BRN Discussion Ongoing

[Tothemoon24]

having a Xmas beer with some fellow mates who I’ve got on the chip .
It was all aboard on Chipper

 

[Stable Genius]

Wow, wow. wwwwow.

After listening to this EEJournal interview one doesn't need a PhD in neuromorphic inferencing to appreciate that Brainchip/Akida is light years ahead of Intel/Loihi for commercial implementation of neuromorphic whatever. I'm in a state of gobsmackery.......That is my favorite Australian state.

Curious to know how Intel handles their new formal relationship with Brainchip. Do they tout 'rival' technology. It will only bring more focus on how far behind they are. Do they buy Brainchip for what would have to be an outrageous sum....and fold on their Loihi aspirations. Do they suppress Akida's wonderfulness. What gives?

You can see that I am struggling mightily with this.

Help me find the roadmap.....to peace and prosperity and understanding

On a personal note.....leaving Waiheke Island tomorrow (it has been fabuloso)....for my home .. Mexico.

Feliz Navidad a todos y hasta pronto

TBC

@charles2

I could be wrong but I see it as “If you can’t beat them; join them!”

Intel knows they are way behind in commercialising a product which is currently inferior and more expensive.

Edge Ai is on the cusp of exploding: within the next 12 months the first products will be hitting the market and there will be exponential growth. They don’t want to get left behind!

Intel needs to make money as do all businesses therefore they would rather make/sell products and make a slightly less profit by including Akida than no profit at all.

This is great for Brainchip as Intel can put their vast no. of engineers into creating products and their resources into advertising etc which will benefit Brainchip!


Also, I’m pretty sure Intel are building FABs and are going to be making a lot of profits from producing SOC’s as a supplier for companies such as Dell. So having them onboard the Brainchip Bus is a win for them and a win for us!

Merry Xmas!

 

[Mccabe84]

?

 

[Mccabe84]

View attachment 25230
?

Axelera launches RISC-V AI platform for edge vision

Axelera AI has launched a platform that combines its Metis AI Processing Unit (AIPU) chips, RISC-V controller and Voyager SDK software stack.

www.eenewseurope.com

 

[Fact Finder]

Axelera launches RISC-V AI platform for edge vision

Axelera AI has launched a platform that combines its Metis AI Processing Unit (AIPU) chips, RISC-V controller and Voyager SDK software stack.

www.eenewseurope.com

Seem to use different architecture and far too much power. I will await the verdict of @Diogenese

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Mccabe84]

Seem to use different architecture and far too much power. I will await the verdict of @Diogenese

My opinion only DYOR
FF

AKIDA BALLISTA

Thanks for your reply, it’s well outside my understanding of technology. I only know how to turn a computer on and off

 

[Stable Genius]

Mike drop from AM:

 

[TopCat]

200 million cameras at $10.00 each is a 2 billion dollar market but highly likely much larger as some cameras will be much more expensive but just one percent of a 2 billion dollar market is 20 million dollars.

Because some here are new to why I keep bringing up one percent a little bit of history.

Back at the HC cesspit a couple of the more intelligent worms attacked the idea that Brainchip would capture one hundred percent of the market faced with competition from the tech giants Nvidia, Intel, Samsung, Qualcomm, Google and Apple.

I approached this argument from the perspective that the Edge market both existing and still to be thought of was so huge that even capturing a tiny little one percent of that market was hugely rewarding for Brainchip with a profit margin of 97%.

I took this argument then to another level by issuing a challenge which was to put a compelling argument as to why Brainchip with the first commercial neuromorphic chip and IP in the market and a three year lead over its nearest competitor would not capture one percent of the addressable market.

Strangely no one ever took up this challenge.

Now one further point to consider and which I add to my one percent argument.

Development Time Lines:

We have seen from Sony and Prophesee, Brainchip and Renesas, Brainchip and Socionext that it takes approximately 3 years to develop a product from the point of technology adoption.

