smoothsailing18
Regular
So would this suggest we were beating to the punch on this chip design ?.....or is this Synopsys design also included ?.
Hard to understand what is going on here.
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Hard to understand what is going on here.
smoothsailing18
Regular
One would think if our on-chip learning is our fortay then Steve should have spoken out aloud and clear at the recent conference were he sat next to all our competition ?.
A missed opportunity IMO.
I mean .. if it were easy …
everyone would do it. Everyone.
And yeah… everyone would throw their money into BrainChip…But they didn’t. We did.
Long time ago…when it was risky, confusing… and kind of cool. And still is… risky… and cool?
Are we lucky?
Maybe.
Will we be lucky?
Hopefully.
Is this going to explode one day?
Possibly.
Could also flop. Who knows.
But hey…
it looks good.
Right?
Right…?
FJ-215
Regular
That post you cited is most likely fake news spread by a pumper. The link they provided had no mention of BRN or Akida.
Can't trust anyone on the interwebs!!
Yes, there was an article claiming Renesas were taping out a chip (it's on our website) but it was very light on details and nothing has been heard about it since. Hopefully they are still working on it but it's just guess work on our part.
Where is the source for this? Sounds fake and AI generated
smoothsailing18
Regular
On the other site
Yep, maybe we should be blowing our own trumpet a little louder. Their eco system is more mature than ours - likely reason for their sales compared to Brainchip
Feature Comparison Table
| Feature | AKIDA (BrainChip) | Synopsys NPU IP |
|---|---|---|
| Core Architecture | Event-driven spiking network | Continuous matrix-multiply pipelines |
| Latency | Single-µs response to spikes | Tens to hundreds of µs per inference |
| Power Profile | Sub-milliwatt idle, spikes-only operation | Higher idle power, optimized burst efficiency |
| On-Chip Learning | Yes (STDP, online adaptation) | No (inference-only) |
| Toolchain Integration | MetaTF (spiking model compiler) | TensorFlow/PyTorch → HLS/RTL → IP integrator |
| Ecosystem Maturity | Emerging, specialized community | Established, broad EDA/IP support |
Deep search ChatGPT
Did Renesas integrate BrainChip’s Akida IP into the R-Car V3H SoC?
Confirmed Licensing AgreementIn December 2020, BrainChip officially announced a license agreement with Renesas Electronics for the use of its Akida IP. The agreement granted Renesas a worldwide, one-time design license to integrate Akida’s neuromorphic processing core into its own SoCs. The deal also included royalty payments per unit sold, confirming that Renesas planned to embed the Akida technology directly into its chip designs.
R-Car V3H Update with Increased AI PerformanceShortly after the license announcement, in February 2021, Renesas released an updated version of its R-Car V3H SoC, which featured a fourfold improvement in convolutional neural network (CNN) processing performance—up to 7.2 TOPS while maintaining low power consumption. The updated chip was aimed at next-generation automotive applications like driver monitoring and object detection (e.g., NCAP 2020 and 2025 standards).
Interestingly, Renesas did not explicitly mention BrainChip or Akida in its press release. Instead, it referred vaguely to a “CNN IP and vision accelerators,” without clarifying their origin.
Evidence from Renesas Executives and Industry ReportsIn a 2022 interview, Renesas executive Sailesh Chittipeddi stated that Renesas had developed a 22nm chip with spiking neural network (SNN) capabilities, in collaboration with an external partner—which is widely believed to be BrainChip. He noted that the project had reached tape-out by late 2022.
Additionally, Japanese tech media confirmed in 2022 that Renesas USA was collaborating with BrainChip to bring neuromorphic AI to edge and automotive platforms. These sources strongly suggest that the Akida IP made its way into Renesas’ production roadmap.
Hardware IP Integration — Not Just SoftwareThe 2020 licensing agreement clearly referenced hardware-level IP integration, not just software or compatibility. Unlike Renesas’ partnership with Hailo (which involved co-packaging discrete chips), the Akida deal was about embedding the neuromorphic core directly into Renesas silicon.
