http://www.engadget.com/2006/09/05/wireless-hdmi-over-uwb-heck-yes/

from engadget.....

"Some buzzwords just sound good together. Like robots and flamethrowers -- those two just hit it off right from the start. Now Tzero Technologies and Analog Devices are teaming up for a new standards-based wireless HDMI tech that marries UWB with everybody's favorite home theater plug. Tzero is bringing the UWB to the table, while Analog Devices' is sharing their JPEG2000 video codec. A UWB transmitter compresses all the HDMI data on the way out of a video device and a receiver converts it back to HDMI on the way back into your display. With a range of 30 meters, and no line of sight requirements, all sorts of diabolical home theater configurations become possible. For instance, we can envision a ceiling mounted projector without all the wiring, or a noisy HTPC hidden away without similar hassles. According to Tzero, the first adapters should be out in a couple of weeks, and the price should be "similar to other WiFi devices." We'll be sure to test out the performance before we jump in with two feet, but if the price is right we're liking where this wireless HDMI thing is headed."

from cnet...

"Tzero Technologies and Analog Devices have teamed up for a new way to watch HD content at home. The companies announced on Tuesday that they have combined their signature technologies into the first standards-based wireless High-Definition Multimedia Interface (HDMI) that will eliminate the need for connector cables between HD media devices.

Tzero, a leader in ultrawideband (UWB) technology, and chipmaker Analog Devices seem to have recognized the demand for high-definition TVs, DVD players, set-top boxes and game consoles and the inevitable tangle of wires that comes with them.

Utilizing Tzero's UWB chipset and Analog Devices' JPEG2000 video codec, a transmitter plugged into a DVD player, game system, DVR or media PC compresses video data using JPEG 2000, combines it with audio, packetizes and encrypts it and sends it via the Tzero chips. The receiver attached to the HDTV decompresses the audio and video before it reaches the display. Tzero said it will announce the manufacturer that will produce the first adapters for purchase in approximately two weeks. The cost will be "similar to other Wi-Fi devices."

The most enticing feature of wireless HDMI is the power to eliminate all wires, except for the power source. Even better, the companies' wireless interface doesn't need a line of sight between any of the devices. That means you can squirrel away your media cabinet to one room and stream a TiVo, satellite, DVD player, Xbox and video iPod to three other TVs in the house up to 10 meters away. Tzero promises that Analog Devices' JPEG2000 codec has a high level of resilience, only dropping the wireless connection every two to three hours. That would theoretically enable you to get through a whole movie without any interruption at all. And, since it's standards-based, it will play nice with other WiMedia devices and is immune to interference from other wireless networks or household appliances, the companies said.

By the time the Consumer Electronics Show rolls around in January, Tzero says we can expect a top-tier TV manufacturer to announce an HDTV with this chipset integrated into the display."

from the Tzero website...

"Gefen, Inc, a top provider of consumer and professional audio/video connectivity solutions, plans to base future products off of the Tzero/ADI design. "We have investigated dozens of alternative solutions over the past years, and finally have found one that meets our stringent requirements for wireless HD video distribution," said Hagai Gefen, president and CEO of Gefen, Inc. "We believe wireless HDMI will revolutionize the way people connect their high definition displays, and are fully prepared to harness this technology to enable a reliable extension of high definition video without cables."

This is great. I only put cabling in the back of the room for the projector. Now I can have a plasma in the front for ordinary viewing. Thanks oh Flawless One.

OK,

This is pretty big news, and for once wireless connectivity in the HT world makes sense for the following reasons:

1. JPEG2000 is a recognized standard in the movie industry mainly on the distribution side of D-cinema. It is a very configurable standard and can achieve "visually lossless" picture quality. Most of the "golden eyes" in Hollywood agree it delivers a quality image at a reasonable bitrate.

