USB 2.0 For Linux

SilentTone writes: "PCWorld is reporting that USB 2.0 or high speed USB will be hitting Linux first half 2002. Intel is already providing space on its Pentium 4 motherboard for the USB 2.0 controller. With a transfer rate of 480Mbps (more than firewire's 400Mbps) it seems promising." Update: 09/04 23:02 PM GMT by H : So, somewhere between my preview and going live, I seem to have "lost" the link - if you find it, please post below. I'm looking - in the meantime, this is a good Linux and USB tutorial, and Blue Cat Linux is supporting USB 2.0. HA! Found it - story updated.

Comments

once again.....

[jeffy124]

The linux Community has beaten Microsft to something big. First Itanium, now USB 2. (for those who dont know: USB 2 will not come w/ WinXP, it requires a separate download)

mmm

[Yablo]

mmm, usb 2.0... now if only my devices supported it...

Apple

[docstrange]

Apple will probably beat major pc manufacturers to include it as a standard option on all their systems as well. Just like they did with USB, Firewire, and even Gigabit ethernet. Too bad their hardware is so expensive.

Re:Apple

[Anonymous Coward]

There's no reason to include USB 2 on Apple (or any computer manufacturer's boards). There are no USB 2 products worth buying. And FireWire products are just plainly better (faster and with higher QoS). USB inherently slows down your CPU whenever you use it. Whereas FireWire is peer-to-peer. FireWire is the future now. No way Apple's going to touch this without some compelling reason. There is none - USB sucks.

Re:Apple

[Knobby]

Thanks.. For a second there I thought about compiling my latest altivec optimized turbulence simulation, but now I realize that I clearly lack the intelligence to do so... I better sign-on to AOL and make sure my DARPA contract monitor knows that I lack the computer literacy to complete my work..

You fucking schmuck.. Mac users are no less computer literate than you and your geek brethren..

Re: Apple

[Aqua OS X]

Paaaaalease. Grow up before you start to rant about "intelligence"...you seem to lack it.

Re:Apple

[Anonymous Coward]

Mac users are no less computer literate than you and your geek brethren.

Then why can't you recognize a troll when you see one?

Device Drivers

[Anonymous Coward]

Who cares??!!!???

Without device drivers, it will be useless and it is tough to get USB device drivers.

We need to organize some effort to convince manufacturer's to produce open specs for USB device's that they build.

Sophitry

[Anonymous Coward]

Is that really indicative of anything? Linux remains absurdly difficult to learn for most computer users, and it will probably remain that way.

Doh

[Anonymous Coward]

400mbps.

400 millibaud!!! Damn that's slow!

Re:Doh

[Anonymous Coward]

Um, firewire 2 = 800mbps

Re:Doh

[kubrick]

That's OK... with USB 2.0 we can step up to 480 millibaud. That should solve all our problems!

(mb could be millibits, Mb would be megabits and MB megabytes... the article uses megabits, but then abbreviates as mb. I don't know much about PCWorld, but throwing measurements around like this is never a good sign. They use MBps for Serial ATA earlier in the article... maybe they're just cutting and pasting from press releases?)

Re:Doh

[Anonymous Coward]

Mb = Megabit
MB = Megabyte
mb = millibaud

Re:Doh

[Anonymous Coward]

Writing mbps to mean Mbps may be moronic, but you can't tell "bps" from "baud", which is no less moronic.
Baud rate means symbol rate, where each symbol may carry a bit, more than a bit or less than a bit, whereas bps always means the actual number of bits per second.

But what

[AirLace]

will Linux driver developers reverse engineer if there are no Microsoft Windows USB2.0 drivers?

Re:But what

[DaRiachu]

Hehe. Aren't they open-source this time? Hehe. I don't know... If it's USB 2, then I guess it would just build on the original USB drivers, putting additional pieces of code in there to make it have faster throughput...

But hey, I don't like the idea of Intel's muscle-ing in on the high-bandwidth bandwagon, anyway. 12Mbps is good enough for a lot of things (Hey, I still have my Iomega Predator [which works on Linux] on USB...), and Firewire's just as fast, and has the ability to work with other devices without a central computer. So, what's the point? The advantages of IEEE1394 far outweigh those of Intel's new product. ::shrug::

Snootchie-Bootchies.

USB 2.0

[Anonymous Coward]

Well whooopie fucking do. That's fantastic.

To bad none of these USB 2 devices are gonna have drivers.

Sure maybe the USB will work, but since none of these kind of devices ever have linux drivers, it's pointless.

Smart Move

[docstrange]

Smart one: now all they need to do is ban your entire proxy list as well.

And where might we read about this?

[Cutriss]

Umm...WHERE exactly are they reporting this? DAMN...This time the Slashdot editors didn't even bother checking to see if there WAS a story to read...

USB 2.0 is already here...

[dew]

At Fry's Electronics here in Silicon Valley, low-cost USB 2.0 PCI controllers have been on the shelves for well over a month now. Funny thing is, I haven't seen any devices on the shelves that could speak 2.0. Maybe I just didn't happen to see them, but it seems that we might run into the kind of time-delay catch-22's that plagued the original USB: it wasn't important to get a USB board or have USB-support because there weren't any peripherals, and there wasn't any impetus to manufacture USB peripherals, since the install base of computers with USB controllers was small. So USB took quite some time to actually achieve widespread penetration. The same fate may befall 2.0; it may be at least a year before 2.0 is truly compelling. In the interim, Firewire will do quite well. (It's more widespread and is also a more interesting, peered protocol with QOS-like features.)

At any rate, Linux support for these next-generation devices is still important; better for it to come sooner (before it's popular) than later (at which point people wonder why Linux is lagging behind).

Re:USB 2.0 is already here...

[Anonymous Coward]

Yes, it should be noted that FireWire soundly kicks USB 2's ass and that getting 480 MBps out of a USB line is a near impossibility. FireWire is still faster than USB 2 in every implementation available on the market.

Re:USB 2.0 is already here...

[srvivn21]

The nice thing about the transition from USB to USB2 is that devices are compatable. You can run USB2 devices through a USB interface (albeit at lower speeds). I don't think that the incubation period is going to be as long this time.

Then again, what the hell do I know? =o)

Re:USB 2.0 is already here...

[Megane]

Funny thing is, I haven't seen any devices on the shelves that could speak 2.0.

Maybe they've been shipping them all to their new (just opened this past weekend) store in Austin? I saw 'em there, but I'm already sold on Firewire as the better technology. Hell, even Intel is hedging their bets and getting 1394 into their chipsets.

Re:USB 2.0 is already here...

[jacobito]

Actually the problem with acceptance of the original USB was lack of OS support. USB-equipped motherboards were available for quite some time, but Win95 and WinNT had extremely poor or no support respectively. I don't think USB devices took off until Win98 and then the iMac came out. As for Linux, well I love it and I use it, but it's probably fair to say that the USB support in 2.2 had little influence on the number of USB devices that were manufactured.

My $0.02, anyway.

Re:USB 2.0 is already here...

[foldedspace]

The Iomega Predator and a CDRW drive made by Plextor are both out. With the old USB controllers they worked, but were limited in speed. Now they're basically IDE speed drives.

The controllers just came out like 2 months ago. I think Belkin or Orange Micro was first and then Adaptec.

Devices:
Where are the internal USB 2.0 and IEEE 1394 devices? I know we have "Firewire" camcorders. Where are the drives with a NATIVE 1394/USB 2.0 interface? Why is there an adpater from IDE slapped on if 1394/USB is so freaking great? Why not go native?

Mucho kudos to the programmers that did the work on Linux USB 2.0. I'll see it someday. Right now Linux is my server and Windows is my client. I'm not sure what would make me throw a USB 2.0 card in that machine?

I still don't see the missing link. It's way too early for April fools. It's a bad mood kind of day.

Re:USB 2.0 is already here...

[vought]

Que! makes a variant of their USB CD-RW drive that uses USB 2.0.

Of course if they ever get that piece of junk to do 4X sustained writes, I'll be a monkey's uncle.

Re:once again.....GNU/Linux lags BSD

[mr]

And FreeBSD (not to mention Apple) had support for USB before GNU/Linux.

Do not get too high on your horse about how superior Linux is, cuz well, someone will knock ya down.

Windows: Where do you want to go today
Linux: Where do you want to go tomarrow
BSD: Hey! Are you guys comming or what?!?!?!

Umm

[Anonymous Coward]

Shouldn't you wait until it's available on Linux before you decide who won the race? It's still vaporware right now. That's what "first half 2002" means.

Who cares about USB? FireWire is where it's at.

[Cutriss]

"PCWorld is reporting that USB 2.0 or high speed USB will be hitting Linux first half 2002. Intel is already providing space on its Pentium 4 motherboard for the USB 2.0 controller. With a transfer rate of 480mbps (more than firewire's 400mbps) it seems promising."

That's nice. Now our USB peripherals will be unsupported in Linux much more quickly than they are in Windows!

Isn't this putting the cart before the horse? Let's get some useful peripherals out there first.

Oh yeah. FireWire v1 runs at 400Mbps, which is nothing to sneeze at, but I'm pretty sure they're testing v2 and v3 now, which double and quadruple that rate.

Re:Who cares about USB? FireWire is where it's at.

[Cutriss]

Oh...before the grammar Nazi comes along...I meant to say "Now our USB peripherals will be unsupported in Linux much more quickly than they were in USB v1.1!"

Re:Who cares about USB? FireWire is where it's at.

[Anonymous Coward]

FireWire is tomorrow's standard today.

Electronic News article:
IEEE 1394b: Faster, Longer and Simpler

Re:Who cares about USB? FireWire is where it's at.

[Anonymous Coward]

You know... my disk is faster, longer, and simpler (just plug and play :)

Coincidently, it's 13.94 inches....

Nice post, but

[TeknoHog]

A few things:

1. How about including a link to the original story? To say that PCWorld can be found at http://www.pcworld.com/ isn't terribly enlightening.
2. If the readers of /. need to be informed that 480 is more than 400, I'm probably jacked in the wrong board.
3. Huh? It's less than a bit per second? 'm' is for milli, one thousandth.
4. Somebody is being sarcastic.

Re:Nice post, but

[bark76]

3.Huh? It's less than a bit per second? 'm' is for milli, one thousandth.

Hold on there cowboy, you're talking crazy metric talk there. In the world of bits and bytes isn't little m = 1000 and big M = 1024 (or was it the other way around, I can't remember). Maybe I'm the one who's crazy. In fact, maybe I should stop posting to slashdot whenever I've spent more than 8 hours in one day working on nothing but pointers and memory.

Re:Nice post, but

[beanball75]

A little off.

Mb = mega bits
MB = mega bytes

So, the original poster was correct (if not a little anal).

USB 2.0, too late...

