One significant but not very well publicized casualty in the troubled Middle East is the Israeli high-tech industry, which is suffering probably its greatest capital crisis ever. Israel, like Silicon Valley, relies on a continual stream of startup companies, none of which are profitable, and so all of which require outside funding. But such funding dried-up a couple years ago when suicide bombers started terrorizing Israel, scaring away venture capitalists in the process. It might get even worse if there is a war in Iraq — that is, if it is even possible for the situation to get worse. So a probably unintended result of all the bombing is that many existing Israeli startups have folded for lack of funding, and very few new startups have come along, which will have a significant impact in future years. This is not good for either Israel or high-tech in general, and will probably have a negative impact on both for the next decade no matter what happens in Iraq. This is not good for anyone, anywhere, so I hope that my writing about tech opportunities in Israel just might help attract investors back to the country before permanent damage is done.

Since I haven't visited Israel in years, I have to rely on my friends there to tell me what is going on. One of these friends is June Dilevsky, who does a lot of things, including writing business plans for startups like Mirabilis, the ICQ company. Long before she moved to Israel, June came from a high-tech household in San Diego, where her father invented the modern fax machine, eventually selling that technology to Ricoh. Imagine what that must have been like growing-up: "My Dad invented the fax machine." Yeah, right.

I also hear about opportunities in Israel from Avram Miller, my neighbor here in the Wine Country. Avram, who is retired from running Intel Capital and is returning to his roots as a jazz musician, used to live in Israel. One could say, I suppose, that Avram has done just about everything in high-tech, since his name pops up at many points in the history of personal computers. Avram ran Franklin Computer when that Philadelphia company was making clones of the Apple II. Later, when he worked at DEC, Avram helped CEO Ken Olsen disassemble one of the first IBM PCs on Olsen's desk. Avram recalls that Olsen was appalled at the poor design, which probably set the stage for DEC's own disastrous entry into the PC business with the ill-fated Rainbow 100.

It was Avram who first told me about the subject of this column, an Israeli startup called Enbaya. The word "enbaya" means "no problem" in Hebrew, and the company has evidently had no problem at all making a significant advance in 3-D graphics and animation software. Now if only they had some money to turn that product into a business.

Enbaya's product is called ProGATE and it is, in essence, a 3-D graphics and animation compressor for video games. ProGATE can compress 3D graphics and animation by an average of 50-to-1. The technology is finished and working on PCs, Sony's PlayStation2, Microsoft's xBox, and Nintendo's GameCube. A total of five patents have been applied for.

But it is probably a mistake to call ProGATE a "compressor." It is really just a more efficient way of describing graphics and animation, much like Adobe's PostScript is a more efficient way of describing 2-D graphics. Nobody calls PostScript a graphics compressor, yet that's what it is. Then think of Enbaya's ProGATE as 3-D PostScript.

What's truly great about this technology is that it can be applied to completed games. Developers can write their games using any tools they like, and then this compression technology is applied after the game is finished, even years after it is finished. The result is a significantly smaller version of the game that loads faster, even runs a bit faster, yet is indistinguishable from the original game. After using ProGATE, games are dramatically smaller. This means that developers can fit more scenery and characters on a CD or DVD, and that networked games can load many times faster.

I see the main opportunity for ProGATE in networked games. The opportunity is clear. It is a chance to set a de facto standard for online gaming.

The ProGATE architecture is built on two major components — an off-line encoder and a real-time decoder. It all starts with a completed game. The encoder analyzes the game, breaking it into graphic elements. After analyzing each element and figuring out what is the best new mathematical representation for it, ProGATE produces a tiny file that has a low memory footprint while loading and during runtime. The decoder then works in memory with the new tiny bit stream and decodes, in runtime, whatever the application needs from it. A typical decoding process may take 10 to 50 microseconds, which means the decoding overhead is very low.

Note that this is quite a different technique from traditional compression and decompression techniques — especially after decompressing. Though the output APPEARS identical to the input, it is radically different. So rather than "compression," maybe the correct term for ProGATE ought to be "re-expression."

The ProGATE decoder libraries and API are seamlessly integrated as a layer into the game engine. Currently versions of ProGATE are available for the very popular RenderWareby Criterion Software and also for Alchemy by Intrinsic. A native real-time library and API is also available for proprietary game engines. This is how ProGATE could easily handle the PC games and other games not covered by the two tools above.

ProGATE's effective compression ratio for animation and graphics is 50 to 70:1, while the effective compression of the entire game is on the order of 20:1, since not all game code is graphics or animation. These numbers will vary from game to game, but are typical.

Think about it. This means 20 games could be put on a single CD or DVD, rather than one. It would be possible, I suppose, for a game company to save money by putting their entire game library on a single disk, then selling keys to unlock individual games as they are purchased.

Because the re-expressed animation and graphics are generally rendered faster than the original code, the game runs faster by an average of 10 percent compared to its original version. To gamers, 10 percent is a perceptible improvement comparable to a memory or processor upgrade or buying a new video card. This is a big deal.

So games using this process require five percent of the bandwidth, five percent of the storage, load 20 times faster and run 10 percent faster. The games can be for consoles or PCs — anything with an Internet connection — and can be run over a broadband or dial-up link. The result is that dial-up gamers can have performance equal to or better than current broadband users and broadband users, while their gaming experience will be even better.

From a systems standpoint, the same server farm ought to be able to serve 20 times as many users with no increase in bandwidth.

This technology is finished and ready to go. It is defensible. And whoever owns it will have a decided advantage in online gaming. Any takers?

I couldn't end this column without mentioning Fred Rogers, who died this week. Mr. Rogers is known to every child and adult in America and everyone at PBS was touched by him in some way. You would think that I'd be the least likely person at PBS to know Fred Rogers, but we were friends. Fred learned to fly a Piper Cub in 1946, and I once flew him in my plane to his summer home on Nantucket Island. Fred Rogers was not only the nicest man I ever met, he was the nicest man ANYONE ever met. What you saw is what you got. Fred Rogers was a very positive moral influence at PBS, constantly reminding us by his example why we do what we do, which is to make life better for our children.

Let's not forget that.