BRN Discussion Ongoing

[FabricatedLunacy]

Tapeout is when its ready to be sent off and made,
We have 1500 back already, ........ its been recieved by BRN

Thanks, so I did somewhat get it right then in my head

 

[mrgds]

Thanks, so I did somewhat get it right then in my head

Its like a UFC fight,
Once BRNs competitors realise our supreme dominancy, ...................... they too will "Tape-out"

 

[Rise from the ashes]

Seems like artificial sell stacking still happening to try force price back down. We started off really good with the buying but it's so clear there are bigger entities manipulating. Look at the stacking of that 0.285, I have my camera on you. Screenshots have been taken and I'll send to ASIC. Warning to manipulators.

Good luck with that many have tried and failed before even with more damning evidence of epic proportions.
Asic known as the toothless tiger.

 

[FabricatedLunacy]

Its like a UFC fight,
Once BRNs competitors realise our supreme dominancy, ...................... they too will "Tape-out"

Let us hope they actually do realise it... and not just drive past the bus on the side of the road

 

[Esq.111]

Afternoon FabricatedLunacy ,

My understanding is BrainChip has the PHYSICAL silicone in hand ,......Let the games commence.

Exciting times.

Regards,
Esq.

 

[Rise from the ashes]

Afternoon FabricatedLunacy ,

My understanding is BrainChip has the PHYSICAL silicone in hand ,......Let the games commence.

Exciting times.

Regards,
Esq.
View attachment 43318

Cue dildonics corporation.
You just made me laugh harder than I have laughed in ages.

 

[Esq.111]

Afternoon Rise ,

Yes , the opportunity was to good to miss .....

Regards,
Esq.

 

[FJ-215]

Only skimming past posts today.

How/what have they gotten wrong? Thanks.

G'day Fab,

Seems someone has posted the press release from January about the Tape Out of AKD1500 instead of the lastest press release about BRN receiving the AKD 1500 chips and their availability to customers in the next quarter.

Is it JPR Communications handling this on our behalf?
They need to do better!

 

[Bravo]

Afternoon FabricatedLunacy ,

My understanding is BrainChip has the PHYSICAL silicone in hand ,......Let the games commence.

Exciting times.

Regards,
Esq.
View attachment 43318

But why are her eyes crossed? And why does she have a band-aid on her thumb? So many questions...

 

[FabricatedLunacy]

Afternoon FabricatedLunacy ,

My understanding is BrainChip has the PHYSICAL silicone in hand ,......Let the games commence.

Exciting times.

Regards,
Esq.
View attachment 43318

That's the excited face we make everytime there is perceived good news... only we are yet to see the happy ending

 

[Bravo]

Well that's not half bad!

 

[Esq.111]

Afternoon Bravo ,

One would presume Bench testing new physical silicone is a long, tedious & some times dangerous process...
I wish all our partners well... Its now time to drive it home to our competitors.

Hope we all got a little laugh ......certainly need one.


Regards,
Esq.

 

[Diogenese]

Depends on which type of products I think, automotive definitely needs us now whereas e.g. cameras that perform simple tasks such as counting or simple types of security could probably get by using a software+cpu solution like Plumeai provide. Whitegoods ai might be in the same boat, the current hardware is probably good enough for the use cases out there now. As use cases get more sophisticated, then we become a more attractive proposition. As others have said we should remember the likes of Renesas and Socionext bringing out MCUs that can slot into products in the not too distant future. AIMO.

That's right. As long as we don't have a drop-in chip, we exclude a large share of the market who can't afford the IP licence fee and royalties.

The Renesas chip is very light weight with only 2 nodes (8 NPUs), so this may pick up some of the lower end of the market.

 

[Rise from the ashes]

Hi @Learning,

It's interesting that the platinum partners are Cadence, Siemens and Synopsy.

Remember the BrainChip job ad for an IP Verification Engineer in Hyderabad before it was taken down which also mentioned Cadence, Siemens and Synopsys. Probably just some random and irrelevant co-incidence.

In addition to this there's also this to consider...

• #44,987
• #40,385

View attachment 43305

I'm glad you put those two together otherwise it would have come to me during sleep then wake myself up screaming out cadence , Siemens and synops.

 

[mrgds]

Interesting "turn in market today"

After seeing Management/BOD admit in recent Report .............. ""THIS IS NOT ACCEPTABLE"

I wonder if they have taken my suggestion of;
PUT A BLOODY STOP TO THIS S/P DESIMATION AND ALL OF YOU START BUYING ON MARKET,
AND DONT FORGET TO INCLUDE "OL MATE MANNY" ..........

Probably not,

 

[FJ-215]

Afternoon Bravo ,

One would presume Bench testing new physical silicone is a long, tedious & some times dangerous process...
I wish all our partners well... Its now time to drive it home to our competitors.

Hope we all got a little laugh ......certainly need one.


Regards,
Esq.

Was that a new "Medical Device".

Human trials to commence immediately

 

[Rise from the ashes]

Was that a new "Medical Device".

Human trials to commence immediately

They are looking to penetrate markets globally.

 

[jtardif999]

Hi Cartagena,

I don't believe there is any evidence that Qualcomm uses Akida, although I wish they did.

