BRN Discussion Ongoing

[IloveLamp]

Apologies if already posted

https://www.eetimes.com/can-we-trust-ai-in-safety-critical-systems/

Stuart Feffer on LinkedIn: Can We Trust AI in Safety Critical Systems? - EE Times

There are significant issues in using statistical/probablistic models like #AI and #ML in safety-critical systems. This article does a great job in laying…

www.linkedin.com

 

[chapman89]

There has never been a time in Brainchip’s history when so many positive facts are in full public view.

I won’t list them all but Renesas, NASA, Valeo, Mercedes Benz, Intel, ISL, US Airforce Research, ARM, SiFive, Edge Impulse, MegaChips, MOSCHIPS, VVDN, Prophesee, Nviso, Numera, Tata, SOCIONEXT, Biotome……….

The CTO Head of Accenture stating publicly on a Brainchip Podcast that Brainchip is already successful commercially.

Highly credentialed people in the semiconductor space electing to join Brainchip during difficult global financial and political times.

A new AKD1500 chip taped out with GlobalFoundries in response to customer demand.

My off the wall theory on this one is SiFive, DARPA & NASA.

SiFive is ARM’s major competitor. ARM is SiFive’s major competitor.

Would SiFive really want to be promoting AKD1000 with it using an ARM CPU? It is illogical to my mind so the solution is to produce an AKD1500 without an ARM CPU using GlobalFoundries a US Defence and Space preferred foundry.

There is clear evidence that AKD2000 is still on track as temporary staff are being employed for the stated purpose to facilitate its launch in the next 2 to 3 months.

Healthy cash reserves and an early call on LDA Capital for purpose of funding dedicated sales staff in Japan and Korea and the funding of AKD1500 which has clearly been accelerated.

More sales enquires in Brainchip’s history.

Staff members being publicly awarded for their research.

It truly is exciting times despite what some will try preach.

 

[GStocks123]

Keep an Eye on Jabra- I remember Mr.Dinardo liking a heap of their stuff. Could also be zilch..

Check out this job at Jabra:
https://www.linkedin.com/jobs/view/3393163549

 

[Dhm]

Maybe one of our 1st lines of true revenue

Panasonic to Start Mass Production of Its Nicobo Robot in 2023

Nicobo is a small ball-shaped cat-like robot about 20 cm in diameter that cannot do much, but it can turn around, wag its tail, blink its eyes, and change its facial expressions in response to human actions.

internetprotocol.co

When I spoke to NVISO last year after the AGM there wasn’t much immediate thought as to Akida although it would be identified for future renditions. The news item just published mentions nothing about the minimum standards that Akida would bring.

This is typical of the frustrating reality for us shareholders. Whilst Akida is ready to go into products, our partners themselves aren’t ready for Akida yet. They will be, but not now. Prophesee is a good example of this. Presently they are a ‘technology demonstrator’ for us. That will certainly change but that is the fact for us and explains our small revenue numbers presently.

 

[AARONASX]

There has never been a time in Brainchip’s history when so many positive facts are in full public view.

I won’t list them all but Renesas, NASA, Valeo, Mercedes Benz, Intel, ISL, US Airforce Research, ARM, SiFive, Edge Impulse, MegaChips, MOSCHIPS, VVDN, Prophesee, Nviso, Numera, Tata, SOCIONEXT, Biotome……….

The CTO Head of Accenture stating publicly on a Brainchip Podcast that Brainchip is already successful commercially.

Highly credentialed people in the semiconductor space electing to join Brainchip during difficult global financial and political times.

A new AKD1500 chip taped out with GlobalFoundries in response to customer demand.

My off the wall theory on this one is SiFive, DARPA & NASA.

SiFive is ARM’s major competitor. ARM is SiFive’s major competitor.

Would SiFive really want to be promoting AKD1000 with it using an ARM CPU? It is illogical to my mind so the solution is to produce an AKD1500 without an ARM CPU using GlobalFoundries a US Defence and Space preferred foundry.

There is clear evidence that AKD2000 is still on track as temporary staff are being employed for the stated purpose to facilitate its launch in the next 2 to 3 months.

Healthy cash reserves and an early call on LDA Capital for purpose of funding dedicated sales staff in Japan and Korea and the funding of AKD1500 which has clearly been accelerated.

More sales enquires in Brainchip’s history.

Staff members being publicly awarded for their research.

It truly is exciting times despite what some will try preach.

