As if the world didn’t need further evidence of my nerddom, I have gone and purchased video glasses: Vuzix iWear AV230 [site is down as of time of writing], as sold by woot in a bundle with an iPod adapter for $99 (plus $5 shipping). I’ve wanted to try out a pair of these things since they were monochrome and $500+ in yuppie product magazines – at this price point and feature set I’m pretty happy. With the new job I’ll be doing considerably more traveling by plane (and potentially commuting great distances by bus and/or train) – which, combined with video via an iPod touch, makes these ideal for passing the time in fun and educational ways (yes, educational – I’m a big fan of the TED video podcast, the MIT Open Courseware lectures, etc.).

The specs are a little on the light side as far as the current offering of video glasses are concerned – 320 x 240 (old school VGA), equivalent viewing distance of 44″ 4:3 screen at 9′ (which, for my arm length, looks about like an 8½ x 11” piece of paper held at full reach). Nothing super-stellar or all encompassing from the viewer’s perspective, but certainly comfortable for standard definition video. It accepts RCA video (and audio) in, making it versatile and accommodating for a broad range of devices and applications. It will also automatically demux field-sequential stereoscopic signals onto the independent screens, making full 3D a possibility (though poorly supported by available media).

There are Zero controls or configuration options for the main unit electrically speaking (though the individual lenses can be adjust by +2 to -5 diopters via focus wheels in the bottom): it turns on when it detects a video signal and turns off when it doesn’t. No volume choices, brightness, contrast, etc. The speakers (not headphones) can be bent lightly into position above or near the ear in order to change the effective volume, but there are no options in-line. Given that many media devices output a fixed volume on their audio, this may be less-than-ideal in many situations, although what I’ve demoed so far seems to run too loud (easily corrected with position) rather than too soft. It’s also possible to remove the speaker stalks, enabling head-phone use. This is helpful for those situations where whatever device has its own gain handling and separate headphone jack, which for the iPod is perfect and also very necessary for in-flight use with the high ambient noise and the high-gain audio required to overcome it.

The adjustable (and removable) nose piece has large soft rubber pads which make it not-uncomfortable to wear for extended use, but which will leave marks on a person’s bridge. I don’t recommend trying to use these for very long without the lanyard either, which helps to secure it to one’s face and preserve the viewing angle (which otherwise can be tricky for reasons I’ll get into later). It also helps prevent them from slipping – 4oz. doesn’t sound like a lot, but when all of the weight is forward and on an inclined slope, physics does tend to take a hand and they will slip.

The image quality is good, although photographing it turned out to be one of the most technically challenging pictures I’ve ever attempted. The result here is a little blurry, but that’s an artifact of the difficulty of maneuvering the camera into place and getting the focal depth set just right – the comparatively long shutter speed didn’t help either. These were not designed to be photographed, but they do work excellently with the optics of the human eye (other species mileage may vary). Now here’s the tricky part: these are based on high resolution LCDs (320 x 240 may not sound very high, but fitting 320 x 240 x 3 [RGB] addressable LCD cells within a ½” is not a trivial feat). LCDs have optimal viewing angles in 2 axes, which left the designers up to figuring out a balance between determining best average pupillary distance, or best main-unit tilt. The outcome of this decision dictates where the broadest axis will be placed and whether the displays will be addressed from the top or the side and the resulting structure of the interior and lay-out of the circuit. So let’s take a look.

Several competing factors and decisions are illustrated in this image. First, nearly ¾ of the depth of the unit (and who knows how much of the overall weight) is dedicated to the optics in front of the actual LCD (little clear & white box under the yellow tape on the right). If there were ever an argument for high-resolution Fresnel lenses, this would be it (assuming that the etching resolution doesn’t inadvertently turn it into a diffraction grating). They also chose to mount the displays symmetrically, with their input regions both oriented directly toward the main circuit board.

This is actually Bad News. While LCDs don’t have a top or bottom per se, they do have common properties in terms of the breadth and bias of the viewing angles they support. Since they chose to put the widest range in support of the pupillary variation, that means the supported vertical tilt range is dramatically reduced. These are also identical displays – they were not manufactured in mirror image of one another, so essentially what’s been done here is to rotate one of them 180° from the other and use some sleight-of-circuitry to render one display upside-down in order to correct for that rotation. This seems more costly in the development and design to me than simply running the generously long ribbon cable to the far side of one of the displays and using a combined signal wherever possible – I wonder what cost savings they actually realized from of the symmetrical physical assembly they chose instead, if any. The real reason this is a problem is that in the right-eye display, the bias angle to see the best contrast is a degree or two below horizontal, while the left-eye display is a similar offset above the horizon. If one wears the device perfectly level across the face, either A) both displays will be suboptimal and a little washed out in their contrast, or B) one image will appear brighter than the other, creating an uncomfortable viewing experience. From what I’ve read and personally experienced, B seems the more common option.

It is possible to correct for this in 1 of 2 different ways. First, the wearer has the option to arrange the glasses slightly askew – enough to better align the two disagreeing angles but not enough to throw off the stereoscopic reconciliation and produce a double image, or secondly: to tilt just one of the LCDs within the unit until it agrees with its sibling. The latter option is preferable but problematic, in that the displays are affixed to the frame with a mild adhesive. Glues and electronics are a nasty combination even without throwing optics into the mix – I might be able to perform the kind of delicate surgery required to correct this, but not with the few crude tools I brought with me to the apartment (the rest being left behind in Utah pending the sale of the best house ever and subsequent family move). So for now I’m opting to downgrade from “Nerd” to “Complete Tool” when making use of them by wearing them lopsided – and at these power consumption and battery specs, for 4-5 hours at a time.

Totally worth it.

« A Good Geek in a Bad EconomyLessons Learned »

Latest Comments:

1. Hey Paul, I’ve always wanted to try those. Welcom home!