So consider this Brainchip stands with a three year technology lead soon to become five years with AKIDA 2.0.

If a company wants to implement a neuromorphic solution today it will be three years before they can buy an alternative solution to AKIDA and then a further three years to develop the desired product.

What this means is that failure to take up an AKIDA solution today by a company will mean a potential delay of six years or more until they can bring their product idea to market.

A company in the technology space that waits six years to capitalise on a product idea will most likely have badly missed the boat.

So while I use the one percent argument I personally like many here have much higher expectations than one tiny percent of the addressable market.

Indeed realistically if you consider Renesas, Socionext, MegaChips and Prophesee as standalone engagements I believe it is reasonable to argue they have secured for Brainchip already a percentage of market share over the next year or so far greater than one percent.

I will leave it to others to ponder the percentage points Edge Impulse, ARM and Intel will add to this secured percentage over the same time frame before Mercedes Benz, Valeo and Ford start to contribute in 2024.

It all comes together to make having a plan for ‘when not if’ compulsory.

My opinion only DYOR
FF

AKIDA BALLISTA

Hi FF , further to your post, I really like where nViso are taking us with their Neuro Models apps. To me they seem that they are easily deployable and quick to get started. The one thing I don’t understand is how revenue would come in from the apps though. Would it be a one off license fee perhaps , or an amount for each app downloaded? Anyway, wishing everyone a safe and happy Christmas

 

[equanimous]

I have notice in the last few days on Sydney Commercial TV, that DELL and INTEL are promoting each other in a joint commercial ........... Seems IMO as though Intel wants to improve their net working / associations of late.

Its all about Transforming..

 

[Fact Finder]

Mike drop from AM:

View attachment 25236

That is such sad news the day before Christmas. Anil how could you say this now.

Did you not think of how many tech engineers in the US and EU are going to trudge home through the snow and sludge without a Christmas turkey having left it in the window of the butcher shop because they know despite years of making GPUs more powerful for the automotive industry custom IP from Brainchip has lept so far ahead it is no longer a competition.

Surely you could have let them have one last Christmas living in blissful ignorance with Von Neumann.

 

[Moonshot]

Mike drop from AM:

View attachment 25236

Love this, also he liked an SoC Engineering job a Nvidia a month ago… partners on automotive SoCs?

« Creating power models of key SoC units to evaluate architectural tradeoffs across complex CPU, Memory, GPU/ML, accelerators and high speed serdes subsystems »

 

[Fact Finder]

Hi FF , further to your post, I really like where nViso are taking us with their Neuro Models apps. To me they seem that they are easily deployable and quick to get started. The one thing I don’t understand is how revenue would come in from the apps though. Would it be a one off license fee perhaps , or an amount for each app downloaded? Anyway, wishing everyone a safe and happy Christmas

Hi @TopCat
No idea but what I do know is if anyone wants to maximise the performance of Nviso Apps they need to acquire AKIDA technology.

This means the more success Nviso has the more engagements it brings Brainchip’s way.

My opinion only DYOR
FF

AKIDA BALLISTA

PS: If Nviso fails then Brainchip has backstops in ARM, Intel and a couple of others.

 

[Diogenese]

View attachment 25230
?

Thank you Mccabe84 - the best Christmas presentan AI-NN-ML geek could have in the Alexera presentation:
https://www.axelera.ai/digital-in-memory-computing-for-deep-learning-acceleration/

I'm not sure how far along the Road to Damascus Alexera CTO Evangelos ELEFTHERIOU was when he was struck by the analog/digital divide, but the slide presentation is a wonderful assemblage of the pros and cons.

The patents below were filed in April and October 2020 in the name of IBM naming Alexera CTO Evangelos ELEFTHERIOU as an inventor.

There are no published patent applications in the name of Alexera AI.