Logically if and hypothetical we're the last company to the market but still slightly offered something different and or inferior to the hypothetical existing market we would still have a piece of the pie left for us greater than where BRN is today... However we are not the last, we're the first and superior. How someone people cannot fathom this is mind boggling.

 

[equanimous]

Logically if and hypothetical we're the last company to the market but still slightly offered something different and or inferior to the hypothetical existing market we would still have a piece of the pie left for us greater than where BRN is today... However we are not the last, we're the first and superior. How someone people cannot fathom this is mind boggling.

Renesas and Megachips alone can easily give us 2% market share and more which would be enough to make BRN successful and profitable.

These are bonuses: NASA, Valeo, Mercedes Benz, Intel, ISL, US Airforce Research, ARM, SiFive, Edge Impulse, MOSCHIPS, VVDN, Prophesee, Nviso, Numera, Tata, SOCIONEXT, Biotome

 

[AARONASX]

Renesas and Megachips alone can easily give us 2% market share and more which would be enough to make BRN successful and profitable.

These are bonuses: NASA, Valeo, Mercedes Benz, Intel, ISL, US Airforce Research, ARM, SiFive, Edge Impulse, MOSCHIPS, VVDN, Prophesee, Nviso, Numera, Tata, SOCIONEXT, Biotome

100% and this is only what we know.

 

[MDhere]

Logically if and hypothetical we're the last company to the market but still slightly offered something different and or inferior to the hypothetical existing market we would still have a piece of the pie left for us greater than where BRN is today... However we are not the last, we're the first and superior. How someone people cannot fathom this is mind boggling.

And not forgetting that we are a mere 1000 eyes and how many eyes are there in this world.???
We are a small group but a very lucky small group of individuals to be one of the 1st to witness history in the making soon and all the other eyes will join us in this new era of neuromorphic computing but wont be on this little forum. Can't remember who it was that gave a good graph of the sp tech momemtum - its going to be explosive. Sometimes we get caught up in our little bubble prob why FF is atm absent as we all know the bigger picture is WORLD proportion! and not just this small forum of shareholders. Still i like this forum and the people im connected to here. Happy Sunday fellow brners.

 

[Getupthere]

How many chips are in a Range Rover? “A modern car can have anywhere from 1,500 to 3,000 chips in up to 80 separate electronic control units with the luxury vehicles built by Jaguar Land Rover being at the higher end of the spectrum.

How many computer chips does an electric car have?


3,000 chips

During his remarks, Biden said the average electric vehicle uses about 3,000 chips, meaning EV will need more than double the amount of what's in a non-electric car.

 

[Stable Genius]

How many chips are in a Range Rover? “A modern car can have anywhere from 1,500 to 3,000 chips in up to 80 separate electronic control units with the luxury vehicles built by Jaguar Land Rover being at the higher end of the spectrum.

How many computer chips does an electric car have?


3,000 chips

During his remarks, Biden said the average electric vehicle uses about 3,000 chips, meaning EV will need more than double the amount of what's in a non-electric car.

I recall an interview where PVDM stated he could do a whole car with 100 chips. That was several years before Jerome Nadal during an interview speaking of 70 sensors in a car and Akida nodes with each one to reduce the data flow to the “main computer” within the vehicle. Also stating it would reduce the overall bill of materials to build it.

This is another reason why Akida is likely to be taken up by vehicle manufacturers. They have the choice to stay with old technology and use 3000 chips or move to the latest and greatest technology and also reduce their costs. Socionext are promoting us for vehicles and MB is pushing ahead to advance the technology as a leader in the field. Valeo are trusted partners.

And that’s just cars. You would think trucks/buses etc being larger will have more sensors so the no. increases.

The vehicle industry alone will be awesome for Brainchip!

 

[Glen]

There has never been a time in Brainchip’s history when so many positive facts are in full public view.

I won’t list them all but Renesas, NASA, Valeo, Mercedes Benz, Intel, ISL, US Airforce Research, ARM, SiFive, Edge Impulse, MegaChips, MOSCHIPS, VVDN, Prophesee, Nviso, Numera, Tata, SOCIONEXT, Biotome……….

The CTO Head of Accenture stating publicly on a Brainchip Podcast that Brainchip is already successful commercially.

Highly credentialed people in the semiconductor space electing to join Brainchip during difficult global financial and political times.

A new AKD1500 chip taped out with GlobalFoundries in response to customer demand.

