toddestan]

It won't matter much if Apple doesn't stick with Firewire. They've already eliminated it for the iPods, and for a while they didn't even offer FW800 on their laptops (I believe it's made a re-appearance on the Macbook Pros again).

Re:Yeah -- so what?

[toddestan]

That's not his point. If all the Firewire drives at Best Buy are $30 more than their USB2-only cousins, people aren't going to be buying. The fact that us geeks put together our own external drives using off-the-shelf ATA/SATA drives and a $25 case off of newegg is irrelevant.

Re:Yeah -- so what?

[Anonymous Coward]

Several drive manufacturers are selling FW800/FW400/USB2 hard drives that come Mac-formatted and everything. The peripherals exist for it. Every iMac and Mac Pro has a FW800 port, and Mac users are suddenly wanting big drives for Time Machine.

Speaking of Time Machine, it actually consumes way less space on my backup drive than I was expecting it to. This thing is a really well engineered system.

Does it matter?

[jeramybsmith]

Firewire was mostly used for DV cameras and external hard drives. These devices worked pretty well under XP, Linux, and OSX. However, the 800 pound gorilla has turned on firewire.

Vista's firewire implementation is the pits. I think TI spec controllers basically didn't work at all even though the cards were recognized (maybe it was the other spec). MS recently released a hotfix that remedied some of the problems, but the controllers were then only working up to 100mbps and not 400 even with registry settings set to 400. Getting 12MB/s to an external hard disk instead of 48MB is pretty ghetto.

Also, MS recently released a technote saying that IP over Firewire wasn't an oversight in Vista. It is a feature that will never be re-implemented.

The 800lb gorilla has left the building and I don't think Linux and OSX computers will be enough to keep the market for firewire devices robust except.

Re:Does it matter?

[Anonymous Coward]

So you have hardware that works at 400Mbps in XP (MS Driver) but only 100Mbps in Vista (again, MS driver). After a SW patch by MS, 400Mbps is restored in Vista, yet somehow this is fault of the TI hardware?

Trust me, MS is the one who screwed the pooch on this one.

Oh, and IPv1394 will be made available for Vista. You have to trust me again.

Re:Does it matter?

[RedBear]

The 800lb gorilla has left the building and I don't think Linux and OSX computers will be enough to keep the market for firewire devices robust except.

Except what? During the last two years since Apple moved to Intel hardware, they've sold enough Macs that there are more Intel Macs than PowerPC Macs, even though they were selling PowerPC Macs for several years before that and during the introduction of the Intel Macs. Apple is well on its way to breaking 7% market share for the first time in, well, ever? People are more disappointed with Microsoft than they've ever been before. I don't know, but I'm pretty sure that there's a pretty big and growing chunk of people who really appreciate FireWire and will be very excited to see the FireWire 3200 spec finally arrive in some real hardware. Of course, it will arrive first in Apple hardware, and FireWire networking will be functional as it always has been, and easy to set up. Somehow I just don't see FireWire going away anytime soon even if USB 3.0 doesn't suck as bad as USB 1.1/2.0. The Mac market is nothing to sneeze at anymore, and even when it was there were enough FireWire fanatics around that FireWire drives have been commonly available for years. That's not going to change.

Re:Does it matter?

[Anonymous Coward]

Of course FireWire isn't going away; it's just not that necessary. For one thing, it has three mutually incompatible connectors: 4-pin, 6-pin, and 9-pin (not to mention the possibility of Cat5). USB just has one connector, as does eSATA. I can't just get a FireWire device and expect to be able to plug it in to something else because I need to have a variety of cables and adapters on-hand. What if my device has a 6-pin port and my computer has a 4-pin port? Will it work, or will it fail because there's no way to power the device without those 2 extra pins?

And what's the point in using FW for networking? Are there any computers with FW that don't already have Ethernet? Odds are I already have a Cat5 cable laying around I can use to connect the computers. If it's not long enough, I can make one in a matter of minutes. I can get a gigabit switch and connect 5 computers for the price of a single FW800 cable!

Quite frankly, USB is fine for almost all peripherals, and eSATA provides native speeds for hard drives. FireWire is just for things that need to go fast and don't have an eSATA port, like video and older hard drives. In other words, FW is just a niche market now.

dom

Co-ax

[serviscope_minor]

Finally I can use all the 10 Base 2 crap which I squierelled away. I guess I'll have to disassemble the model Eiffel tower I made out of left over T pieces.