[Anonymous Coward]

It has already been reported that Firewire will be shipping soon at 800Mbps, so why bother comparing? royalities to Apple or Intel(?) we still pay someone.

too many standards = no standard at all

[Anonymous Coward]

Let's just stick w/ FireWire/iLink/ieee1394 shall we? When there are so many different 'standards' (like this which isn't even out yet!) it comes to a point when there are no real standards at all. USB, USB 2.0, Firewire, and the many surrounding names. If we could just settle on something and go w/ it we will be much better off. The world would be a better place if we would have one interface for Input devices, audio/video, fixed media, removable media, webcams, digital cameras, scanners, etc. Too much to ask I suppose.

Re:too many standards = no standard at all

[MassacrE]

you wouldn't like it if I agreed with you, because the input devices would knock out FireIeee1394LinkWire out of the running.

Dude...Where's the article?

[bark76]

Thank god someone else can't find it, I thought I was going crazy. Has anyone found it yet, or did some mysterious poster from the future post this before all the other kharma whores even had a chance?

USB networking?

[Yorrike]

How long before we can set up a USB2.0 network?

Sounds like a good step if USB2.0 doesn't cost too much.

Re:USB networking?

[Anonymous Coward]

That's a friggin idiot idea. USB hub != networking hub. If you need faster than 100Mb networking, upgrade to gigabit, so you can use proper switching equipment.

But leave it to some lame linux fucker to kludge together a USB LAN, totally pointless and inefficient but HEY, why not? It isn't like that time couldn't be better put to use, say, writing drivers for USB devices, eh?

Re:USB networking?

[Anonymous Coward]

Oh trust me a USB LAN is just the kind of gay shit linux lamers love. It would probably get front page slashdot to. "hey check out this hack, two dorks hooked up their dorm with a USB network, they attach webcams in the hall to it, so they don't get cuaght masterbating". USB LAN is the ultimate in gay. Only a fucking total wanker would even suggest that shit. Seriously are these people smoking cock. Suggest a USB lan is the same as saying "i'm a moron who dones't know shit about networks". God damn slashfags.

Re:USB networking?

[Anonymous Coward]

werd, homie!

Re:USB networking?

[DavidJA]

A USB lan would be a great idea for simple home networking. Imagine it, just take this USB cable to connect your two computers together and your away!

Re:USB networking?

[Anonymous Coward]

A simple A to A cable won't cut it, there hsa to be switching in between, which is too complex to homebrew. Face it, USB was not made for this, PERIOD.

Re:USB networking?

[Anonymous Coward]

Firewire works fully well in both directions and can work autonomously without a computer in the loop.

Therefore, while no real support for this has been thought out, it is completely technically feasible and even desirable to create a Firewire network.

can the same be said of the USB 2.0 standard, working solely from a technology standpoint? is USB 2.0 as well-suited to networking as Firewire, even though networking with either would be silly?

hell, i've heard people talk about firewire monitors. what about usb 2.0 monitors? could you do that?

Re:USB networking?

[Wesley Felter]

Neither 1394 nor USB 2.0 have enough bandwidth to drive a monitor.

Re:USB networking?

[Anonymous Coward]

USB was made to transfer data from one point to another. What's accomplished with it after that is up to the public. It doesn't matter what YOU think it was made for. I can almost guarantee someone will come up with a simple networking solution that uses USB. Sure it won't replace Ethernet, but it's much more useful than a parallel cable.

Re:USB networking?

[Anonymous Coward]

Well, they've shown more knowledge when it comes to networking than you have. All I see is crap filtering from your posts. No knowledge or useful info whatsoever.

Re:USB networking?

[Overrated Nazi]

How have you been tracking his posts? I'd like to know, seriously. Seeing that he is an AC, as are you................

Re:USB networking?

[erc]

What? Are you on crack? That's what Ethernet is for! Imagine, just plug this crossover cable to connect your two computers together ...

Re:USB networking?

[Anonymous Coward]

And what does Ethernet require? NICs. Every computer in the world has USB ports. Not every computer has a NIC, and not everyone wants to spend the 10 bucks to get one, and time to install it.

The problem with USB 2.0

[James Ray Kenney]

After reading the comments so far, I figured that there should be something on topic, SO...

The one problem with USB 2.0 is that it needs a computer to function. That makes it useless for many consumer uses.
Firewire does not need a computer in the loop. Each device is intelligent enough to talk to other devices in and of itself.

While USB 2.0 does not market itself for those purposes, it does market itself for purposes that firewire has worked fine for, for the last few years. Purposes like video transfer, high-speed data connection, etc. Fire wire is cheep enough these days that interface boards are being bundled along with low-end video editing software.

If more motherboards would provide it onboard, there would be NO need for USB 2.0, except in the few situations where a hub topology was really needed.

James Ray Kenney

Re:The problem with USB 2.0

[Jeffrey Baker]

Whenever I compare the two, I always arrive at the analogy that FireWire is to USB2 as SCSI is to EIDE. FireWire is simply a very elegant and appealing solution that already exists and is already better than USB 2.0. The only way to push USB is by putting a huge marketing force (Intel) behind it. Eventually, I think FireWire will own the high-end while USB owns the random Taiwanese junk market. Compare SCSI and ATA today.

Re:The problem with USB 2.0

[James Ray Kenney]

Yes, I know what you mean. If companies had started using SCSI in bulk, then the prices would have been MUCH lower, MUCH faster.
It is a shame that SCSI and Firewire are not included as standard equipment on everything except the VERY cheapest motherboards. If they were, the market for SCSI and Firewire devices would be MUCH larger and the prices would plummet, creating MORE demand, causing MORE price decreases, causing .... you get the idea.

James Ray Kenney

Re:The problem with USB 2.0

[standards]

Alright, finally someone who knows what he's talking about:

Firewire is a bit more powerful than USB 2.0, given it's expandability and it's hostless capabilities.

Firewire has already become a huge advantage in my industry - the video editing industry. USB 2.0 cannot and will not replace Firewire in terms of performance and flexibility.

USB is a great, inexpensive interface for mice and other desktop devices connected to a computer. Firewire is great for predictable, high-speed data transfers between many classes of high speed devices.

USB 2.0 is an improvement in technology, but it does not improve customer relations. More incompatibile hardware and moving-target standards only results in more costs to the end users.

Re:The problem with USB 2.0

[macinslak]

Speaking of which, which of the two pull lower CPU utilization? As in would somebody be completely out of their mind to try high end video tricks on a USB 2.0 device?

Re:The problem with USB 2.0

[nd]

I've heard that USB 2.0 is considerably more taxing on the processor compared to FireWire.

Re:The problem with USB 2.0

[mimbleton]

Why bother ?
Current generation IDE devices are more than enough for everyday computer use.
Most people wouldn't even notice even someone switched their disks to SCSI.

Re:The problem with USB 2.0

[stripes]

Current generation IDE devices are more than enough for everyday computer use. Most people wouldn't even notice even someone switched their disks to SCSI.

Yeah, but we could have been there 10 years ago (well 7) if SCSI had been pushed to the mass market. Now we would have way way faster systems, plus it wouldn't be such a pain to get a whole bunch of IDE controlers in a machine to build a huge MP3 jukebox (since SCSI can handle 7 or 15 drives on the bus).

Plus I'm not convinced that IDE is as easy on the system as command tagged SCSI, and the IDE command tagging is still kinda buggy. It is definitly much closer then ever though, and very seldom is SCSI worth paying the 3x price jump for disks... (or whatever the conversion ratio is today)

Re:The problem with USB 2.0

[Anonymous Coward]

I thought you might appreciate the article found at this link:


IEEE 1394b: Faster, Longer and Simpler

Re:The problem with USB 2.0

[John Whitley]

And not to put too fine a point on the excellent comment that "it needs a computer to function" -- USB puts *ALL* intelligence in the USB host.. i.e. the computer. If you're Intel, this sounds like a good thing. If you're an embedded developer with anything resembling a processor, it can actually rather suck.

You see, the USB bus basically requires that the host POLL the client devices, with all of the problems that entails. E.g. the USB device's driver polling frequency determines the latency by which the host can accept an update of client state, and since polling itself requires client processing bandwidth to handle, polling faster sucks more client bandwidth. Moreover, polling sucks cycles even if there is nothing to tell the host!

USB is great as a simple desktop bus facility since it does permit cheap implementation of things like keyboards, mice, etc. But for high-rate communication between embedded devices USB is just awful.

Re:The problem with USB 2.0

[Grishnakh]

>The one problem with USB 2.0 is that it needs a computer to function. That makes it useless for many consumer uses.

What kind of crappy bus wouldn't require a high-powered computer to function? Without a computer required, Intel wouldn't make as much money. We wouldn't want that, now would we?

Intel wouldn't have much incentive to push a new high-speed bus standard if it didn't require a computer. Don't like it? Stop supporting the Wintel duopoly.

It's all well and good...

[CtrlPhreak]

But I don't see USB 2.0 succeding Firewire. With the second generation Firewire 2.0 comming out soon with 800Mb/s, I would think that many manufacturers already with products connecting with Firewire(orig) would more easily switch to Firewire 2. I see each product filling the respective roles right now, but just with a speed bump. Digital cameras with USB 2.0 and high end digital camcorders with Firewire 2.0.

more than firewire..?

[Anonymous Coward]

aren't newer and fully backward-compatible firewire interfaces going to start showing up soon that run much faster than the original ones?

does USB 2.0 have any advantages over firewire other than that you're paying licensing fees to Intel and not Apple? (and isn't it free to conform to the firewire spec as long as you call it "IEEE 1394" or whatever instead of "firewire"?) Is there any reason for this USB2 standard to exist other than ensuring that nothing that involves an Apple patent becomes industry standard?

gay

[Anonymous Coward]

uh theres no fucking article. seriusly do these fucking fags even look if theres an article there or what? they just zee some pro linux crap and post it so they can get back to cneosaring everyone.

Getting USB support first is poinltess becuase

1) There are no devices for it yet, by the time there are everyone and there moms stupid os will support it.

2) None of these devices will actually have drivers anyway.

So once again fuck you slashdot for a stupid article from a fucking faggot ass editor.

If a camel is a horse designed by committee

[alewando]

Then USB 2.0 is a duckbilled platypus.

Everyone, sing after me:
Let's slap together a bunch of features onto a product never intended to provide them!

Hey!

Let's win this battle on the marketing field rather than the technical merits!

Hey!

Let's leverage our existing monopolies to create new ones!

Hey!

What's the "SB" in USB stand for? Serial Bus? No! Super Bandwidth!

Hey!

Microsoft isn't going with USB 2.0; that alone should give pause. And what's the roadmap for the future? A present negligible superiority is all well and good for the moment, but how much can they expect to increase it as IEEE 1394 plods ahead? Not terribly much.