My off the wall theory on this one is SiFive, DARPA & NASA.

SiFive is ARM’s major competitor. ARM is SiFive’s major competitor.

Would SiFive really want to be promoting AKD1000 with it using an ARM CPU? It is illogical to my mind so the solution is to produce an AKD1500 without an ARM CPU using GlobalFoundries a US Defence and Space preferred foundry.

There is clear evidence that AKD2000 is still on track as temporary staff are being employed for the stated purpose to facilitate its launch in the next 2 to 3 months.

Healthy cash reserves and an early call on LDA Capital for purpose of funding dedicated sales staff in Japan and Korea and the funding of AKD1500 which has clearly been accelerated.

More sales enquires in Brainchip’s history.

Staff members being publicly awarded for their research.

It truly is exciting times despite what some will try preach.

It just came out the other day, Global foundries will supply chips to GM. https://www.prnewswire.com/news-rel...duction-of-semiconductor-chips-301743257.html

 

[Getupthere]

It just came out the other day, Global foundries will supply chips to GM. https://www.prnewswire.com/news-rel...duction-of-semiconductor-chips-301743257.html

I like the sounds of this.

This agreement supports GM's strategy to reduce the number of unique chips needed to power increasingly complex and tech-laden vehicles. With this strategy, chips can be produced in higher volumes and are expected to offer better quality and predictability, maximizing high value content creation for the end customer.
Semiconductors are the foundation of the technologies that are powering the electrification, autonomous driving and connectivity of the auto industry, and they have been center stage in the global chip shortage that has impacted automakers the last couple of years.

"We see our semiconductor requirements more than doubling over the next several years as vehicles become technology platforms," said Doug Parks, GM executive vice president of Global Product Development, Purchasing and Supply Chain. "The supply agreement with GlobalFoundries will help establish a strong, resilient supply of critical technology in the U.S. that will help GM meet this demand, while delivering new technology and features to our customers."

 

[equanimous]

It just came out the other day, Global foundries will supply chips to GM. https://www.prnewswire.com/news-rel...duction-of-semiconductor-chips-301743257.html

Reliance on China diminishing..

GlobalFoundries Inc. is a multinational semiconductor contract manufacturing and design company incorporated in the Cayman Islands and headquartered in Malta, New York. Wikipedia
Stock price: GFS (NASDAQ) USD 64.59 -0.31 (-0.48%)
10 Feb, 4:00 pm GMT-5 - Disclaimer
Headquarters: Santa Clara, California, United States
CEO: Thomas Caulfield (9 Mar 2018–)
Revenue: 6.59 billion USD (2021)
Subsidiaries: AMD FAB 36 LLC, AMD FAB 36 Holding GmbH, MORE
Number of employees: 14,600 (2021)
Founded: 2 March 2009
Total assets: 15.03 billion USD (2021)

 

[Rskiff]

Renesas and Megachips alone can easily give us 2% market share and more which would be enough to make BRN successful and profitable.

These are bonuses: NASA, Valeo, Mercedes Benz, Intel, ISL, US Airforce Research, ARM, SiFive, Edge Impulse, MOSCHIPS, VVDN, Prophesee, Nviso, Numera, Tata, SOCIONEXT, Biotome

Agree and it will be great to see small companies such as NVISO succeed with Brainchips technology and grow to a sizable company. They have been so enthusiastic about Akida, so wish them all the success.

 

[Rach2512]

I recall an interview where PVDM stated he could do a whole car with 100 chips. That was several years before Jerome Nadal during an interview speaking of 70 sensors in a car and Akida nodes with each one to reduce the data flow to the “main computer” within the vehicle. Also stating it would reduce the overall bill of materials to build it.

This is another reason why Akida is likely to be taken up by vehicle manufacturers. They have the choice to stay with old technology and use 3000 chips or move to the latest and greatest technology and also reduce their costs. Socionext are promoting us for vehicles and MB is pushing ahead to advance the technology as a leader in the field. Valeo are trusted partners.

And that’s just cars. You would think trucks/buses etc being larger will have more sensors so the no. increases.

The vehicle industry alone will be awesome for Brainchip!

Also, thinking if every gram of weight reduced increases efficiency, how much weight could be saved going from 3000 to 70 chips? I have no idea? Just a thought.

Have a great day everyone

 

[Diogenese]

Agree and it will be great to see small companies such as NVISO succeed with Brainchips technology and grow to a sizable company. They have been so enthusiastic about Akida, so wish them all the success.