They have their in-house Hexagon:

https://www.qualcomm.com/news/onq/2...ile-computing-performance-for-windows-laptops


https://www.theregister.com/2022/11/15/qualcomm_snapdragon_8_gen_2/?td=readmore

Qualcomm pushes latest Arm-powered Snapdragon chip amid bitter license fight

The Snapdragon 8 Gen 2 system-on-chip features eight off-the-shelf cores from Arm, which is locked in a bitter legal fight with Qualcomm over licenses and contracts.
...
This includes an AI acceleration engine that is, we're told, up to 4.35 times faster than the previous generation, and with a potential 60 percent increase in performance-per-watt, depending on how it's used. This unit can be used to speed up machine-learning tasks on the device without any outside help, such as object recognition, and real-time spoken language translation and transcription. The dual-processor engine can handle as low as INT4 precision for AI models that don't need a lot of precision but do need it done fast on battery power, which the 4-bit integer format can afford developers, according to Qualcomm.

Qualcomm is pushing the INT4 capabilities as a precision ideal for modern mobile apps. It said a cross-platform Qualcomm AI Studio is due to be made available in preview form in the first half of next year that will optimize developers' models for this precision as well as other formats. This studio looks like a typical IDE in which programmers can organize their training workflows.
...
The SoC supports up to 200MP image capture and 8K HDR video capture in 10-bit HDR, according to the specifications. Qualcomm said it worked with Samsung and Sony to develop large sensors that the 8 Gen 2 can handle. There are direct paths in the chip that link the Hexagon AI engines to, say, the image-processing units so that pictures and video can be manipulated more efficiently.

The processor, according to Qualcomm, can also be made to reduce the amount of data read and written during neural network inference – which saves power – by breaking input data into not just tiles as other chipsets do, but micro tiles that apparently do a better job of cutting down information transfer.

This is from a quote in a now-deleted Stable genius post on TSEx:

Qualcomm uses all of the Snapdragon SoC’s processing elements for AI processing and calls the combination of these processing elements the “AI engine.” Among the enhancements incorporated into the AI engine was a dedicated power plane and a doubling of the tensor processing cores within the Hexagon processor. The result is a 4.35x improvement in performance and an equally impressive 60% improvement in performance per watt efficiency. Qualcomm also added support for transformer neural network models which are critical for applications like natural language processing (NLP) for speech-to-text and text-to-speech translation. The Hexagon can splice the neural NLP model into smaller elements to run on micro tiles allowing for more efficient use of the processing cores. Additionally, Qualcomm added support for Int4 data structures. In many cases, this lower precision data structure can be used by neural network models like computational photography image enhancement without a noticeable loss of accuracy while improving speed and power efficiency. The end result is faster and more efficient processing of neural network models.


This is from a 2021 Qualcomm patent application:

WO2023049655A1 TRANSFORMER-BASED ARCHITECTURE FOR TRANSFORM CODING OF MEDIA 2021-09-27


View attachment 43051

Systems and techniques are described herein for processing media data using a neural network system. For instance, a process can include obtaining a latent representation of a frame of encoded image data and generating, by a plurality of decoder transformer layers of a decoder sub-network using the latent representation of the frame of encoded image data as input, a frame of decoded image data. At least one decoder transformer layer of the plurality of decoder transformer layers includes: one or more transformer blocks for generating one or more patches of features and determine self-attention locally within one or more window partitions and shifted window partitions applied over the one or more patches; and a patch un-merging engine for decreasing a respective size of each patch of the one or more patches.

...
[0075] The SOC 100 may also include additional processing blocks tailored to specific functions, such as a GPU 104, a DSP 106, a connectivity block 110, which may include fifth generation (5G) connectivity, fourth generation long term evolution (4G LTE) connectivity, Wi-Fi connectivity, USB connectivity, Bluetooth connectivity, and the like, and a multimedia processor 112 that may, for example, detect and recognize gestures. In one implementation, the NPU is implemented in the CPU 102, DSP 106, and/or GPU 104. The SOC 100 may also include a sensor processor 114, image signal processors (ISPs) 116, and/or navigation module 120, which may include a global positioning system.



Snapdragon8_2Hexagon

Snapdragon 8 Gen 2 deep dive: Everything you need to know (androidauthority.com)

Snapdragon 8 Gen 2 deep dive: Everything you need to know

All you'll ever need to know about Qualcomm's latest Snapdragon 8 Gen 2 mobile platform for next-gen phones can be found here.

www.androidauthority.com

...
Qualcomm doubled the physical link between the image signal processor (ISP), Hexagon DSP, and Adreno GPU, driving higher bandwidth and lowering latency. This allows the Snapdragon 8 Gen 2 to run much more powerful machine-learning tasks on imaging data right off the camera sensor. RAW data, for instance, can be passed directly to the DSP/AI Engine for imaging workloads, or Qualcomm can use the link to upscale low-res gaming scenarios to assist with GPU load balancing.

Yeah but would their Hexagon tech provide the capability for end devices to function efficiently in the generative AI sense? Akida2 is ready made as a solution to localised generative AI - I don’t see that there is an alternative atm that is better or even good enough for this. Management have claimed that Akida2 would make a compelling case for adoption in cases where there is no alternative, now all this has to come to fruition with the release of the platform and then hopefully some big name contracts. We live in the hope..

 

[Rise from the ashes]

But why are her eyes crossed? And why does she have a band-aid on her thumb? So many questions...

Err that's not a bandaid. And the finger looking thing is part of that contraption

 

[equanimous]

Was that a new "Medical Device".

Human trials to commence immediately

BrainChip Vibrational Analysis Tactile Sensing and Hand gesture recogition with Prohesee