Fullmoonfever
Top 20
I suspect they are using the Maxim7800 at a guess. Bottom info and they seem to have an affiliation there and also runs on the M4 also.
Happy to be wrong obviously and be Akida haha
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Happy to be wrong obviously and be Akida haha
alwaysgreen
Top 20
This would need to be a price sensitive announcement on the ASX.
Afternoon VictorG,I hope so JK but I doubt it, at least for this shipment.
Actually the reason I wanted someone to ask ChatGPT this question " Who are the shareholders of AIZIP Inc" or similar questions is because I kept getting the following answers,
They are worthy of a full body laugh
Q: is AIZIP located in Silicon Valley
AIZip is a software tool developed by BrainChip, a company with offices and operations in several countries, including the United States. As of my knowledge cutoff in September 2021, BrainChip had an office in San Jose, California, which is located in the heart of Silicon Valley.
It's possible that AIZip is primarily developed and managed out of BrainChip's San Jose office, but without more specific information it's difficult to say for certain. It's worth noting that AIZip is a software tool that can be used by anyone with a compatible computer or device, and it's not tied to any specific physical location.
Q: who are the shareholders of AIZIP Inc
As far as I know, AIZip is not a standalone company, but rather a software tool developed by BrainChip Holdings Ltd, a publicly-traded company listed on the Australian Securities Exchange (ASX) under the ticker symbol "BRN". Therefore, the shareholders of AIZip would be the same as the shareholders of BrainChip Holdings Ltd.
The shareholders of BrainChip Holdings Ltd can include institutional investors, retail investors, and company insiders such as directors and executives. The specific shareholders of the company can change over time due to buying and selling of shares, as well as other corporate actions.
As an AI language model, I don't have access to real-time information about the current shareholders of BrainChip Holdings Ltd. However, the company is required to disclose its major shareholders to the ASX, and this information can be found on the ASX website or the company's official website under the "Investor Relations" section.
Q: when was aizip incorporated
As far as I know, AIZip is not a standalone company but rather a software tool developed by BrainChip Holdings Ltd, a publicly-traded company listed on the Australian Securities Exchange (ASX) under the ticker symbol "BRN". Therefore, AIZip was not incorporated as a separate entity and does not have its own incorporation date.
BrainChip Holdings Ltd was originally incorporated in the United States in 2011 as "Aziana Limited" and was later re-registered in Australia in 2015. The company has since expanded its operations and developed various AI technologies, including the Akida neuromorphic processor and AIZip. The specific date when AIZip was first released or made available is not known to me, but it appears to have been developed and released by BrainChip in recent years as part of its suite of AI tools and technologies.
Q: who is the ceo of aizip
As far as I know, AIZip is not a standalone company, but rather a software tool developed by BrainChip Holdings Ltd, a publicly-traded company listed on the Australian Securities Exchange (ASX) under the ticker symbol "BRN". Therefore, AIZip does not have its own CEO.
The CEO of BrainChip Holdings Ltd is currently Louis DiNardo, who was appointed to the position in 2016. DiNardo has a background in engineering and has held various executive positions in the technology industry, including serving as CEO of several other companies before joining BrainChip.
It's worth noting that AIZip is just one of several AI tools and technologies developed by BrainChip, and the company has a broader management team and staff who are involved in its development, marketing, and support.
Certainly a interesting response from ChatGPT.
Below extract from Brainchips last Annual Report. 2021
No mention of AIZip under subsidiary companys, so yes....strange answer.
Regards,
Esq.
Attachments
BaconLover
Founding Member
Thanks AG.He has as much value as any other shareholder, whether anybody here likes it or not.
And unfortunately the value is 54 cents per share at the moment
If you look at the poll created in one of the other threads, one can see many people do agree with the idea that BRN needs to communicate well to the market.
At the moment it is all podcasts saying how they will bring AGI to 2030. Honestly, they cannot even predict what will happen late this year, how do they predict so far out? 7 years is a long time.
What I want to see is better sales, better communication to the market and better revenue. And if they can do this with respecting their time lines that will be fantastic.
