Yeah but there's a difference between voice recognition and good voice recognition. It's never been used much in the past because it doesn't really work all that well. It seems as though MS is investing a lot in it for Durango though so I'm expecting a much more natural experience. Between the high quality microphones in Kinect, relatively massive processing power and dedicated bags of RAM we will hopefully and finally have truly voice controlled "computer" in the living room.
- Status
Yeah but there's a difference between voice recognition and good voice recognition. It's never been used much in the past because it doesn't really work all that well. It seems as though MS is investing a lot in it for Durango though so I'm expecting a much more natural experience. Between the high quality microphones in Kinect, relatively massive processing power and dedicated bags of RAM we will hopefully and finally have truly voice controlled "computer" in the living room.
He probably read about all that stuff here on B3D...
Kinect is more than DragonVoice on a console, kinect2 will probably be even better at isolating voices
Interesting read for those with more understanding,it was some spanish article translate to english.
http://www.microsofttranslator.com/...013/02/08/durango-nos-hace-coger-el-delorean/
http://www.microsofttranslator.com/...013/02/08/durango-nos-hace-coger-el-delorean/
Megatextures/megameshes.
Here is the thing. Everyone is dismissing this because he is not a dev, which is a big mistake. MS WAS planning a 3 soc machine per the Yukon leak. That design plainly described what was essentially a 3 SKU strategy - Set top-box / tablet with APU, Base 720 with APU plus a second, more powerful SOC, and Premium 720 with hardware BC. That was a strategy that made a lot of sense.
However, nothing about current Durango hardware makes sense if you view from a "traditional" box view. MS has been working on this since probably at least 2009, if not 2008. They have hired massive numbers of engineers and spent untold resources including priority projects at AMD designing the thing, and the leaked specs are what we get? I can't think of any corporate strategy scenarios where that makes sense beyond incompetence or overconfidence.
Until recently, I also favored the Yukon strategy because it was simple, effective, and the most likely outcome was utter domination by MS, at least in North America and UK, with vastly expanded penetration. The leaked specs point to a different design/strategy than Yukon winning the internal MS battle. Therefore, the new strategy had to offer advantages over Yukon.
There is also Xbox surface. It appears that Xbox surface is coming barring changes in strategy. The available information points to Xbox surface being WiiU done right. However that doesn't make sense either given Durango. Some vital pieces are missing. One thing that might unify some of the rumors is if the power7 IBM chips yields were so bad that MS had to switch to the Jaguar cores.
Everything points to Durango espousing a different design philosophy. As soon as I read about DME, display planes and tiling, along with relevant background material, I independently came to the conclusion as the teamxbox guy.
Here are a few scenarios, in order of decreasing likelihood, that make some sense given the available information.
1. Durango's overall architecture creates a novel graphics environment yielding massive increase in output per unit compared to the more "traditional" environments, i.e. uber-efficient design.
2. Some variation of Yukon multi-SKU ecosystem is in play and we don't have the details yet.
2a. Xbox surface and Xbox surface server play into this somehow, either with or without durango as a component, although it seems inefficient to not use durango as the server.
3. MS is implementing an Apple-style upgrade cycle - some evidence for this exists in the MSnerd timelines. Or cloud really plays a big part of the strategy.
4. Sony surprised MS with summer 2012 upgrades and durango is exactly what it seems, an underpowered box.
SlickShoesRUCrazy
Newcomer
^I believe in the multi-machine scenario. Especially after Paul Thurrot said there were multiple Xbox machines on the way.
Far more likely than your first option IMO is that MS simply wanted a machine that could be sold at a profit from day 1 and incorporated sufficient hardware to make it the dominant media and casual gaming machine. This has necessarily come at the expense of high end gaming performance but Microsofts gamble is that as long as the high end gaming performance is "good enough" - which it appears to be - then the system will capture the market through its vast array of other functions. I think it's a damn good gamble.
Hi, since ppl here are evidently speaking on my behalf I opted to go ahead and sign up since I read these forums a lot anyhow.
1) I never thought nor claimed there would be an RT chip in Durango. I DID argue in support of the idea it was possible to do real time RT (even better actually, path tracing!) and that real time RT chips most certainly exist in the real world and had no shortage of details/evidential support in those arguments.