2. JPEG2000's algoritm is on a per frame basis. The announcement above alludes to the fact that you could lose a frame here and there. But with JPEG2000, you would only lose one frame not a full second (or so) of video like you would with other standards. A missed frame will mean that the display will have to repeat the previous one. You won't have lingering macro blocks or huge skips as in a corrupted MPEG2/MPEG4/VC-1/H-264 stream.

3. JPEG2000 is easy to encode and decode. This means the latency will be minimal at most and you are going to have a 1/60th of a second frame of delay at the encode and decode of the chain at most. This is critical for lip sync purposes (and gaming if you are into that).

4. UWB can handle the bitrate necessary to pull this off. Since JPEG2000 doesn't do any temporal compression, it's only weakness is it's limited compression ratio. "Flawless" JPEG2000 requires no more than 10:1 compression. Is that going to be an issue?

According to the UWB forum: "current candidate proposals are based on UWB, that will provide in excess of 110 Mbps at a 10m distance and 480 Mbps at 2m." Well, for your typical 4:2:0 24 fps 1080p HD-DVD/Blu-ray movie we are looking at an encode rate of about 700 Mbps raw which becomes 70 Mbps under JPEG2000; UWB will have no problem with that. Now let's take a look at the extreme case. For 4:4:4 RGB 1080p at 60Hz the raw bitrate becomes about 3Gbps. So JPEG2000 is going to need a 300Mbps bandwidth from UWB to pull it off. According to the above, your source and receiver would need to be within about 3 meters or 10 feet from each other. In this extreme case, things might be a little dicey, but still doable.

5. UWB isn't another 2.4 GHz standard. It will be done at different range of frequencies. So you don't have worry about someone in the house stealing your bandwidth everytime they get on your wireless intranet.

6. It's Analog Devices. Being an electrical engineer, I can think of no other chip manufacturer that delivers on their claims like AD. I have never had a serious issue with any of their chips. Every other chip company I used has given me very mild to massive headaches from time to time. I'm confident that AD's confident that this will work. They were the first to do chip level JPEG2000 compression at reasonable power.


We have heard numerous wireless projector ideas come and go. This is the only one I really like. We finally have the bandwidth and a robust codec needed to pull this off. An UWB JPEG2000 wireless link will be able to do a lot more than Powerpoint presentations in the companies boardroom.

-Mr. Wigggles

here's an online demo of jpeg2000 w/ compression performance:

http://www.aware.com/products/compression/j2kmaindemo.html

and...

"UWB isn't another 2.4 GHz standard. It will be done at different range of frequencies. So you don't have worry about someone in the house stealing your bandwidth everytime they get on your wireless intranet."

this is the most important point, imo. most of the other devices use a 2.4GHz solution.

...

2. JPEG2000's algoritm is on a per frame basis. The announcement above alludes to the fact that you could lose a frame here and there. But with JPEG2000, you would only lose one frame not a full second (or so) of video like you would with other standards. A missed frame will mean that the display will have to repeat the previous one. You won't have lingering macro blocks or huge skips as in a corrupted MPEG2/MPEG4/VC-1/H-264 stream.

Does it treat each field of a 1080i signal as a frame? You did say "per frame" not "per field" basis.

3. JPEG2000 is easy to encode and decode. This means the latency will be minimal at most and you are going to have a 1/60th of a second frame of delay at the encode and decode of the chain at most. This is critical for lip sync purposes (and gaming if you are into that).

Does easy == inexpensive too? The two aren't always related.

According to the UWB forum: "current candidate proposals are based on UWB, that will provide in excess of 110 Mbps at a 10m distance and 480 Mbps at 2m."

That's not much. Sounds almost like Wireless N speeds away from the 2.4 GHz band.

I'm probably not the only one who made this mistake. This is straight line, not "in the wall" wire distance. So 10m (about 33 feet) is a fair distance.

Well, for your typical 4:2:0 24 fps 1080p HD-DVD/Blu-ray movie we are looking at an encode rate of about 700 Mbps raw which becomes 70 Mbps under JPEG2000; UWB will have no problem with that.

