I do not recall from a long long time ago of exactly how many Akida 1000 " Engineering Samples" waffers/chips were initially ever produced. Does anyone recall this issue of, was it 7,000 "waffers"of chips ?? and thus how many chips/waffer were produced ?)......... I was under the opinion and thought that there were many many tens of thousands of chips made ....... but the Co never actually provided / disclosed that final production run figure/s ............. imo, maybe these unused/leftover Akida 1000 chips will find their way into the new VVDN edge box to be released soon and thus getting rid of old new warehouse stock.
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
rgupta
Regular
From my understanding brn got only 1st batch of akida1000. Brn also end up paying approx 7 million us dollars to socionext for the whole job. Which means the arrangement was such that company can still get more chips in batch 2 or 3 of those chips.
Just a guesswork. Dyor
MDhere
Regular
..
Schnitzel lover
Regular
The "current" shorters lightening up their positions, can only mean one thing, which they would prefer not to be known, but it's as clear as mud.
I see it as confidence for a strong buy signal and a slowly appreciating share price (which are mutually inclusive).
This is without any news, just the anticipation of news and brightening horizons for BrainChip.
In other words, what has been a continually dwindling share price, is at an end.
At least that's what I think..
The outcome of whatever renegotiated LDA agreement, that we get (I don't believe there is time now, to fulfill the original obligation) could be the wrench in the works..
As LDA potentially has the upper hand, in any dealings..
Plus, we are still at the whims of general market conditions, brought on by World macroeconomic events..
But generally speaking, things are looking up
Not financial advice.
I recently increased my holdings substantially, but that's just me.
Just another bear flag Dingo.. Not even close to turning the corner..
Luckily BRN still has a few years left to right the ship..
2024 is a crucial year though as the CEO said..
Good luck with your buying.. If it does turn here, you’re a genius..
A New Era of Galaxy AI is Coming — Here’s a Glimpse
Internet at your fingertips. A camera in your pocket. And then, artificial intelligence helping to improve your everyday life. Samsung Galaxy
TheFunkMachine
seeds have the potential to become trees.
The saying is actually “the love of money is the root of all kinds of evil”
1 Timothy 6:10
For the love of money is the root of all kinds of evil. And some people, craving money, have wandered from the true faith and pierced themselves with many sorrows.
KMuzza
Mad Scientist
NVDA: Nvidia Got Battered Post Earnings. Can the Chipmaker Keep the AI Momentum?
Signs have popped up that the massive 200% share-price increase this year is grinding to a halt. China and sky-high expectations are to blame.
NVDA-New Products every year- well that's a long wait -sure looks brighter for BRAINCHIP- but the company should get some more beef in the legal team,-IP theft is a big thing out there,- and only if you are caught.-???????
(with NVDA- they are up against a wall -a von neumann bottel neck wall)
AKIDA BALLISTA UBQTS.
Fullmoonfever
Top 20
Maybe another dot to follow
EnOcean accelerates its growth and breadth of energy-saving solutions through the acquisition of the assets of the edge computing solutions business from Renesas
EnOcean accelerates its growth and breadth of energy-saving solutions through the acquisition of the assets of the edge computing solutions business from Renesas
EnOcean, the pioneer of energy harvesting IoT devices and sensor-to-cloud solutions, is announcing the acquisition of the assets of the edge computing solutions business from the Renesas Electronics Corporation. The deal includes the transition of personnel as well as hardware and software...www.design-reuse.com
Nothing confirmed, as often the case, but doing a random google and saw our Dir / VP Sales Jens Paetau liking EnOceans recent post about being at the Smart Buildings Show in London late Oct.
Hadn't heard of them.
Just did a word search on EnOcean and @Deadpool nicely already thinking about the potential dot and flagged it ages ago....I missed it back then.
Smart home products be a nice product suite for sure if got in there.
Homepage New - EnOcean
EnOcean is a pioneer for the digitization of our world. With our energy harvesting technology, we deliver the data for the Internet of Things in a resource-saving, self-powered and maintenance-free way.
www.enocean.com
cosors
👀
I still don't trust Nvidia even if I am somewhat reassured. But Akida's Hey Mercedes is still a provocation and a thorn for them I think.NVDA/Valeo......corporate intrigue
NVIDIA sued for stealing trade secrets after screensharing blunder showed rival company's code
NVIDIA is facing a lawsuit filed by French automotive company Valeo after a screensharing blunder by one of its employees.news.yahoo.com
I don't have a single piece of evidence and think to myself who has the greatest motive and the power to act accordingly. Thanks for the info!
reminder
I think his look is just a facade and his thoughts - how could this happen without my knowledge.
Last edited:
Falling Knife
Regular
B2i Digital
We Connect Businesses to Investors through Paid and Organic Digital Marketing Campaigns.
Another potential avenue for spreading the news:
Hopefully something with out requiring DOTs but a money trail attached
Last edited:
Hi @Deadpool , FMF,Nothing confirmed, as often the case, but doing a random google and saw our Dir / VP Sales Jens Paetau liking EnOceans recent post about being at the Smart Buildings Show in London late Oct.