*Sigh*

Re:If a camel is a horse designed by committee

[yukonbob]

I had trouble singing your song. Of what tune should I sing it to?

-yb

Re:If a camel is a horse designed by committee

[dan_bethe]

Good points. Furthermore I don't know how long mass adoption will take because the consumer electronics industry's chief R&D and mass marketing shop, Apple Computer, is already committed to Firewire. :)

Re:If a camel is a horse designed by committee

[dangermouse]

I had trouble singing your song. Of what tune should I sing it to?

Oh, you can sing it to the tune of about 49 billion.

Re:If a camel is a horse designed by committee

[Chagrin]

Sometimes you don't need that kind of horsepower or bandwidth - putting a mouse on firewire seems a little rediculous to me.

USB (well, USB 1.0) microcontrollers can be found for under $1 each. I don't think you're going to be able to swing that with FireWire anytime soon.

Re:If a camel is a horse designed by committee

[Salamander]

USB 2.0 is a duckbilled platypus.

I resent that metaphor. The platypus is unique and special, whereas USB 2.0 is a rehash of an existing standard in a failed effort to make it competitive with another existing standard. If we must pick a mascot for USB 2.0, I suggest the frilled lizard - a small lizard that uses fakery to make itself look like a much larger lizard.

Firewire will be at 800mbps this fall...

[dnorman]

It's just waiting for the rubber stamp. Initial hardware is almost ready to go.

I really don't need my keyboard running at 480mbps, so USB2 doesn't really appeal to me.

800mbps of firewire, now THAT is nice...

Re:Firewire will be at 800mbps this fall...

[Anonymous Coward]

Hrm... I thought I saw in InfoWorld (or one of the *World mags) that the new Firewire spec was pushing for 2.98 GHZ. But then again, at the outsourcing I was at when I read this, I just could have been dreaming. :)

Re:Firewire will be at 800mbps this fall...

[dnorman]

Frequency != Speed.

It could be running at 50THz for all I care. All I know is that 800mbps are gonna flow...

Gee

[Anonymous Coward]

Thanks so much for linking to the article. By the way, CNN is reporting that George W. Bush has been assassinated.

Re:Gee

[Anonymous Coward]

If only it were true. Drat!

Re:once again

[fobbman]

Mind you I'm unable to quote whatever article that Hemos is referring to as there is no link to the story and I've searched the PCWorld website and found nothing about Linux and USB 2.0, but just going off of the quote it says that Linux won't have support until the first half of 2002 while this story quotes that Microsoft already has beta drivers and final WinXP drivers will be available by either the end of this year or the first part of next.

Sure there was no linked article, but at least read the freaking POST before you go trumpeting Linux beating M$.

ok, as you wish :-)

[Anonymous Coward]

USB is common.

FireWire isn't, unless you use Macs or DV.

So, looks like we pick USB and the USB 2.0 revision of it to remain backwards compatible whilst upping the speed.

Re:ok, as you wish :-)

[Anonymous Coward]

Bullshit. FireWire is ubiquitous on the line of currently manufactured PC boards. FireWire peripherals actually exist (unlike USB 2). It's also the basis for HAVi (the next-gen digital home audio/video interface standard and the basis for automobile a/v interconnects. And the next-gen version with 800 Mbps has already been accepted. Try peer-to-peer with USB. You just can't do it.

Did anyone else attend WinHEC? USB 2.0 is dead...

[rarose]

Listen folks... anyone who attended WinHEC in Anaheim this year couldn't have missed the writing on the wall. USB is low speed keyboard/mice/joystick/ticker-tape stuff. High bandwidth video/storage/networking is Firewire.
And we're all grown up enough to realize that Windows (and WinHEC) drive the volumes that hardware manufacturers look at to determine what to make.

Put a fork in USB 2.0... it's toast.

But, will anyone use it?

[myov]

Intel wants to expand USB so that it can be used for high bandwidth devices, which traditionally has been the role of FireWire. Yet, Firewire's already established. There are many devices already on the market supporting this standard (camcorders, hard drives, etc). USB 2.0 is still basically vaporware - it's not shipping, and I don't know of any usb 2.0 devices. I really don't see any reason to use USB 2.0 over FireWire.

Firewire can also be used for data transfer without a computer (digital camcorder to vcr, as an example). Since Intel's behind USB and they're interested in selling boxes, you need a computer in every usb chain.

Besides, the next generation of firewire will be ready soon, doubling the transfer rate to 800 mbps.

Re:But, will anyone use it?

[hbackert]

There are devices for USB 2.0 Lots of them. Everywhere here in Akihabara in Tokyo. Check out this webpage: Tsukumos CD RW etc. page and you will find (currently) 4 external CD-RW & Co for USB 2.0/1.1, including brand names like Plextor and Yamaha. But I have yet to see a USB 2.0 hub.

Re:once again.....GNU/Linux lags BSD

[Anonymous Coward]

Actually, NetBSD had USB support before any other free operating system, including FreeBSD.

I believe that FreeBSD based their USB code off of NetBSD.

Not much of a story

[Quikah]

After a few minutes of searching I think I finally figured out exactly where PCWorld has "reported" that Linux will have USB 2.0 drivers in first half of 2002. It is located in this story .

Here is the information they give:

"But don't count USB 2.0 out. Microsoft has announced that it will offer downloadable USB 2.0 drivers for Windows 2000 and for the upcoming Windows XP operating system. Linux support for USB 2.0 should come in the first half of 2002.

Silicon behemoth Intel currently provides space for a USB 2.0 controller chip on its Pentium 4 motherboards, and Gateway has announced that it will put the chips in some PCs beginning this fall. Intel and Acer Labs plan to put USB 2.0 into at least some chip sets by mid-2002; Via Technologies, on the other hand, will add IEEE 1394 support to its chip sets before turning to USB 2.0. AMD says it will support USB 2.0, but not how or when."

Sounds like speculation to me on the Linux drivers. Do any Linux USB devs have any actual info about this?

Re:once again.....GNU/Linux lags BSD

[garett_spencley]

Will someone please hurry up and mod this freak a troll please, I'm all out of mod points! I'm very sick and tired of hearing BSD zealots bash linux.

I have nothing against BSD. As a matter of fact I LOVE BSD. I have deployed all the major BSD variants (Free, NET and Open) and their merits are undisputable. But for crying out loud this flaming is FUCKING ANNOYING!

And the funny thing is that what you guys accuse Linux users of you are guilty of yourself! I'm thinking in particular of the 1337 h4x0r attitude. While a number of Linux lusers have been guilty of this in the past I'm seeing more and more BSD lusers doing the same thing. By bashing Linux! "I'm so 1337 u 1inUx users 5ucK! Switch to BSD! It's awesome. It doesn't suck like linux!"

So please shut up and stop being hypocrits. BSD is great but so is Linux. Get over it! No one wants to hear your whining.

Now someone please mod this post as offtopic.

P.S Oh and for the record. Regarding my first paragraph: I'm also sick of hearing Linux users bash MS.

--
Garett

your logic boggles...

[levl289]

If you were to read the headline, Linux gets this in Q1 2002. WinXP is gold *now*, meaning that it'll be out before then. Beyond that, I'm running 2.2.18 right now, so yeah, that too will require a separate download.

If you're gonna bash MS, do it with proper logic.

(I'm sure this'll win me a modding-down)

Re:your logic boggles...

[Pooua]

If you were to read the headline, Linux gets this in Q1 2002. WinXP is gold *now*, meaning that it'll be out before then.

I would guess that Microsoft is preparing their USB 2.0 support as a update to the gold release, instead of including it in the gold release, based on the following story:

"Windows XP won't support USB 2.0"

By Joe Wilcox

Staff Writer, CNET News.com

April 10, 2001, 9:10 a.m. PT

"The Redmond, Wash.-based software maker said it will not include support for USB 2.0, the latest iteration of the universal serial bus connection technology, in Windows XP, its next-generation operating system expected later this year. Microsoft will instead throw its support behind IEEE 1394, also known as FireWire, which was developed by Apple."

Just a quick question

[JohnHegarty]

Will an exisiting usb device , say a mouse usb1 (that will really need 480mb) work on usb2.

If not , then it will never take off.

Anyway , is there really that much need for this transfer rate with home users... most hard drives would have trouble reading 480mb in a second)....

I like USB, but...

[srvivn21]

USB2 is FINNALY catching up to IEEE 1394 (firewire, iLink, whatever) in terms of speed. Have you heard of 1394b? Coming to devices near you starting at 800mbps, ramping up to over 3.2gbps by 2003.

Ain't technology grand?

Re:I like USB, but...

[Anonymous Coward]

Repeat after me: USB is not meant to compete with "FireWire". It is meant to be a replacement providing plug-and-play for serial, parallel, and PS/2 devices.

Re:I like USB, but...

[lup23]

Agreed...

People keep saying that USB 2.0 is faster then 1394a and it is, slightly. 1394b has a number of other benifits that USB doesn't have.

1. 1394 is non-computer centric. There can be any number of computers in a chain from zero to 63.

2. 1394 provides an isosyncronous mode of transmision. This is required for streaming video.

3. 1394 has better methods of bandwidth enforcement then USB

Beyond thoes benifits, 1394b supports speed up to 3.2Gbits/sec at up to 100 meters over fiber.

Another sign from the industry is that theLucent spinoff, Agere Systems, has scrapped plans to produce a USB 2.0 chipset and in order to speed development of it's 1394b chipset http://www.lucent.com/press/0701/010716.mea.html

USB 2.0 looks to be too little, too late.

Re:I like USB, but...

[Anonymous Coward]

Sorry, none of the normal serial, parallel, and PS/2 devices require 480mbs bandwidth.

USB 2.0 is a replacement for SCSI and an upgrade of USB 1 -- CD burners, ethernet cards, that sort of thing that only sorta works over current USB 1.0.

As pointed out, it doesn't have any consumer electronic industry backing, so it probably won't displace firewire.

Re:I like USB, but...

[stripes]

1394 provides an isosyncronous mode of transmision. This is required for streaming video.

USB1 (and I assume 2) has isosyncronous transmision, I think it was for the speakers that didn't catch on. You don't really need isosyncronous transmision to do streaming you need a buffer twice the size of the jitter. However to sync up multiple event streams isosyncronous transmision makes life way simpler.

That said, I continue to enjoy my FireWire CD-RW, I'm not going to speed right out and buy a USB2 anything.

Suggestion for a use of this

[JohnHegarty]

At this type of speed, you would have a server computer (with hard driver), connected to clients (no hard drive).... 480mb is more than enough speed to run most programs....

how many computers could you support on such a system ?

Re:Suggestion for a use of this

[Anonymous Coward]

I know! I know!