Out of curiosity, which HD-DVD player is outputting at 24fps right now ;) I don't even think the Samsung outputs 1080p24.

So now we have to think about a future player outputting /p24.

Now let's take a look at the extreme case. For 4:4:4 RGB 1080p at 60Hz the raw bitrate becomes about 3Gbps. So JPEG2000 is going to need a 300Mbps bandwidth from UWB to pull it off. According to the above, your source and receiver would need to be within about 3 meters or 10 feet from each other. In this extreme case, things might be a little dicey, but still doable.

10 feet is pushing it for larger FPTV environments. But you could use a cable on one side of a wall in the EQ closet and a shorter HDMI cable run to minimize that wireless hop to accomplish the job.

I'm assuming the device for the PJ end pulls power from the HDMI interface? If not, here comes another use for PoE ;)


5. UWB isn't another 2.4 GHz standard. It will be done at different range of frequencies. So you don't have worry about someone in the house stealing your bandwidth everytime they get on your wireless intranet.

Don't forget interference from analog and digital 2.4GHz wireless phones and Microwave ovens.

6. It's Analog Devices. Being an electrical engineer, I can think of no other chip manufacturer that delivers on their claims like AD. I have never had a serious issue with any of their chips. Every other chip company I used has given me very mild to massive headaches from time to time. I'm confident that AD's confident that this will work. They were the first to do chip level JPEG2000 compression at reasonable power.

Power as in power consumption or compute? I think you mean power consumption.

We have heard numerous wireless projector ideas come and go. This is the only one I really like. We finally have the bandwidth and a robust codec needed to pull this off. An UWB JPEG2000 wireless link will be able to do a lot more than Powerpoint presentations in the companies boardroom.

I liked the JVC solution with that "Frickin' laser beam" ;)

http://arstechnica.com/news.ars/post/20060906-7681.html





Wireless HDMI solution eliminates cords, at a cost

9/6/2006 4:38:03 PM, by Jacqui Cheng

Ultrawideband chipmaker TZero has partnered with Analog Devices to announce an HDMI wireless technology that would allow high-definition video and audio to be transmitted between devices at up to 480 Mbps without the use of cords. With the use of a transmitter and receiver, the user would be able to transmit high-definition content from any HDMI-enabled device, such as a DVD player or a game system, to the receiver attached to the HDMI display from up to 30 meters away. No line-of-sight is required, which would allow much more flexibility in placement of devices, the ability to transmit to multiple HDMI sets around the house from one device, and the elimination of the greatly-feared cord gnome.

Analog Devices is contributing its JPEG2000 compression to compress the stream, which will then be transmitted via RF from a TZero chip to the HDMI receiver on the other end. At 480 Mbps (60MBps), there's just enough bandwidth to watch uncompressed 720p in an ideal "no packet loss, no interference" wireless networking environment. Reality suggests that once traffic conditions are accounted for, 720p video and higher will effectively be compressed, with 1080i requiring more than a 50% compression level to stream. As such, the convenience of wireless HDMI is made less tasty by the loss of what HDMI is supposed to bring, namely pristine HD signals.

Aside from HDMI content, though, the transmitters can take input from component or composite video and analog audio as well. TZero and Analog Devices claim that this new wireless HDMI solution "meets the world’s leading consumer electronics manufacturers’ requirements for link reliability and packet error rate" and is guaranteed to work with all WiMedia devices as well as full-support for HDCP. They even claim that their solution is "immune" to traditional types of wireless interference, such as interference from other wireless networks or appliances in the house.

What would be even more useful in the future would be if HD systems and other HDMI devices would come prebuilt with HDMI wireless technology so that we could go a step further and eliminate the need for external transmitters and receivers. There is no pricing information available from TZero or Analog Devices yet, although a showing of the devices is expected at the Consumer Electronics Show next January.

