byte.com
Pentium 4 : A Coming Revolution
By Jerry Pournelle
January 08, 2001
I've also got a Pentium 4 box. I haven't had much chance to put it through its paces. It runs at 1.5 gigahertz, has an nVIDIA GeForce 2 board, and 256 megs of memory. The OS is Windows 2000 Professional.
It came completely built with Windows 2000 pre-installed (as it would be if I had got it from a store rather than Intel). Telling it that it was now named Forrie and was part of the Chaos Manor domain took about half a minute and a reboot. Reboot is FAST.
It also has an LS-120 "superfloppy." I never had one of those before. It's the D: drive, and looks like removable media, but when I put a normal floppy in it reads it just fine. Fortunately, not much software has the installation drive hardwired as A:, so this shouldn't be a problem. I don't have any LS-120 media on hand at the moment, so I'll have to report on that next month after I make a trip to Fry's. Meanwhile, it works OK as a floppy, and in fact the whole system works just fine.
It plays Quake something wonderful. I'm seriously tempted to get Crimson Skies, which Roland says I'll like; it's a little like the first Wing Commander. But whatever simulation game you put in there, this thing will scream, as expected.
They tell me the Pentium 4 is not as fast for ordinary software as a Pentium III of the same speed would be. I wouldn't know. It seems fast enough for me. However, where the Pentium 4 shines is in processing video images in software optimized for the SSE2 instruction set introduced with the Pentium 4. We saw demonstrations of this at Comdex, and the result is awesome. The Pentium 4 won't come into its own for several months until there is more development software making use of its features; but when it does, this is going to be one powerful system, and we will begin to see a new revolution in audiovisual software.
Today, the electronics for audio studios are still expensive enough that it takes real money to build one. This means that start-up performer groups have to make deals with major recording studios and publishers, and get skinned alive on their contracts. Whereas book authors own all rights not given to the publisher, with recording artists it's the other way around, and performers often don't see any royalties from their records.
That's all about to change. New computer systems like the Intel Pentium 4 have the capacity to let you simulate a 200-channel audio mixer in software. Hard-disk prices have fallen through the floor, so there's plenty of room for recordings on hard drives. Sound Forge, the best of the audio-editing software, is cheap in comparison to renting a sound studio or making a deal with a recording company.
Good microphones and other physical equipment are also cheap in comparison to what studios charge, while sound baffle material is sold all over for practically nothing. It will take some serious fooling around with foam wall cones and baffles to make a garage have seriously good acoustics, but it can be done. My late, mad friend MacLean did it 20 years ago with a lot worse materials. For a discussion of other equipment now available see Alex and David's MediaLab columns. The point is that good equipment is out there at prices you can afford.
In other words, it's now possible to put together a studio capable of making a professional-quality audio master for well under $50,000, a figure within reach of many start-up groups if several of them pool their resources.
"But digital recordings are harsh, and sound wrong," a semi-pro country and western singer told me the other day. She was dead serious, but I think she's wrong. The charge of harshness and distortion may have been true enough in the past when sampling rates were low. It may even be true today, although I can't tell the difference between two recordings, one digital and one analog, of a recent concert my wife sang in. That could be just me -- I have been partially deaf since 1950. On the other hand, at least twice I have noticed musically trained people remarking at how good a digital recording was, and how it "sounds so much better than those horrible digital records." They were astonished when told they weren't listening to an analog record.
Digital Audio standards today -- what's on the digitally recorded CD you buy at Tower Records -- is 44.1 kilohertz sampling rate of 16-bit samples. The new DVD Audio standard is 192-kilohertz sample rate of 24-bit samples. That's a LOT more information, and I question anyone's ability to distinguish DVD Audio from the very best analog recordings.
Look, information is information, and a high enough digital-sampling rate can emulate analog recording to an exactness far beyond human ability to detect differences. People just don't hear sounds above 30,000 cycles. The physiology books say 20,000, but music experts say they can tell the difference in recordings that reproduce high-frequency overtones and those that don't. No matter, 30,000 Hertz is small potatoes for modern computers that run at a full gigahertz, and 100 channels of DVD Audio won't come anywhere near using up the capacity of a 1.5-GHz Pentium 4.
I am quite certain that new systems like the Pentium 4 are the start of a new revolution in the performing arts, bringing the capability to do professional-quality recordings into the hands of nearly anyone who seriously wants to do them. Once that gets established for audio, video will be next. And of course once you have the performance properly recorded, publishing is pretty simple. For good or ill, the stranglehold the big studios have on recordings and movies is going to come to an end.
Intellectual-Property Problems
The revolution in capabilities is going to force some real changes in intellectual-property rights.
So far, there has been surprisingly little effect on the fortunes of artists and writers. That won't last. Microsoft Reader puts up easily read displays, and new small computers -- larger than Palm Pilot but smaller than a clip board -- with bright new screens and the capability of holding a hundred books and downloading more right off the airwaves are already appearing on the scene. Books are nice and we all like paper, but realistically the age of the pocket computer we described in The Mote In God's Eye 30 years ago is nearly at hand. Most people underestimate the time before a major change begins to take off, and greatly underestimate the time that it takes for that change to consolidate and replace what came before. Who, in 1980, would have believed that the next generation would see typewriters only in museums? Or that vinyl records would vanish in favor of CDs? (In 1982, the first commercially produced CD (in Japan) was Billy Joel's 52nd Street. In 1984 Bruce Springsteen's "Born in the USA" was the first CD produced in the United States. Within a couple of years, there were essentially no mainstream vinyl albums being made.)
New technology will give artists far more control over their own works. For some that will be good, others will not realize they needed editors far more badly than they thought. Residual rights will be worth considerably less, and writers will have to insist on more money up front. Revenue models will change. It's all up in the air now, and we are still on the rapidly rising part of the curve. It's going to be one heck of a ride. |
