BRN Discussion Ongoing

[Diogenese]

Some details of Qualcomm Snapdragon 8.3 with Hexagon:

https://www.notebookcheck.net/Qualc...-Processor-Benchmarks-and-Specs.762133.0.html

The cryo-CPU is based on ARM's v9.2 architecture and consists of three clusters: The first includes a fast prime core (ARM Cortex-X4) with up to 3.3 GHz. The second consists of three power cores with up to 3.2 GHz and two additional cores clocking at up to 3.0 GHz (all Cortex-A720). The efficiency cluster consists of two power-saving cores (Cortex-A520) that operate at up to 2.3 GHz. The SoC uses the integrated Adreno 750 for graphics acceleration.

View attachment 77051

Compared to the Snapdragon 8 Gen. 2, Qualcomm promises a 25 percent higher GPU clock rate, whose power consumption is said to have been reduced by 25 percent at the same time. Similarly, the CPU's performance is announced to be now a whopping 30 per cent faster and requires 20 per cent less power.

A special focus is also placed on machine learning (AI). The integrated Hexagon NPU is 98% faster than its predecessor and promises a 40% improvement in efficiency. Qualcomm's AI engine is the first to support multimodal generative AI models, including LLM, LVM and ASR, and can process up to 10 trillion parameters directly on the SoC. Image manipulation is said to be possible within fractions of a second. In addition, the NPU of the Snapdragon 8 Gen 3 is also capable of INT4 and Meta Llama 2.

Qualcomm have a big portfolio of AI patents so it's difficult to tell which ones are smokescreens. This is a fairly recent one relating to ML. It uses capacitive voltage accumulation (analog) blended with digital processing.

US2022414443A1 COMPUTE IN MEMORY-BASED MACHINE LEARNING ACCELERATOR ARCHITECTURE 20210625


View attachment 77052


If this is what Qualcomm are using, clearly they have found some work-arounds to negate the manufacturing variability problem of analog. However, such work-arounds usually diminish the theoretical advantages of analog.

Snapdragon has a fistfull of ARM processors, so it should be possible to run TENNs on Snapdragon. Wouldn't that be a dainty flower arrangement to place before the customer?

Some more on Qualcomm's Hexagon AI which is pretty promiscuous, sharing the AI workload selectively between GPU, CPU, and NPU:

https://www.qualcomm.com/content/dam/qcomm-martech/dm-assets/documents/Unlocking-on-device-generative-AI-with-an-NPU-and-heterogeneous-computing.pdf

6 Heterogeneous computing:

Leveraging all the processors for generative AI Generative AI models suitable for on-device execution are becoming more complex and trending toward larger sizes, from one billion to 10 billion to 70 billion parameters. They are increasingly multi-modal, meaning that they can take in multiple inputs — such as text, speech, or images — and produce several outputs.

Further, many use cases concurrently run multiple models. For example, a personal assistant application uses voice for input and output. This requires running an automatic speech recognition (ASR) model for voice to text, an LLM for text to text, and a text-to-speech (TTS) model for a voice output. The complexity, concurrency, and diversity of generative AI workloads require harnessing the capabilities of all the processors in an SoC. An optimal solution entails:

1. Scaling generative AI processing across cores of a processor and across processors

2. Mapping generative AI models and use cases to one or more cores and processors

Choosing the right processor depends on many factors, including use case, device type, device tier, development time, key performance indicators (KPIs), and developer expertise. Many tradeoffs drive decisions, and the target KPI could be power, performance, latency, or accessibility for different use cases. For example, an original equipment manufacturer (OEM) making an app for multiple devices across categories and tiers will need to choose the best processor to run an AI model based on SoC specs, end-product capabilities, ease of development, cost, and graceful degradation of the app across device tiers.

As previously mentioned, most generative AI use cases can be categorized into on-demand, sustained, or pervasive. For on-demand applications, latency is the KPI since users do not want to wait. When these applications use small models, the CPU is usually the right choice. When models get bigger (e.g., billions of parameters), the GPU and NPU tend to be more appropriate. For sustained and pervasive use cases, in which battery life is vital and power efficiency is the critical factor, the NPU is the best option.