Will the cost and convenience of implementing an entirely wireless HDMI solution in one's household be worth the elimination of cords? For users who don't mind watching a compressed video stream and who want the convenience of being able to broadcast HDMI content throughout the house, it could be an ideal solution, but for hardcore high-def geeks, HDMI wireless still has a long way to go before providing the kind of video quality they've come to expect.

Three attempts to reach TZero by phone to inquire about operating frequency of the devices were fruitless.

Alright John,

I really don't even know how to respond to your response. I don't know if you are trying to be argumentative or simply trying to nail down the details of my reply.

In other words, you are either being nit-picky for whatever reason or my post is so good that you want me to elaborate on every detail. Only because you are a prestigous WSR reviewer will I will try my best to enlightened on the video world:

(The bold quoted items are your "questions".)

2. JPEG2000's algoritm is on a per frame basis. The announcement above alludes to the fact that you could lose a frame here and there. But with JPEG2000, you would only lose one frame not a full second (or so) of video like you would with other standards. A missed frame will mean that the display will have to repeat the previous one. You won't have lingering macro blocks or huge skips as in a corrupted MPEG2/MPEG4/VC-1/H-264 stream.

Does it treat each field of a 1080i signal as a frame? You did say "per frame" not "per field" basis.
Doesn't really matter. If it is 1080i stuff captured at 60 individual fields per second (sports etc.) you would encode each 1080i field separately so the result would be a series of 1920X540 images at 60Hz.

3. JPEG2000 is easy to encode and decode. This means the latency will be minimal at most and you are going to have a 1/60th of a second frame of delay at the encode and decode of the chain at most. This is critical for lip sync purposes (and gaming if you are into that).

Does easy == inexpensive too? The two aren't always related.
Look, most microchips are cheap. They are basically software. If wireless HDMI becomes really popular, Analog Devices or whoever will charge as much as the market will bear. Will the solution be $10 or $100 per device? I don't know; it could easily be either.

According to the UWB forum: "current candidate proposals are based on UWB, that will provide in excess of 110 Mbps at a 10m distance and 480 Mbps at 2m."

That's not much. Sounds almost like Wireless N speeds away from the 2.4 GHz band.

I'm probably not the only one who made this mistake. This is straight line, not "in the wall" wire distance. So 10m (about 33 feet) is a fair distance.
Well, the 10m at 110 Mbps would be effectively for "low" data rate. That would be enough bandwidth for 24fps movies but probably not enough for 1080p fast action games from your PS3.

Well, for your typical 4:2:0 24 fps 1080p HD-DVD/Blu-ray movie we are looking at an encode rate of about 700 Mbps raw which becomes 70 Mbps under JPEG2000; UWB will have no problem with that.

Out of curiosity, which HD-DVD player is outputting at 24fps right now I don't even think the Samsung outputs 1080p24.

So now we have to think about a future player outputting 24p
Alright, now you are being silly. I would assume that for 24p content there will be some sort of "repeat last frame" flag in the JPEG2000 stream to keep bandwidth low - they could send stuff at 24p or 60p with a lot of "repeat last frame" flags and the required bandwidth would be the same. The point of my comment was to underline the minimum required bandwidth for quality 1080 content using JPEG2000. Film will be the easiest thing to do at approximately 70Mbps and 1080p gaming will likely require twice that amount if not more.

Now let's take a look at the extreme case. For 4:4:4 RGB 1080p at 60Hz the raw bitrate becomes about 3Gbps. So JPEG2000 is going to need a 300Mbps bandwidth from UWB to pull it off. According to the above, your source and receiver would need to be within about 3 meters or 10 feet from each other. In this extreme case, things might be a little dicey, but still doable.

10 feet is pushing it for larger FPTV environments. But you could use a cable on one side of a wall in the EQ closet and a shorter HDMI cable run to minimize that wireless hop to accomplish the job.