Zero. Because such a network setup doesn't exist.

Re:once again.....GNU/Linux lags BSD

[mr]

Hey, self-taught programming freak-boy......do you actually have proof that FreeBSD's annoucne suppro wans't days before Apple announced USP supprt, and BOTH parties had USB support before any Linux Distro shipped w/USP support?

You are quick to whine, quick to ask for moderators, but where is your proof that I'm wrong?

Links, whiner boy....links.

More background Re:The problem with USB 2.0

[jdunlevy]

An article ("USB 2.0 versus FireWire") related to this ran on MacCentral back in April.

It predicts firewire and usb will coexist, with firewire probably remaining dominant in audio/video.

The point-to-point aspect of firewire seems like a huge advantage for these applications, and it will be interesting to see if the predicted speed bumps of firewire 2 and 3 really are double and triple current speeds, as expected (and way faster than USB 2).

Re:once again.....GNU/Linux lags BSD

[mr]

NetBSD may have been 1st. I don't track NetBSD. I do know wihin 2 weeks of FreeBSD's announcement, Appple supported USB....and Linux lagged in this case.

USB was designed by Intel to waste cpu power

[lupine]

USB is a low cost I/O interface that relys on the cpu. Intel wanted a standard that was cheap and would promote cpu sales. As devices are added processing goes up and available cycles goes down. Other I/O interfaces like scsi utilize a seperate processor to allow the cpu to perform more important tasks.

I like usb, its fairly fast and cheap but intel has its own reasons for pushing the standard.

Re:USB was designed by Intel to waste cpu power

[Johannes]

I'm not sure where you got the information from, but USB is actually CPU efficient. All of the hard processing is done by the Host Controllers (UHCI, OHCI and EHCI).

In fact, USB from a CPU perspective is simpler than SCSI, IP and in fact, is roughly as complex as Firewire.

Re:USB was designed by Intel to waste cpu power

[stripes]

All of the hard processing is done by the Host Controllers (UHCI, OHCI and EHCI).

Either UHCI or OHCI is a pretty dumb part and doesn't do a whole lot of work. The other is a bit better, but I don't think it does scatter gather like many SCSI controllers (and gigabit ethernets). I have no idea about EHCI, it might be pretty bright.

In fact, USB from a CPU perspective is simpler than SCSI, IP and in fact, is roughly as complex as Firewire

Simper is not the same as more efficient. Copying all the data to a fixed location, fixed size buffer, or even doing OUTB in a loop is quite simple. Setting up a scatter gather ring buffer and letting the (non-CPU) hardware do all the hard work is frequently much much faster.

Re:USB was designed by Intel to waste cpu power

[Johannes]

UHCI does scatter gather because of the simplistic processing it does. It splits everything into the endpoint sized chunks (8,16,32 or 64 bytes).

OHCI and EHCI are very similar and can do scatter gather as well, but differently than UHCI.

I really suggest you read some of the specs, because UHCI, OHCI and EHCI all do all of the hard work you mention. All the OS does is the higher level USB protocol stuff, which is simple.

We just tell the HC where the data is and where it should go, it does all of the CRC's, bitstuffing, retries, etc.
USB was designed to be simple and it really is. The work the CPU does is negligble.

All of the problems with USB are either a) device specific (don't adhere to the specs, etc) or b) poor HC design (UHCI can eat a bit of PCI bandwidth and is poorly specified in some cases), but nothing is fundamently wrong with USB or requires complex work.

usb 2 isn't the second coming

[kilgore_47]

First: It's Cool that linux has support for a buzzword before windoze.

That being said, usb2 is just Intel wishing they had gotten on the firewire bandwagon early on. It's on 80mbps faster than FireWire, and doesn't have any serious advantages. FireWire is here for a while, and when it is replaced it will be by something a lot faster than usb2.

But thats just my opinion, I could be wrong. ;-)

Re:once again.....GNU/Linux lags BSD

[garett_spencley]

His post was the troll, not mine. Explanation:

My post offered a honest opinion (in a purposefully-loud tone) and gave facts and explanations to back it up.

All he did was bash Linux. He mentioned how BSD had USB before Linux in an attempt to make linux look inferior. Then he offered a rude joke that bashed MS and Linux and made BSD come out favourable. What he essentially did was offered an opposed opinion with nothing to back it up in an attempt to aggravate his "target".

IMHO his post was troll-like. Mine was rude and flameful, but it was well though-out, had a point and was not meant to offend anyone. It was meant to simply to point out how annoyed I was at his post.

--
Garett

Re:once again.....pOSAROFJ Ajdsf/sl/dsmg /

[Anonymous Coward]

"The linux Community has beaten Microsft to something big. First Itanium, now USB 2. (for those who dont know: USB 2 will not come w/ WinXP, it requires a separate download)"

Fuck off, USB 2 is dead in the water...besides it's vapor..2002, shea ok just like mozilla was supposed to come out in 2000...

uh i violated the postersubj compression filter, what ever the fuck that is, i thought this web site has been around long enough that all the shoddiness had been elimanted from the software, apparently not.

Re:once again.....GNU/Linux lags BSD

[garett_spencley]

I never said that you were wrong. I don't believe that you were wrong. Where did you get that impression from?

Did you even read my post?

--
Garett

Re:once again.....GNU/Linux lags BSD

[mr]

Hey Troll whiner.....where is your links to DISPROVE MY STATEMENT that FreeBSD had USB support before Apple and BOTH had it before Linux? Oh, wait, you have no facts....just linux hype and got a moderator to back it up.

(and for the record, NetBSD had it Before FreeBSD)

USB 2.0 problems

[Auckerman]

1. Firewire is ALREADY ubquitious and plans to move it to your TV already in place.

2. The moment you put a mouse (or anyother low speed device) on that USB 2.0 port you loose the 480MB/sec max throughput.

3. Microsoft supports Firewire instead of USB 2

4. Firewire is looking to move to 800MB/sec in the near future.

Re:USB 2.0 problems

[Lord of the Fries]

Firewire today (1394a) supports:

100 Mb/s
200 Mb/s
400 Mb/s

1394b will support:

800 Mb/s
1.6 Gb/s
3.2 Gb/s

Which should soundly spank USB :)

Re:USB 2.0 problems

[rreay]

2. The moment you put a mouse (or anyother low speed device) on that USB 2.0 port you loose the 480MB/sec max throughput.

Nope. 2.0 hubs do a lot of clever things and full and low speed devices (12 mb/s and 1.5 mb/s respectively) are rate matched at the hub. The hub then transfers data to the host at the full 480 mb/s.

-Rob

Re:USB 2.0 problems

[Auckerman]

"Nope. 2.0 hubs do a lot of clever..."

Irrelevent. It has nothing to do with putting a USB 1.0 device on the chain. It has to do with the fact that USB 2.0 uses time slices, rather than bandwidth sharing. Move your mouse and half the time your USB 2 port will ONLY do mouse stuff, even if you are trying to import video. Firewire has no such problem...

firewire first

[2ms]

Will someone explain to me why anyone should be excited about USB 2.0 when we already have Firewire? Look how long it's taken to get decent support for Firewire. Let's get widespread Firewire first for christ's sake.

bits?

[metalhed77]

is it bits or bytes? i assume bits but i can't really tell since every said mbps and i doubt we're talking about millibits here since that doesn't exist!

Will USB Become a Superstar?

[Jeremiah Cornelius]

Do you think that USB 2.0 stands a chance of winning an Emmy ?

Gosh I hope so! I want to visit the USB 2.0 star on Hollywood's "walk of fame"! Maybe it could replace one of the older, less visited dedications, like Ida Lupino or Jack Lord...

Jeremiah

Pardon me for...

[Burning1]

...sounding like a troll, but... who cares?

Does it really matter weather Linux supports USB 2.0 before manufacturer X?

Do we really need to act like a bunch of kids arguing back and forth about our toy being better than someone else's?

• Linux gets USB2 support? Great.
• Linux isn't going to be behind the curve on this new technology? Excellent.

But seriously people, do you seriously feel that it's only possible for Linux to be considered a viable operating system if we parade our wins?

Microsoft will support USB 2.0 when the situation demands it.

Hey, If you want to do an item by item comparason between operating systems, and are willing to admit failures as well as success: go for it.

...But this kind of comment just makes us look like children.

Re:Pardon me for...

[jedidiah]

>Microsoft will support USB 2.0 when the situation demands it.

Sure, then they'll use it as a club to get their way on something else like an extra bit of bundleware or an upgrade (like NT5 & Win98).

Re:Pardon me for...

[PONA-Boy]

I hate to admit it but soon all our precious iNTEL-based motherboards are going to come sans PS/2.

I have already seen it. I deal stock-in-trade with Acer AOpen products and it is plain undeniable fact that "EVERYthing"(TM) is going USB. PS/2 is going the way of the legacy serial bus; buh-bye!! My own personal opinion is let's keep PS/2 and serial COM ports but technology must move ever onward, I suppose.

It sorta makes you wonder why, though. I mean, how much MORE bandwith and speed do you really need for your keyboard and mouse? Is there anything to be gained by eliminating the PS/2 port altogether save getting rid of a couple of headers off the motherboard? USB is still nothing more than a quicker serial bus after all...

Linux users like myself are only going to benefit from integrated USB support, be it in the kernal or as a loadable module. With the hardware moving on, we can't help but move with it regardless of its righteousness.

-PONA-

The old Firewire vs. USB argument...

[unconfused1]

I applaud Linux developers for supporting USB2 before certain other companies.

There are two things to remember about USB2 as compared to Firewire though...

1. 480Mbits/sec is only a possible maximum. Ultimately your getting that speed depends largely on the topology that is employed.

2. Firewire packets have a time code. USB2 packets do not. That makes USB2 inappropriate for use as a dependable high-quality media conduit.

Re:once again.....GNU/Linux lags BSD

[mr]

So what you are saying is you are a linux bigot, you whine when someone posts a FACT that FreeBSD had usb support before Linux.

What is amazing is that moderators are such mindless sheep to follow your whining. March off the cliff Linux Lemmings! Dance to the puppet master spencley-sales!!!!

The Link

[XBL]

I think maybe he means http://www.pcworld.com/news/article/0,aid,60124,00 .asp.

stop comparing mbit/sec so closely

[Anonymous Coward]

That's like comparing the clock frequency of a pentium 4 and an athlon to determine which one is faster. Its pointless.

USB uses bit stuffing in its signalling as part of its serial bus signal encoding so the effective bandwidth is always 6/7 of the total listed.

your car has 1000 in^3 of cylinder? that's nothing, mine as 2400 ml^3 so it must be better.

oranges are better than apples too!

One problem with that...

