| zorg1000 said: Wouldnt it make more sense to just go with the existing Tegra X2 instead of redesigning the Tegra X1? X2 has the same performance at half the power draw or twice the performance at the same power draw. X2 could be used for both a Lite & Pro revision. |
Actually it would. Forgot abut the X2. Its already a on a 16nm process to and been in circulation since 2016.
Its almost identical to the X1 barring a smaller fab process which allows for higher clocks and lower power draw. GPU wise they have the same number of cores.
But I don't think it will be used for a lite+pro model. Thats just bad form, putting the same APU in two different devices but telling people that in one device it performs better and in another its worse.
The X2 however does check all the right boxes for fr a revision more in line with what I mentioned. But then there is the price. We dont know if availability is why nintendo went with the X1 since the X2 didnt really come around till 2016 or if the option was there but was more expensive.... being Nvidia.
If the issue is a price thing and the X2 costs more than the X1 then there is no way nintendo will go with that. And the only real way to drop prices of these chips is to move onto a smaller fab process, and this is something that is more out of necessity because it will actually cost more to keep insisting for a 20nm processor when the foundry making them has mostly shifted its business to 14nm and 7nm chips.
That revision when it comes will offer slightly better performance (think XB1s improvements over XB1) and all round better efficiency (less heat, less power draw, longer battery life..etc) and may even be cheaper if they go to a smaller (14nm/7nm) fabrication process which is something nvidia MUST do eventually for their tegra line. But this notion of two models...... doesn't take, especially if it would entail using the same processor clocked at different speeds across different skus.