I'm assuming the device for the PJ end pulls power from the HDMI interface? If not, here comes another use for PoE
My guess is that the link will automatically adjust. If you can only get 200Mbps reliably, then the datarate will go down to that and some minor video degradation will occur for some sources; I don't think picture quality for movies will be effected for wireless links up to say 25 feet.

5. UWB isn't another 2.4 GHz standard. It will be done at different range of frequencies. So you don't have worry about someone in the house stealing your bandwidth everytime they get on your wireless intranet.

Don't forget interference from analog and digital 2.4GHz wireless phones and Microwave ovens.
Microwave ovens create tremendous interference in the 2.4GHz range but I don't know if they do as much in the higher GHz. Bottomline: pop your popcorn before the movie starts if it is a problem. (I don't think it will be.)

6. It's Analog Devices. Being an electrical engineer, I can think of no other chip manufacturer that delivers on their claims like AD. I have never had a serious issue with any of their chips. Every other chip company I used has given me very mild to massive headaches from time to time. I'm confident that AD's confident that this will work. They were the first to do chip level JPEG2000 compression at reasonable power.

Power as in power consumption or compute? I think you mean power consumption.
Does it really matter? For the record I mean power consumption. Some JPEG2000 solutions out there are quite bulky and power hungry.

We have heard numerous wireless projector ideas come and go. This is the only one I really like. We finally have the bandwidth and a robust codec needed to pull this off. An UWB JPEG2000 wireless link will be able to do a lot more than Powerpoint presentations in the companies boardroom.

I liked the JVC solution with that "Frickin' laser beam"
I think we are finally in agreement.

-Mr. Wigggles

Alright John,

I really don't even know how to respond to your response. I don't know if you are trying to be argumentative or simply trying to nail down the details of my reply.

In other words, you are either being nit-picky for whatever reason or my post is so good that you want me to elaborate on every detail. Only because you are a prestigous WSR reviewer will I will try my best to enlightened on the video world

Well my intent was to get you to expound on a few points where I wanted a little more information. I imagine it might be useful to other people as well. If you want to take this exchange as being argumentative that's certainly your perogative.


Does it treat each field of a 1080i signal as a frame? You did say "per frame" not "per field" basis.
Doesn't really matter. If it is 1080i stuff captured at 60 individual fields per second (sports etc.) you would encode each 1080i field separately so the result would be a series of 1920X540 images at 60Hz.

Would the "half" fields be capable of the same level of compression without artifacts as the full frames? I'm trying to figure out if there's a penalty in interlaced vs progressive input.

Does easy == inexpensive too? The two aren't always related.
Look, most microchips are cheap. They are basically software. If wireless HDMI becomes really popular, Analog Devices or whoever will charge as much as the market will bear. Will the solution be $10 or $100 per device? I don't know; it could easily be either.

I'm trying to determine if it's going to be cost competitive with physical cables. Probably not with the Monoprice ones, but definitely cheaper than the high-end ones. What about versus optical cables? I'm interested to see what the Gefen solution comes in at. I'm somewhat disappointed with the cost of their Cat-5 extender @ $500. Will it be cheaper than that? I hope so, but I don't know yet.

I have the answer in my e-mail this AM. The MSRP for the Gefen product is $500.

That's not much. Sounds almost like Wireless N speeds away from the 2.4 GHz band.

I'm probably not the only one who made this mistake. This is straight line, not "in the wall" wire distance. So 10m (about 33 feet) is a fair distance.
Well, the 10m at 110 Mbps would be effectively for "low" data rate. That would be enough bandwidth for 24fps movies but probably not enough for 1080p fast action games from your PS3.

I'm not a gamer, so I won't own a PS3. It'll have to be someone else's PS3 ;) But the point is no less valid.

My point about the straight line was to remind everyone that you don't have to think about your current cable length and say "this won't work". My current cable length to my projector is 20m. But that includes a lot of length that doesn't need to be counted for a wireless device since it's a straight run -- transmitter to receiver.

Out of curiosity, which HD-DVD player is outputting at 24fps right now I don't even think the Samsung outputs 1080p24.