[BoarderPhreak]

It's nice that USB 2.0 is faster than FireWire and all, but show me one camcorder that has a USB port. ;)

I don't see FireWire getting replaced by USB 2.0 any time soon, if ever. What's the point? USB 2.0 is not THAT much faster to warrant retooling.

It will speed up scanners, cheap webcams and other such things though. As for your keyboard, well, if you can type THAT fast... Upgrade!

Facts?

[Penguinoflight]

Scince noone else seems to be posting facts, I will.
USB2 is backward compatible to USB1. Firewire isnt.
Firewire 2 is not 800Mbps, it's 3.2Gbps.

So, for the most part, USB2 will takeover the low-end high-band market, and Firewire2 will replace gigabit ethernet, and possibly SCSI.

USB2 is not meant to replace 1394

[Anonymous Coward]

USB2 is not JUST Intel. The USB2 standard is a consortium of OEMs such as Compaq, HP, Lucent, NEC, Philips in addition to Intel.

I think that /.'s are just keen on Intel bashing.

Re:USB2 is not meant to replace 1394

[rarose]

Umm... let's look at that list:

Compaq: in the ditch

HP: halfway into the ditch (with Compaq now pulling)

Lucent: the water in the bottom of the ditch

NEC: what do they actually make again?

Philips: Anything that sells their components. (i.e. Philips had near 0 marketshare in the sound card chip business. Change the target to "USB speakers" and now Phillips can leverage their chip making expertise (see I2C designs) to equal the footing with entrenched sound card competitors)

USB 2.0 is pointless.

[Mr. Spleen]

As has been said many times already among these responses, there is no reason to use USB 2.0 when FireWire exists. Why would you want to use a technology that will, when it is released, be slightly faster than FireWire has been for over 2 years? FireWire is also much more versatile. FireWire devices can communicate with each other without a PC. And with 1394b coming out *very* soon, there is just no reason to switch from FireWire to USB 2.0.

Now if Intel wasn't pushing USB 2.0 as a replacement for FireWire, I would be all for it. I think it would be great if USB 2.0 replaced USB 1.1. Having digital still cameras with a 480MB connection will make downloading all those images really fast. However, billing USB 2.0 as a replacement for FireWire is just insane. There are already millions of camcorders with built in FireWire, and millions of PCs are shipping with it. FireWire has won the war and it hasn't even started.

Mr. Spleen

Re:USB 2.0 is pointless.

[Glytch]

As has been said many times already among these responses, there is no reason to use USB 2.0 when FireWire exists.

Yes, there is a reason to use USB2. As far as I know, USB devices won't work on a Firewire port, and will work on USB2 ports. In effect, USB2 will already have it's user base, and Firewire will have to build one up.

Firewire may be a superior technology, but customers will find USB2 more convenient because they've already got the peripherals to go with it. Firewire will remain a high-end professional video tool and nothing more.

Of course, if you'd be willing to show me lots of PC owners who are eager to buy expensive new Firewire peripherals to replace their expensive old USB peripherals, I might change my mind. :)

Re:USB 2.0 is pointless.

[Midnight Thunder]

Firewire already has a user base, albeit not in the low end PC market. Most high speed storage devices and multimedia devices use firewire. Also, now that Intel is going to add IEEE1394 to their mother boards things may change in the low end aswell. Each has their market and it is a question of the hardware manufacturers deciding which technology best fits their application.

Re:USB 2.0 is pointless.

[stripes]

Yes, there is a reason to use USB2. As far as I know, USB devices won't work on a Firewire port, and will work on USB2 ports [usbworkshop.com]. In effect, USB2 will already have it's user base, and Firewire will have to build one up.

Um, haven't FireWire cameras (still and video), CD-RWs, Hard Disks, RAID arrays, and what not been shipping for at least 18 months?

Firewire may be a superior technology, but customers will find USB2 more convenient because they've already got the peripherals to go with it. Firewire will remain a high-end professional video tool and nothing more.

High end video? All of the digital video cameras I have seen had FireWire. That's $800 cameras at the local 1 hour film shop!

You do have to go pretty far up the food chain on computers to find FireWire though, like Viaos and Macs.

So I don't see a big installed base of USB2 devices. I won't be upset to see USB2 drive down the cost of fast external drives or whatever. As long as it doesn't turn out to suck.

Re:USB 2.0 is pointless.

[Ziviyr]

Having digital still cameras with a 480MB connection ...

Sorry, thats 480 Mb/s, or 60 MB/s.

USB 2.0 technologically inferior to IEEE 1394

[guygee]

With a transfer rate of 480mbps (more than firewire's 400mbps) it seems promising."

The theoretical transfer rate of USB is misleading. Overall, USB remains an inferior technology to USB for applications requiring high bandwidth with deterministic, isochronous transmission. This article provides a good explanation of some of the issues involved. In one of the projects I have been leading, we have been involved in developing the Linux IP over 1394 drivers, and have obtained over 150 Mb/s point-to-point bandwidth using IEEE 1394 asynchronous mode, with room for left for further optimization . The increased function call overhead of USB makes even this modest performance level unlikely.

We are saddled with this inferior technological solution due to the recalcitrance and greed of Intel, who, as usual, are elevating their hidden agenda borne of backroom deals and "strategic partnerships" above the interests of their customers.

Re:USB 2.0 technologically inferior to IEEE 1394

[guygee]

Just to be clear,
150 Mb/s is meant to be 150 Mbit/s.

Re:USB 2.0 technologically inferior to IEEE 1394

[Anonymous Coward]

USB remains an inferior technology to USB

Woah.

Re:USB 2.0 technologically inferior to IEEE 1394

[guygee]

Arghh! Where is that "Slow down cowboy" warning when you really need it?

Let's try "Overall, USB remains an inferior technology to IEEE 1394 for applications requiring high bandwidth with deterministic, isochornous transmission".

Apologies to readers for the hurried post.

why not speed USB up?

[Anonymous Coward]

I find it interesting that everyone keeps bashing USB2. USB2 was not created to replace 1394. Intel already came out and said that 1394 is the hot thing for multimedia/connectability's future.

Why keep USB living in the dark ages (circa. 1993) speeds when it can be upgraded to something that is MUCH faster (12 Mbs vs. 480 Mbs). Are you guys all saying that USB2 is stupid because performance is going to be 40x faster? That's pretty dumb if you ask me.

Why not give OEMs the option of building either USB2 or 1394 devices?

There's nothing to prevent you from running both/either in any particular system.

From what I understand, the actual construction of a 1394 device is more expensive than that of USB2 because any 1394 device has to have the capability of being bus master, whereas USB2 devices have to be 'dumb' (slaves). Seems like computer peripheral OEMs (such as web cam manufacturers) would benefit from USB2, because they would get the high-bandwidth they need and not have to be expensive price-tag associated with developing a more complicated 1394 device.

hey fucknuts

[Anonymous Coward]

bps (bits per second) is not necessarily the same as baud (signal changes per second), and in fact is almost never the same.

Re:hey fucknuts

[mrfiddlehead]

Hey nillnuts, he was referring to the 'm' for milli; as opposed to the M for mega. Pinhead.

who said USB2 would replace 1394?

[Anonymous Coward]

I dont think ANYONE has come out and said that USB2 will replace 1394.

I think /. is just showing their bias again by bashing Intel without facts.

Evolution Theory Halted Again...

[sfe_software]

Just as the Evolution Theory was about to be proven to the scientific community once and for all, Slashdot Editor Hemos mis-places the Missing Link. Conspiracy theorists believe this is the result of an international coverup funded by religious organizations in an effort to keep the Evolution Theory down.

In related news, Linux is rumored to support the new high-speed USB 2.0, which should begin to appear in kernel version 2.6, due out by the end of 2044.

horray!

[Anonymous Coward]

three cheers for my former employer!

Re:once again.....GNU/Linux lags BSD

[Anonymous Coward]

Of course Apple did, that's becuase FreeBSD uses a little girl license that lets apple and microsoft steal anything useful they might actually be able to create inspite of the egomaniacal power games that BSD developement is all about.

OK.. I say USB2.0 was intended to replace 1394

[Anonymous Coward]

And let's look at how alike USB 1.1 and USB 2.0 are...

Last time I looked Intel was putting 4 (count them!) USB controllers in their southbridge. Three USB 1.1 controllers (supporting 2 USB 1.1 connections each) and one USB 2.0 controller (supported 6 USB 2.0 connections ONLY). It's then up to the BIOS or OS to crossbar the external connector to the proper version controller.

If they weren't intent of displacing 1394 (Ha!) they would have tweaked the existing USB 1.1 instead of pulling such a major redesign.

Re:OK.. I say USB2.0 was intended to replace 1394

[pierrego]

Um, doesn't sound like they redesigned anything to me. Sounds like they left their USB1.1 controller alone that they know works and has worked for years, and designed from scratch their USB2 controller. That makes perfect sense to me. I don't understand your logic. Intel puts in a USB2 controller, and that says 1394 goes away? That don't jive...

Pay attention...

[Anonymous Coward]

..to what is going on all around us. Intel comes out today with USB 2.oh support. The apple website (the inventors of firewire) start promoting all the merits of firewire (it won an oscar or something like that.) Intel is just trying to catch up on the marketing program apple has lanched a week ago. Support for usb 2.0 for what? There are no devices. External hard drives, not a one. And if they ever do come out in the distant future (summer 2002), they will be way more than expensive for 80 mb. The people that would be attracted to USB 2.0 the graphic designer (video people), are already using a mac anyway. So this is just intel trying to say "hey, look at us, we have cool things" when they don't really look at if the consumer really need this product they are going to try and force down there throats.

A valid on topic question

[slashdot2.2sucks]

I know that from a licensing standpoint USB is cheaper than Firewire, but

Seeing that USB is reliant on a processor, and that it seems to be more simple . . . is USB also cheaper, physically, to use/make?

In other words, is this going to be IDE versus Scuzzy all over again?

Re:A valid on topic question

[rarose]

Certainly a valid question. Looking at the USB 1.0 and 1.1 designs it's obvious that Intel's goal was to make peripheral development as cheap as possible.

They made the hardware cheap by putting every thing they could into software. If you could save a gate by making it a "soft function", they did it. Why? Two big reasons:

1. Mfgs pay for each unit of hardware produced while software is a flat cost... whether you sell 100 widget joysticks or 10 million widget joysticks you only pay the software guys once, unlike an extra transistor that you 100 versus 10 million of.

2. More software functionality equals more CPU bandwidth used. And chipzilla loves CPU hungry designs.

I don't think it'll be IDE vs. SCSI though. Yes Firewire is more expensive than USB... but it's being incorporated into relatively expensive devices (camcorders, hard drives, etc) as opposed to keyboards and mice. If firewire adds $1 to the cost of a $2000 camcorder, that's no big deal. If that $1 was instead on a $10 keyboard it'd be a totally different story.