Re:Salvia

[Pojut]

Screw the mod that put this at offtopic. I am aware that Slashdot isn't Erowid.org, but this was a well written Salvia experience.

eSATA

[BestNicksRTaken]

Can't we just standardise on eSATA rather than bothering updating Firewire and USB?

The drives in external enclosures are limited by SATA's speed, so why not just use an external version of SATA? 3Gbps, cabling up to 2 metres......

Frankly Firewire should be used for video cameras and USB should be used for flash card readers. eSATA should be used for external SATA drives (the clue is in the name!)

Re:eSATA

[99BottlesOfBeerInMyF]

Frankly Firewire should be used for video cameras and USB should be used for flash card readers. eSATA should be used for external SATA drives (the clue is in the name!)

Ahh, but can you plug an eSATA drive into your (dumb) video recorder and transfer data straight to it? That's one of the big reasons Firewire is still here, because it works for both computers and other device types without CPUs like drives, cameras, stereo receivers, etc. I mean if I already have a Firewire port on my laptop for my video camera, why do I want a eSATA port as well?

Re:eSATA

[TheRaven64]

Does eSATA let you daisy-chain devices? If not, it's of far less interest to me than FireWire 800. I have two FW800 drives on a small chain and can access both by plugging a single cable in to my laptop.

Re:eSATA

[FireFury03]

I'm left wondering what the advantage is of FireWire3200 over 10Gbps Ethernet (possibly using something like ISCSI for hard drive access). I know that 10GigE is pretty pricey, but I imagine the same will apply to FireWire3200.

Re:eSATA

[dangitman]

3Gbps, cabling up to 2 metres......

Gosh, a whole 2 meters? Did you ever stop to think that many people need a lot more than that? P{articularly in things like entertainment and video production. You said it yourself, eSATA is really only good for one thing. Firewire is far more versatile. So, if anything should die, it should be eSATA for the sake of Firewire. eSATA is just yet another not-widely-supported protocol that is helping take oxygen away from superior standards. The end result of this is that shit like USB ends up winning, because the less-well-known standards keep fighting among themselves, so USB has nothing to contend with.

I guess eSATA is like the Ralph Nader of protocols - not so bad in itself, but a pain in the ass for everyone else. It's a pity there wasn't more cooperation and suppport for Firewire in the first place, because it is a really well thought-out protocol - with many more applications and a lot more room for growth than most others.

Re:eSATA

[Anonymous Coward]

not-widely-supported-say-what-again? eSATA is just an extension of the SATA bus straight out, so unlike firewire it doesn't require special drivers and you get the same speed as an internal device. As a computer builder you don't have to think about licensing special chips, you don't have to think about driver support, all you have to do is put the physical connector on there, and it'll work.

Things don't get much simpler than that./p

Re:eSATA

[dangitman]

Right. But only a small minority of computers and external HD cases have the physical connector. Therefore, eSATA not widely supported. There's a reason it has that name - it refers to a specific connector, which is not widely found on today's equipment. It doesn't refer to making your own ad-hoc connector to extend the ATA connection outside the box, it refers to a specific standard.

Re:eSATA

[jandrese]

Has anybody had any good experiences with eSATA yet? Thus far my experiences have been that the cables tend to fall out easily and it's hit or miss if an eSATA port on an external enclosure actually works at all. When it does work the hotplugging is a bit touchy and the driver support could be better. Oh, and the cables that ship with the enclosures tend to be hilariously short (like a foot or two long) and rather stiff.

Given all of the praise I've seen in this article about eSATA I figure people must be having better luck with it than me, but I've had enough trouble to swear off of it for at least a year or two at this point.

Re:eSATA

[Anonymous Coward]

For non-laptops all you need is a an eSATA bracket, which for new MBs you probably got in the package, and if not, it's really cheap. Laptops are another thing, but Firewire isn't exactly standard there either. Not to mention, there's a joke in security circles: "How do you fix a firewire port? With a hot glue gun."

Getting an external chassi with eSATA isn't exactly hard, I recently bought one with both eSATA and USB support (externally) and both ATA and SATA connectors internally. Convenient.

Re:eSATA

[Anonymous Coward]

The xSATA standard specifies upto 8m cables, but I don't know if anybody supports it. But who cares?