Lots to like about nViso (soon to list on ASX):
https://www.ir.nviso.ai/site/investors/investor-fact-sheet

Plenty of BrainChip related material:

Presentations

Download Latest General Company Presentation (Dec 2022) Download BrainChip Akida Performance Results Presentation (July 2022) Watch Recording of Neuromorphic Performance Results Webinar (July 2022) Watch Recording of Neuromorphic Update Webinar (July 2022)

• Data-centric AI software for edge devices designed for ultra-low power, always-on, battery powered devices. Partnerships with leading neuromorphic hardware providers (BrainChip) and commercial availability of neuromorphic EVKs.

• Highly scalable technology platform with wide application to the extreme edge: NVISO’s AI platform is built for scalability as the business grows and can be applied to wide range of industries and applications, creating exponential growth opportunities as both the number of AI Apps (data driven) and AI Solutions (platform and industry driven) scale. Partnerships with first-to-market neuromorphic computing providers such as BrainChip allow NVISO to release first-to-market neuromorphic computing SDKs accelerating the adoption of its Human Behaviour AI SDK in low latency and low power applications such as robots, electric vehicles, wearables, drones, medical devices, and any other battery powered devices.

• Interoperable with neuromorphic computing: NVISO software is interoperable with neuromorphic computing hardware reaching throughput speeds of over 1000 FPS, latencies below 1ms, and model store sizes below 1MB. Neuromorphic processors are expected to accelerate the adoption of AI at the extreme edge where AI software is run at the sensor level opposed to transmitting data off-device to the cloud for processing.

• NVISO and BrainChip partner on Human Behavioral Analytics in automotive and edge AI devices.

• NVISO releases first Neuromorphic AI App and Neuromorphic AI Solution using BrainChip Akida platform.

• 6th July 2022 the new Regulation (EU) 2019/2144 of the European Parliament and of the Council has mandated driver drowsiness and attention warning (DDAW) systems for all new type-approved motor vehicles of categories M and N from 6 July 2022 and from 7 July 2024 for all new vehicle registrations. This regulation is expected to accelerate the adoption of AI within interior monitoring systems of drivers and occupants.

• Neuromorphic hardware commercial availability to accelerate Edge AI software uptake is expected to accelerate in the next 12-months. A coming #neuromorphic hardware revolution is brewing with first-to-market BrainChip neuromorphic processor, and soon to follow GrAI Matter Labs, and Axelera AI who have announced the availability of silicon with neuromorphic computing. These new AI hardware platforms offer NVISO customers additional price/power/performance points which can accelerate the adoption of its AI App and Solution offerings to major global manufacturers. NVISO has started discussions and evaluation with Semiconductor partners looking to accelerate the adoption of neuromorphic computing through combined software / hardware offerings.

 

[JK200SX]

My son received an email today for the pre-order of the Sony Playstation VR2 that launches on February 22. Here is a link to the online version of the email:

Click here:

PlayStation

PlayStation®VR2 | The next generation of VR gaming on PS5 | PlayStation

PlayStation VR2 and PlayStation VR2 Sense controller: the next generation of VR gaming on PS5. Introducing Horizon Call of the Mountain for PS VR2.

www.playstation.com

Hands-on with Gran Turismo 7’s free PS VR2 update, out February 22

Four reasons racing fans should look forward to playing the entire campaign in PS VR2, including the all-new VR Showroom.

blog.playstation.com

"From eye tracking and headset feedback, to finger touch detection, adaptive triggers, and so much more* – see how new PlayStation VR2 games can use exciting headset and controller features to immerse you deeper than ever into heightened sensory and emotional experiences."

Looks like the technology covers some of the modalities that Brainchip talks about. What are your thought?

 

[equanimous]

Lots to like about nViso (soon to list on ASX):
https://www.ir.nviso.ai/site/investors/investor-fact-sheet

Plenty of BrainChip related material:

Presentations

Download Latest General Company Presentation (Dec 2022) Download BrainChip Akida Performance Results Presentation (July 2022) Watch Recording of Neuromorphic Performance Results Webinar (July 2022) Watch Recording of Neuromorphic Update Webinar (July 2022)

• Data-centric AI software for edge devices designed for ultra-low power, always-on, battery powered devices. Partnerships with leading neuromorphic hardware providers (BrainChip) and commercial availability of neuromorphic EVKs.