Another key distinction is identifying whether the AI model is memory bound — performance is limited by memory bandwidth — or compute bound — performance is limited by the speed of the processor. Today’s LLMs are memory bound for the text generation, so focusing on memory efficiency on the CPU, GPU, or NPU is appropriate. For LVMs, which could be compute or memory bound, the GPU or NPU could be used, but the NPU provides the best performance per watt.

A personal assistant that offers a natural voice user interface (UI) to improve productivity and enhance user experiences is expected to be a popular generative AI application. The speech recognition, LLM, and speech models must all run with some concurrency, so it is desirable to split the models between the NPU, GPU, CPU, and the sensor processor. For PCs, agents are expected to run pervasively (always-on), so as much of it as possible should run on the NPU for performance and power efficiency.

As we know, Akida can multi-task, running different models on the one SoC.

Running AI on CPU or GPU necessarily entails the use of software.

 

[JDelekto]

Some more on Qualcomm's Hexagon AI which is pretty promiscupos, sharing the AI workload selectively between GPU, CPU, and NPU:

https://www.qualcomm.com/content/dam/qcomm-martech/dm-assets/documents/Unlocking-on-device-generative-AI-with-an-NPU-and-heterogeneous-computing.pdf

6 Heterogeneous computing:

Leveraging all the processors for generative AI Generative AI models suitable for on-device execution are becoming more complex and trending toward larger sizes, from one billion to 10 billion to 70 billion parameters. They are increasingly multi-modal, meaning that they can take in multiple inputs — such as text, speech, or images — and produce several outputs.

Further, many use cases concurrently run multiple models. For example, a personal assistant application uses voice for input and output. This requires running an automatic speech recognition (ASR) model for voice to text, an LLM for text to text, and a text-to-speech (TTS) model for a voice output. The complexity, concurrency, and diversity of generative AI workloads require harnessing the capabilities of all the processors in an SoC. An optimal solution entails:

1. Scaling generative AI processing across cores of a processor and across processors

2. Mapping generative AI models and use cases to one or more cores and processors

Choosing the right processor depends on many factors, including use case, device type, device tier, development time, key performance indicators (KPIs), and developer expertise. Many tradeoffs drive decisions, and the target KPI could be power, performance, latency, or accessibility for different use cases. For example, an original equipment manufacturer (OEM) making an app for multiple devices across categories and tiers will need to choose the best processor to run an AI model based on SoC specs, end-product capabilities, ease of development, cost, and graceful degradation of the app across device tiers.

As previously mentioned, most generative AI use cases can be categorized into on-demand, sustained, or pervasive. For on-demand applications, latency is the KPI since users do not want to wait. When these applications use small models, the CPU is usually the right choice. When models get bigger (e.g., billions of parameters), the GPU and NPU tend to be more appropriate. For sustained and pervasive use cases, in which battery life is vital and power efficiency is the critical factor, the NPU is the best option.

Another key distinction is identifying whether the AI model is memory bound — performance is limited by memory bandwidth — or compute bound — performance is limited by the speed of the processor. Today’s LLMs are memory bound for the text generation, so focusing on memory efficiency on the CPU, GPU, or NPU is appropriate. For LVMs, which could be compute or memory bound, the GPU or NPU could be used, but the NPU provides the best performance per watt.

A personal assistant that offers a natural voice user interface (UI) to improve productivity and enhance user experiences is expected to be a popular generative AI application. The speech recognition, LLM, and speech models must all run with some concurrency, so it is desirable to split the models between the NPU, GPU, CPU, and the sensor processor. For PCs, agents are expected to run pervasively (always-on), so as much of it as possible should run on the NPU for performance and power efficiency.

As we know, Akida can multi-task, running different models on the one SoC.