The important thing is that eSATA is essentially free. Your computer already has a SATA host port and your hard drive already speaks SATA. The eSATA specification just describes how to connect them when they're in different boxes. Every external HD enclosure will eventually include an eSATA connector because all HDs will be SATA natively. Every desktop will eventually include an optional eSATA connector because it is just a cable to a header on the motherboard. Buy a 2-port eSATA adapter for $10 and your Mac Pro or PC compatible machine has 2 eSATA ports with no additional software or hardware to install.

Why should I pay for a SATA-FW adapter for every external HD I buy? Why should I have to put up with the software overhead? The problem with FW is that it is so versatile, making it expensive to implement. Because it was designed to have all those great features for the video/entertainment industry, it's primarily only used by the video/entertainment industry.

Why do you think Apple took the FW interface off the iPod and put on USB? Only 10% of the market has a FW port, while 100% of the market has a USB port. If you have an old USB port, your songs may download slowly, but it will still charge. If you tried to plug your FW iPod into a 4-pin FW port, not only would you have to buy an adapter cable, the iPod would drain its battery while syncing.

dom

Re:eSATA

[dangitman]

Why do you think Apple took the FW interface off the iPod and put on USB?

Space. The iPods were getting smaller, so less room for extra chipsets, so they had to pick the most common, even if it was inferior.

moron

[Anonymous Coward]

don't mistake my stupidity for laziness.

Its like a VCR...

[KoshClassic]

USB = VHS
USB 2.0 = Super VHS
FireWire = Beta - technically superior but doomed due to lack of marketplace penetration.

Seems to me that the iPod was the first real killer Firewire app for the masses (yeah, video and audio pro's had their own killer apps for Firewire, but they didn't represent enough of the marketplace for that to matter. If Apple had kept the iPod's Firewire only devices (as were the first generations) something would have had to give. Either the iPod would have been DOA (in the PC world at least, since every new Mac has had at least one Firewire port for years), or PC manufacturers would have been forced to start making Firewire the standard due to demand.

Re:Its like a VCR...

[Joe The Dragon]

more like DVD vs blue ray / HD-DVD

DVD = usb 2.0 good

firewire = blue ray / HD-DVD better but costs more and is not as common.
AKA most pcs come DVD very few have blue ray or HD-DVD.

Re:Its like a VCR...

[Anonymous Coward]

Your analogy sucks. Super VHS is far superior to Beta. Super VHS even surpasses laserdisc quality.

Re:Its like a VCR...

[dangitman]

That analogy sucks even more than the sucky VHS/Beta analogy. Wait a minute, why are people trying to make analogies in this case, when there is no analogy needed?

Oh right, this is slashdot. I forgot that you need to reach your daily analogy quota. BMW vs. Chevy FTW!

Re:Its like a VCR...

[Anonymous Coward]

If Apple had kept the iPod's Firewire only devices (as were the first generations) something would have had to give. Either the iPod would have been DOA (in the PC world at least, since every new Mac has had at least one Firewire port for years), or PC manufacturers would have been forced to start making Firewire the standard due to demand.

Apple would have lost, plain and simple. Apple wanted the PC market so they went USB 2.0, and only then did the iPod become a runaway success. If they had stayed Firewire only the iPod would be a niche product with a fraction of the share it has now.

Really, think about it. They wanted to sell to Windows PC owners. These potential customers weren't going to install Firewire cards (if that's even an option for many laptop owners) to use iPods. They wanted something that would work on their computer as-is and right now. I don't think there was even a choice on Apple's part, and I certainly don't think Firewire adoption is a cause they would sink their billion dollar baby for.

The real fault lies with camcorder manufacturers. Firewire should have been the standard for DV input, it's a perfect match and that's what it was designed for. But the manufacturers used three different names for it (Firewire, iEEE, iLink, WTF), and naturally customers got confused, unsure if they were even compatible; and then there's the odd cameras using USB 2.0 (fine for data transfer, criminally wrong for video) or God knows what else Sony dreamed up (chokes at the thought of Mini-DVDs). It's almost criminal IMO.

Re:Its like a VCR...

[Gilmoure]

I put a twin turbo big block chevy into my BMW M3. It now does back flips.

Re:Its like a VCR...

[Anonymous Coward]

dude, you're post is beyond clueless. I think you just posted what seemed right in your limited exposure to the standards.

