By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Gaming - Exposed: Natal Milo smoke and mirrors by Peter Molyneux - Gamer Blorge

 

Project Natal’s Milo was the star of Microsoft’s E3 press conference this year. The Milo demo shown via a video by Peter Molyneux surprised the audience and many gamers around the globe by the promise of what the technology could offer. Molyneux did bring a select number of journalists to demo Milo and it was found to be mostly smoke and mirrors.

I commented the other day that I was extremely impressed with the Milo project in comparison to the other basic motion detection demos the old vision cam was capable of. However, I also commented that I wasn’t sold on the voice recognition technology as my expensive Lexus Navigation system can’t even recognize what I am telling it to find half the time.

Project Natel’s Milo is what sparked the imagination of many around the globe and those at attendance at the E3 conference. The video shown by Peter Molyneux of the Milo demo was presented to indicate that the Natal project was capable of advanced AI and voice recognition. However, attendees who live demoed Milo behind curtain doors were not as impressed after finding out that there was some smoke and mirrors involved.

Is Pinocchio a puppet or a real boy? 

According to Spike’s VP of Content and editor-in-chief of Gamestrailers, Shane Satterfield indicated recently along with his peers that there was a bit of deception with Milo behind the scenes after demoing it. He stated that the select few to demo Milo live behind closed doors were instructed by Microsoft to follow exactly what was shown in the video. Satterfield also stated that there definitely is someone “controlling” (influencing) Milo behind the scenes.

He states:

All I’m going to say about Milo is that its not legit. there were a lot of smoke and mirrors going on there, not the case with the ball demo [Ricochet] and the other stuff but with Milo there was definitely a little bit of deception going on.

Satterfield also revealed that the assumed voice recognition shown in the demonstration video for Milo was a little misleading. He indicates that Peter Molyneux told him that Milo does not actually recognize the player’s voice but reacts to the tone of your voice. He also further explains that Milo says rudimentary open ended statements that will fit into whatever you say.

Apparently, Natel is not capable of voice recognition currently but in theory only. However, this does go back to what I was stating originally - is Natal’s technology or application groundbreaking?

I am not saying that the new Natal cam is not better than the old Vision cam as the depth and motion detection is superior as well as a dedicated mic that is more sensitive to picking up sounds despite distance.  

I believe the idea and its application is what could be revolutionary. The technology itself is not all that impressive when you consider the fact cams like the Xbox Live Vision Cam or the PS3 EyeToy is capable of motion, depth and sound recognition through the built in microphone. The software just needs to be developed to match the appropriate application.

I do want to say that I really do respect Peter Molyneux as I think that he is a visionary but at times is restricted to the technology we have available today. However, I do believe people like him will eventually help propel the gaming industry to new planes. It will be interesting to see what kind of progress Moleyneux makes with the Milo project as time goes by.



Around the Network

that picture is creepy.



                                                                                                  

What is interesting from all this, tho is M$'s never ending jeliousy of SONY. To constantly try to out-do SONY, but is always only just falling behind. After E3, Eyepet looks to be the real ground breaker in this field.



Playing and finishing games first>>>>>>>>>>Then talking!

Opinions are subjective and just like moods, can change.

TOP 12: Deus Ex, Shadow Man, Castlevania: Symphony of the Night, Shin Megami Tensei: Nocturn, Castlevania: Lords of Shadow, Metroid Prime, Zelda (series), Uncharted (series), FF Tactics, Persona (series), Demons Souls, Vagrant Story.

MOST WANTED: Deus Ex: Human Revolution, The Last Guardian, ICO/Shadow OTC HD

Why are we posting a thread that just got locked? If you're the same guy I'll be worried about a ban...



Well , microsoft have a bad reputation of doing the "Smoke and mirrors" lie

i am thinking this game is going to be at least as good as this.

http://www.youtube.com/watch?v=2Y_Jp6PxsSQ



Around the Network

Why did the other get locked?

Also... Milo is a fake confirmed.



                            

Didn't MikeB post this thread already?







VGChartz♥♥♥♥♥FOREVER

Xbone... the new "N" word   Apparently I troll MS now | Evidence | Evidence
pbroy said:
Didn't MikeB post this thread already?

Yep, and it was completely destroyed with logic.

Here is an interview that libellule found:

french ITW of Peter Molyneux (translated) :

http://translate.google.com/translate?prev=hp&hl=fr&js=n&u=http%3A%2F%2Fwww.gamekult.com%2Farticles%2FA0000077061%2F&sl=fr&tl=en&history_state0=

With regard to artificial intelligence, I suspect that Milo does not really all that said, right?

Well, the honest answer - and you know I love being boastful enough to tell you that we have resolved one of the most complex of all time understanding exactly the human voice, but of course it is not the case, otherwise we'd all be billionaires riding in Rolls-Royce in gold - is that Milo does not include the entirety of what you say, it gives you the impression, the illusion, he understands. The way it works - and it is really good that you explain it to all those who continue to believe a hoax - is that we work with a company that specializes in speech recognition called Tellme Networks (acquired by Microsoft in March 2007 editor's note) to build databases of words Milo really understands. The value of these databases is that you can talk to Milo speech without using a split-type "Come, stop, left, right" but rather normal sentences to be analyzed so that Milo can recognize one or more words before to use them. This is the first thing important. The second is that over time Milo makes a profile of the user to assimilate emotions, timbre, tone and nature of her voice to give us clues about the topic of discussion. The third important point is that the controlling party in the conversation you have with Milo: if something happens to Kate - Kate is the dog Milo - just before you begin talking, then it can be assumed without too much risk that the discussion will look at the dog. All those things mixed together with a very complicated thing called neuro-linguistic programming persuade the player that he can really talk to Milo.

 

And here is the link to the locked thread.

http://www.vgchartz.com/forum/thread.php?id=76080&page=1



pbroy said:
Didn't MikeB post this thread already?

yep, and it got locked.  However, maybe if somebody other than mikeB posts it the thread won't be locked?



Why did it get locked? Did MikeB request it to be locked after he got the last word in?







VGChartz♥♥♥♥♥FOREVER

Xbone... the new "N" word   Apparently I troll MS now | Evidence | Evidence