Agreed

[marcsiry]

I worked for a major toy company that was basically subsidized by Intel to put a USB port on every one of our products. Additionally, we were encouraged to bundle CPU intensive software in order to drive computer upgrades.

We all knew that USB was a poor choice for anything but momentary inputs, but we were pumping video, sound, all sorts of crap through the lines, and watching the signals degrade and the software sputter to a halt. This was USB-1, of course, but IMO, regardless of the bandwidth, it's a poor choice for the sort of tasks FireWire is ideal for, precisely because it's CPU dependent.

what are you going to use it for?

[Anonymous Coward]

honestly, what will USB2.0 be used for?! right now plain old vanilla USB is plenty enough for your mouse, keyboard, printer, scanner etc.. do you really think you can TYPE at a rate of 480Mbps? fact remains that usb is a Serial Bus and rocket fast speed isn't needed. use SCSI or Firewire for high bandwidth peripherals, there are allready plenty of devices that support these interfaces and i think both technologies (SCSI & FW) have proven their worth time and time again. i think for once Micro$oft did the right thing and dropped USB2 in favor of FW.

support for USB, but what about the devices?

[garcia]

yeah great, we will have support for USB 2.0 but don't we have to be able to have support for each device?

Linux supports USB just fine it is the individual drivers for the devices that we are lacking. There are plenty of people putting in some hardcore work into making things work w/Linux but for the most part I see a lot of red X's next to just about every device (including every USB device I have ever had)

I am glad to see that Intel is going to push for Linux support but we need to have the drivers written too!

Re:support for USB, but what about the devices?

[Ziviyr]

I am glad to see that Intel is going to push for Linux support but we need to have the drivers written too!

Really, my USB Intel webcam won't even work decently on Windows 2000!

Good luck getting support from Intel for Linux!

24 hour time + PM?

[Anonymous Coward]

I know this isn't on topic but a correction:
Update: 09/04 23:02 PM GMT -- there is no 23:02 PM.

I had a USB 1.0 Mobo in 97

[alexhmit01]

I bought in January of '97 a Tyan Tomcat motherboard with a P75 chip, it had a USB port. The spec changed and USB was busted, so they relabeled the parts (and replaced the motherboards if you complained enought).

The problem wasn't motherboards. By mid-97 all machines had the ports. Every machine my high school picked up that year (first half of 97) were P133s-P166s and had USB ports. These were Dell computers with vanilla mobos.

The problem was originally software support, MS didn't support USB until Win98, the Win95 support was busted. Additionally, the market for mice and keyboards died around then. The computers shipped with them AND the market was only supporting $20 replacements, not the highend ones that were popular through 96 and early 97.

Without software support, there was no interest in the hardware. People were pushing parallel port solutions instead. The parallel port scanners, zip drives, etc., dominated the low-end, and SCSI still ruled the high end.

Apple made USB a reality. They used it to replace ADP when they needed something to replace the external SCSI-1 port they used for expansion forever. With their move to IDE hard drives, the SCSI port was rediculous.

Anyone selling addons for Apple built them as USB devices, including mice and keyboards. As the standard was the same, there was no reason to not write Win98 drivers and open up the PC market.

Apple's ability to make something a standard on a segment of the industry is powerful. While Dell and Compaq (soon to be HP) ship lots of machines, nobody is interested in a Dell-only or Compaq-only option on the consumer level. The PC world is commodity only now, so only MS/Intel can add things to the standard. There is no room for vendors to improve the experience, since we scream and yell that it is propriatary.

USB 2.0 is a bad hack. If you don't use a USB 2.0 hub, then any USB 1 device (which keyboards, mice, scanners, etc., should always remain) drops the whole thing to USB 1. In addition, the bus is split up, so the 480 MB theoretical is a real joke. The bus uses time slices, not bandwidth slices. So when the keyboard and mouse grab their fractions of a second, they take bandwidth that could go to the video camera.

Furthermore, Firewire 2.0 brings Firewire up to 800 MB, and its reality is much closer to the theory.

OTOH, I agree that it is good for Linux to support it. As Linux distributions/kernels in the wild don't get upgraded as often, having the support now means that in 2 years, everyone will have it. Better to have the software beat the hardware.

Adding support in Windows is more user-painless (insert CD, press setup, watch this application you got from a no-name vendor to save $3 overrights basic operating system files...) then Linux, so it is good to see Linux beat the curve.

Alex

Re:I had a USB 1.0 Mobo in 97

[Anonymous Coward]

...so when is a PC vendor besides Sony gonna ship 1394 support?

Re:I had a USB 1.0 Mobo in 97

[Yakman]

...so when is a PC vendor besides Sony gonna ship 1394 support?

I was looking at Dell notebook prices recently and noticed that the high end Dell notebooks come with onboard 1394. So their high end desktops might have it as well.

Re:I had a USB 1.0 Mobo in 97

[Anonymous Coward]

...so when is a PC vendor besides Sony gonna ship 1394 support?

Why do you need a PC vendor to ship it? Buy a 1394 card and install it yourself. That's the beauty of a PC, it's expandable.

Re:I had a USB 1.0 Mobo in 97

[Anonymous Coward]

Lots of Compaq consumer models do.

Re:I had a USB 1.0 Mobo in 97

[morzel]

I have a Dell Latitude C800, with built-in 1394 support thank-you-very-much.

They also have firewave (no typo) cards as options in their desktop PCs, and they come pretty cheap (I think it's about 25 EUR for the card).

Re:I had a USB 1.0 Mobo in 97

[Secret Coward]

USB 2.0 is a bad hack. If you don't use a USB 2.0 hub, then any USB 1 device (which keyboards, mice, scanners, etc., should always remain) drops the whole thing to USB 1. In addition, the bus is split up, so the 480 MB theoretical is a real joke. The bus uses time slices, not bandwidth slices. So when the keyboard and mouse grab their fractions of a second, they take bandwidth that could go to the video camera.

Furthermore, Firewire 2.0 brings Firewire up to 800 MB, and its reality is much closer to the theory.

This reminds me of the Token Ring vs. Ethernet battle. Token Ring was an elegant protocol which provided nearly 100% efficiency. Ethernet was a sloppy protocol which provided about 40% efficiency.

There is a lesson in this. In a mass market, the cheap product always wins.

Re:I had a USB 1.0 Mobo in 97

[n_reduce]

USB 2.0 is a bad hack. If you don't use a USB 2.0 hub, then any USB 1 device (which keyboards, mice, scanners, etc., should always remain) drops the whole thing to USB 1.

Well, implicitly your root hub will have to be a USB2.0 hub to have any chance of a USB2.0 acting at full speed wouldn't it?

USB2 is a bit of a hack, but it is absolutely paramount that it has backwards compatibility. So it isn't faster than FireWire? Who cares? Not the average home user. It is perhaps just fast enough for home use to squeeze uncompressed mpeg streams down the wire.

With regard to the time slicing, that is an implicit deficiency however USB2 hubs can work around slow devices with NYET token i.e the host controller is now not simply polling devices (and let the slow ones slow everyone else down). The host controller can skip slow devices, and come back to them later in the schedule.

Re:I had a USB 1.0 Mobo in 97

[mmol_6453]

USB 1 device (which keyboards, mice, scanners, etc., should always remain)

Not true...

USB 2 chipsets will soon be produced in far greater quantity than USB 1 chipsets, and therefor will be cheaper. There's no difference between the cost of printing a USB 2 chipset vs a USB 1 chipset, but you have to think about bulk ordering, etc.

I can see it now...every time they run low on USB 1 chipsets, they'd have to special-order a new batch of outdated ICs.

Would you rather pay extra for the special priviledge of having an antique chipset in your brand-spanking-new keyboard with builtin USB soundcard (and stereo support!)?

Remember, as antiques get older, they cost more to buy. Even in the IC manufacturing market.

Re:once again.....GNU/Linux lags BSD

[mr]

I'd be happy to call you....Oh, wait you are an AC, and most /. users don't hand out their number.

So, how am I supposed to call you?

IEEE1394b already supports 3.6Gbps

[aibrahim]

OK OK so there are no real devices at that data rate yet, but 1394b supports 800Mbps, 1.6Gbps and 3.6Gbps. The standard is inplace and you can expect devices rather soon at 800 and 1.6Gbps with just around the corner.

Yes and no

[Johannes]

There is an EHCI driver and USB 2.0 core patches available right now for Linux.

The current plan is to merge them into the 2.5 kernel, and perhaps backport into the 2.4 kernel once it is deemed stable.

The problem holding back USB 2.0 under Linux is device availability. We've had a couple of vendors donate some USB 2.0 Host Controllers but only 1 device. There are a couple of devices available for purchase now and they work with the aforementioned patches.

The story on pcworld.com is speculation. We have USB 2.0 support, but it's not finished and it will only be finished when we have devices to test against.

That's a good analogy

[Johannes]

Do you know why EIDE is still around and popular today? Because it's cheap. EIDE is simpler than SCSI.

Do you know why USB will still stay around? Because implenting USB hardware is much easier and cheaper than Firewire.

I don't think either will die.

Re:That's a good analogy

[Baki]

What makes you think that USB(2) is easier than Firewire?

It isn't really cheaper either. The price difference is nowhere like SCSI vs. EIDE.

And, Firewire has some important property that makes it way easier from the viewpoint of many simple end-users: it can connect consumer devices directly, without the need for a computer as controller. This, IMO, shall be crucial in the demise of USB2. Firewire has a place and won't go away because of this, for the rest they are mostly the same (from a simplistic POV).

If you must have firewire anyways (because all video cameras have it or shall have it) and for the rest firewire and USB2 are almost the same, then why bother with USB2?

*cough*Bullshit*cough*

[mojo-raisin]

While the ATA interface currently maxes out at 133 megabytes per second, serial ATA will debut at about 150 MBps and is expected to quadruple to 600 MBps in the next few years.

Right... they said the same thing about RDRAM speed, DVD capacity and countless other technologies. The only computer part that reliable gets that much faster is the CPU. Even the good old mainstay of increasing ATA harddrive capacities has it a brick wall for increasing capacity in the past year.

Firewire works.

[Futurepower(tm)]

My experience with Firewire is that it works. No excuses. No problems with confusing the operating system, unlike some situations with USB.

Actually, SCSI is slower...

[Futurepower(tm)]

Actually, SCSI is slower than the current IDE unless used in a multiuser system.

Re:Actually, SCSI is slower...

[Anonymous Coward]

Or a single user system with more processes than just the benchmark program running.

Re:Actually, SCSI is slower...

[Wdomburg]

>Actually, SCSI is slower than the current IDE
>unless used in a multiuser system.

Care to substantiate this? SCSI is up to 320MB/sec compared to ATA at 100MB/sec (or 150MB/sec if you count the yet to appear on market SATA).