If you think I am downramping for saying this, so be it.
What I won't do is give into these bullies. Brainchip is still a good investment, otherwise I wouldn't be here, but management needs to improve on the above things. If not, the next AGM would be an interesting one.
Don't like my opinion? Tough luck, I don't care, I don't like yours either
JK200SX
Regular
Attachments
BaconLover
Founding Member
Well, that is his opinion Damo, and I do agree with him.
He understands that is what we need to convince markets about BRN's potential. What I am saying is, he acknowledges this is what market wants. Happy to edit it if it is read otherwise.
I am glad that company sees it from a shareholder perspective too, because being a public company, they do have a duty of care to look after the shareholders.
M_C
Founding Member
Future computer systems, said Hinton, will be take a different approach: they will be "neuromorphic," and they will be "mortal," meaning that every computer will be a close bond of the software that represents neural nets with hardware that is messy, in the sense of having analog rather than digital elements, which can incorporate elements of uncertainty and can develop over time.
"It'll be used for putting something else: It'll be used for putting something like GPT-3 in your toaster for one dollar, so running on a few watts, you can have a conversation with your toaster."
Damo4
Regular
"even Tony Dawe acknowledged that Brainchip company needs sales and revenue"
This is misleading and you know it I've realized
BaconLover
Founding Member
Edited it to make it more sense.
But Tony Dawe has ackowledged that Brainchip needs sales and revenue.
How is this misleading? That is exactly what he has said in the email.
FJ-215
Regular
Can I return the favour? Don't bite.
Rise from the ashes
Regular
Possibly mate. And I also would like to know what "I’m eatincx" means
Tothemoon24
Top 20
So many dots going on certainly looking like a master stroke joining GlobalFoundries
MENU
DIVE BRIEF
Kate MagillEditor
Listen to the article2 min
“We see our semiconductor requirements more than doubling over the next several years as vehicles become technology platforms,” Doug Parks, GM executive vice president of global product development, purchasing and supply chain, said in a statement. “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.”
GM signed three supplier deals last summer to secure components needed for battery production, including lithium, nickel, cobalt and cathode active material. And in January, the automaker announced plans to invest $650 million in a Nevada lithium mine.
GlobalFoundries, meanwhile, has been investing in its manufacturing capacity. The company announced the extension of its partnership with QualComm Technologies in August, including investing $4.2 billion in chip manufacturing and expanding capacity at GlobalFoundries’ Malta facility.
For New York, the GM-GlobalFoundries news builds on its efforts to become a hub of semiconductor manufacturing activity. The state passed the Green CHIPS legislation in August, aimed at attracting chip investment and job creation in the state.
“We’re making New York State not only the semiconductor capital of the country — but of the globe,” New York Gov. Kathy Hochul said in a statement.
MENU
DIVE BRIEF
GM signs exclusive chip supplier agreement with GlobalFoundries
Published Feb. 15, 2023
Kate MagillEditor
-
A Chevrolet Volt extended-range electric vehicle. GM and GlobalFoundries signed an exclusive semiconductor supplier deal as the automaker looks to expand its EV production capacity. Bill Pugliano via Getty Images
Dive Brief:
- GM signed an exclusive supplier agreementwith semiconductor maker GlobalFoundries, another arm of the automaker’s growing electric vehicle supply chain network.
- Through the agreement, the chip manufacturer aims to reduce the number of unique chips needed to power GM’s EVs by producing higher quality supply in greater quantity and with more predictability, according to the release.
- GlobalFoundries will produce the chips at its manufacturing facility in Malta, New York.
Dive Insight:
GM has spent the past year fortifying its EV supplier network, looking to secure ample supply of electric vehicle and battery components as it races to produce 1 million EVs annually by 2025.“We see our semiconductor requirements more than doubling over the next several years as vehicles become technology platforms,” Doug Parks, GM executive vice president of global product development, purchasing and supply chain, said in a statement. “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.”
