Nvidia Offers Brief Introduction to Project Logan, Next Generation Mobile Architecture



+ Add a Comment


If this Kepplerette is really as powerful as they're hyping it to be (the face demo thingy looked awfully good), how do they justify their high-end $1000, 250W Titan board? Also, doesn't this render their own Tegra 3-based handheld obsolete before it even ships?



Comparing the desktop graphics cards with this SOC is like comparing a ford fiesta with Hitler, they are two completely different things for two extremely different purposes.

Also, as much as I love nVidia, they do love to over hype their new products, I think this will be a great chip, but I don't think it will be as amazing as they make it out to be.

I am not sure when the new device is set to ship, but these wont be ready until the first half of next year, Probably late in that first half, and that's barring any delays. Even then they will have to trickle from the fabs to the OEM's, so how ever long it takes them to adopt them into new devices, if they even decide to.

They are cutting things close with the drop to 20nm process while these are 28nm. They could be up against much stiffer competition than they are comparing against now when these release.



I was at a launch event for Citrix and NVidia was there to demonstrate how they are virtualizing their GPU's and it really was impressive. They were able to manipulate GB size CAD drawings in real time over a Citrix session using virtualized video cards. The technology behind that was truly impressive. As for the chips in this piece, I dunno... lol



I won't lie I am EXTREMELY Interested to see what they are doing with all of their success in visualization. It all does look truly amazing.

Because if I could stream over 4g games from my desktop/their servers with low latency at 720p (I think thats the shields resolution, can't remember off the top of my head.) I would buy the shield in a heartbeat.

I just don't really understand how they intend to beat the latency of poor internet connections which plague the US either because of there being no hookups, or just fee's that are outrageous.

Most people where I live run at about 3-4 mbps, we can get 150+, but the fee is just soo high its imaginary...



Tegra4 based, not 3


Julian Reiche

Logan will be using 1 Kepler SMX while the Titan has 13, also the Titan is geared towards a GPU compute professional audience.

The GTX 780 is the highest end gamming GPU and has 12 SMX, at $650 that is $54 a SMX.

SMX = 192 cores