Looking at the timeline, ELEFTHERIOU joined Alexera as CTO a year after the patents were filed, yet within a few months Alexera, which was founded in July 2021 (having been in stealth mode since Q4 2019) was taping out its test chip in December 2021.

https://www.axelera.ai/our-story/

WO2021220069A2 CROSSBAR ARRAYS FOR COMPUTATIONS IN MEMORY-AUGMENTED NEURAL NETWORKS
IBM 20200429


Inventors: BOHNSTINGL THOMAS [CH]; PANTAZI ANGELIKI [CH]; WOZNIAK STANISLAW [CH]; EVANGELOS ELEFTHERIOU (CH)

In a hardware-implemented approach for operating a neural network system, a neural network system is provided comprising a controller, a memory, and an interface connecting the controller to the memory, where the controller comprises a processing unit configured to execute a neural network and the memory comprises a neuromorphic memory device with a crossbar array structure that includes input lines and output lines interconnected at junctions via electronic devices. The electronic devices of the neuromorphic memory device are programmed to incrementally change states by coupling write signals into the input lines based on: write instructions received from the controller and write vectors generated by the interface. Data is retrieved from the neuromorphic memory device, according to a multiply-accumulate operation, by coupling read signals into one or more of the input lines of the neuromorphic memory device based on: read instructions from the controller and read vectors generated by the interface.




US11430524B2
Method for designing an initialization function for programming a memory element
IBM 20201030

The present disclosure relates to a storage device comprising a memory element. The memory element may comprise a changeable physical quantity for storing information. The physical quantity may be in a drifted state. The memory element may be configured for setting the physical quantity to an initial state. Furthermore, the memory element may comprise a drift of the physical quantity from the initial state to the drifted state. The initial state of the physical quantity may be computable by means of an initialization function. The initialization function may be dependent on a target state of the physical quantity and the target state of the physical quantity may be approximately equal to the drifted state of the physical quantity.

from Alexera's Technology page:
https://www.axelera.ai/technology/
In-memory computing is a radically different approach to data processing, in which crossbar arrays of memory devices can be used to store a matrix and perform matrix-vector multiplications “in-place” without intermediate movement of data. Our proprietary Digital In-Memory Computing (D-IMC) technology is key to delivering high energy efficiency and outstanding performance. Based on SRAM (Static Random-Access Memory) combined with digital computations, each memory cell effectively becomes a compute element. This radically increases the number of operations per computer cycle (one multiplication and one accumulation per cycle per memory cell) without suffering from issues such as noise or lower accuracy.

https://www.axelera.ai/digital-in-memory-computing-for-deep-learning-acceleration/
Our CTO and Co-Founder Evangelos Eleftheriou, presented at the ESSCIRC – ESSDERC 2022 event about In-memory computing for deep-learning acceleration.
In-memory computing (IMC) is a novel computing paradigm, where certain computational tasks are performed in the memory itself using analog or mixed signal computation techniques.
In his presentation he shares a broad overview of the recent progress of IMC for accelerating deep learning workloads, highlighting the strengths and weaknesses of the various approaches.
Learn all about it in his presentation.


I've attached a few selected slides, but the Grinch @Zeebot limits the number of files that can be attached, so I recommend you skim through the slides:
https://www.axelera.ai/digital-in-memory-computing-for-deep-learning-acceleration/

Edit: Sorry I forgot the original question: This is not Akida.

 

[White Horse]

I found this youtube discussion whilst hyperventilating over the proliferation of announcements and TSE research.
Good watch, if you want something to take your brain to another place, other than nirvana.



PS, Eat, Drink and be Merry, and above all, enjoy the company of grandchildren.

 

[Mccabe84]

Thank you Mccabe84 - the best Christmas presentan AI-NN-ML geek could have in the Alexera presentation:
https://www.axelera.ai/digital-in-memory-computing-for-deep-learning-acceleration/

I'm not sure how far along the Road to Damascus Alexera CTO Evangelos ELEFTHERIOU was when he was struck by the analog/digital divide, but the slide presentation is a wonderful assemblage of the pros and cons.