Hadn't heard of them.
Just did a word search on EnOcean and @Deadpool nicely already thinking about the potential dot and flagged it ages ago....I missed it back then.
Smart home products be a nice product suite for sure if got in there.
Homepage New - EnOcean
EnOcean is a pioneer for the digitization of our world. With our energy harvesting technology, we deliver the data for the Internet of Things in a resource-saving, self-powered and maintenance-free way.
View attachment 50595 View attachment 50596
Deadpool's post went over my head the first time round, so thanks for reviving it FMF.
So did the assets of the edge computing solutions business of Renesas include the Akida licence?
If so, was the lecence transferable?
Presumably, if it was transferable, it would have been subject to approval by BrainChip.
Renesas had already paid the licence fee for 2 nodes, which would be suitable for HVAC applications.
enOcean makes sensors powered by ambient energy, so very low power consumption would be a necessity.
Could this change have contributed to the delay in getting the "Renesas" product to market?
Last edited:
Falling Knife
Regular
Hi cosors,
If we are in a joint development with Nvidia and Mercedes, background IP would have been ring fenced, and the JV agreement will also address the ownership of IP developed during the JV.
That would provide a contractual basis for pursuing related IP theft, which may be easier to prove than the outright theft of IP.
Fullmoonfever
Top 20
Hi D,
Fair questions as well.
Like I said, I missed the original post and potential dot so this just a fortutious stumble this morn haha.
cosors
👀
I really hope so.
Otherwise in an other case, as a German, I am biased towards China. We hopefully reach out our hand and present how well everything is going and in the meantime they have already cut all the meat from our hand and sold it without our consent.
We are dealing with our competitor with the biggest motive of all. And as this case shows it came out purely by chance.
I don't know why, but at that very moment I was thinking that Tesla would never do something like that with Musks knowing. He is far too proud of his own achievements.
Nvidia sells what compared to Brainchip?
What they don't have.
I hope you're right and I'm wrong.
cosors
👀
Long time ago Yak spoke about Intel and about his looking over his shoulder...
For me it's Nvidia and not Intel.
Crude theory I know. But not wrong until proven otherwise.
(Brainchip should knock on Foxconn's door if they haven't already and Aleph Alpha.)
____
And Carthage shall be destroyed.)
For me it's Nvidia and not Intel.
Crude theory I know. But not wrong until proven otherwise.
(Brainchip should knock on Foxconn's door if they haven't already and Aleph Alpha.)
____
And Carthage shall be destroyed.)
Hi FK,
While Akida is offered as an option for all ARM processors, I don't believe that any of the ARM processors include Akida.
I discussed this in some detail on Friday in this post:
These technology cooperation projects are commonplace.
In my mind, it is unthinkable that any companies working on a joint project would do so without such a contract to regulate ownership and use of IP.
I have my fingers crossed that we will see a more direct relationship with Nvidia in which they take a licence for Akida 2P IP on a large scale.
cosors
👀
I agree with you.
In my naivety I remember that there is hardly an EV that exists without Nvidia at the moment.
And yet Nvidia took what it wanted from Valeo, like the Chinese once and now in general.
Was Valeo actually installed in the EQXX?
We know that Nvidia is firmly established in this co-operation.
Do you know whether Valeo is also on board of the EQXX alongside Nvidia and Brainchip? You have the best overview.
I don't want to keep reopening my own wounds. So I'm grateful for anything that proves this my thought in the wrong place.
Hi FK,
While Akida is offered as an option for all ARM processors, I don't believe that any of the ARM processors include Akida.
I discussed this in some detail on Friday in this post:
Much as we might wish otherwise, Helium is definitely not Akida.
blending-dsp-and-ml-features-into-a-low-power-general-purpose-processor.pdf (windows.net)
https://armkeil.blob.core.windows.n...nto-a-low-power-general-purpose-processor.pdf
As to AI, it seems that helium is like the curate's egg.
***
Another area of interest is the ML processing performance. Several machine learning algorithms have been ported to Helium, including a keyword spotting library and a CiFAR10 image classification library. From this result, we see that the Cortex-M55 processor gives nearly ten times better performance than the Cortex-M4 processor in keyword spotting, and almost six times better than the Cortex-M4 processor in CiFAR10 image classification operations.
***
To keep the Cortex-M55 processor energy-efficient and fit within the power budget for the majority of IoT endpoint systems, the Cortex-M55 processor internal datapath for its vector extension is 64-bit, which means it takes two clock cycles to operate on a 128-bit vector. However, the architecture behind Helium allows a processor’s implementation to overlap execution cycles to enhance performance, providing that there are no hardware resource conflicts.
