History 2002: USB 2.0 Revolution

The change from USB 1.1 to USB 2.0 isn’t setting the world on fire, nor even the world of IT connectivity. And yet it should.

USB stands for Universal Serial Bus, and in its 2.0 form could well create a small revolution for PCs, simply by becoming a universal interface eliminating the need for all others.

Version 1.1, at 12Mb/s, first appeared in 1995, when it was integrated in all PCs, but its speed limited it to slow peripherals (keyboard, modem, external FDD, digital camera, and ultimately scanner, printer or CD-R/RW drive, for users who weren’t in a hurry).

With USB 2.0, at 480Mb/s, i.e. 60MB/s, in one fell swoop the speed has been multiplied by 40.

From now on, all peripherals can be satisfied, including HDD and tape drives, with one caveat.

Don’t expect to see the maximum transfer rates of 60MB/s right away. Initial testing of USB 2.0 drives has yielded averages around 10 to 12MB/s.

We’re still missing the equivalent of the Oxford 911 chipset, which boosts 1394.

Until now, it was fairly complicated to design one’s own computer configuration. You had to buy the shell, then integrate various cards and peripherals, screwdriver and migraine not included.

Imagine, however, that tomorrow a shoe-box sized case arrives on the market with, on the inside, merely a power supply, motherboard and a dozen USB 2.0 ports in front. Period. No Windows, no software applications, no peripherals.

Such a USB/Intel computer would cost only $150 or $200. Gone are the days of the PC with a tangle of wires and connectors in the back. Instead the user chooses the features he or she wishes depending on need and the best available price: OS and application software, or on the hardware side: keyboard, screen, scanner, printer, storage devices, ADSL modem, webcam, speakers, etc.

For many of these peripherals, no additional power supply would be necessary, since they can use that of the serial interface.

Currently, consumer electronics companies have shown a preference for 1394, but this probably won’t last, whether for audio Walkmans or video cameras, or eventually cell phones, TVs, stereos or the range of home video devices. And why not USB in the car? Or USB smart cards in key-chain form (several companies, including Rainbow Technologies, have already conceived of USB security devices)?

USB 2.0’s success is guaranteed, not only because of its speed, but also its low cost. Will IEEE 1394 be able to survive? The latter interface is just as fast (400Mb/s), but costs considerably more, at least $10 to implement on a PC, with additional royalties, albeit small ($0.25), for Apple. A new version, 1394b, should offer 800Mb/s, but it better not be late. The only incontestable advantage of 1394 is its peer-topeer functionality, i.e. that 2 FireWire devices can be connected to each other without the intermediary of a PC, for example, a digital camera connected directly to an HOD or printer.

A new form of USB, called USB OntheGo, however, is in the works, to correct the gap.

It is not inconceivable that in the future, USB could replace Ethernet to build a small local network.

Already, SCSI has been replaced by 1394 on PCs, particularly Apples. Caught between USB at the bottom, S-ATA in the middle and FC on the high end, SCSI’s future looks bleak. We certainly can’t see the point behind the current project for serial SCSI, which seems like little more than a marketing scheme to extend the interface’s life artificially.

An ATA-to-USB controller card is still indispensable to run a HDD drive, since currently there are no drives with embedded ATA-to-USB cards. HDD makers are still waiting for stronger consumer demand before taking that plunge. With USB 2.0, however, the volume will no doubt make it worthwhile. Is it possible even ATA would then fall by the wayside? In any case, don’t be surprised to see a flood of USB 2.0 announcements at the upcoming CeBIT in Hanover, Germany. Taiwanese companies will likely be the first to jump on USB 2.0-cards, hubs and peripherals of all shapes and sizes.

This article is an abstract of news published on issue 178 on November 2002 from the former paper version of Computer Data Storage Newsletter.