FrederikSchack
Regular
True, wait until ChatGPT 4.0 is released, it's better than the average law student, but always check up the facts. I never take ChatGPT's answers for granted, but I do use them for inspiration.
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
FrederikSchack
Regular
Oh, I just had a revelation.
So, Brainchip uses very little power. You would likely not only be able to connect it as chiplets, but stack it as chiplets without heating issues!
I have no idea how many neurons Akida-P can emulate, but probably 6-7 times the amount of the original Akida.
Thinking a little wildly, stacking 100 on top of each other and let's say organizing stacks in a 10x10 grid. Then you would have 10.000 x Akida-P with let's say 5 million 4-bit neurons each. That would be 50 billion neurons, half of what the human brain has
I have no Idea if this is feasible, just a wild thought If they don't emit very much heat each of them, the heat may be able to transfer through the 100 layers, before being cooled at the top of them. Of coulrse there would be some requirements to the NOC (Network on Chip) and interconnects.
So, Brainchip uses very little power. You would likely not only be able to connect it as chiplets, but stack it as chiplets without heating issues!
I have no idea how many neurons Akida-P can emulate, but probably 6-7 times the amount of the original Akida.
Thinking a little wildly, stacking 100 on top of each other and let's say organizing stacks in a 10x10 grid. Then you would have 10.000 x Akida-P with let's say 5 million 4-bit neurons each. That would be 50 billion neurons, half of what the human brain has
I have no Idea if this is feasible, just a wild thought
If they don't emit very much heat each of them, the heat may be able to transfer through the 100 layers, before being cooled at the top of them. Of coulrse there would be some requirements to the NOC (Network on Chip) and interconnects.BaconLover
Founding Member
My post was reported for profanity and personal attacks against management.
@zeeb0t can you boot the person who reports the posts wrongly? Feel like I'm being targeted here, since 3 of my posts got reported yesterday, which eventually got reinstated. So the poster who abuses this system needs to go.
Just because someone doesn't like a post doesn't mean it's violating forum rules.
I stated facts, and if I want to report posts I can do too, hundreds of posts here because most are off topic to BRN discussion. But I don't do it, because I'm not that low.
D
Deleted member 118
Guest
My post was reported for profanity and personal attacks against management.
@zeeb0t can you boot the person who reports the posts wrongly? Feel like I'm being targeted here, since 3 of my posts got reported yesterday, which eventually got reinstated. So the poster who abuses this system needs to go.
Just because someone doesn't like a post doesn't mean it's violating forum rules.
I stated facts, and if I want to report posts I can do too, hundreds of posts here because most are off topic to BRN discussion. But I don't do it, because I'm not that low.
Xhosa12345
Regular
Agree 100%. I had one modded yesterday and the reasoning given was incorrect. Someones just being a dick...My post was reported for profanity and personal attacks against management.
@zeeb0t can you boot the person who reports the posts wrongly? Feel like I'm being targeted here, since 3 of my posts got reported yesterday, which eventually got reinstated. So the poster who abuses this system needs to go.
Just because someone doesn't like a post doesn't mean it's violating forum rules.
I stated facts, and if I want to report posts I can do too, hundreds of posts here because most are off topic to BRN discussion. But I don't do it, because I'm not that low.
HopalongPetrovski
I'm Spartacus!
Some interesting insights and worth a listen as two smart Tech guys have a somewhat philosophical conversation about how AGI may eventuate.
7 minute listen.
Sorry if previously posted.
7 minute listen.
Sorry if previously posted.
BaconLover
Founding Member
No.
Can you share/provide evidence where I have made a Brainchip misinformation comment before you accuse me of this?
Otherwise you're welcome to retract your statement and apologise.
My previous post was reported again, simply for replying to you. What's going on @zeeb0t ?
HopalongPetrovski
I'm Spartacus!
Different people from different backgrounds with different interests providing their viewpoints is healthy debate and more likely to throw up and perhaps even generate new knowledge.
That this may involve others having a different view from you, I or even a majority should not be seen as threatening or even as a negative.
Rather, I see it as adding depth to our discussion.
This, the BRN DISCUSSION ONGOING thread is from what I can see the General Thread for all things BRN and so should have the broadest reach and be the least proscribed and restricted.
Mere polite criticism should be able to be discussed and tolerated as part of this discussion ongoing as a relevant concern to an investor and should be able to be argued politely by those who wish to prosecute a view either way.
If some find others viewpoints disturbing I understand they have a right to report them, as I have done on a couple of occasions when new posters launched blatantly obvious attacks or invoked vile accusations towards other posters here, or on the company or its officers.
However, I hope and trust that the mechanism is not too sensitively triggered and has enough intelligence built in to it to also monitor the reporters so its power is not pervertedly used to persecute upon the mere whim of the aggrieved.
One would hope a reasonable threshold has to be reached before action is taken and if any singular or small group of repeat reporters becomes evident this then is made known to the larger group.
I don't think any version of a Star Chamber or other fascist paraphernalia is welcome to the majority here.
That this may involve others having a different view from you, I or even a majority should not be seen as threatening or even as a negative.