As I understand it, the NPU (which is easily confused for neuromorphic because of the 'N') is another dedicated processor that parallelizes computational operations with the CPU and GPU at lower power requirements.

Qualcomm's Hexagon NPU is designed to offload those computations from the other two and is optimized for vector, matrix, and tensor processing (basically a lot of matrix math). I ran across an interesting thread on Hacker News, where someone benchmarked the NPU and found that it was not quite as good as the CPU itself. Again, the NPU intends to achieve performance through parallelization at lower power. They have the code for their benchmarks here on GitHub.

I believe Akida could still be a strong competitor to Qualcomm's AI offerings, or even potentially a replacement for Hexagon for better real-time processing on power-constrained devices.

 

[TECH]

Intel lost almost $19bn last year

Intel’s revenue was down 7pc to $14.3bn (€13.78bn) in its fourth quarter, the company has said in its first earnings report since Pat Gelsinger was ousted as CEO in December.

www.independent.ie

Intel cancels AI chip, talks painful past and simplified future

Nearly double the foundry losses of 2023.

www.theverge.com

Like I say, I personally don't see Intel derailing our company anytime soon, do you Donald Duck ?

We're targeting the Edge, Intel is targeting the Edge of a Cliff...

 

[Diogenese]

Some more on Qualcomm's Hexagon AI which is pretty promiscuous, sharing the AI workload selectively between GPU, CPU, and NPU:

https://www.qualcomm.com/content/dam/qcomm-martech/dm-assets/documents/Unlocking-on-device-generative-AI-with-an-NPU-and-heterogeneous-computing.pdf

6 Heterogeneous computing:

Leveraging all the processors for generative AI Generative AI models suitable for on-device execution are becoming more complex and trending toward larger sizes, from one billion to 10 billion to 70 billion parameters. They are increasingly multi-modal, meaning that they can take in multiple inputs — such as text, speech, or images — and produce several outputs.

Further, many use cases concurrently run multiple models. For example, a personal assistant application uses voice for input and output. This requires running an automatic speech recognition (ASR) model for voice to text, an LLM for text to text, and a text-to-speech (TTS) model for a voice output. The complexity, concurrency, and diversity of generative AI workloads require harnessing the capabilities of all the processors in an SoC. An optimal solution entails:

1. Scaling generative AI processing across cores of a processor and across processors

2. Mapping generative AI models and use cases to one or more cores and processors

Choosing the right processor depends on many factors, including use case, device type, device tier, development time, key performance indicators (KPIs), and developer expertise. Many tradeoffs drive decisions, and the target KPI could be power, performance, latency, or accessibility for different use cases. For example, an original equipment manufacturer (OEM) making an app for multiple devices across categories and tiers will need to choose the best processor to run an AI model based on SoC specs, end-product capabilities, ease of development, cost, and graceful degradation of the app across device tiers.

As previously mentioned, most generative AI use cases can be categorized into on-demand, sustained, or pervasive. For on-demand applications, latency is the KPI since users do not want to wait. When these applications use small models, the CPU is usually the right choice. When models get bigger (e.g., billions of parameters), the GPU and NPU tend to be more appropriate. For sustained and pervasive use cases, in which battery life is vital and power efficiency is the critical factor, the NPU is the best option.

Another key distinction is identifying whether the AI model is memory bound — performance is limited by memory bandwidth — or compute bound — performance is limited by the speed of the processor. Today’s LLMs are memory bound for the text generation, so focusing on memory efficiency on the CPU, GPU, or NPU is appropriate. For LVMs, which could be compute or memory bound, the GPU or NPU could be used, but the NPU provides the best performance per watt.

A personal assistant that offers a natural voice user interface (UI) to improve productivity and enhance user experiences is expected to be a popular generative AI application. The speech recognition, LLM, and speech models must all run with some concurrency, so it is desirable to split the models between the NPU, GPU, CPU, and the sensor processor. For PCs, agents are expected to run pervasively (always-on), so as much of it as possible should run on the NPU for performance and power efficiency.