And beyond that, the available devices vary wildly - the fastest IDE drives I've seen are 7200rpm w/ 2mb cache, whereas I can get myself 15000rpm w/ 8mb cache in Ultra/160.

Matt

Firewire 2 = 800mbps

[Anonymous Coward]

asdf.

PS/2 port!?

[Midnight Thunder]

One intersting thing to note is that Apple is probably the only computer company offering USB mice and keyboards standard. Sure things are starting to change, but too many manufacturers are still pushing the PS/2 port !?

3.6Gbps extra

[Midnight Thunder]

The only catch will be that most motherboard manufacturers will be making 800Mbps standard and forcing you to buy 3.6Gbps cards separatly if you really need that speed.

BTW isn't the 3.6Gbps rate only available via optical connections?

Re:3.6Gbps extra

[Peter La Casse]

There's another catch, too: most motherboards' PCI busses don't/won't support 3.6 Gbps of throughput.

Honestly, though, if you need that kind of throughput, I can't really see you going with commodity hardware; you're more likely to go with something like this.

Disclaimer: I think that the linked product is really neat, because I work on something similar.

Useless Technology

[Bren]

How useless is this technology? My keyboard uses USB©

At 400 Mbps, that is 400/8 = 50 MBps©
Suppose that one keystroke is translated by the keyboard and sent to the bus as one byte©
1024 bytes is one megabyte©
50 megabytes is equal to 51,200 bytes©

So that is equal to 51,200 keystrokes per second©

Additionally, say a keyboard can handle a maximum of 5 keys held down at one time, before a key must be released to accept more data©
That means 5 concurrent keypresses/second = 10,240 ¥5keys/sec
Say each key press travels a distance of ©5cm downwards, or ©005m downwards©
Using the equations F=ma and a = d/t, along with the fact that time = 1/10240 seconds per ¥5 keypresses©©©
The acceleration can be solved such that each ¥5keys must be accelerated at 51©2 m/s

Then solve for F, we find that each ¥5keys must also have a force of 51©2 Newtons behind it© ¥assuming an average key has a mass of 1g

Obviously not a practical technology for the likes of me©

Bren©

Re:once again.

[Anonymous Coward]

the only thing the linux community has beaten is their collective dicks, seeing as not one in twenty linux users could get laid in a monkey whorehouse with a bag of bananas.

Apple Technology Rocks!!!!!

[Anonymous Coward]

Do not even try to compare Firewire to USB2!

Please read carefully the technical specs before making any comments!

Microsoft supports USB 2.0

[sheldon]

Microsoft is planning on supporting USB 2.0 under Windows XP. It's just that drivers were not ready in time for the RTM version.

This link talks about their support for USB 2.0:
http://www.microsoft.com/hwdev/usb/

Oh man, im gonna Frag in CS

[BrookHarty]

480mbps bandwidth for my USB2 Mouse! (-;

Microsoft already supports USB 2

[sheldon]

At least in beta form, they've drivers available for WinXP already. It's just that USB 2.0 wasn't ready for their feature complete date for WinXP RTM.

I'm puzzled by the /. commentary. The PCWorld article I read doesn't mention Linux at all. When they say having drivers available for Linux, do they mean production stable, or beta?

USB 2.0 transfer rates

[iso]

With a transfer rate of 480mbps (more than firewire's 400mbps) it seems promising

Let me guess: you also buy processors based soley on the megahertz rating. If you seriously believe that the "480mb/s" rating of USB 2.0 (chosen only because it appears on paper to be faster than FireWire), then I have a bridge to sell you.

USB was meant to be a replacement for serial ports; for low-speed devices that could tolerate high-latencies, like keyboards and mice. It was never meant for devices like digital camcoders; that's FireWire's specialty. USB 2.0 is a hack. A wide adoption of USB 2.0 over FireWire would be a very bad thing. Thankfully FireWire 2.0 will reach very close to real and sustainable speeds of 800mb/s, cleanly beating even the highly exaggerated speeds of USB 2.0.

People that buy on "specs" really piss me off. Learn something about the underlying technology before you go making rampant generalizations.

- j

Re:USB 2.0 transfer rates

[aunchaki]

This is exactly the same issue with Bluetooth and 802.11b. I've been waiting for Bluetooth to be reality for months, but hear how 802.11b is making it redundant. The power usage and range requirements of Bluetooth and 802.11b are completely different. They're really apples and oranges.

Bluetooth is meant to make serial/USB-style devices wireless. It's not a wireless networking protocol. Do you need Ethernet to connect your keyboard and mouse? NO!

Ah well, we'll see how it plays out in the market -- if Bluetooth ever gets a chance.

Re:USB 2.0 transfer rates

[Anonymous Coward]

USB was meant . . . for low-speed devices that could tolerate high-latencies, like keyboards and mice

What do you mean by "high-latency?" Do you mean this only in a relative sense? To me, latency means, essentially, the time it takes for data to "get through." If I pressed a key on my keyboard or moved my mouse and then had to wait for a character to appear on the screen or my pointer to move, I'd be quite frustrated.

Re:USB 2.0 transfer rates

[iso]

Yes, in a relative sense. If a device (RAM or a bus) has "high latency" it can take many clock ticks to send a piece of data. This isn't too much of a big deal if you have a large burst (say a DMA transfer) as the latency is only a small fraction of the data transfer time, but if you have many small packets of data, or many different devices sending data at the same time (switching between them for instance) then the latency can really start to add up. These latencies will severely affect your throughput in non-burst conditions.

- j

USB is Intel's ploy to use more CPU...

[Anonymous Coward]

While it is only marginally faster than Firewire, it consumes much more processing cycles from the CPU than Firewire does with three signifcant implications:

1) the devices it can connect are more limited since they must have a powerful CPU consuming power which leaves out many embedded devices. Ever think you'll see USB 2.0 DV cameras instead of Firewire? And porcine quadrapeds might fly.

2) The extra CPU cycles come at an opportunity cost: you need more powerful CPUs for the same throughput as Firewire while processing other tasks and for a given CPU power, you have less to do other things at the same time. Linux users should dislike such inefficiency jsut to sell Rambus components and P4s at a premium.

3) the trend is to distribute processing power, but USB 2.0 reverses that trend by consolidating processing on the CPU. Bad move.

Great for Intel since, if they can shift more burden to the CPU, they can justify the P4 upgrade from the P3. But bad for consumers since there are smarter distributed processing solutions like Firewire that save power and CPU cost.

Just say no: go Firewire, the intelligent high-speed serial bus and one of the few things Apple did right.

Is Apple....

[slashdaughter]

going to have to pull USB's butt out of the fire AGAIN?

Re:once again.....GNU/Linux lags BSD

[Anonymous Coward]

You moderate *MY* post as off-topic but leave the original +2 posting bonus abuse remain untouched? Well fuck YOU, cocksmoking motherfuckers. See you in metamod.

Idiots

[Anonymous Coward]

Firewire is now at 800 Mbps. When can we expect USB 3.0?

Re:once again.....GNU/Linux lags BSD

[Anonymous Coward]

Warning, karma whore alert (just check out his posting history for proof). And he expects an interested employer to hire the likes of him. Bwahaahah.
Moderators, please adjust this travesty of previous moderation abuse and misuse of +2 posting bonus.

I don't think so

[Baki]

Since the difference between Firewire and USB2 is not only, unlike SCSI vs. EIDE, performance, but also real functionality that shall be very visible to end-users.

Whereas non-demanding end-users didn't have a real reason to prefer SCSI over EIDE (on the contrary, SCSI was a bit more complicated to set up) they do have a reason to prefer Firewire: It can function without a computer directly between electronic consumer devices. This makes USB2 more complex and less functional even in the eyes of simple end-users.

Therefore I believe that in this case, not even Intels marketing and pushing to make a computer indispensible for working with video etc. will succeed in letting the inferior solution prevail.

USB polling? Depends on how you look at it

[XNormal]

If you look at the low-level traffic on the bus USB is really a polling architecture. But from the host processor's point of view USB is interrupt-driven since the USB host controller does the polling itself and interrupt the host only if it needs special attention. In reality, a well-written USB driver puts a very light load on the host. Microsoft's USB drivers are not so well-written. The audio drivers are particularly aweful. This is NOT a limitation of the USB standard.

Re:once again.....GNU/Linux lags BSD

[world_citizen]

Who cares anyway USB2 sucks. Firewire /IEE1394 rules. But the worst will probably win like with SCSI.

I don't know why we need USB2 as IEE1394 can do the same. (What i understood USB2 is based on IDE protocol and IEE1394 is more like the SCSI protocol?)

Regards,

Joop.

FUD - "USB wastes CPU cycles"

[XNormal]

Other I/O interfaces like scsi utilize a seperate processor to allow the cpu to perform more important tasks.

The USB host controller is about as smart as many SCSI controllers. It uses bus-mastering DMA based on control structures prepared in memory by the CPU and intepreted by the USB host controller. It puts a very light load on the CPU. What wastes CPU cycles is the type of devices that people build - USB WinModems that rely on the CPU for the modulation, or simple, DAC-only USB audio devices that use the CPU for all sample rate conversion, mixing, software synthesis etc.

True, Intel has been pushing to move more and more of the value in a PC from the peripherals to the motherboard where it can monopolize it. In order to do that they needed an EFFICIENT serial I/O bus. USB is not wasteful in itself.

The fact that USB is a low-cost interface makes it possible to build such devices that use (abuse?) the CPU power. The cost savings of a WinModem compared to a DSP-based modem, for example, would not have such a big effect on the price tag if the interface were much more expensive.

The problem with cars

[Anonymous Coward]

After reading the comments so far, I figured that there should be something off topic, SO...

The one problem with cars is that they need a road to function. That makes the car useless for many offroad uses. A helicopter does not need a road in the loop. A helicopter is useful enough over both oceans and land where there are no roads.

It's like Blue Tooth and 802.11x

[Anonymous Coward]

We've been there before with Blue Tooth and 802.11x. Where Blue Tooth was named the winner before it ever saw daylight, 802.11x proved itself in real life and is growing in popularity now.

With the current IEEE1394 and new upcoming FireWire 2 (wasn't it Texas Instruments ? Correct me if I'm wrong) I IMHO think USB 2.0 will be a dead born technology.

BTW My PPC/Linux distro is now running on a PCI PowerMac. But does anyone known if IEEE1394 is yet supported under Linux ?

Ultimate serial standard beats all of them!

[Anonymous Coward]

Why are PC manufacturers even bothering with all of these various serial proctocols that just keep increasing in speed every few years to the Gig range? Usually they are proprietary and require a license/royalty to the originator of the protocol/hardware technology.

