Wednesday, February 21, 2018

timing... yet this Q......


We are also on schedule to introduce a new low voltage version of EOS later this quarter that will be called EOS S3LV. With this, we will lower our power consumption for always-on/always-listening use cases by as much as 30% and extend our already significant competitive advantage over other MCU based solutions.


comments... thanks to PMCW for the QUIK quarterly documents are so well done. I can see the time it takes to craft them.

So for the casual readers....

1. Voice focus- it specifically says lower power for Voice..... they are delivering the power savings for what is in huge demand right now.

2. QUIK is NOT just sitting around twiddling its thumb while it waits for that Tier 1....they are busy, They have this coming our way soon.
I noted a while ago that there has been a subtle change in QUIK slide deck.... A yr or so ago the algos for MCHR ( motion comensated heart rate )
were more compue intense than voice...now they have changed places....thats good.

So I put this here also.....


A new solution we will add later this quarter leverages our dual microphone inputs by adding beam forming technology and advanced noise cancellation. The goal here is to enhance user experiences in noisy environments. An example of this would be cancelling wind noise to enable accurate voice commands while driving a car with the window down.


These are algos, they add compexity and compute intensity, and also make for a better voice input into the AI....

So the IoV has them bringing a Eos S3LV, adding the algos for noise.


So let it percolate....

Me, I'm a news addict. We have incremental improvements with a focus on VOICE that are coming our way NEAR term. QUIK cannot control when a tier 1 makes up its mind, it cannot think for the Japanese carrier on when/ or if they will mandate always listening voice. 




They can execute to make Voice easier to adopt, perhaps get a few more cents / unit sold.
QUIK is very careful so I think it IS significant that in Q2 they could say that the feature in question was always on voice




the good news is that the design has expanded from a simple sensor hub function to one that now includes a carrier-driven requirement for always-on voice as well. This means the design will leverage several elements of the EOS S3 silicon platform including ARM Cortex M4F MCU, LPSD, and FFE.



, the next Q it changed to " certain features",
as though they had said too much already? :cool:



PS snips of text....

a concept smartphone that it intends to showcase at Mobile World Congress in Barcelona later this quarter. 


neither Kyocera, or Casio are listed as exhibiting at MWC....so I have taken them off of my list....leaving
2 names for me..... and as I handicap things for only FUN.

Sharp- at MWC they have thier S3, the reporting says that this device can be purchased in Japan already, its on the market there for a month or so.

It has shifted the probabilities a little for me. Sure they could have another version on tap for the Carrier in question.


Finally for the casual readers on my personal approach to QUIK. After reading S Johnson's chapter on the Adjacent possible I wrote in my journal that I would use his lens on QUIK. I would NOT use the price listed on the ticker much, as its not useful for me. It gave me a green light to read digressively ALL I want for anything I can find. I really liked the eFPGA IP initiative as the common question of, " what in the world does this have to do with QUIK." is gone for the most part. I try not to have an opinion, just use the adjacent possible and what opinion it would have.

So EOS S3LV, and the coming algos for noise are exactly what we would expect.

No comments:

Post a Comment