So now we have to think about a future player outputting 24p

Alright, now you are being silly. I would assume that for 24p content there will be some sort of "repeat last frame" flag in the JPEG2000 stream to keep bandwidth low - they could send stuff at 24p or 60p with a lot of "repeat last frame" flags and the required bandwidth would be the same. The point of my comment was to underline the minimum required bandwidth for quality 1080 content using JPEG2000. Film will be the easiest thing to do at approximately 70Mbps and 1080p gaming will likely require twice that amount if not more.

That's an assumption you're willing to make that I'm not, ie the repeat last frame will actually be out there. It does increase the per unit cost as it now has to do a store and compare in addition to crunching down the input to a JPEG2000 all within the allocated time window. Just because it's logical to do so doesn't mean it's a reality in the product.

10 feet is pushing it for larger FPTV environments. But you could use a cable on one side of a wall in the EQ closet and a shorter HDMI cable run to minimize that wireless hop to accomplish the job.

I'm assuming the device for the PJ end pulls power from the HDMI interface? If not, here comes another use for PoE

My guess is that the link will automatically adjust. If you can only get 200Mbps reliably, then the datarate will go down to that and some minor video degradation will occur for some sources; I don't think picture quality for movies will be effected for wireless links up to say 25 feet.

I still think using a cable to get the transmitter to the highest possible data rate while remaining hidden is the best option. If you're going to the expense of a high-end display, you'll want to keep the data rate as high as possible.


Don't forget interference from analog and digital 2.4GHz wireless phones and Microwave ovens.

Microwave ovens create tremendous interference in the 2.4GHz range but I don't know if they do as much in the higher GHz. Bottomline: pop your popcorn before the movie starts if it is a problem. (I don't think it will be.)

I was just expanding on why staying away from 2.4GHz was good. Two huge sources of "pollution" besides 802.11a/b/g.

Power as in power consumption or compute? I think you mean power consumption.

Does it really matter? For the record I mean power consumption. Some JPEG2000 solutions out there are quite bulky and power hungry.

Yeah, it does matter. You'll want as low a power consumption solution as possible for the receiver end if we're talking about an FPTV environment. Outlets are sometimes scarce up on ceilings ;)

The article I read on it said JPEG2000 had a lossless compression option that would be used. Maybe they meant visually lossless.

No mention of 1.3 spec compliance?

The article I read on it said JPEG2000 had a lossless compression option that would be used. Maybe they meant visually lossless.

JPEG2000 does have a lossless compression option but it is no different than the type of coding you use to make a zip file on your computer. On average lossless will be about 2:1 to maybe 3:1. That is not enough compression for the super high datarate images.

-Mr. Wigggles

On the topic of JPEG2000- Does anyone know if/how JPEG2000 defines the black and white points? i.e., is the notion of reserving codewords for head/foot room defined or is it eliminated as it seems to be for still images and print/press?

Dave

Is there a possibility of timing/lip sync issues?

On the topic of JPEG2000- Does anyone know if/how JPEG2000 defines the black and white points? i.e., is the notion of reserving codewords for head/foot room defined or is it eliminated as it seems to be for still images and print/press?

Dave
If you want to play around with JPEG2000, you can download XnView from downloads.com and it allows you to import/convert/export JPEG2000 images.

JPEG2000 will work with whatever the input format defines as black and white. It is an encoder/decoder. So ideally the value for every pixel on the input will be the value for every pixel on the output. So if you put in a video white image of (235,235,235) into a JPEG2000 encoder, you will get a (235,235,235) image on the final output. The codec doesn't know what white or black is and it doesn't care.

This is sort a discussion for another thread but I personally think using the full 0 to 255 for each color is the way to go. If you use the full 8 bits, you get 16.7 million color total. If you only use the reduced video range you only get 10.6 million colors for RGB and only about 3 million colors for component video YCrCb.

