Qualcomm, Samsung Hit 10nm On Snapdragon 835 SoC

[IInuyasha74]

Qualcomm and Samsung announced that the two companies have collaborated to create the new Snapdragon 835 SoC on Samsung’s 10nm FinFET process.

Qualcomm, Samsung Hit 10nm On Snapdragon 835 SoC : Read more

 

[jasonelmore]

The base layer should be 10nm, and that's where the majority of the CPU and GPU cores will be. the 14nm parts will most likely be cache, modem, and we are probably still looking at a 20nm metal interconnect, because i've yet to see a foundry get below 20nm on that.

 

[targetdrone]

After the exploding Note 7s and washing machines, I for one will not be rushing out to buy one of these new Samsung "thing" with these 10nm chips.

 

[ZolaIII]

"The Snapdragon 835 will contain 10nm FinFET transistors when it is released, but that doesn’t necessarily mean it will be a complete 10nm design" naturally not all components will on a FinFET at all. CPU's, GPU cluster & hopefully DSP blocks would be on 10 nm FinFET all RF and probably mixed signal components (cellular broadband radio, WiFi, GPS, Bluetooth...) certainly won't (on FinFET at all). I would be more interested & presently surprised if those RF components would be built using (now mature) 28nm Samsungs FD-SOI as I am certain that would much more contribute to battery life (especially cellular radio & GPS) & there for user experience. Certainly that would be a much better than retarded MHz war in something that should be power consumption optimized. We will see but I am not a big believer when it comes to this. Up to date we have seen only a Sony GPS module built on Samsung 28nm FD-SOI and only in couple of smart watches. Results are about half power consumption compared to regular 20 nm planar proces. When you took into consideration the how much Google location service diminish battery life on your current Android phone that you certainly felt you will get a pretty good picture what I am talking about. Best regards.

 

[TheHeisenberg]

I hope that this new Qualcomm Snapdragon 835 does reduce the number of phone explosions that has happened recently.

With Quick Charge 4, "5 hours of use from just 5 minutes of charging.”
It is something i doubt, like how Google said at first how the Pixel could charge "7 hours in 15 minutes"

Here is a more indepth article on Qualcomm's new Snapdragon 835.

http://saintlad.com/the-new-qualcomm-snapdragon-835-quick-charge-4/

Further Reading

USB Type-C

• https://plus.google.com/+BensonLeung/posts/cEvVQLXhyRX


FinFET manufacturing of nano-sized processors

• https://people.eecs.berkeley.edu/~tking/presentations/KingLiu_2012VLSI-Tshortcourse

• http://www.samsung.com/semiconductor/foundry/process-technology/14nm/

• http://www.radio-electronics.com/info/data/semicond/fet-field-effect-transistor/finfet-technology-basics.php

• http://www.intel.com/content/dam/www/public/us/en/documents/pdf/foundry/mark-bohr-2014-idf-presentation.pdf

 

[targetdrone]

5 hours of phone use in 5 minutes is dumping a heck of a lot of watts into the phone. Either they have to up the voltage beyond 12V or dump a HUGE about of amps which is going to lead to some serious battery stress.

 

[anbello262]

Actually, most Li-Ion batteries, if well built with enough quality and space for the inner divisions (simplified version) should be able to whitstand with no praactical issues up to 1C/h up to ~40-50% charge. This means that you can theoretically charge up to 50% on half an hour (and slower afterwards) with no issues.

The problem was most likely not the quick charge itself, but the quick charge applied on mediocre batteries, not built to whitstand the current.

But this has little to do with the article. This is about a new processor, not a new quick charge battery. It's a mostly unrelated issue.