  Firewire is not even close being doomed. Its a thriving and successful standard, the only one that provides enough throughput bandwidth for many multi-track recording peripherals and that isn't due to change any time soon. eSATA is encroaching on its domination of throughput for external hard disks, but eSATA is for storage alone and peripheral makers who need high packet throughput without the overhead of USB will continue to use IEEE-1394a/b for the foreseeable future. While there are quite a few 2-4 channel USB audio interfaces, there are no 24 channel USB audio interfaces for good reason. Its not only superior, but it is not at all doomed by the threat posed by USB. The design of the USB protocol itself handicaps it with overhead that prevents you from using the full capacity of the connection. So a 480 USB 2.0 vs a Firewire 400 connection still favors the Firewire connection. But everyone has started moving to Firewire 800, which completely marginalizes USB 2.0.

With this 3.2 GBps announcement, USB is unfortunately a standard that is doomed to be used for only low performing consumer peripherals. Professionals will go for the lower latency, higher throughput standard. And with USB 3 requiring a completely different fiber optic cable, and the S3200 working with the same low-cost cables, I don't see much hope of USB 3 "dooming" firewire any more than DVI has doomed HDMI.

To say the least, your analogy was severely flawed.

cheapness

[OrangeTide]

Firewire 400 cards are like $10. which is pretty much what a USB2 card costs although you don't need a USB2 card since every PC has one built-in now, so you can call it $0.

I hope that Firewire-3200 quickly becomes price competitive, and that chipset vendors get smart and just bundle a 3200 controller in with their prosumer chipsets.

Re:cheapness

[vux984]

Firewire 400 cards are like $10. which is pretty much what a USB2 card costs although you don't need a USB2 card since every PC has one built-in now, so you can call it $0.

When I said a penny saved is a penny earned I was serious.

Firewire maybe less than a dollar more expensive per port than USB, but it adds up. And the bean counters designing hardware care about the pennies.

Not too mention we couldn't cut over to pure firewire even if wanted to. Firewire versions of low bandwidth devices like keyboards, mice, etc simply don't exist...

Re:cheapness

[jandrese]

Nor does it make sense to use what is designed to be a high speed data streaming pipe for something like a keyboard or a mouse. I don't see why people think these serial buses are like the Highlander. I also don't think it was some massive conspiracy to keep Firewire ports off of regular computers (although Intel certainly didn't do anything to reduce the appearance of some major conspiracy), rather USB ports were common because people almost always need to hook up keyboards and mice where external HDDs and camcorders are defiantly more specialized and not something everybody (or even the majority of people) are going to have. Plus, if you find you need a port later you can buy a Firewire card for like $20 at Best Buy and less online.

Re:cheapness

[Anonymous Coward]

Apple also screwed up by not putting Firewire in the original iMac. That really helped to popularize USB and extend its use to things it was not supposed to provide for, like storage.

Firewire2USB?

[Doc Ruby]

Why aren't there any Firewire-<USB(-lt;PC) adapters? It's not a cheap kind of part, but why not a $50 part that can connect Firewire gear to USB ports, even if it's not full bandwidth?

Re:Firewire2USB?

[dangitman]

1. Why would you want to connet your Firewire gear to a USB port?
2. USB doesn't have the features necessary to provide Firewire functionality.
3. When an actual Firewire card costs less than $50, why would you use a silly adapter?

Re:Firewire2USB?

[Doc Ruby]

Because I already have a USB port, and no Firewire port, but I have some Firewire gear. I don't want to use an extra PCI slot for the Firewire (some cheap little PCs I'd like to use don't even have any), or deal with the driver complexity.

The Firewire gear is just a CD carousel, so any extra Firewire features aren't needed. Couldn't a Firewire/USB adapter just allow those Firewire devices that use only the USB features to run?

Re:Firewire2USB?

[dangitman]

i think you'd find more driver complexity in supporting such a device, as well as sub-optimal performance. I'm not sure of what driver complecity you are referring to, as all major Operating Systems have native Firewire support these days, don't they?

Re:Firewire2USB?

[Doc Ruby]

How am I supposed to use Firewire devices on my USB-only PS3/Ubuntu?

And that's not the only device that can't use Firewire.

But I didn't ask why my requirement was required. I asked why it can't be met.

Re:Firewire2USB?