The patents below were filed in April and October 2020 in the name of IBM naming Alexera CTO Evangelos ELEFTHERIOU as an inventor.

There are no published patent applications in the name of Alexera AI.

Looking at the timeline, ELEFTHERIOU joined Alexera as CTO a year after the patents were filed, yet within a few months Alexera, which was founded in July 2021 (having been in stealth mode since Q4 2019) was taping out its test chip in December 2021.

https://www.axelera.ai/our-story/

View attachment 25234

WO2021220069A2 CROSSBAR ARRAYS FOR COMPUTATIONS IN MEMORY-AUGMENTED NEURAL NETWORKS
IBM 20200429


Inventors: BOHNSTINGL THOMAS [CH]; PANTAZI ANGELIKI [CH]; WOZNIAK STANISLAW [CH]; EVANGELOS ELEFTHERIOU (CH)


View attachment 25232



In a hardware-implemented approach for operating a neural network system, a neural network system is provided comprising a controller, a memory, and an interface connecting the controller to the memory, where the controller comprises a processing unit configured to execute a neural network and the memory comprises a neuromorphic memory device with a crossbar array structure that includes input lines and output lines interconnected at junctions via electronic devices. The electronic devices of the neuromorphic memory device are programmed to incrementally change states by coupling write signals into the input lines based on: write instructions received from the controller and write vectors generated by the interface. Data is retrieved from the neuromorphic memory device, according to a multiply-accumulate operation, by coupling read signals into one or more of the input lines of the neuromorphic memory device based on: read instructions from the controller and read vectors generated by the interface.




US11430524B2
Method for designing an initialization function for programming a memory element
IBM 20201030


View attachment 25237

View attachment 25235



The present disclosure relates to a storage device comprising a memory element. The memory element may comprise a changeable physical quantity for storing information. The physical quantity may be in a drifted state. The memory element may be configured for setting the physical quantity to an initial state. Furthermore, the memory element may comprise a drift of the physical quantity from the initial state to the drifted state. The initial state of the physical quantity may be computable by means of an initialization function. The initialization function may be dependent on a target state of the physical quantity and the target state of the physical quantity may be approximately equal to the drifted state of the physical quantity.

from Alexera's Technology page:
https://www.axelera.ai/technology/
In-memory computing is a radically different approach to data processing, in which crossbar arrays of memory devices can be used to store a matrix and perform matrix-vector multiplications “in-place” without intermediate movement of data. Our proprietary Digital In-Memory Computing (D-IMC) technology is key to delivering high energy efficiency and outstanding performance. Based on SRAM (Static Random-Access Memory) combined with digital computations, each memory cell effectively becomes a compute element. This radically increases the number of operations per computer cycle (one multiplication and one accumulation per cycle per memory cell) without suffering from issues such as noise or lower accuracy.

https://www.axelera.ai/digital-in-memory-computing-for-deep-learning-acceleration/
Our CTO and Co-Founder Evangelos Eleftheriou, presented at the ESSCIRC – ESSDERC 2022 event about In-memory computing for deep-learning acceleration.
In-memory computing (IMC) is a novel computing paradigm, where certain computational tasks are performed in the memory itself using analog or mixed signal computation techniques.
In his presentation he shares a broad overview of the recent progress of IMC for accelerating deep learning workloads, highlighting the strengths and weaknesses of the various approaches.
Learn all about it in his presentation.


I've attached a few selected slides, but the Grinch @Zeebot limits the number of files that can be attached, so I recommend you skim through the slides:
https://www.axelera.ai/digital-in-memory-computing-for-deep-learning-acceleration/




View attachment 25238


View attachment 25239






View attachment 25245




Edit: Sorry I forgot the original question: This is not Akida.