Rather, I see it as adding depth to our discussion.
This, the BRN DISCUSSION ONGOING thread is from what I can see the General Thread for all things BRN and so should have the broadest reach and be the least proscribed and restricted.
Mere polite criticism should be able to be discussed and tolerated as part of this discussion ongoing as a relevant concern to an investor and should be able to be argued politely by those who wish to prosecute a view either way.
If some find others viewpoints disturbing I understand they have a right to report them, as I have done on a couple of occasions when new posters launched blatantly obvious attacks or invoked vile accusations towards other posters here, or on the company or its officers.
However, I hope and trust that the mechanism is not too sensitively triggered and has enough intelligence built in to it to also monitor the reporters so its power is not pervertedly used to persecute upon the mere whim of the aggrieved.
One would hope a reasonable threshold has to be reached before action is taken and if any singular or small group of repeat reporters becomes evident this then is made known to the larger group.
I don't think any version of a Star Chamber or other fascist paraphernalia is welcome to the majority here.
Bravo
If ARM was an arm, BRN would be its biceps💪!
Speaking of smart cameras.
In March, Plumerai worked with Renesas to run Plumerai People Detection on the Arm Cortex-M85 processor.
By sheer co-incidence
, in March, BrainChip validated it incorporation with Arm Cortex-M85 Processor.
This blog below states "The Plumerai People Detection runs with an incredible 13 fps on the Renesas Helium-enabled Arm Cortex-M85, greatly outperforming any other MCU solution. It speeds up the processing of Plumerai People Detection by roughly 4x compared to an Arm Cortex-M7 at the same frequency. The combination of Plumerai’s efficient software with Renesas’ powerful MCU opens up new applications such as battery-powered AI cameras that can be deployed anywhere."
And it also states, "This performance uplift will enable our customers to use larger and more accurate versions of Plumerai's People Detection AI, add additional product features, and extend battery life. Our customers have an insatiable appetite for adding new and more accurate AI features that run on a microcontroller"
And...thanks to @Sirod69
In March, Plumerai worked with Renesas to run Plumerai People Detection on the Arm Cortex-M85 processor.
By sheer co-incidence
, in March, BrainChip validated it incorporation with Arm Cortex-M85 Processor.This blog below states "The Plumerai People Detection runs with an incredible 13 fps on the Renesas Helium-enabled Arm Cortex-M85, greatly outperforming any other MCU solution. It speeds up the processing of Plumerai People Detection by roughly 4x compared to an Arm Cortex-M7 at the same frequency. The combination of Plumerai’s efficient software with Renesas’ powerful MCU opens up new applications such as battery-powered AI cameras that can be deployed anywhere."
And it also states, "This performance uplift will enable our customers to use larger and more accurate versions of Plumerai's People Detection AI, add additional product features, and extend battery life. Our customers have an insatiable appetite for adding new and more accurate AI features that run on a microcontroller"
And...thanks to @Sirod69
Last edited:
FlipDollar
Never dog the boys
It’s borderline sickening how this place can be at times. Some people need to explore outside of the echo chamber.
Rooting for BRN, but a lot of posters that were criticised this weekend for venting frustration haven’t deserved it.
There is a slide in one of the very recent Investor presentations that say how many Patents have been issued and how many pending.
BRN employ 2 full time Patent Attorneys'
Bravo
If ARM was an arm, BRN would be its biceps💪!
If we put things into perspective, I consider that we're tracking along quite nicely. I'm going to use MegaChips latest Corporation Briefing FY2022 Q3 (year ending March 2023) as an example.
The first slide shows the MegaChips mid to long-term plan which stretches to 2025 by which time business is "in full swing". Starting and growing any new business is time consuming. The commercialisation process alone involves a series of steps including research, assessment, patenting, prospecting, licensing, ect.
Excluding disclaimers, this presentation totals 15 pages. The last page of the presentation is devoted to the strategic partnership with BrainChip and Quadric.
The first slide shows the MegaChips mid to long-term plan which stretches to 2025 by which time business is "in full swing". Starting and growing any new business is time consuming. The commercialisation process alone involves a series of steps including research, assessment, patenting, prospecting, licensing, ect.
Excluding disclaimers, this presentation totals 15 pages. The last page of the presentation is devoted to the strategic partnership with BrainChip and Quadric.
Last edited:
The Pope
Regular
Thanks for the update manny, much appreciated.
I do recall those but it doesn’t generally answer the questions I have raised. I see FF, Dio and others are reading posts but not replying which is fine as it’s my understanding they are taking a holiday
I’m going to approach Tony Dawe via email and see if he can answer these general questions I have raised.
Bravo
If ARM was an arm, BRN would be its biceps💪!
Lakshmi Narayana Kethamreddy - Vice President & Head - Foundry BU at Samsung Semiconductor India R&D Centre
As well as the BrainChip benchmark post (as above) Lakshmi also liked our 2nd Gen post 1 month ago.Murty Gonella on LinkedIn: BrainChip Readies 2nd Gen Platform For Power-Efficient Edge AI
BrainChip Readies 2nd Gen Platform For Power-Efficient Edge AI
Last edited:
jtardif999
Regular
25 watts is too much for this to be Akida based I think.I find it really interesting that Rob likes that interview @goodvibes. Take a look at Recogni's 1000 TOPS solution below.