[dangitman]

Well, it is probably possible (almost anything is possible with enough effort) - but the market for it is likely to be so small that it's not economically feasible. As for technical reasons, Firewire and USb work on completely different paradigms when it comes to sharing bandwidth, streaming data, etc. So you'd basically need a very complex translator to manage it - a programmed CPU, basically.

Re:Firewire2USB?

[toddestan]

Why would you want to connet your Firewire gear to a USB port?

Because you have a USB port and not a Firewire port?

USB doesn't have the features necessary to provide Firewire functionality.

We know that.

When an actual Firewire card costs less than $50, why would you use a silly adapter?

The same reason I have a USB-Ethernet adaptor. I don't have the luxury of installing whatever PCI cards I want in every computer I work on, as not all of them are my computers (or have available PCI slots for that matter).

Probably the main reason that such a device doesn't exist is that pretty much every Firewire device out there also speaks USB anyway, and the ones that don't would be in the 2nd category above and wouldn't work with the adaptor properly.

HDTVs and set-top boxes people

[OxFF52]

Unfortunately, many of you are doing just what the article predicts...

The new FireWire release will likely be compared to USB 3.0, which is still under development. Read the article again!

The new spec also will let users interconnect various home-networking appliances via coax cable, linking HDTVs with set-top boxes...

Why would I pay $100 for an HDMI cable to connect a Blu-ray/HD-DVD to my HDTV when I could use a $4 coax cable instead! PLUS, you can send the signal over 100 meters... with any hope, there will be splitters so that I can send a movie/broadcast from my cable providers set-top box to EVERY tv in my entire house.

Step away from your PC a moment and consider the possibilities this brings to the Home Entertainment industry!

Re:HDTVs and set-top boxes people

[Anonymous Coward]

That would be nice, except the coax connection doesn't connect to the HD hardware in the TV and you will just get the SD quality picture.

Oh, and no sound over it. You'll need to get a new TV that supports this fancy schmancy protocol that has hardware to decode the A/V in all its glory. But wait, you'll need a new tuner too, can't have you using the optical audio out on your TV when using protected content, so get a Coax FireWire 3200 tuner. Yeah yeah, that's the ticket!

Oh, wait, my PS3 and x-box360 don't have that output, and even though my Coax FireWire3200 uber-l33t receiver has HDMI and Coax, it'll need the HDMI to the TV for HDMI sources.

Somehow I don't see this panning out.

Re:HDTVs and set-top boxes people

[Anonymous Coward]

Why would I pay $100 for an HDMI cable to connect a Blu-ray/HD-DVD to my HDTV when I could use a $4 coax cable instead!

I dont know where you buy your cables, but you should try monoprice.
$6.43 HDMI 1.3a Category 2 Certified Cable 28AWG - 6ft w/Ferrite Cores (Gold Plated) - YELLOW

Re:HDTVs and set-top boxes people

[JaySSSS]

| Step away from your PC a moment and consider the possibilities this brings to the Home Entertainment industry! That is until the entertainment industry forces everyone to put some kind of brainf***ed DRM on devices!

Better yet.... 3.0

[hellfire]

Technically it is 3.0. 400 is 1.0 and 800 is 2.0, if you were to renumber retrospectively.

Geeks get the 400 vs 800 reference, but I think nongeeks get it completely. Sure 400 is not as good as 800, but what does that mean compared to USB?

USB is 1.0 and 2.0. Firewire should be 1.0, 2.0 and 3.0. Why? Because the general masses see version digits as newer = better. USB is only on 2.0? But Firewire is on 3.0? Gee, that must mean Firewire is more advanced!

Geeks know better, but you don't tell only to the geeks, do you? Besides, versioning for the geeks just makes it easier to support.

Re:Better yet.... 3.0

[sarahbau]

Technically, no, it isn't 2.0 for Firewire 800. It is either IEEE 1394b or Firewire 800. Apple has never called it "Firewire 2"

Re:Better yet.... 3.0

[eh2o]

This will be officially IEEE 1394c. FWIW, USB 3.0 will be ratified at about the same time (mid 2008) and have a speed of up to 5 Gbps on an optical link. The USB people are claiming to have found some workaround for their historically crappy performance (high interrupt overhead) as well, but this remains to be seen...

plus camcorders are going USB2 for video

[schwaang]

It seems like newer camcorders (ones with hard disks or flash storage for video) are using USB2.0 exclusively.

Good. Now I wish they retrofit on some Apple's.