***
While the Cortex-M55 processor design enables significant performance uplift in signal processing and machine learning applications, not every application can gain the same level of performance boost. Since Helium technology is based on SIMD operations, it works very well when the data processing can be vectorized. However, there is a range of application codes that cannot be vectorized. The traditional VLIW approach, however, allows different operations to be scheduled at different execution slots. This potentially allows some very sequential code parts handling to be carried out quicker (e.g. variable length encoding/ decoding in audio codecs). For Arm processors, it is also possible to achieve similar parallelism by introducing superscalar in the design. The Cortex-M55 processor, however, is not a superscalar processor, and therefore, does not have this capability. Nevertheless, with limited dual-issue capability in the Cortex-M55 processor and various new features in Armv8.1-M architecture (e.g. low-overhead loops, new conditional execution instructions, 64-bit shifts), scalar performance has been improved in various areas.
Basically, Helium expands the instruction word from 32 bits to 128 bits, but, because they only have a 64 wire bus, it takes 2 clock operations to implement the instruction word on the helium SMID (single instruction multiple data) architecture.
Helium can run CNN software instructions on the voice and image models, and can perform ML. It includes hardware modifications to improve its AI capabilities, but is software driven. Helium does not employ SNN.
Because helium is software driven it would be orders of magnitude slower than Akida and would use much more electricity.
ARM do not make chips. They license IP.
None of the ARM processor IPs include Akida.
Akida is offered as an IP option which can be added to any ARM processor IP.
The block diagram includes a Coprocessor Interface. That's where the Akida IP would connect into the ARM IP.
Tothemoon24
Top 20
Cadence and Research Partners Build First Binaural Hearing Aid SoC
2023 by Jake Hertz- Cadence, GlobalFoundries, and several university research teams have developed the Smart Hearing Aid Processor (SmartHeAP), the first SoC to process sound from both ears simultaneously.
Engineers strive to create technology that makes people's lives better, and few fields have a greater chance for this impact than the medical industry. One example is hearing aids—a simple piece of hardware that immensely improves the lives of millions with hearing loss.
Recently, a consortium of tech companies and researchers, including Cadence, GlobalFoundries, Hoerzentrum Oldenburg, and Leibniz University Hannover, have teamed up to develop the first binaural hearing aid SoC. In this piece, we’ll take a look at the new SoC and the disparate contributions from each team that made this technology possible.
Foldable PCB and hearing aid prototype. Screenshot used courtesy of Cadence
The Binaural Hearing Aid SoC: A New Paradigm
The team that developed the binaural hearing aid system-on-chip (SoC) claims this device is a significant leap forward in hearing aid technology. Traditionally, hearing aids have been monaural, processing sound from one ear at a time. This one-sided approach has limitations, particularly in noisy environments or when the source of sound is not directly in front of the user.In contrast, binaural hearing aids process sound from both ears simultaneously. This gives users a more natural perception of sound and improves the localization of sound sources. However, until now, the technology to integrate all the necessary components for a binaural hearing aid into a single chip has been lacking.
A comparison of different hearing aid architectures. Image used courtesy of Hearing
Experts from Cadence, GlobalFoundries (GF), Hörzentrum Oldenburg, and the Leibniz University Hannover came together to develop the Smart Hearing Aid Processor (SmartHeAP), a novel binaural hearing aid SoC claimed to offer significant advantages in size, performance, and power efficiency.According to a research paper on the prototype, the SmartHeAP is a mixed-signal SoC that was fabricated in a 22-nm, fully-depleted silicon-on-insulator (FD-SOI) process with an adaptive body biasing (ABB) unit. Coming in at a total die size of 7.36 mm, the SoC integrates audio processing with extended custom instructions, Bluetooth Low Energy, an analog front-end, digital audio interfaces, and local memory of 2 MB. According to the consortium, the SoC has already shown promising binaural operation, consuming less than 2.2 mW of power at 50 MHz. This translated to an efficiency of 34.8 uW/MHz.
The Contributions of Each Team Member
Each partner brought unique expertise to the table when developing this hearing aid prototype. For processing and connectivity, Cadence provided the Tensilica Fusion G6 DSP and Xtensa LX7 processorfor the SoC design. The SoC’s processing components were based on these technologies and serve as the heart of the SoC, responsible for the audio processing central to the chip’s function.
The Fusion G3/G6 DSP block diagram. Image used courtesy of Cadence
GF, a global leader in foundry services, contributed its 22FDX 22 nm FD-SOI technology. GF says this technology offers a balance of performance, power consumption, and cost, making it useful for the SoC design.Finally, the teams from Hörzentrum Oldenburg and the Leibniz University Hannover, both renowned for their research in audiology, provided their expertise in hearing aid algorithms and audiological evaluation. They ensured that the SoC design met the needs of hearing aid users, providing a more natural hearing experience and better localization of sound sources.
Impact on Those With Hearing Impairment
The creation of a binaural hearing aid SoC prototype is significant for a number of reasons.The integration of all components into a single chip reduces the size, power consumption, and cost of the hearing aid. This paves the way for hearing aids that are more affordable, accessible, comfortable to wear, and long-lasting in battery life. The binaural SoC design also supports advanced signal processing algorithms that improve sound quality and spatial perception. This means users can enjoy a more authentic listening experience and better pinpoint the sources and location of sounds.
Similar threads