GM signed three supplier deals last summer to secure components needed for battery production, including lithium, nickel, cobalt and cathode active material. And in January, the automaker announced plans to invest $650 million in a Nevada lithium mine.
GlobalFoundries, meanwhile, has been investing in its manufacturing capacity. The company announced the extension of its partnership with QualComm Technologies in August, including investing $4.2 billion in chip manufacturing and expanding capacity at GlobalFoundries’ Malta facility.
For New York, the GM-GlobalFoundries news builds on its efforts to become a hub of semiconductor manufacturing activity. The state passed the Green CHIPS legislation in August, aimed at attracting chip investment and job creation in the state.
“We’re making New York State not only the semiconductor capital of the country — but of the globe,” New York Gov. Kathy Hochul said in a statement.
RECOMMENDED READING
Tothemoon24
Top 20
Did someone say KFC
Fullmoonfever
Top 20
Just further to AIZip.
Does look like the MAX78000.
Maxim, Aizip Partnering to Develop IoT Person Detection Device - News
April 22, 2021PERRY COHEN MAXIM INTEGRATED AIZIP
Maxim Integrated and Aizip announced the MAX78000 neural-network microcontroller detects people in an image using Aizip’s Visual Wake Words (VWW) model at just 0.7 millijoules (mJ) of energy per inference
Taproot
Regular
Sorry if i missed the connection along the way, but can someone fill me in on when / how Aizip are doing an Akida demo ?
I am now officially not going to criticize anyone and I am asking for a truce. That the the focus of this thread once again becomes dedicated to great research and that those wanting to voice their fears or complaints about management, lack of sales etc use the other thread that was created as it seemed to be working well until things got derailed. In fairness, I think it is reasonable to allow the next 24 hours a time when all that want to can throw their insults and accusations at me and I guess keep fighting each other if you want to. I will accept what anyone dishes out and I promise I won’t respond. Or alternatively screw it, let’s keep the status quo of this thread and accept the in fighting. It would however be a shame if we lose any members which has happened before.
Fullmoonfever
Top 20
I know Renault struck up partnership with Chronocam which went on to become Prophesee.
Just saw this paper from mid last year with the LEAT / Renault team.
They speak of Loihi but from what I found earlier they obviously now looking at testing Akidas CNN2SNN capabilities and I'm tying that in with us going to GF 22nm FDSOI which from that NASA solicitation I found indicated was the chip often used in automotive.
I'm starting to wonder if the tape out could be part of supplying some gear to LEAT as well for this upcoming project
DingoBorat
Slim
I’ve always just assumed neuromorphic computing was mainly going to benefit EV’s but here’s a research paper using SNN’s to improve the efficiency of ICE vehicles cheaply and quickly.
www.osti.gov
INTRODUCTION
Neuromorphic computers provide an intriguing platform for low power artificial intelligence (AI) at the edge [1]. However, there are several key hurdles to developing neuromorphic com- puting solutions to real-world applications, including finding sufficiently low power neuromorphic hardware, implementing the appropriate algorithms used to train a spiking neural network (SNN) to deploy onto the hardware platform, and ensuring that the trained SNN fits within the constraints of the pre-determined neuromorphic hardware system.
In this work, we present a complete pipeline for training and deploying a low size, weight, and power (SWaP) neuromorphic hardware solution for a real-world application. The real-world application of interest is an engine control unit (ECU) of a spark-ignition internal combustion engine. To improve engine efficiency and reduce greenhouse gas emissions, transportation researchers have developed advanced combustion strategies that minimize the amount of fuel needed to run the engine. However, as the engine efficiency is pushed to the practical limit, instabilities in the combustion process cause sporadic misfires and partial burns. Such events increase the cycle- to-cycle variability (CCV) of the combustion process which causes undesired levels of noise, vibration, and harshness (NVH). In order to mitigate combustion instabilities, the ECU should command a higher amount of fuel during combustion cycles where misfires or partial burns would otherwise occur. In this case, we would like to utilize a neuromorphic hardware system that will take as input information about the engine combustion’s state at each cycle and provide as output the amount of fuel to inject during the next cycle. The goal is to keep the engine running with few or no engine misfires and partial burns in order to minimize combustion CCV and keep the engine running smoothly, but also to minimize the amount of extra fuel needed to stabilize the combustion events.