[Neanderthal Ninny]

I like my FireWire 800 but would love it at S3200. Backups with this FireWire on my MacBookPro only take minutes.

FireWie over RJ-45 / Cat5e

[Anonymous Coward]

This could be interesting because there's already an extension (1394c) that allows for FireWire to be transported over Cat5e cable (and co-exist with Ethernet).

I'm curious to know if the new 3.2 Gbps document allows for Cat5e, or whether you can only use it over the FW800 cables.

Oh no, please, no...

[Chabil Ha']

The new spec also will let users interconnect various home-networking appliances via coax cable... Have we forgotten so quickly about vamp clamps, terminators, and finding the break in the damn coax cable???

Re:Oh no, please, no...

[Detritus]

In my experience, thicknet was very reliable. It was installed by people who knew what they were doing and was never touched by the average cow-orker. Thinnet was the reliability nightmare. Many points of failure coupled with random physical abuse and reconfiguration errors.

Re:Oh no, please, no...

[Windom Earle]

Yes, and everybody will be happy drilling the holes to run thicknet throughout their home. heh.

Why not HDMI?

[__aanjtz122]

This may sound like a stupid question, but why would we see the implementation of the new Firewire spec in HDTVs when we already have HDMI using equivalent or greater bandwidth? Would it not be just as likely to see HDMI ports appearing in PCs for connections to HDTV and AV equipment? Thanks

Re:Why not HDMI?

[LordGlenn]

you can buy a dvi-to-hdmi cable right now. I have my pc connected to my hdtv with one...

Re:Why not HDMI?

[silas_moeckel]

HDMI is just one one way uncompressed data cable.

Firewire can send the compressed data to the massive upscaller/decoder that can send the data to the TV. The TV can also send back remote codes to the decoder and storage medium. There does not have to be a one to one ratio on firewire you could have two TV's watching the same thing or something different think one set top box with DVR the can serve every TV in the house, adding a HTPC is trivial as connecting it to firewire. Pretty much it's designed to be the one digital cable moving all the video and audio around your home. Unfortunately they are not in the pocket of the mpaa so there is not all this mandatory encryption like HDMI/HDCP. Add that device classing is mandatory so you should never need drivers and things just work.

USB cheap no enforcement of mandatory classing so everything needs it's own driver. It's a dumb network so everybody can only talk to the controlling PC.

Re:Why not HDMI?

[Joseph_Daniel_Zukige]

While you're at it, why don't we see Freescale/xStreme Spectrum's UWB wireless, which could make all of these cables superfluous around the house and around the office?

Good for innovation...?

[droopycom]

Competing technologies are good for innovations in some cases.

It seems to me that improvement in Firewire speed didnt really happen until USB 2 came along, and that now there is a speed-contest between USB and Firewire that cause the technologies to improve rapidly.

By contrast, the pace at which ethernet speeds evolved seems rather low...

USB is NOT good for webcams unless they're close.

[Ungrounded Lightning]

USB is great for small devices -- thumb drives, mice, webcams and such --

I beg to differ.

I wired my new (2001) house with a couple runs of cat-5E from the comp room to each corner, expecting inexpensive single-chip cameras to become available to be used for security cams.

Well they became available, all right. But all the cheap ones were USB, not Ethernet, and USB has a distance limit suitable for a workstation's desk rather than a house.

If it was just an electrical issue I could have built suitable level-shifters, baluns, etc. to extend the reach. But the limit is apparently timing of the poll/response rather than just signal integrity, so I'm hosed.

In fairness:
  - There are ethernet webcams - but they're not cheap.
  - There are active USB extenders to bridge USB to cat-5/5e and back for long runs - but they cost more than the cameras and I'd need one for each corner.
  - I haven't had the spare round-TUITs to look into whether the timing issue is programmable in the driver or hardwired into the chips...

USB has a nice connector at least

[BlueParrot]

Fix the damn rj45 connector. You know just what is wrong with it. Bah, same thing happened to my thermostat the other day... Somewhere some engineer is going to go straight to hell for making a screw that had to hold force out of plastic.

At least USB deserves that much credit, the connector is quite nice. it plugs in easily, doesn't break unless you do something really stupid, and it is easily removed. Only thing I'll complain about is the tiny version you find in cameras and cellphones. I'm just wondering why they didn't make it a slightly asymetric shape however. You still end up guessing which side goes which way.

