Intel Launches Wi-Di 172
Barence writes "Intel has launched a new display technology called Wi-Di at CES. Intel Wireless Display uses Wi-Fi to wirelessly transmit video from PCs running Intel's latest generation of Core processors to HD television sets. Televisions will require a special adapter made by companies such as Netgear — which will cost around $100 — to receive the wireless video signals. Intel also revealed its optical interconnect technology, Light Peak, will be in PCs 'in about a year.'"
will the cable/ satellite industry fight this? (Score:5, Interesting)
if you can broadcast a signal to every set in your house, or even your entire apartment floor, then there goes a bunch of lucrative descrambler box fees. then again, they can all only show one channel at a time. however, media companies seem to all be losing income nowadays, and have all taken a hostile attitude towards new technology. they seem to need very little reason, however slim and irrational, to pick a fight with new technology
of course, the future is all streaming media over the internet, mostly on demand and mostly free, so they're all fucked
All this wireless is Class 'C' ? (Score:1, Interesting)
So, if you live next to an Air Force Base, Airport, I don't know - all of your home entertainment gadgets won't work? And if you have a Kindle, you better have a shit load of books stored on the machine, because you won't be able to download any.
'Luddite' may not be a derogatory term in the near future.
Something else I'll probably never need (Score:2, Interesting)
Yet another kind of connection from PC to TV?
Why not just watch on the monitor of the PC, or use a projector?
Televisions will require a special adapter.... (Score:3, Interesting)
Yeah, because we all know how completely difficult it is to connect a DVI to HDMI cable and an 1/8" cable from your computer to your TV.
Of course someone will say, "Most people don't keep their PCs near their TVs."
If people were willing to spend $600 on a PS3 that sits in their living room, I don't see why they can't spend a few hundred for a PC. Heck, if you subtract the $100 "special adapter" from the price of the PC, you can get one real cheap.
Of course someone else will say, "Who wants a noisy PC in their living room?" And to that I'll say, "Have you ever been in the same room with an Xbox 360?" Mine is much more noisy than my PC by a wide margin.
Compared to the 90s, I think retail desktop PCs are pretty quite nowadays. (Of course I built mine myself.)
So many questions... (Score:3, Interesting)
... and none of the articles I've read about 'Wi-Di' seem to answer them.
How about sound? Transmitting video directly to my tv sounds nice, but how does this tech account for transmitting sound to a HT receiver? Potential for audio/video de-sync? How will this be handled?
Potential for latency issues? This could be a big one, especially for gaming.
Re:Why wouldn't... (Score:3, Interesting)
Speaking purely about networked appliances (NOT notebooks or phones), why on earth would you ever need unique IP addresses (in the global sense?).
Well, I can give you one fairly good example. I'm typing this on a laptop, a Macbook Pro. It came with a web server, which I enabled, and I routinely use it to test assorted web stuff locally. However, I can't use it as a "real" web server, because I can't get a fixed IP address for it.
I use it for its major function, a portable computer. Under the current IP regime, this means that when I carry it around, its IP address either doesn't exist, or is constantly changing. This means that even when it's exposed on the Net, not behind a firewall, it still can't be found, because the DNS system has no way of tracking a host with a rapidly changing IP address. Imagine that you had a cell phone whose phone number changed as you drove around, always having a number that belonged to the "local" exchange that you were driving through. Do you think that anyone could call you on such a phone? The same problem exists with portable computers of any sort. They can call you, but you can't call them. Two phones of this nature couldn't call each other at all. Similarly, two moving laptops with browsers can't find each others' web servers with the current IP setup.
Actually, I have a G1 "google" phone, and it has the same problem. Since it's running a linux OS, it could easily support a web server, and could "serve" things like pictures that I've just taken. Software on my home machine could automatically download files from the phone as they're created, using "HTTP GET" or scp or rsync or whatever. But this can't be made to work, because the phone's IP address changes rapidly. I've verified that, even sitting here at home, a web server's log shows successive HTTP requests from the phone as coming from different IP addresses. So, even if I did run a server on it, my home machine (or your smart phone) couldn't get to it, because there's no way you could ask the DNS system for its instantaneous IP address. Even if you could, by the time you did a connect(), it's address could have changed. The same sort of thing happens with my wife's iPhone.
If you don't see how this kills a lot of very useful network apps, you don't have much imagination. Maybe it would help to consider: You and a friend both have laptops. Fire up web servers on both of them. Then try to get a browser on each to connect to the web server on the other. Try this while carrying them around. Do you know a way to make this work? Do you understand why it would be useful?
Until we can give every net-enabled gadget its own IP address, there are a lot of things that simply can't be made to work right. IPv4 was designed with the idea that every "host" would have at least one fixed IP address. The NAT stuff is a huge kludge to get around the fact that this isn't possible (and even when it was possible, it wasn't allowed by the ISPs ;-). I've seen a number of claims that there are already many more than 2^32 IP-enabled gadgets in existence, most of them with no access to the public Internet. As long as this state remains, there are a lot of useful things that those gadgets can't do.
An interesting one that I worked on a few years ago is IP-enabled medical monitors and implants. A major problem that arose is that, in practice, putting even one of these on or in the body of all existing patients would require more IP addresses than were available then. We did talk about a design of a single wireless device per patient with NAT used to hide the other devices. We had no problems finding objections to this. One was that it would make it difficult for software back at the hospital to connect to a specific device on/in the patient; the device would have to connect to the hospital. This could be fixed by the usual "polling" technique, but the polling messages would quickly overload the low-bitrate channels available to wireless devices. Also, it