Thanks for your reply but I didn’t understand anything you said until I got to the picture of the thumbs down

 

[Foxdog]

It will in my opinion be impossible for Intel to suppress "AKIDA's wonderfulness" given that they do not have an exclusive relationship.

If they tried to downplay its applications any semiconductor customer would clearly ask the question "Well if that's the case why does ARM promote the use of AKIDA across virtually every use case you can possibly imagine???"

They may even ask "Well why does Edge Impulse say that it outcompetes a GPU running at 900 gigahertz."

They may even ask "Why does Mercedes Benz use it for voice and claim that it is up to 10 times more efficient than competitor solutions???"

They may even ask "If it is like you say why do Valeo and NASA trust them???"

They may even ask "Why is MegaChips so heavily promoting Brainchip and its AKIDA solution???"

They may even ask "Why is Prophesee partnering with them for event based sensors???"

They may even ask "Why is Renesas taping out an intelligent AKIDA powered MCU for automotive???

They may even ask "Why has Socionext launched a new AKIDA powered semiconductor range???

Personally how Intel handles the problem is irrelevant the fact that they have brought Brainchip and AKIDA technology into the fold clearly indicates they understand that this was essential if they are to remain relevant in the neuromorphic chip space.

While Mike Davies is trying to talk the talk Brainchip and AKIDA are actually walking the walk and as I said earlier increasing the technology lead they have with every single step.

Loihi 1 & 2 with the release of the Renesas AKIDA MCU falls six years behind. Three years of technology lead plus three years of product development.

Loihi 1 & 2 with the release of the Socionext AKIDA offering falls six years behind. Three years of technology lead plus three years of product development.

Loihi 1 & 2 with the Mercedes Benz and Valeo AKIDA powered product offerings in 2024 falls six years behind. Two years of technology lead plus four years of product development. Product development could be even longer as we have some uncertainty around the exact date they became involved with Brainchip.

Whether they are speaking about it openly or not Blind Freddie is of the opinion that if he knows these obvious facts that Intel must know and have a very large file of documents in which they have debated their present course of action leading to the obvious conclusion that we cannot beat them so we just have to join them or go the way of Kodak where neuromorphic computing is concerned.

Have a very Merry Christmas back in Mexico.

My opinion only DYOR
FF

AKIDA BALLISTA

All I could think of while reading your post FF:
And you may find yourself living in a shotgun shack
And you may find yourself in another part of the world
And you may find yourself behind the wheel of a large automobile
And you may find yourself in a beautiful house, with a beautiful wife
And you may ask yourself, "Well, how did I get here?"
… Letting the days go by, let the water hold me down
Letting the days go by, water flowing underground
Into the blue again, after the money's gone
Once in a lifetime, water flowing underground
… And you may ask yourself, "How do I work this?"
And you may ask yourself, "Where is that large automobile?"
And you may tell yourself, "This is not my beautiful house"
And you may tell yourself, "This is not my beautiful wife"
… Letting the days go by, let the water hold me down
Letting the days go by, water flowing underground
Into the blue again, after the money's gone
Once in a lifetime, water flowing underground
… Same as it ever was, same as it ever was
Same as it ever was, same as it ever was
Same as it ever was, same as it ever was
Same as it ever was, same as it ever was
… Water dissolving and water removing
There is water at the bottom of the ocean
Under the water, carry the water
Remove the water from the bottom of the ocean
Water dissolving and water removing
… Letting the days go by, let the water hold me down
Letting the days go by, water flowing underground
Into the blue again, into the silent water
Under the rocks and stones, there is water underground
… Letting the days go by, let the water hold me down
Leting the days go by, water flowing underground
Into the blue again, after the money's gone
Once in a lifetime, water flowing underground
… You may ask yourself, "What is that beautiful house?"
You may ask yourself, "Where does that highway go to?"
And you may ask yourself, "Am I right, am I wrong?"
And you may say to yourself, "My God, what have I done?"
… Letting the days go by, let the water hold me down
Letting the days go by, water flowing underground
Into the blue again, into the silent water
Under the rocks and stones, there is water underground
… Letting the days go by, let the water hold me down
Letting the days go by, water flowing underground
Into the blue again, after the money's gone
Once in a lifetime, water flowing underground
… Same as it ever was, same as it ever was
Same as it ever was, look where my hand was
Time isn't holding up, time isn't after us
Same as it ever was, same as it ever was
Same as it ever was, same as it ever was
Same as it ever was, same as it ever was (I couldn't get no rest)
Same as it ever was, hey let's all twist our thumbs
Here comes the twister
… Letting the days go by (same as it ever was, same as it ever was)
Letting the days go by (same as it ever was, same as it ever was)
Once in a lifetime, let the water hold me down
Letting the days go by, water flowing underground