As we know, Akida can multi-task, running different models on the one SoC.

Running AI on CPU or GPU necessarily entails the use of software.

When you find yourself in a hole ... keep digging:

So Qualcomm's Hexagon NPU evolved from a DSP:

Building our NPU from a DSP architecture was the right choice for improved programmability and the ability to tightly control scalar, vector, and tensor operations that are inherent to AI processing. Our design approach of optimized scalar, vector, and tensor acceleration combined with large local shared memory, dedicated power delivery systems, and other hardware acceleration differentiates our solution. Our NPU mimics the neural network layers and operations of the most popular models, such as convolutions, fully-connected layers, transformers, and popular activation functions, to deliver sustained high performance at low power.

So naturally there's a side tunnel to DSPs:

Digital signal processor - Wikipedia

en.wikipedia.org

Such performance improvements have led to the introduction of digital signal processing in commercial communications satellites where hundreds or even thousands of analog filters, switches, frequency converters and so on are required to receive and process the uplinked signals and ready them for downlinking, and can be replaced with specialised DSPs with significant benefits to the satellites' weight, power consumption, complexity/cost of construction, reliability and flexibility of operation. For example, the SES-12 and SES-14 satellites from operator SES launched in 2018, were both built by Airbus Defence and Space with 25% of capacity using DSP.[6]


I wonder if Airbus intends to swap its DSPs for SNNs?

 

[Diogenese]

As I understand it, the NPU (which is easily confused for neuromorphic because of the 'N') is another dedicated processor that parallelizes computational operations with the CPU and GPU at lower power requirements.

Qualcomm's Hexagon NPU is designed to offload those computations from the other two and is optimized for vector, matrix, and tensor processing (basically a lot of matrix math). I ran across an interesting thread on Hacker News, where someone benchmarked the NPU and found that it was not quite as good as the CPU itself. Again, the NPU intends to achieve performance through parallelization at lower power. They have the code for their benchmarks here on GitHub.

I believe Akida could still be a strong competitor to Qualcomm's AI offerings, or even potentially a replacement for Hexagon for better real-time processing on power-constrained devices.

Yes. The Qualcomm white paper above states that their NPU evolved from the DSP, and it looks like they are sticking with it.

"Building our NPU from a DSP architecture was the right choice for improved programmability and the ability to tightly control scalar, vector, and tensor operations that are inherent to AI processing. Our design approach of optimized scalar, vector, and tensor acceleration combined with large local shared memory, dedicated power delivery systems, and other hardware acceleration differentiates our solution."

"You know, I reckon if we put a supercharger on this Model T, it'll be as good as the rest."

 

[Getupthere]

Apple’s Next iPhone Could Be The Best iPhone Yet — Forbes

With Apple expected to launch the iPhone SE in the next few months, why could this iPhone be one of the best smartphones to buy in 2025?

apple.news

While Apple is late to generative AI, it continues to innovate with other techniques that contribute to the iPhone experience. You’ll find Apple’s Neural Engine built into the Axx chipsets, which allow for on-device language processing, image recognition, and data processing through machine learning.


When there is so much talk of moving personal data into the cloud or processing (as well as the intense energy demands of generative AI), many will look at Apple’s efforts to keep data on the device and find it a more personable decision.

 

[Frangipani]

IFS Direct Connect 2025 April 29 San Jose, CA

IFS Direct Connect 2025 April 29 San Jose, CA BrainChip is excited to participate in Intel Foundry Direct Connect 2025, where industry leaders will explore the latest advancements in semiconductor design, packaging, and resilient supply chains. As a key innovator in ultra-low power AI...

brainchip.com

 

[FiveBucks]

So Nintendo. You know, Megachips biggest customer....

https://gamerant.com/switch-2-joy-cons-patent-player-move-prediction/

New controllers predicting a player's input.