• Highly scalable technology platform with wide application to the extreme edge: NVISO’s AI platform is built for scalability as the business grows and can be applied to wide range of industries and applications, creating exponential growth opportunities as both the number of AI Apps (data driven) and AI Solutions (platform and industry driven) scale. Partnerships with first-to-market neuromorphic computing providers such as BrainChip allow NVISO to release first-to-market neuromorphic computing SDKs accelerating the adoption of its Human Behaviour AI SDK in low latency and low power applications such as robots, electric vehicles, wearables, drones, medical devices, and any other battery powered devices.

• Interoperable with neuromorphic computing: NVISO software is interoperable with neuromorphic computing hardware reaching throughput speeds of over 1000 FPS, latencies below 1ms, and model store sizes below 1MB. Neuromorphic processors are expected to accelerate the adoption of AI at the extreme edge where AI software is run at the sensor level opposed to transmitting data off-device to the cloud for processing.

• NVISO and BrainChip partner on Human Behavioral Analytics in automotive and edge AI devices.

• NVISO releases first Neuromorphic AI App and Neuromorphic AI Solution using BrainChip Akida platform.

• 6th July 2022 the new Regulation (EU) 2019/2144 of the European Parliament and of the Council has mandated driver drowsiness and attention warning (DDAW) systems for all new type-approved motor vehicles of categories M and N from 6 July 2022 and from 7 July 2024 for all new vehicle registrations. This regulation is expected to accelerate the adoption of AI within interior monitoring systems of drivers and occupants.

• Neuromorphic hardware commercial availability to accelerate Edge AI software uptake is expected to accelerate in the next 12-months. A coming #neuromorphic hardware revolution is brewing with first-to-market BrainChip neuromorphic processor, and soon to follow GrAI Matter Labs, and Axelera AI who have announced the availability of silicon with neuromorphic computing. These new AI hardware platforms offer NVISO customers additional price/power/performance points which can accelerate the adoption of its AI App and Solution offerings to major global manufacturers. NVISO has started discussions and evaluation with Semiconductor partners looking to accelerate the adoption of neuromorphic computing through combined software / hardware offerings.

Commercial Achievements Past 12-months

US Patent Awarded granted a US Patent for Human Behavioural Profiling from image processing using Artificial Intelligence. NVISO and Tobii to collaborate to accelerate innovation in Interior Monitoring Systems for Smart Mobility. NVISO and Panasonic sign license agreement to embed Human Behaviour AI in companion robots for Smart Living. NVISO and BrainChip partner on Human Behavioral Analytics in automotive and edge AI devices. NVISO and Siemens Healthineers collaborate to apply Human Behaviour AI in medical imaging applications for Smart Health. NVISO releases first Neuromorphic AI App and Neuromorphic AI Solution using BrainChip Akida platform. Japanese Semiconductor Manufacturer starts evaluation of NVISO Human Behaviour AI SDK for Neuromorphic Computing NVISO starts discussions with strategic commercial partners for distribution of its Human Behaviour AI SDK. NVISO announces interoperability with Neuromorphic Processors

 

[JDelekto]

Lots to like about nViso (soon to list on ASX):
https://www.ir.nviso.ai/site/investors/investor-fact-sheet

Plenty of BrainChip related material:

Presentations

Download Latest General Company Presentation (Dec 2022) Download BrainChip Akida Performance Results Presentation (July 2022) Watch Recording of Neuromorphic Performance Results Webinar (July 2022) Watch Recording of Neuromorphic Update Webinar (July 2022)

• Data-centric AI software for edge devices designed for ultra-low power, always-on, battery powered devices. Partnerships with leading neuromorphic hardware providers (BrainChip) and commercial availability of neuromorphic EVKs.

• Highly scalable technology platform with wide application to the extreme edge: NVISO’s AI platform is built for scalability as the business grows and can be applied to wide range of industries and applications, creating exponential growth opportunities as both the number of AI Apps (data driven) and AI Solutions (platform and industry driven) scale. Partnerships with first-to-market neuromorphic computing providers such as BrainChip allow NVISO to release first-to-market neuromorphic computing SDKs accelerating the adoption of its Human Behaviour AI SDK in low latency and low power applications such as robots, electric vehicles, wearables, drones, medical devices, and any other battery powered devices.