Thanks Talking Heads

 

[Fact Finder]

The world’s first commercial producer
of neuromorphic IP.​

BrainChip brings common sense to the processing of sensor data, freeing machines to do more with less, enabling them to infer the big picture from the basics.
Accurately. Elegantly. Meaningfully.
We call this Essential AI. Essential is optimizing compute. Maximizing performance. Minimizing power. In the real world. And in real time. We’re proving that on-chip AI, close to the sensor, has a sensational future, for our customers’ products, as well as the planet.

AkidaTM is the world’s first commercial neuromorphic processor.​

It mimics the brain to analyze only essential sensor
inputs at the point of acquisition—rather than through
transmission via the cloud.

• Efficient – with unparalleled precision and economy of energy.
• Independent – keeping learning local to the chip.
• Scalable – Universally deployable to support any sensor data.

BrainChip is the antidote to more. To more devices. More data. More demands on the data center.
Our technology brings commonsense to the processing of sensor data, freeing machines to do more with less, enabling them to infer the big picture from the basics.
About BrainChip
Our Products
Our Technology

You’re in good company​

Trusted by:​

​

News:​

• 

• 

University AI Accelerator Program​

BrainChip is bringing its neuromorphic technology into higher education institutions via the BrainChip University AI Accelerator Program, which shares technical knowledge, promotes leading-edge discoveries and positions students to be next-generation technology innovators.
BrainChip’s University AI Accelerator Program provides hardware, training, and guidance to students at higher education institutions with existing AI engineering programs. BrainChip’s products can be leveraged by students to support projects in any number of novel use cases or to demonstrate AI enablement. Students participating in the program will have access to real-world, event-based technologies offering unparalleled performance and efficiency to advance their learning through graduation and beyond.
Current university participants include:

By partnering with BrainChip’s AI Accelerator Program, universities are able to ensure that students have the tools and resources needed to encourage development of cutting-edge technologies that will continue to usher in an era of essential AI solutions.
Have your university become AI smarter, join the University AI Accelerator Program today."

Just could not help myself. These are the highlights on the updated Brainchip Home page. I know we know it but I suppose its like the first time a performer sees their name in lights on Broadway. Once you see it in orange highlights you know Brainchip can make it anywhere.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[Bravo]

Mike drop from AM:

View attachment 25236

Very interesting @Stable Genius. But I wonder who Slade is? How come that name sounds so familiar to me?

 

[Dozzaman1977]

Looks like SOCIONEXT is spreading the word about its fantastic products featuring Akida IP around the WORLD
Big Tech event in China 26/27th DEC
Big market CHINA!!!!!!!!!!

Socionext will present at ICCAD 2022​

• Home
• Socionext will present at ICCAD 2022

Socionext Inc. will be exhibiting at the ICCAD 2022 event to be held in Xiamen, China on December 26 and 27, 2022.
Demand for improvements in the energy efficiency and the computational efficiency and capacity of integrated circuits continues to grow, driven by dramatic advances over recent years in applications such as artificial intelligence, autonomous driving, and large data centers.