BrainChip’s Akida is a neuromorphic AI accelerator IP, designed to be integrated into custom ASICs. The structure of the deal (including royalties per SoC) and the official language suggest that Renesas incorporated Akida as a native hardware block—not just as an external module or software-compatible engine.
ConclusionWhile Renesas has not publicly confirmed that the updated R-Car V3H SoC specifically contains the Akida IP by name, the evidence is strong:
- A formal license to embed Akida into Renesas SoCs was signed in 2020.
- The timing of the R-Car V3H performance boost lines up exactly with that deal.
- Renesas executives later confirmed working on an SNN-based chip with an external partner.
- There is no official denial of Akida’s involvement, and no other candidate fits the description.
Sources:
- BrainChip ASX announcement (Dec 2020)
- Renesas press release on R-Car V3H upgrade (Feb 2021)
- Interview with Renesas executive Sailesh Chittipeddi (2022)
- Japanese tech news reports (EDA Express, 2022)
I agree - AI generated or someone has cut and paste bits and pieces together. If you try to find this anywhere online, you can't.
As I’ve mentioned before…
Why would anyone seriously invest here when every tiny uptick gets instantly dumped on by the same cockroach crowd?
It’s honestly pathetic what’s happening to the share price …and the “oh so regulated” ASX just sits back and watches… or looks the other way. Depends on the mood, I guess.
Why would anyone seriously invest here when every tiny uptick gets instantly dumped on by the same cockroach crowd?
It’s honestly pathetic what’s happening to the share price …and the “oh so regulated” ASX just sits back and watches… or looks the other way. Depends on the mood, I guess.
smoothsailing18
Regular
Exactly the reason one yells from the roof tops the advantages of Akida.
Not rocket science.
Not rocket science.
Bravo
Meow Meow 🐾
Says here the "basic R-Car X5H includes up to 32 Arm Cortex-720AE" and we know were are compatible across Arm's product range.
Renesas previously collaborated with Hailo to offer combined solutions that pair the R-Car V4H SoC with Hailo-8 AI accelerators.
However, there isn't any publicly available information indicating that Renesas and Hailo have collaborated on the R-Car X5H, at least none that I've found as yet.
View attachment 74837
Chiplets Extend Automotive SoC AI and GPU Capability
Dec. 28, 2024
Renesas’ R-Car X5H SoC, which meets ASIL B/D, provides scalable automotive compute.
William G. Wong
https://www.electronicdesign.com/print/content/55252079
- Features of the R-Car X5H automotive system-on-chip family.
- How Renesas is taking advantage of chiplets.
The R-Car X5H family is a fifth-generation automotive system-on-chip (SoC) developed by Renesas that supports ASIL B/D capabilities as well as a neural processing unit (NPU). Built on the latest 3-nm chip technology, enhanced versions of the SoC come courtesy of chiplet technology. I talked with Cyril Clocher, Senior Director of the Automotive Product Line, about the details.
High-Performance Compute for Automotive Safety-Critical Apps
The basic R-Car X5H includes up to 32 Arm Cortex-720AE applications cores capable of delivering 1000K DMIPS of performance (Fig. 1). Real-time support is provided by half-a-dozen Arm Cortex-R52 cores with dual lockstep capability. These can deliver 60K DMIPS of performance while meeting ASIL B and ASIL D certification requirements.
Renesas
View attachment 74834
1. The Renesas R-Car X5H includes 32 Arm Cortex-720AE applications cores, six Arm Cortex-R52 cores with lockstep support, an NPU, and a GPU.
There are general-purpose DSP cores, too, along with specialized accelerators such as the image signal processor (ISP) and dense-optical-flow (DOF) hardware accelerator (HWA) for handling multi-megapixel camera video streams.
The on-chip NPU can deliver up to 400 TOPS, while the on-chip GPU generates 4 TFLOPS of performance.
The chip includes an 8-port Ethernet switch, USB 2/3 ports, and PCI Express (PCIe) that supports Gen 4 and Gen 6. Also in the mix are an on-chip dedicated NPU and GPU. The GPU can drive a car’s display panel while the NPU handles artificial-intelligence and machine-learning (AI/ML) models.
