En el Reddit de Magic Leap
hay un par de hilos con comentarios de otras gentes que lo han probado.
Como solución gráfica parece que tiene buena pinta ya mismo y de cara a mejorar en el futuro. Aunque solo han visto una breve demo y es pronto para decir cuales son sus puntos fuertes y sus limitaciones. Por ejemplo, el FOV parece que es superior al Hololens pero inferior a Meta2, quizás sobre los 60-70 grados, quizás menos.
FoV was better than HoloLens and ODG, less than Meta. It was honestly really hard to ask questions about the display because things just reacted the way your brain would expect to in terms of focusing which was a really bizarre experience but very pleasant. Not perfect yet and not super refined as it's still a prototype, comparisons to any other headset are kind of unfair as it's unclear how expensive this rig was so it could've been research-grade. I work for TechCrunch so it was a press demo, waiting to detail full experiences in a video we shot during the demo which we should be pubbing next week.
ianott 1 punto hace 10 horas
Great write up. Can you comment on what you felt the approximate FOV was on the demo unit?
[–]kguttagKarl Guttag, kguttag.com 2 puntos hace 9 horas
I wish I could but I was blindsided in getting to see the demo. Also the content I saw was not conducive to telling the FOV (very sparse pixel content). As I wrote I was expecting something totally different. I spent the time I had trying to both see the demo and look for tell-tall artifacts as to how it worked.
It is a 720p device which means they only have 1280 pixels horizontally. That is not a lot to support a wide FOV. At 30 degree FOV, you have about 1.4 arminutes/pixel. At 50 degrees you are at 2.3 a-m/pixel which is starting to get chunky. An advantage of the "birdbath" optics they are using is that it will support a wide FOV (much easier than waveguides). In the end the FOV is more limited by eye relief (the distance from the eye to the displays) which is particularly important if you want to allow people to wear their own glasses.
They should be able to switch to a 1080p device and get about a 50 degree FOV with about a 1.56 a-m/pixel resolution which will be good enough for most people if not "ideal" (ideal is considered 1 a-m/pixel or better).
Como es el reddit de Magic Leap se comenta que esto va a meter más prisas a la gente de Florida para que muestren algo palpable pronto.
También comentan que los gráficos son muy importantes, pero solo una parte del complejo puzle que supone crear un dispositivo y un ecosistema de realidad aumentada comercial. Lo lógico es esperar que los compre alguna empresa muuuy grande que ya tenga esa otras piezas del puzle encaminadas.