FUTURE WORK AND CONCLUSIONS
In this work, we have demonstrated a complete neuromor-
phic workflow, from application to hardware, for training and
deploying low size, weight, and power neuromorphic solutions
for real-world applications. We demonstrated the results of this
workflow on a control task to improve fuel efficiency in spark-
ignition internal combustion engines. By utilizing low power
AI hardware such as neuromorphic systems, we can potentially
enable more efficient engines and reduce greenhouse gas emissions. We show that it is feasible to utilize an SNN approach as deployed on neuromorphic hardware to control an engine in simulation. We show that even this preliminary SNN approach can outperform current, open-loop control strategies. Moreover, we demonstrate the resulting SNNs are very small and sparse and can be deployed onto an inexpensive and low size, weight, and power commercial FPGA, providing the opportunity for rapid deployment.
There are several avenues that we intend to pursue for future work. First, the engine simulators we use in this work are based on a single set of engine operating conditions. We intend to apply this same workflow to train SNNs for other engine characteristic settings.
Low Size, Weight, and Power Neuromorphic Computing to Improve Combustion Engine Efficiency (Conference) | OSTI.GOV
The U.S. Department of Energy's Office of Scientific and Technical Information
www.osti.gov
INTRODUCTION
Neuromorphic computers provide an intriguing platform for low power artificial intelligence (AI) at the edge [1]. However, there are several key hurdles to developing neuromorphic com- puting solutions to real-world applications, including finding sufficiently low power neuromorphic hardware, implementing the appropriate algorithms used to train a spiking neural network (SNN) to deploy onto the hardware platform, and ensuring that the trained SNN fits within the constraints of the pre-determined neuromorphic hardware system.
In this work, we present a complete pipeline for training and deploying a low size, weight, and power (SWaP) neuromorphic hardware solution for a real-world application. The real-world application of interest is an engine control unit (ECU) of a spark-ignition internal combustion engine. To improve engine efficiency and reduce greenhouse gas emissions, transportation researchers have developed advanced combustion strategies that minimize the amount of fuel needed to run the engine. However, as the engine efficiency is pushed to the practical limit, instabilities in the combustion process cause sporadic misfires and partial burns. Such events increase the cycle- to-cycle variability (CCV) of the combustion process which causes undesired levels of noise, vibration, and harshness (NVH). In order to mitigate combustion instabilities, the ECU should command a higher amount of fuel during combustion cycles where misfires or partial burns would otherwise occur. In this case, we would like to utilize a neuromorphic hardware system that will take as input information about the engine combustion’s state at each cycle and provide as output the amount of fuel to inject during the next cycle. The goal is to keep the engine running with few or no engine misfires and partial burns in order to minimize combustion CCV and keep the engine running smoothly, but also to minimize the amount of extra fuel needed to stabilize the combustion events.
FUTURE WORK AND CONCLUSIONS
In this work, we have demonstrated a complete neuromor-
phic workflow, from application to hardware, for training and
deploying low size, weight, and power neuromorphic solutions
for real-world applications. We demonstrated the results of this
workflow on a control task to improve fuel efficiency in spark-
ignition internal combustion engines. By utilizing low power
AI hardware such as neuromorphic systems, we can potentially
enable more efficient engines and reduce greenhouse gas emissions. We show that it is feasible to utilize an SNN approach as deployed on neuromorphic hardware to control an engine in simulation. We show that even this preliminary SNN approach can outperform current, open-loop control strategies. Moreover, we demonstrate the resulting SNNs are very small and sparse and can be deployed onto an inexpensive and low size, weight, and power commercial FPGA, providing the opportunity for rapid deployment.
There are several avenues that we intend to pursue for future work. First, the engine simulators we use in this work are based on a single set of engine operating conditions. We intend to apply this same workflow to train SNNs for other engine characteristic settings.
Similar threads