• The Nintendo Switch 2 Joy-Con controllers may predict player movements based on a patent filing.
• The patent describes a system for tracking finger movement to predict player inputs.
• The Switch 2 will get a proper unveiling in April, potentially revealing new controller features.

From our website:

 

[uiux]

https://patentscope.wipo.int/search/en/detail.jsf?docId=US447184420


US20250025779 - SYSTEM, INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM

Abstract
A system includes a controller operated by a user and one or more processors. The controller includes a plurality of pressable buttons that are independent of each other, a first sensor that detects pressing of at least one of the buttons, and a second sensor that detects approach or contact of a finger of a user to at least one of the plurality of buttons. When a movement operation to sequentially approach or contact at least two buttons with the finger of the user is performed, the one or more processors perform first processing based on an order of the buttons approached or contacted in the movement operation.

---

From GPT:

Nintendo's patent (US20250025779) describes an innovative controller system that integrates gesture-based input recognition using a combination of mechanical buttons and sensors. This system is designed to detect both traditional button presses and touch-based gestures, allowing users to perform sequential interactions without physically pressing the buttons.

Technology and Sensors Involved​

1. First Sensor (Mechanical Press Detection)
   • Detects button presses through mechanical switches or conductive rubber contacts.
   • Likely implemented using standard electrical contacts (e.g., conductive carbon pads on a PCB).
2. Second Sensor (Touch/Proximity Detection)
   • Detects when a finger approaches or lightly touches a button without pressing it.
   • Possible implementations include:
     • Capacitive Touch Sensors (similar to smartphone touchscreens).
     • Infrared Sensors (measuring reflected IR light for proximity detection).
     • Ultrasonic Sensors (using sound waves to sense distance).
3. Gesture Recognition via Sequential Button Interaction
   • The system detects ordered button touches to determine the intended input.
   • Supports clockwise vs. counterclockwise gestures for different actions.
   • Filters out accidental touches using timing thresholds.

Neuromorphic Potential​

A neuromorphic processor like BrainChip Akida could significantly enhance this system. Unlike conventional processors, neuromorphic chips process events asynchronously, only activating when needed. This provides:

• Power efficiency: No need for continuous polling of sensors.
• Adaptive learning: The system could learn user-specific input habits over time.
• Ultra-low-latency processing: Faster and more accurate gesture recognition.
• Improved filtering: Better distinguishing between intentional gestures and accidental touches.

 

[Meatloaf]

So Nintendo. You know, Megachips biggest customer....

https://gamerant.com/switch-2-joy-cons-patent-player-move-prediction/

New controllers predicting a player's input.

• The Nintendo Switch 2 Joy-Con controllers may predict player movements based on a patent filing.
• The patent describes a system for tracking finger movement to predict player inputs.
• The Switch 2 will get a proper unveiling in April, potentially revealing new controller features.

From our website:

View attachment 77092

It would be absolutely fantastic if Akida was in Nintendo switch 2.

According to Wikipedia:

As of September 2024, the Nintendo Switch has shipped over 146 million units worldwide. It is Nintendo's best-selling home console and the third-best-selling game console of all time, behind the PlayStation 2 and Nintendo DS.

This would be something that would see Brn profitable over night.

Fingers crossed

 

[Slade]

The problem with copying and pasting large amounts from ChatGPT is that it creates posts that are full of inaccuracies.

 

[uiux]

The problem with copying and pasting large amounts from ChatGPT is that it creates posts that are full of inaccuracies.

Are you talking about my post ?

 

[Food4 thought]

Here we go again
The start of a new month
The start of a new week
Are we moving closer to seeing the ink
Will February be as exciting as January? or will it prove to be our best month this year
Well so far this year it’s been pretty good I am still green, but last week was a shocker.
I am thinking that India will be the next big player to release something to rattle the market, but hopefully it will have some brains in it
Have a great week
And get to the gym

 

[DingoBorat]

https://patentscope.wipo.int/search/en/detail.jsf?docId=US447184420


US20250025779 - SYSTEM, INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM

Abstract
A system includes a controller operated by a user and one or more processors. The controller includes a plurality of pressable buttons that are independent of each other, a first sensor that detects pressing of at least one of the buttons, and a second sensor that detects approach or contact of a finger of a user to at least one of the plurality of buttons. When a movement operation to sequentially approach or contact at least two buttons with the finger of the user is performed, the one or more processors perform first processing based on an order of the buttons approached or contacted in the movement operation.