2) I've no idea where you got the 2nd thing there. I never said any of that.
3) I'm not a dev, nor do I have any any insider access to anything, nor have I ever presented myself in a way to suggest otherwise. I'm just collecting the various pieces of this puzzle just like the rest of you guys here (who do a fantastic job btw).
It's interesting you bring up this possibility because Arthur Gies has made comments before how some publishers weren't happy that a next Gen console wasn't released in 2012 (outside the Wii-U) as if they were expecting it. No idea how true this is, but I always imagined a 2013, or even 2014, release was more probable.
I know it's too early to determine this, but I wonder how likely it would be for them to profit on day one at the price points I think most expect ($400-$500).
Here's a link for those interested: http://research.microsoft.com/pubs/176610/foveated_final15.pdf
MSR used a Tobii TX300 camera to track the eyes with 10ms latency. Not sure why you'd need 'microscopic' level precision here, just millimeter scale should be fine which iirc the patents suggest the new Kinect device can do.
In any case, it's a pretty similar concept to what the patents suggest about the display planes, which is why I mentioned it in my TXB post. The display plan patents suggest they want to lock down things like framerates and resolution for the HUD/foreground stuff potentially via dynamically changing the IQ background stuff is being rendered at. For games with tessellated models this is pretty much exactly what is being done to get the performance jumps in their foveated approach with eye tracking. That implementation is just much more ambitious obviously.
I was really just openly pondering what connections there are between the two concepts and if meaningful estimates for performance gains could be expected from the display planes in Durango via digging more into the foveated stuff.
It's speculation based on various chunks of information. Again, not a dev nor do I have insider info to be clear about that. Plenty of that speculation even originated here, indeed some even from you Shifty.
The part quoted is couched in a speculation thread and the first several words pretty clearly show it is speculation derived from fitting together a puzzle, with the pieces stemming from the rumors/leaks.
I do have a question for you though Shifty, you've mentioned something about TBDR. I'm a physicist and know some tech areas but very little when it comes to computer engineering. What is TBDR and how might it fit into the puzzle of Durango's strange design decisions?
Welcome to the forums. Thank you for signing up and sharing your thoughts while keeping the discussion on topic. Hopefully we'll be able to piece together this puzzle without too many insanities happening.
They' ll go for $400-$500 ? With what ?
Even though this article is old, the basic premise still applies. This is B3D's article from Kristof that discusses the TBDR designs used in Imgtec PowerVR - http://www.beyond3d.com/content/articles/38/.
bitsandbytes
Newcomer
I think it was true. In March 2011 a poster on Gaf who works for AMD said that they were told to push to complete the CPU for launch in fall 2012. It was posted in a Microsoft next-gen thread....
http://www.neogaf.com/forum/showpost.php?p=26392219&postcount=446
I have no idea why they seemingly changed their mind or if specs have been boosted since.
I believe in a multi-SKU strategy from a consumers point of view but what are you going to pull out to make the box $100-200 cheaper? The trouble with building a set-top-box SKU is what happens if someone wants to upgrade to the real thing? Should they actually have to put down $399 to be able to play 720 games? That's idiotic.
What I see happening is a $299 SKU with 120GB drive, $399 SKU with 500GB drive (aka the one to get) and a $99 version of the $299 SKU with a $15 monthly fee for 2 years and includes Live. Any actual gamer already has Live so it doesn't make sense to pay for it again plus they are far more likely to want the high end SKU. Makes perfect sense.
I'm sorry I don't quite understand what you mean with what. It's just the opinion I formed over the last year based on posters responses on the net. It seems, at least IMO, that there are more people expecting these consoles to launch in the $400-$500 range than something cheaper like $300.
I totally forgot about that post. Never really thought an original 2012 was out of the realm of possibility, I just always thought a 2013 launch made more sense. Especially considering the launch of Halo 4 last year where you wouldn't want your new system and high profile holiday release competing for consumer dollars.
Also considering how disappointing the specs sound, again IMO, it wouldn't surprise me if the specs weren't boosted at all. I'm not even sure a year delay would give enough time for a major hardware changes to allow such a boost.
- Status
Similar threads
- Sticky