Oh well, thou I'm wandering off-topic here my favourite connector is still the 3-pin plug for British household wiring ( and no, I'm not British ). It is solid, it does its job, it is damn obvious how to connect it, and I've never found one you have to force. Why can't more connectors be like that? I'm sick of sockets that are impossible to connect without a flashlight, pins and clips that break if you do something more violent than sneeze on them, quasi-standard plugs that are just different enough so you have to force them, tiny screws that get dented and then stuck.

Really, it shouldn't be this hard. Use solid pins ( giving good contact and stopping them from bending ). Make it easy to see how to connect it, and make it easy to position one pin right and then adjust the other ones. The British power plug has it right, the PCI bus has it right. The stereo jack and phono-plugs have it right. The BNC contacts have it right. The SCART, rj45 and VGA plugs have it wrong. Whoever designed ps2 deserves to spend an eternity in hell trying to connect round 13765-pin contacts with randomly spaced and easily bendable pins, in the dark, using one hand, rebooting the system every time it ends up wrong.

Re:USB has a nice connector at least

[Jeremy Erwin]

The stereo jack and phono-plugs have it right.

I hate the minijack with a passion.

Re:USB has a nice connector at least

[toddestan]

Why? The biggest flaw I see with it is that the typical minijack is held in place by soldered electrical connections to a circuit board, but that's not a fault of the plug itself, but cost cutting by manufacturers. Otherwise, you can insert it any direction, the plug doesn't slide out accidently but at the same time doesn't get stuck in the jack, and I don't have any problems with electrical contact either with the plug and socket itself.

Re:USB has a nice connector at least

[Jeremy Erwin]

I've just had problems with them. Mechanical wear and tear, I suppose. I like a clean stereo signal, and all too often, the plug loosens, resulting in crosstalk.
Incidentally, RCA plugs are not preferred by professionals for video applications-- it's easier to make a 75ohm BNC connection.

Firewire is a dead end technology

[PingXao]

Firewire is good technology, it just has no future beyond a small number of products. There are 3 big reasons for this. First, the consortium that controls the specs for audio-visual delivery over firewire (specs 61883 and AV/C) will not release information about their protocols unless you pay many hundreds of dollars for the specs. USB, an inferior technology IMO, has everything any programmer could ever want available for writing drivers and software. Good luck on that score with firewire.

The second big reason is most cable TV companies refuse to provide working firewire outputs from their set top boxes. The FCC mandated these, but most cable companies did not bother to comply. There is virtually no enforcement of this requirement. Lately more cable co's have been supplying set top boxes with firewire, but copy-prevention encryption forbids premium content from going out over firewire. There's a reason there is no Windows software to receive such a stream. There is software for Apple machines that will do this, but even that is limited to what the cable company lets you do. Moreover, many cable companies turn on the copy-prevention controls EVEN FOR CHANNELS THAT SHOULD NOT BE protected.

The third reason, only now becoming relevant, is the move to HDTV as opposed to SD. Video over firewire is typically encoded as a DV stream, and that is Standard Definition only.

So a new standard is probably needed at this point, but reasons 1 and 2 will ensure no decent software ever gets written to do anything useful with it. Not using Microsoft products. There are firewire drivers in XP that were crippled on purpose by MS to prevent your PC from broadcasting video out to firewire devices. Vista must be much, much worse in this regard, as will every future product and OS from Microsoft.

USB, on the other hand, is eating firewire's lunch even though it is an inferior technology. For that there is nobody to blame but the firewire consortium that controls the protocol specs. Don't believe me? Check it out. If you're developing shareware, freeware or even plain old open source software you have a high barrier to entry right there. They say everything is available on the internet. Not those specs! Not even on the p2p networks and that alone speaks volumes about the lack of interest in firewire.

Re:Firewire is a dead end technology

[Jeremy Erwin]

The third reason, only now becoming relevant, is the move to HDTV as opposed to SD. Video over firewire is typically encoded as a DV stream, and that is Standard Definition only.

I've recorded a few Transport Streams off my HD Tuner, before it died. 1080i and 720p video multiplexed with AC3 audio.

Re:Firewire is a dead end technology

[russellh]

The third reason, only now becoming relevant, is the move to HDTV as opposed to SD. Video over firewire is typically encoded as a DV stream, and that is Standard Definition only.

You don't know what you're talking about. Every single professional HD camera supports firewire.