---

From GPT:

Nintendo's patent (US20250025779) describes an innovative controller system that integrates gesture-based input recognition using a combination of mechanical buttons and sensors. This system is designed to detect both traditional button presses and touch-based gestures, allowing users to perform sequential interactions without physically pressing the buttons.

Technology and Sensors Involved​

1. First Sensor (Mechanical Press Detection)
   • Detects button presses through mechanical switches or conductive rubber contacts.
   • Likely implemented using standard electrical contacts (e.g., conductive carbon pads on a PCB).
2. Second Sensor (Touch/Proximity Detection)
   • Detects when a finger approaches or lightly touches a button without pressing it.
   • Possible implementations include:
     • Capacitive Touch Sensors (similar to smartphone touchscreens).
     • Infrared Sensors (measuring reflected IR light for proximity detection).
     • Ultrasonic Sensors (using sound waves to sense distance).
3. Gesture Recognition via Sequential Button Interaction
   • The system detects ordered button touches to determine the intended input.
   • Supports clockwise vs. counterclockwise gestures for different actions.
   • Filters out accidental touches using timing thresholds.

Neuromorphic Potential​

A neuromorphic processor like BrainChip Akida could significantly enhance this system. Unlike conventional processors, neuromorphic chips process events asynchronously, only activating when needed. This provides:

• Power efficiency: No need for continuous polling of sensors.
• Adaptive learning: The system could learn user-specific input habits over time.
• Ultra-low-latency processing: Faster and more accurate gesture recognition.
• Improved filtering: Better distinguishing between intentional gestures and accidental touches.

Based on the fact that we have a Strategic Partnership and IP licence agreement with MegaChips and that putting that kind of processing power in a battery operated controller, with standard chips, would likely severely limit its usage time, I declare that this has to be an application of BrainChip's AKIDA technology.

This proclamation has been reached by me, without debate and no debate shall, or will be entered into.

Congratulations to all shareholders, this is a fantastic milestone for us.

The Nintendo Switch 2, is a Premier Product.

 

[uiux]

Why? There is no *specific* mention of a processor, just generic

 

[DingoBorat]

Why? There is no *specific* mention of a processor, just generic

View attachment 77107

Please "re-read" my post Uiux ..

 

[FiveBucks]

Why? There is no *specific* mention of a processor, just generic

View attachment 77107

So Nintendo or Nintendon't?

 

[uiux]

So Nintendo or Nintendon't?

Nintendontknow

 

[Meatloaf]

Based on the fact that we have a Strategic Partnership and IP licence agreement with MegaChips and that putting that kind of processing power in a battery operated controller, with standard chips, would likely severely limit its usage time, I declare that this has to be an application of BrainChip's AKIDA technology.

This proclamation has been reached by me, without debate and no debate shall, or will be entered into.

Congratulations to all shareholders, this is a fantastic milestone for us.

The Nintendo Switch 2, is a Premier Product.


View attachment 77106

The Nintendo Switch retails for approximately $460 AUD. If we were to earn 0.5% royalties on each sold that would be approximately $2.30 AUD.
$2.30 X 150 million=$345,000,000.
Even if it was 0.25% it would still be a decent amount.

Fingers crossed

 

[FiveBucks]

Ni

The Nintendo Switch retails for approximately $460 AUD. If we were to earn 0.5% royalties on each sold that would be approximately $2.30 AUD.
$2.30 X 150 million=$345,000,000.
Even if it was 0.25% it would still be a decent amount.

Fingers crossed

Nintendo is not giving away that much cash. Maybe $30 million. Not 345