The ultimate standard that is already there is to just put a Gigabit ethernet card in the device. The network card manufacturers are struggling and thus it should be cheap. Moreover, the protocol is open -- thus no royalties -- the speed is already at a Gig, and the number of devices it can address is 255 already. Also, if you want priority based communications to your devices, then just use or modify the RTP (real-time streaming protocol) for listening to mouses/keyboards or sending video to the player whiles't printing takes a back seat.

It seems to me that we have four useless standards such as USB,IEEE 1394, Ilink, firewire and probably others that are extracting a never ending upgrade cycle from the consumer.

If you want to make things go faster than Gigabit, then develop WDM (wave division multiplexing) FDDI devices that will take you up into the Terabit range.

What about GigE?

[mrfiddlehead]

Why the hell don't we just use GigE? It's here now. It works. It can be networked. It's supported by all OS's. It's backward compatible with 10 and 100 Mb devices.

Re: No tutorial needed, just plug and play

[Totally_Tux]

Hi,

Get yourself a recent distribution and install it. No further work is necessary, both my optical mice and digital camera work great.

Cheers,
Joseph Tan

USB support - problems

[wowbagger]

The problem with USB in general that I can see is that the USB design does not have the idea of "standard types".

Consider SCSI: any SCSI hard disk ID's itself as a SCSI mass storage device, any SCSI CDROM ID's itself as a CDROM, any SCSI CDR as a CDR (with the new standards....), etc. I don't have to have a special driver to connect my Quantum Fireball to my SCSI bus.

Now look at USB: I have a USB mass storage device (the docking bay for my NEO-35), a SCSI over USB based scanner, and a USB serial port. Do any of those things have drivers under Linux? Only the serial port. Why? Because there is no standard for USB to SCSI adapters, no standard for USB mass storage devices, no standards IN GENERAL.

The USB design committee basically said "Here's how you read a unique ID from the device. From there, you look in C:\Windows\System, A:; and D: for your DLL". In other words, they basically did half the job of coming up with a standard, taking the "Yer gonna use the Winders Drivers, right?" attitude.

Why don't they establish some standards for devices: Any USB mass storage adaptor must provide these commands, any USB CDR must provide these commands, any USB scanner must provide these commands, any USB to SCSI bridge must provide these commands, etc.

Also, on the subject of the USB HID (human interface device): this is nothing but a big MGI (mogolian group intercourse) - an HID is a thing that does stuff. That's about the extent of the USB spec. How about specifying that an HID must provide a descriptor list in a well defined format (XML, anyone?) that defines what inputs and outputs the device has?

Sorry, but until the various standards committees accept that "Supply a GUID, and the rest is up to Windows" is not enough, things like USB, Firewire, Bluetooth, etc. will not be supportable by anything other than Windows.

As an embedded systems developer, I am disgusted by having to either waste my time writing a million drivers for the things people want to hang off my box, or having to embed Windows.

Who cares...

[kaldari]

FireWire 2 will be out before too long. Now that'll be FAST.

Re:USB 2.0 _HARDWARE_ is here, drivers barely

[dazedNconfuzed]

The USB 2.0 PC cards may have been on the shelves, but they did not come with drivers. Only a couple manufacturers are releasing drivers now, and those arrived just a few weeks ago. It's kinda hard to develop, debug & release a device when the software isn't there to let them talk.

USB 2.0

[Anonymous Coward]

Although USB 2.0 is very fast, I think that Firewire will still be the mainly used system for videos, this is because USB 2.0 does not provide a guaranteed Video line to the PC so the speeds fluctuate and therefore cannot be guaranteed, where as Firewire does.

Mod this down for karma whining

[Anonymous Coward]

If I had points I'd hit you for whining about the possibility of losing karma. STOP YOUR CRYING and post your comment...we don't need the lame-ass disclaimers.

Comment applies only to identical hard drives.

[Futurepower(tm)]

My comment was based on a report I read and my own experimentation. It applies only to identical hard drives, of course.

The report said that many SCSI hard drives are IDE internally, and the IDE is translated to SCSI. So, in those cases, SCSI must be slower.

Several years ago, I criticised Adaptec for implying that SCSI was faster. Almost immediately they changed the language on their site. So they apparently agree.

Note that the theoretical transfer speed of ATA 100 is a lie. No devices currently available can sustain speeds like that.

Conceivably there are SCSI devices that are not available in IDE versions that are faster. However, my understanding is that any speed increase in a single-user system is due only to the hardware being faster.

The storage write and retrieval speed of modern computers is limited by the fact that hard drives are mechanical devices that turn very slowly in comparison to the speed of the CPU.

Note that my comment only applies to a SINGLE-USER system. Such systems only have one process active at a time, in almost all cases. This is because the user turns his or her attention from one process to another, and the other processes are essentially idle. If there were more than one process running on a single-user computer, and both processes were competing heavily for hard drive access, SCSI might be faster.

The conclusion is that SCSI is useful only for busy servers.

Re:Comment applies only to identical hard drives.

[Anonymous Coward]

No devices currently available can sustain speeds like that.

Sure, hard drive spin rates and densities aren't that high (yet). But couldn't a burst read ahead and cached at the drive briefly saturate a UDMA/100 bus?

Re:Comment applies only to identical hard drives.

[James Ray Kenney]

Not Single-User systems but single-tasking systems.
If you are using DOS then I agree, but if you are using windows, windows9x/ME, windowsNT/2000, or windowsXP then it is more than likley that you have a LOT of tasks running in the background.
I will admit that most of them do not use the disk a lot, but some(like fastfind) do.
And ALMOST anytime more than one process tries to access the drive at the same time, SCSI will be faster.
Also SCSI(like Firewire) does have the CPU utilitizion problem. of IDE(or USB.)
It is almost the same as putting a win-modem in a system...At work they upgraded my 33k(hardware) modem to a 57.6k(software) modem and since the laptop was an old pentium, my actual online speed decreased. The pages take longer to render than they do to download, because the modem is taking up so many cpu cycles! Downloads are faster, of course, but the online 'experence' has degraded.

James Ray Kenney

Re:Comment applies only to identical hard drives.

[Futurepower(tm)]

I always disable FastFind.

In my experience, when customers switch to another window, usually the processes in the background are not doing disk access.

WinModems do use the CPU a lot. But most other programs in the background do not, I think.

Re:Comment applies only to identical hard drives.

[Wdomburg]

The basis for your claim being anecdotal, save for some unnamed report doesn't give me much faith in your claims. For that matter, the rest of your post doesn't give me much faith either.

I can't claim with authority that some drives don't have a SCSI layer over an IDE one. However, regardless of that, it does not mean that "in those cases, SCSI must be slower".

For example, one of the major advantages of SCSI over IDE is tagged command queueing. In an IDE bus, you can only ever have one command pending at any given time, whereas SCSI allows typically for at least 64 commands.

One advantage of this is that it allows for reads and writes to be reordered for more efficiency. And this speed benefit would be gained even if it was done in a layer above the drives native interface. If you want an analogy, look at the improvements in performance in different kernel versions - hardware is the same, but it is being accessed and utilized more intelligently.

And no, the transfer speed of ATA/100 is not a lie. The bus is perfectly capable of transfering data at that speed, regardless of the speed of the devices. And sustained speed is not the only factor here; there are performance gains to be had even if you are just bursting traffic at that speed.

That does, however, bring up yet another disadvantage of the IDE bus - multiple devices. Aside from only being able to handle two per bus, if you use even that many (!) you seriously degrade performance since only one device can "speak" at a given time. An example of where this could be bad is if you're copying from your hard drive to your CD-RW.

Your claims that single user systems don't claim benefit from the multi-tasking features of SCSI are at best naive and at worst stupid (and more likely is a combination of both). Modern operating systems (i.e. anything post DOS when talking about PCs) never have just one process active at a time under normal use, even when the user is only running a single application. (E.g. Windows is always running Explorer (no, not the browser, the desktop shell) in addition to the foreground application).

And increasingly, users are running multiple applications at once - an mp3 or cd player in the background, a copy of ICQ/AIM/whatever in their taskbar, etc, etc, etc).

If you'd actually run some performance testing instead of relying on a single article and your owen prejudices, you might learn that even under very mild concurency (i.e. not "multiple processes competing heavily for hard drive access") SCSI yeilds a signifigant performance increase (particularly since the cost of high speed CPUs and memory have dropped so drastically, thus making the hard drive all the more of a performance bottleneck.)

Matt

Such rare events make no detectable difference.

[Futurepower(tm)]

My understanding is that, yes, it could. However, such rare events make no detectable difference over several hours of use.

Both the cache in the hard drive and the cache in the operating system are very efficient. My understanding is that, in most cases, it is the memory cache associated with the operating system that would supply the data in cases where it was recognized that further data would probably be needed.

The software that runs both caches would recognize that more data was needed, and the OS cache would read the data from the hard drive cache as the data became available during the slow turning of the disk. The only case where a hard drive cache would provide burst information would be those where the software in the hard drive cache predicted the user's needs better than did the software in the OS cache. Since OS caches are so important to OS benchmark results, they are very well-designed. It is unlikely that a hard drive cache would guess better.

The main purpose of a hard drive cache is to read an entire track into memory.

Eliminating cable clutter

[speedbump]

I started buying USB motherboards when it looked like USB was going to be the Next Great Thing, but it fizzled. I STILL don't have a USB mouse, keyboard, CDROM, or whatever.

But, I do have FireWire devices, and their apparent bandwidth doesn't seem to bog down the I/O subsystem of my motherboard.

USB 2.0? No thanks, I will stick with 1394.

We don't get paid.

[Futurepower(tm)]

We don't get paid to write these comments, so not as much work is put into providing references. In response to my complaint, Adaptec changed the language on their web site, so that the web site did not make claims that SCSI was faster. There is a huge hardware book with a black cover sold in big technical bookstores that explains that most SCSI drives are slower than the equivalent IDE. I don't remember the name. I have experimented with SCSI myself, with an Adaptec 2940 UW SCSI adapter (not the latest). I did not see a performance increase under use that I would consider normal for my customers and myself.

My understanding is that the latest IDE is also able to cache commands.

"SCSI yeilds a signifigant performance increase (particularly since the cost of high speed CPUs and memory have dropped so drastically, thus making the hard drive all the more of a performance bottleneck.)

Note that this sentence is illogical. If the CPU is fast, what is saved by having a lighter CPU load using SCSI?

It is unfortunate that when there is a disagreement or misunderstanding, Slashdot readers often make hostile remarks, such as you did above in calling me "stupid".

Most users would not see a difference it they had SCSI. Most users don't try to run two processes that both do hard drive access at the same time. So, my comment is very helpful, because it saves readers from buying SCSI and wasting money.

People who have very heavy usage know who they are and are unlikely to be led astray by my comments.