Re:Firewire is a dead end technology

[PingXao]

It might use firewire but the video is definitely not DV. It might be storable on firewire connected hard drives but that's hardly an application that's going to become mainstream. I'll stick by my claim that firewire, as a consumer interface, is never going to achieve widespread acceptance or success.

You can already see it in the vast number of external hard drives that are available from the major vendors. The previous generation had firewire as an optional interface on the high-end models. The new generation has jettisoned firewire in favor of eSATA. Another firewire application bites the dust.

Re:Firewire is a dead end technology

[jandrese]

Heh, I was noticing at the local cable company that all of the set top boxes they had on display had firewire ports. I asked about it and they're all disabled, always have been always will be, it's an option the cable company can request and they always do.

more like IDE vs SCSI,

[Scudsucker]

only without the big disparity in cost. Firewire doesn't tax the CPU, and is faster than USB 2.0, even at 400 Mpbs. And it took years for USB to add features like booting and the ability to link devices together without the need for a computer.

Follow-up

[ThousandStars]

Another commenter further down has an excellent post on the subject of why FireWire is a good protocol; it nicely complements my post about why, even though the protocol is good, it's effectively lost in the marketplace and is unlikely to matter very much.

A better question

[Anonymous Coward]

Why would you pay $100 for an HDMI cable when they go for less than $4?

Networking over coax?

[SillyNickName]

The new spec also will let users interconnect various home-networking appliances via coax cable

Networking over coax? Wow! What ever will they think of next?

The _host_ can select?

[Joseph_Daniel_Zukige]

I think you missed something there.

Re:The _host_ can select?

[eh2o]

For an isochronous endpoint, the USB device descriptor can list a number of alternative configurations along with the bandwidth requirements of each one. The host knows how much bandwidth is available, not the device -- remember that USB is a host-centric design, nothing gets to talk until the host controller gives the signal -- so the host is responsible for selecting the appropriate mode that fits within the capabilities of the current bus configuration. A typical use case is a video camera offering several different settings of compression/framerate/size etc having a range of bandwidth needs.

I read this and I'm thinking

[Joseph_Daniel_Zukige]

Wow.

That's kind of like the difference between the typical floppy interfaces way back when and SCSI.

And (was it Xebec?) they extended the floppy interface concepts to provide cheap HD interfaces (IDE).

Which would compare to ...

(And I've always been in favor of making even keyboards be intelligent enough to be independent nodes on the network.)

Even with 16 cores,

[Joseph_Daniel_Zukige]

How many busses between the CPU and peripherals?

(Now I'm imagining iNTEL implementing new multiple IO bus controller silicon so that their multiple CPUs can get at system resources more easily.)

There is actually good reason for designing IO de devices so that the CPU that controls them communicates on a bus that is not shared with a lot of other stuff.

Not to say that multiple IO busses is a bad idea. (I exit stage left to go reread the specs on Sun's processors.)

Was that you I answered before?

[Joseph_Daniel_Zukige]

How many busses between the CPU and IO devices?

Re:USB is NOT good for webcams unless they're clos

[Joseph_Daniel_Zukige]

Well, how about some intelligent USB controllers to convert from USB nodes to ethernet?

If I had the spare time, I'd try to build some, maybe with ARM or MCORE or something similarly cheap and low power.

Can we get rid of this meme?

[Joseph_Daniel_Zukige]

Or do inferior memes rule over superior memes, as well?

(Actually, there is a better way to describe this, but it's a verse of Mormon scripture: D&C 1: 19.)

Firewire is dead

[JackMeyhoff]

considering Apple DUMPED it on the iPods and more.

Protocol

[Joseph_Daniel_Zukige]

It's not just the speed. You'd have to build an intelligent controller into the adaptor to behave like a firewire node on the firewire cable and a USB node on the USB cable.

But I have a USB to SCSI adaptor, so the idea is probably not entirely without merit. It's not as useful as it seemed at first that it would be, but it allows me to read MOs on my old clamshell iBook.

Firewire chaining and networking

[ErkDemon]

I chained my PCs and external harddrives together using firewire, XP automatically recognised it as a potential network connection and gave me "LAN over firewire". Whichever PC was switched on first "got" the peripherals, and the second one got to share them over the network. No LAN cables involved, and no hubs. I also included a redundant connection to make a complete circuit, so that even if only one PC was powered up, whichever one it was could still access all the peripherals directly with no replugging. You can't do that with USB!