Against this background, Socionext has been pursuing its own unique “Solution SoC” business model in its role as an industry-leading supplier of SoCs, building deep relationships of trust with its customers throughout the world and supplying custom SoC services that utilize leading-edge technologies to support development of the differentiated products and services that customers are calling for.

At the upcoming ICCAD event, Socionext will be exhibiting advanced technologies and solutions with a particular focus on autonomous driving, smart cockpits, and smart home appliances, accompanying this with presentations on the company’s ongoing research and development work that is aimed at meeting ever-evolving market needs.

Event Information​

Title	CSIA-ICCAD 2022 Annual Conference & Guangzhou IC Industry Innovation and Development Summit (ICCAD 2022)
Days	December 26 Mon. to December 27 Tue.
Venue	Xiamen International Conference & Exhibition Center, Hall C2-C4
Sponsor	China Semiconductor Industry Association IC Design Branch, etc.
Register	Registration and Participation

Exhibition Information​

Exhibiting Booth	Socionext Booth(No. C3-301)
Exhibits	Next Generation Graphics Display Controller for Automotive AI Accelerator for Deep Learning 24GHz/60GHz Radio-wave Ranging Sensor Smart home appliances developed in partnership with local Chinese companies, etc.

Forum Information​

Time Schedule	December 27 Tue. 10:40 ‐ 11:00 (CST)
Venue	Xiamen International Conference & Exhibition Center, Room 303
Main Title	IP & IC Design Services
Sub Title	High Performance Customized SoC Design --- Enabling a Future that Customers Can Envision
Speaker	Mike Xie, Marketing Director, Socionext Technology (Shanghai) Co., Ltd.

 

[Jefwilto]

Looks like SOCIONEXT is spreading the word about its fantastic products featuring Akida IP around the WORLD
Big Tech event in China 26/27th DEC
Big market CHINA!!!!!!!!!!

Socionext will present at ICCAD 2022​

• Home
• Socionext will present at ICCAD 2022

Socionext Inc. will be exhibiting at the ICCAD 2022 event to be held in Xiamen, China on December 26 and 27, 2022.
Demand for improvements in the energy efficiency and the computational efficiency and capacity of integrated circuits continues to grow, driven by dramatic advances over recent years in applications such as artificial intelligence, autonomous driving, and large data centers.

Against this background, Socionext has been pursuing its own unique “Solution SoC” business model in its role as an industry-leading supplier of SoCs, building deep relationships of trust with its customers throughout the world and supplying custom SoC services that utilize leading-edge technologies to support development of the differentiated products and services that customers are calling for.

At the upcoming ICCAD event, Socionext will be exhibiting advanced technologies and solutions with a particular focus on autonomous driving, smart cockpits, and smart home appliances, accompanying this with presentations on the company’s ongoing research and development work that is aimed at meeting ever-evolving market needs.

Event Information​

Title	CSIA-ICCAD 2022 Annual Conference & Guangzhou IC Industry Innovation and Development Summit (ICCAD 2022)
Days	December 26 Mon. to December 27 Tue.
Venue	Xiamen International Conference & Exhibition Center, Hall C2-C4
Sponsor	China Semiconductor Industry Association IC Design Branch, etc.
Register	Registration and Participation

Exhibition Information​

Exhibiting Booth	Socionext Booth(No. C3-301)
Exhibits	Next Generation Graphics Display Controller for Automotive AI Accelerator for Deep Learning 24GHz/60GHz Radio-wave Ranging Sensor Smart home appliances developed in partnership with local Chinese companies, etc.

Forum Information​

Time Schedule	December 27 Tue. 10:40 ‐ 11:00 (CST)
Venue	Xiamen International Conference & Exhibition Center, Room 303
Main Title	IP & IC Design Services
Sub Title	High Performance Customized SoC Design --- Enabling a Future that Customers Can Envision
Speaker	Mike Xie, Marketing Director, Socionext Technology (Shanghai) Co., Ltd.

We dont need China