• Interoperable with neuromorphic computing: NVISO software is interoperable with neuromorphic computing hardware reaching throughput speeds of over 1000 FPS, latencies below 1ms, and model store sizes below 1MB. Neuromorphic processors are expected to accelerate the adoption of AI at the extreme edge where AI software is run at the sensor level opposed to transmitting data off-device to the cloud for processing.

• NVISO and BrainChip partner on Human Behavioral Analytics in automotive and edge AI devices.

• NVISO releases first Neuromorphic AI App and Neuromorphic AI Solution using BrainChip Akida platform.

• 6th July 2022 the new Regulation (EU) 2019/2144 of the European Parliament and of the Council has mandated driver drowsiness and attention warning (DDAW) systems for all new type-approved motor vehicles of categories M and N from 6 July 2022 and from 7 July 2024 for all new vehicle registrations. This regulation is expected to accelerate the adoption of AI within interior monitoring systems of drivers and occupants.

• Neuromorphic hardware commercial availability to accelerate Edge AI software uptake is expected to accelerate in the next 12-months. A coming #neuromorphic hardware revolution is brewing with first-to-market BrainChip neuromorphic processor, and soon to follow GrAI Matter Labs, and Axelera AI who have announced the availability of silicon with neuromorphic computing. These new AI hardware platforms offer NVISO customers additional price/power/performance points which can accelerate the adoption of its AI App and Solution offerings to major global manufacturers. NVISO has started discussions and evaluation with Semiconductor partners looking to accelerate the adoption of neuromorphic computing through combined software / hardware offerings.

The results from the performance presentation were absolutely amazing. Akida was able to exceed 1000 fps (over 1600 fps in the best case) while the other two (Jetson Nano, and Tegra X1 GPU) did not even pass that threshold. Still yet, it did this at a 1/3 of the clock cycles of the Jetson, while roughly 1/5 of the cycles for the Tegra. It would be interesting to see the power drawn by each for the testing.

I really like the slide in the presentation that shows the transition over time. They've classified BrainChip as "Extreme Edge". It should be pretty clear at this point that BrainChip's goal is to license their IP, and allowing their customers to incorporate it into their own edge solutions. They are taping out with different manufacturing processes as a proof of concept to demonstrate the flexibility of their technology.

A certain individual (who shall not be named) will attempt to convince people that companies could choose any competitor to BrainChip as long as doing so ultimately gives them a working solution. While logically this is true, the statement would indicate to me a lack of formal training in engineering, where the idea is to not only provide a solution, but doing so in a way that requires less power, provides superior performance, and most importantly, keeps costs low. Choosing the technology that can meet both those criteria, while having an extremely strong showing against its competitors in benchmarks makes that argument seem rather silly.

I would think that one working on an edge-based solution that can benefit from AI inference, after looking at the slide below (as well as that of the other benchmarks from NVISO's presentation), it certainly gives the reader an impression as to the best choice of technology for their solutions. It would certainly warrant a closer look and taking time to evaluate the technology.

 

[Diogenese]

The results from the performance presentation were absolutely amazing. Akida was able to exceed 1000 fps (over 1600 fps in the best case) while the other two (Jetson Nano, and Tegra X1 GPU) did not even pass that threshold. Still yet, it did this at a 1/3 of the clock cycles of the Jetson, while roughly 1/5 of the cycles for the Tegra. It would be interesting to see the power drawn by each for the testing.

I really like the slide in the presentation that shows the transition over time. They've classified BrainChip as "Extreme Edge". It should be pretty clear at this point that BrainChip's goal is to license their IP, and allowing their customers to incorporate it into their own edge solutions. They are taping out with different manufacturing processes as a proof of concept to demonstrate the flexibility of their technology.

A certain individual (who shall not be named) will attempt to convince people that companies could choose any competitor to BrainChip as long as doing so ultimately gives them a working solution. While logically this is true, the statement would indicate to me a lack of formal training in engineering, where the idea is to not only provide a solution, but doing so in a way that requires less power, provides superior performance, and most importantly, keeps costs low. Choosing the technology that can meet both those criteria, while having an extremely strong showing against its competitors in benchmarks makes that argument seem rather silly.

I would think that one working on an edge-based solution that can benefit from AI inference, after looking at the slide below (as well as that of the other benchmarks from NVISO's presentation), it certainly gives the reader an impression as to the best choice of technology for their solutions. It would certainly warrant a closer look and taking time to evaluate the technology.

View attachment 29331

Akida: beyond Nvidia.