-Mr. Wigggles

Is there a possibility of timing/lip sync issues?
I kind of touched on this earlier.

In a complete solution where the audio and wireless video are handled by the source, the source can delay the audio as it feels appropriate for the video. (A high-end DVD player with wireless HDMI would have this capability.)

For situations where the audio doesn't know what the video is doing (where you have video going wirelessly to the projector but a cable to the receiver for the audio), there will be added delay.

It is not clear at this point what the delay will be, but it has to be less than 1/60th of a second on the encode and the same on the decode side or the chipset would not be able to keep up. That equals a total of 33.33 ms. When this product finally hits the streets, I bet the delay will be a lot less than that.

-Mr. Wigggles

This is sort a discussion for another thread but I personally think using the full 0 to 255 for each color is the way to go.

I agree- hence my question. I was hopeful to see a new en/decoding standard that uses all the available codewords to actually represent valid image information and not drag the antiquated (IMO) need to reserve a good percentage of them for head/toe room for codewords that can’t/don’t represent valid image information.

Seems like it’s not JPEG2000 then. Thanks for the info.

Dave

extremeuwb.com

On Tuesday, Tzero Technologies and Analog Devices announced that they have created a wireless HDMI interface for HDTVs, next-gen DVD players, and set-top boxes.

The companies' first wireless HDMI setup is a standards-based system, predicated on the standards set forth by Panasonic, Philips, Samsung, Sharp, and Sony in July of 2003. Other wireless HDMI technologies introduced to date have been proprietary, both Tzero and Analog Devices alleged.

Wireless HDMI (high-definition multimedia interface) enables wireless video and audio connections between all types of home entertainment products such as HDTVs, DVD players, set-top boxes, and game systems. It allows HDMI content to stream to several devices without a tangle of wires. The backbone for the technology is ultrawideband, also used as a future replacement for wired USB.

The standard calls for link reliability of at least 95 percent, packet error rate of less than 1 in one hundred million, interference resistances for microwaves and cordless phones, and the ability to process three or more HD streams at 10 meters.

"The primary thing that is the most challenging is linked with reliability," said Dan Karr, vice president of marketing for Tzero. "It needs to be good under all conditions. Wi-Fi serves this market very well, but in a video environment you can see that would be very problematic."

The wireless HDMI system consists of a transmitter and receiver. The transmitter is a black box about the size of a laptop. Karr says that it is "bigger than necessary, and that's because it's new, but it will get smaller."

The transmitter integrates the Tzero TZ7000 chipset and Analog Devices' JPEG2000 compression integrated circuits. The Analog Device compresses data with JPEG2000 video code, which is then packetized and encrypted, and transmitted via the Tzero MAC and PHY chip. The chip transmits over the air to the receiver where the audio/video data with HDMI is decompressed and presented to the display device via the HDMI port.

JPEG 2000 is a relatively new image compression standard that was approved by the JPEG committee. It can operate at higher compression ratios without the characteristic blocky and blurry effects of the original JPEG standard.

"We pass the video through the JPEG2000 and then pass the video and audio through the Tzero chip," said Bill Bucklen, director for the Advanced TV Segment with Analog Devices. "You can't compress a content-protected signal so we've built a repeater function and because we encrypt the link, it's like we've taken a repeater and put them on the other side of the room."

Tzero and Analog executives say that wireless HDMI will make for much more aesthetically pleasing HD systems, which, according to them, will make women happier in the selection of home theater systems.

"One of the things we are hearing more and more now is that the disinterested spouse is taking a more active role in selecting and hanging the television, typically that's the wife," Bucklen said. "That's all well and good until you start dragging cables into the solution. HDI cables are expensive and bulky and we think that a wireless approach can give consumers the flexibility to put televisions where they want them."

The Tzero wireless HDMI system will be demonstrated to electronics manufacturers, press and analysts over the next few months. It will be available for purchase in November. So far the companies have not yet set a final price.