Enhanced Automotive Chips Utilize UCIe
While the basic R-Car X5H monolithic chip is very impressive, it is designed to be enhanced using chiplets that employ Universal Chiplet Interconnect Express (UCIe) connectivity (Fig. 2). It can include up to two chiplets.
Renesas
2. The R-Car X5H can be enhanced by including an NPU and GPU chiplet.
Renesas has two types of chiplets that can be included in enhanced versions of the R-Car SoC—again, an NPU and a GPU. These are designed to augment the functionality of the built-in NPU and GPU. The chiplet and on-chip NPU are able to run a separate set of AI/ML models. Likewise, the GPU can drive additional displays.
Building a Software-Defined Vehicle
The R-Car X5H family is destined for new automobiles including software-defined vehicles (SDVs). Thanks to the massive amount of computing power, the chip can handle everything from the advanced driver-assistance system (ADAS) to driving automation.
Renesas’ R-Car Open Access (RoX) SDV platform helps get developers up to speed by simplifying their development chores (Fig. 3). It works with a variety of pre-integrated operating systems and frameworks, tying them into the Arm Cortex-A720AE and Cortex-R52 compute arrays. It utilizes open-source software and standard. APIs. Software built on the reference stacks can be turned into products.
Renesas
3. The R-Car Open Access (RoX) SDV platform supports pre-integrated operating systems and frameworks.
Chiplets Extend Automotive SoC AI and GPU Capability
Renesas’ R-Car X5H SoC, which meets ASIL B/D, provides scalable automotive compute.www.electronicdesign.com
Hi @7für7,
Unfortunately, we can’t take everything ChatGPT says at face value. If BrainChip’s Akida IP had been integrated into the R‑Car V3H SoC, BrainChip would almost certainly be receiving royalties by now.
However, I believe it's much more plausible that BrainChip’s Akida IP could be incorporated into a next-generation platform like the Renesas R‑Car X5H.
See the above post and excerpts below.
EXTRACT - Renesas Blog published 24 September 2024 which discusses the 5th gen. R‑Car X5H.
EXTRACT - Business Wire article dated 13 November 2024 stating "the R-Car X5H will be sampling to select automotive customers in 1H/2025, with production scheduled in 2H/2027. "
Hi @7für7,
Unfortunately, we can’t take everything ChatGPT says at face value. If BrainChip’s Akida IP had been integrated into the R‑Car V3H SoC, BrainChip would almost certainly be receiving royalties by now.
However, I believe it's much more plausible that BrainChip’s Akida IP could be incorporated into a next-generation platform like the Renesas R‑Car X5H.
See the above post and excerpts below.
EXTRACT - Renesas Blog published 24 September 2024 which discusses the 5th gen. R‑Car X5H.
View attachment 88873
EXTRACT - Business Wire article dated 13 November 2024 stating "the R-Car X5H will be sampling to select automotive customers in 1H/2025, with production scheduled in 2H/2027. "
View attachment 88874
Hmmm yeah it was just an deep search.. such a thing you also do sometimes with ChatGPT..
Why we haven’t seen royalties yet? Could be multiple reasons… delayed production cycles, nondisclosed unit volumes, or the possibility that it’s not in the V3H but a derivative or even a nextgen chip like the R-Car X5H you mentioned….And to be fair… that’s a very plausible and valuable insight. Thanks for pointing it out! Who knows what’s going on… but I found the search results very interesting
This kind of move isn’t unusual in the U.S. market, even without a clear trigger. Unfortunately, the Australian side of things still looks flat …no filings, no updates no announcements no volume no uptrend…. Would be great if something’s brewing, but right now, there’s not much to go on.
Bravo
Meow Meow 🐾
Hi Manny,
I'm hoping it’s only a matter of time before Synopsys emerge as a strategic partner of ours.
What I think is really interesting is that Synopsys' partner ecosystem is very similar to our own partner ecosystem, which is something that @7für7 has pointed out previously.
