So we’ve got Nvidia’s roadmap for the next year or so, but it seems there’s some confusion about this really means on the Internet. But first let’s see what Nvidia has prepared for us in the coming year. We can see that Kal-El as supposed to come around mid 2011, or as Nvidia originally announced – August – but it seemed to have gotten delayed again. Hopefully, it won’t be delayed as much was Tegra 2 was, and we’ll actually start seeing some Kal-El tablet this year.
Then we can see they have prepared a so-called “Kal-El+” at the beginning of 2012, and a Wayne chip, again they say around mid 2012, but my guess is we’ll see it in the fall of 2012. Also there’s a Grey chip, that seems to be meant for lower end smartphones, which is absolutely the smart thing to do for Nvidia. If Nvidia wants to be the “Intel” of mobile chips, they need to cover not only the high-end, but also the low-end.
Right now Qualcomm does that the best. Qualcomm has chips from the 600 Mhz ARM11 range up to dual core 1.5 Ghz coming soon this year. But if Nvidia really wants to “dominate”, they need to cover all the markets. Then manufacturers would take them even more seriously, and they’ll do more business with them.
Now, about this Kal-El+ chip – what is it? Some even think it’s the “real” Kal-el chip, because it will be manufactured at a 28 nm node, as opposed to Kal-El which is 40nm, and considering it’s a quad core chip, it seems a little risky. We’ll see when it arrives if that strategy proved the right one, and if the battery life will still be good, especially when it arrives to smartphones.
I don’t know if Kal-El+ will be made at 28 nm – it’s possible, but I doubt it. I don’t think Nvidia will move to 28nm until Wayne. Kal-El+ is nothing more than a mid-life kicker to Kal-El. It will get a relatively small improvement in performance, possibly being upgraded to a quad core 2.0 Ghz chip, but that’s still like only 33% improvement – it’s not exactly “next-gen”. It’s meant, just like dual core 1.2 Ghz Tegra 2 3D was meant, to prolong the life and brand of “Kal-El” before Way shows up in the market.
Basically, Nvidia wants to release a new chip every 6 months, instead of every year. This isn’t really surprising considering Samsung, Qualcomm and TI are all doing the same. Samsung has a dual core Exynos at 1.2 Ghz, and will have one at 1.4 Ghz soon. Then Qualcomm also has a dual core Snapdragon at 1.2 Ghz, and will have one at 1.5 Ghz this fall, until the dual core 2.5 Ghz Krait arrives.
This is exactly why Nvidia wants a quad core 2 Ghz chip in the market around the same time the dual core 2.5 Ghz Krait arrives. They don’t even want people to consider whether they want a quad core at a significantly lower 1.5 Ghz clock speed, or a dual core at a higher clock speed of 2.5 Ghz. Between a quad core 2 Ghz and a dual core 2.5 Ghz, the choice seems a lot easier. And that’s why Kal-El+ will exist.
Now there’s also something else to consider. Nvidia plans to release the tablet version of the Kal-El chip this fall, and the more energy optimized version for phones in the beginning of the year 2012, so there’s a 3-4 months delay between these 2 versions. What this means is that about the same time we’ll see Kal-El in phones, we’ll also start seeing Kal-El+ in tablets. Then by the same we start seeing Wayne in tablets, we’ll see Kal-El+ in phones, and so on. But this should still give them significant advantage in the market over the competition, and from now on I expect Nvidia’s chips to be the most powerful in the market.
As for Wayne, if you ask me, I think it will be a 28nm manufactured quad core 2.5 Ghz Cortex A15, which should be theoretically about 2.5x faster than Kal-El on the CPU side. And that’s what I meant that Kal-El+ won’t be a significant improvement over Kal-El, because it’s not really next-gen. Next-gen chips tend to be 2.5x faster than the previous gen in the ARM world.