In addition, Recogni has announced the availability of the Phoenix ADAS/AD ECU system, an open-source computing platform for autonomous applications. The Phoenix platform combines Recogni’s Scorpio AI inference processor and the Renesas R-Car V4H SoC for configurations up to 2000 TOPS, the highest TOPS per watt capacity on the market, according to Recogni.
Recogni Announces 1000 TOPS Class Solution for Autonomous Driving
NEWS PROVIDED BY
Recogni Inc.
Dec 13, 2022, 09:03 ET
Microprocessor Report Acknowledges Recogni's Superior High Performance Perception Processing at Very Low Power Consumption Over Competitors
SAN JOSE, Calif., Dec. 13, 2022 /PRNewswire/ -- Recogni, Inc., the leader in AI-based perception for autonomous vehicles, today announced Recogni Scorpio, the world's first 1000 TOPS (Peta-Op) class inference solution for autonomous mobility. The company's vision-inference chip enables superhuman object detection accuracy up to - 300m in real-time under various road and environment conditions, and the ability to process multiple streams of ultra-high resolution and very high frame rate cameras. Microprocessor Report noted that Recogni's solution "performs far better" than other inference engines in leading SoCs on the market.
Recogni Scorpio is the world’s first 1000 TOPS (Peta-Op) class inference solution for autonomous mobility
"Vision is fundamental to accurate perception processing and essential to autonomous driving platforms," said RK Anand, Founder and Chief Product Officer at Recogni. "From the beginning we took a unique approach of processing high resolution images at the edge to achieve near-perfect object detection and classification, and enable autonomous driving stacks to make better driving decisions. Scorpio can process multiple 8 megapixel streams at 30 frames per second in less than 10 milliseconds using only 25 watts. That's a performance order of magnitude greater than anything else on the market and, we believe, will help to accelerate autonomous driving to become a reality."
Currently being evaluated by several top tier automotive manufacturers and suppliers, Recogni's solution can achieve 1000 TOPS with less than 10ms of processing delay and below 25 watts of power consumption. This is not only 10-20 times more power efficient than competing solutions, but it enables the flexible design of autonomous driving vehicle stacks and minimizes the impact on driving range. In addition, with such a short processing time of less than 10ms, the Electronic Control Unit (ECU) has more-than-ample time for taking the necessary driving decisions. High compute capacity, efficient processing, low latency, and low power consumption are the pillars of Recogni's platform.
"Recogni's purpose built architecture is a unique approach to AI perception, allowing customers to perceive the environment in high resolution with very low latency and low power - this is a gamechanger for OEMs and suppliers looking to add new, powerful ADAS and self-driving features to new vehicles," said Marc Bolitho, Chief Executive Officer at Recogni. "Recogni's unique approach in perception processing is truly first of its kind and enables customers to deploy safe autonomous driving functions as well as extending the range for electric vehicles."
Industry Validation
Microprocessor Report's Bryon Moyer completed a thorough evaluation of Recogni's solution and reached the conclusion that, "As the industry moves to greater resolution, Recogni is well positioned, whereas other SoCs will need an upgrade." The report also notes that competing SoCs are likely to be that upgrade, but it's "two to three years behind Recogni's first chip."
The report details Recogni's product superiority in terms of supporting high-resolution cameras at high frame rates, support for red/clear/clear/blue (RCCB) image sensors, the approach of minimizing preprocessing before inference, compression algorithms to reduce the required on-chip memory and the product's self-managing capabilities.
Moyer recognized that "no other stand-alone automotive AI accelerator uses high-resolution cameras and processes each full frame (rather than targeting regions of interest. More common are SoCs that handle inference as one of those many functions." This is one of the biggest differentiators and advantages of the Recogni solution.
To download the full version of the Microprocessor Report, please visit www.linelygroup.com.
About Recogni
Recogni develops AI-based vision perception processing to autonomous driving platforms that achieves high compute and low latency with low power consumption. The company was founded in 2017 with offices in San Jose California and Munich Germany. Lead investors are GreatPoint Ventures, Celesta Capital, Mayfield, DNS Capital, as well as notable automotive OEM and tier 1 suppliers including BMW iVentures, Toyota Ventures, Bosch, Continental, Forvia and FluxUnit-OSRAM Ventures. For more information, visit www.recogni.com
Bravo
If ARM was an arm, BRN would be its biceps💪!
Hi @jtardif999, my point was that I think our IP may be in a Renesas chip this year, via Arm® Cortex®-M85 processor, potentially in the Renesas R-Car V4H SoC, which means it would eventually be incorporated with Recogni's product.
IMO. DYOR.
HopalongPetrovski
I'm Spartacus!
In this one He talks about just that. AI creating AI.
Just what all the sci-fi writings warn against and the general fear of autonomous weapons where people's cognitive abilities and biological reaction times render them a critical bottleneck and pragmatically kick them out of the loop.
Similar threads