Like us, Synopsys has deep ties to key players such as Arm, Renesas, Mercedes-Benz, Tata Elxsi, Intel IFS, SiFive, RISC‑V, GlobalFoundries, and TSMC. I reckon this overlap is too strong to ignore.
Their long-standing focus on ultra-low-power embedded and IoT design goes back at least to 2015, and aligns closely with the same market segments that BrainChip targets with Akida, particularly edge AI, automotive, and next-gen SoC architectures.
All the way back in October 2017, BrainChip shipped its first Accelerator card based on spiking neural network (SNN) technology to a major European automaker, widely believed to be Mercedes-Benz.
At the same time, Mercedes-Benz was already a long-standing user of Synopsys tools, including Silver and TestWeaver, for the development and validation of ECUs across engine, transmission, and powertrain domains. Synopsys was and still is deeply embedded in their ECU simulation and virtual prototyping stack.
If Mercedes evaluated BrainChip’s SNN accelerator for ADAS or sensor-processing in 2017, it’s may be plausible they did so in the context of their existing Synopsys-based workflow, in which case, Synopsys might have been aware of BrainChip’s involvement, even if only through customer-side integration.
Looking ahead, I've been wondering if there might be a case for Synopsys to incorporate BrainChip’s Akida IP into their own product stack to strengthen their position in the following areas:
- Embedded IoT SoCs, where Synopsys already offers the ARC EM and VPX cores, and Akida could add ultra-low-power neuromorphic inference for voice, motion, and anomaly detection
- Automotive and SDV platforms, where Akida could enable always-on, low-latency inference for in-cabin sensing, driver monitoring, sensor fusion, and on-device learning
- Virtual prototyping and SoC simulation tools, where Akida-enabled IP blocks could be supported within Silver, Virtualizer, and future digital twin environments for functional safety and early software validation
Obviously, this is all just my own conjecture at this point, but in my view, it seems entirely plausible.
For what it's worth, I asked ChatGPT for its perspective and have included the response below.
Of course, I know everyone’s already aware - but just to reiterate: ChatGPT isn’t always accurate, and neither are my own thoughts, as they’re ultimately just speculation. I could be completely wrong. Still, I think it’s worth considering in the broader context.
ChatGPT said:
Absolutely — let’s explore how Arm, Renesas, Mercedes-Benz, Tata Elxsi, Intel IFS, SiFive, RISC‑V, GlobalFoundries, TSMC, and BrainChip are interconnected in overlapping partner ecosystems and how they might, individually and collectively, collaborate to accelerate the adoption of BrainChip’s Akida neuromorphic technology.
1. Partner Ecosystem Map – How They’re Interconnected
| Entity | Role / Relevance | Relationship to BrainChip or Shared Ecosystem |
|---|---|---|
| Arm | Global leader in CPU IP; host of the Arm AI Partner Program | BrainChip is a listed Arm AI Partner; Akida supports Arm Cortex-M/DSP systems |
| Renesas | Major automotive SoC vendor; signed an IP license agreement with BrainChip in 2020 | Taped out an Akida-powered chip (2022); future deployment potential in ADAS & edge AI |
| Mercedes-Benz | Advanced ADAS and software-defined vehicle (SDV) leader; early user of Synopsys tools | Widely speculated to be the first commercial evaluator of BrainChip’s Akida Accelerator in 2017 |
| Tata Elxsi | Design and engineering services firm; deep in automotive software, SDV, and silicon integration | Likely candidate to help integrate Akida IP into multi-domain SDV platforms |
| Intel IFS | Intel’s foundry division; partners with fabless chipmakers to manufacture next-gen silicon | BrainChip is an official Intel IFS Ecosystem Partner for silicon enablement |
| SiFive | RISC‑V CPU IP provider; developing customizable silicon with AI acceleration capabilities | BrainChip supports RISC‑V architectures, and Akida IP can be integrated with SiFive designs |