"If we break this down, it's going to be less than an HDMI cable," Karr said. "Those are about $100 plus installation. We think that a TV manufacturer can build this into their system, mark it up with their market and an option to buy would be somewhere around $100. For stand-alone boxes, it will depend on how many things you can pull into that box. It would be nominally in the same range as what a consumer electronics company would integrate it for."

any news for this babe?

so far I only see Asus and Philips having this product and not sure when are they launching them officially......

Gefen one is super overpriced!

From a previous thread:

Hold your horses, folks. This product claims "visually lossless HD compression." If it's truly lossless, then they would say so; saying "visually lossless" suggests that the compression is, in fact, lossy---but that we shouldn't see the difference. Well, I'll believe that when I see it. See this FAQ. (http://www.jpeg2000info.com/what/comp101.html)

The literature claims a data rate of up to 480Mpbs. 1080p30, which is the highest-bandwidth signal they claim to carry, is 1.49Mbps or so. That means they have to achieve at least a factor of 3 compression in order to fit the video through the pipe. (EDIT: I say "at least" because the data rate is up to 480Mbps; your mileage may vary.) Honestly, that's probably doable...

...but they seem to have explicitly left out 1080p60...

I'm not ready to give up my cable yet.

Yeah, to be honest, maybe for source material that is already compressed, a 3:1 JPEG2000 compression isn't going to matter much. The proof will be in the viewing.

But you still can't feed it 1080p60, which means that your projector is likely going to have to do some deinterlacing (for, say, true 1080i60 content).

If the processor in the projector isn't up to snuff, then maybe you can put an outboard HDMI deinterlacer/scaler next to the projector and stream wireless HDMI into that instead. At least then you won't have to worry about trying to pass 1080p60 through the air; you can keep it all at its source resolution.

I saw a Taiwanese ODM displaying this in Computex.

But he told me the current TZero chipset is maxing out at 1080i60.

No way to do 1080p60. But it does seem that the product has still yet to be launched. He promised to get in touch with me when it did. Hopefully it can do 1080p60 as well as 1080p24 by then.

Dealer pricing was expected to be US$600 ish (for a pair), so expect mark ups from there.

I also saw a bunch of HDMI repeaters that worked using Cat5E cables which may be another alternative for some of us to consider. The neat thing is that some of the cable runs I saw were very very long. The downside is that they couldn't confirm if they could handle HDCP over those repeaters.

There is some insider’s information that Tzero radio’s transmit power required to meet FCC regulatory compliance doesn’t not provide sufficient bandwidth to support HD applications. Achievable usable throughput is a par with 802.11g in the best case when FCC requirements are met. Working distance is much less than 802.11g one's.

Analog Devices started working on designs that will support multiple transports, including 802.11n, UWB, Powerline, and other transports. However it may take another few months to complete at least couple of new reference designs.

That would explain why Gefen's product still doesn't seem to be shipping. At the Gefen booth at the last Infocomm they didn't even have it on display. I looked at their site again a few days ago and it seems to have been re-designed compared with earilier photos and specs. New antenna ( with 'sample' clearly stamped on them ). Perhaps they're trying to combat a reduced power enviornment with a higher gain antenna pair and improved RF front end?

I guess we'll see.

Jonathan

How's does this differ from the Avocent (http://www.connectivity.avocent.com/products/emerge/mpx1000.asp) wireless transceivers?

How's does this differ from the Avocent (http://www.connectivity.avocent.com/products/emerge/mpx1000.asp) wireless transceivers?

Efficient bandwidth of Tzero radio in demo design is around 80 Mbit/sec. This allows to get visually lossless transmission: typical PSNR for HD-DVD/Bluray source is greater or equal 40dB, for game source even higher, for text screen slightly less.

Avocent uses radio with efficient bandwidth around 25 Mbit/sec. Of course, heavier compression is required, which results less PSNR. PQ loss is somewhat noticeable on movies and computer games. PQ loss on texts is pretty visible.