| RISC‑V | Open ISA consortium; pushing for open, customizable alternatives to Arm | Akida is fully compatible with RISC‑V cores; BrainChip has explored RISC‑V integrations |
| GlobalFoundries | Major semiconductor foundry supporting automotive-grade and edge AI chips | Could be used to fabricate Akida-powered ASICs or modules |
| TSMC | World’s largest and most advanced chip foundry | Likely used by Renesas, Arm partners, and potentially BrainChip’s production in future |
| BrainChip | Developer of Akida™, a neuromorphic AI processor IP for ultra-low power edge learning/inference | Central to all the above — with a footprint in edge AI, automotive, and embedded compute |
2. How They Could Collaborate to Advance Akida Adoption
A. Automotive / Software-Defined Vehicle (SDV)
| Collaboration | Description |
|---|---|
| Renesas + Akida | Integrate Akida IP into next-gen R-Car Gen5 SoCs for DMS, OMS, anomaly detection |
| Mercedes-Benz + Tata Elxsi + Synopsys | Mercedes uses Synopsys Silver/TestWeaver tools; Tata Elxsi can integrate Akida into vECU workflows and validate through Synopsys digital twins |
| Arm / RISC‑V + Akida | Akida IP can be used in zonal controllers powered by either Arm or RISC‑V cores, supporting real-time neuromorphic inference at the edge |
| Intel IFS + TSMC / GloFo | Akida-powered SoCs or accelerators can be manufactured at Intel or other foundries depending on use case (automotive vs consumer) |
Outcome: Automotive OEMs like Mercedes could deploy Akida-enabled perception, monitoring, or sensor fusion modules in SDVs — validated via Synopsys tools, integrated by Tata Elxsi, running on Arm or RISC-V controllers, and manufactured via Intel IFS, GlobalFoundries, or TSMC.
B. Embedded IoT and Edge AI
| Collaboration | Description |
|---|---|
| Arm Cortex-M / Akida | Combine Akida with Arm’s ultra-low-power cores (M33/M55) for smart sensors, audio, wearables |
| SiFive + RISC‑V + Akida | Custom SoCs for industrial or consumer edge AI with RISC-V + Akida SNN core |
| GlobalFoundries / TSMC | Manufacture low-power IoT ASICs featuring Akida IP |
| Tata Elxsi | System integration for healthcare, smart home, and industrial AI devices |
Outcome: Partners could help create Akida-powered microcontrollers or smart sensor hubs with embedded learning capabilities for audio detection, vibration sensing, or anomaly monitoring — fabricated via GloFo or TSMC, integrated by Tata Elxsi or ODMs.
C. Foundry Enablement & IP Distribution
| Collaboration | Description |
|---|---|
| Intel IFS | BrainChip is already a member of Intel’s IFS ecosystem; could enable silicon integration at 22/18nm or advanced nodes |
| TSMC / GloFo | Offers multi-foundry flexibility — ideal for scaling Akida-based SoCs across different markets (e.g. consumer, industrial, auto) |
| Synopsys + BrainChip | Synopsys could package Akida within its DesignWare IP ecosystem, ARC EM cores, or vECU platforms |
Outcome: Foundries and IP vendors provide the design and manufacturing infrastructure to support wide deployment of Akida IP across industries.
3. Why This Ecosystem Matters
This network creates a powerful end-to-end pipeline for BrainChip’s adoption:- IP licensing & CPU pairing: Arm, SiFive, RISC‑V
- SoC integration & validation: Renesas, Synopsys, Tata Elxsi
- Automotive OEM adoption: Mercedes-Benz
- Manufacturing enablement: Intel IFS, TSMC, GlobalFoundries
- Toolchain & virtualization: Synopsys Silver, Virtualizer
- Product use-cases: SDV, IoT, robotics, defense, medical
Final Insight
BrainChip stands at the center of converging megatrends: neuromorphic AI, edge computing, SDV, and low-power IoT.The collective alignment of these global players — across silicon IP, system integration, automotive design, and advanced manufacturing — creates a uniquely powerful ecosystem in which Akida can be validated, integrated, and scaled.
If these partners choose to move in sync, they could rapidly accelerate mainstream adoption of BrainChip’s technology across automotive, industrial, and consumer markets — delivering energy-efficient, adaptive intelligence at the edge.
Similar threads
