buena suerte :-)
BOB Bank of Brainchip
Hey misslou ... Dodgy Knees will no doubt shed some light on this!!
.... GREEN today please 
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
.... GREEN today please 
uiux
Regular
Misslou
It's interesting that the Grai VIP is "a thirdâgen product" - so we are essentially comparing Akida v1 to Graimatter v3
It's on 12nm, which means they can squeeze a lot more into it
From cursory glance, Akida has 1.2 million neurons and 10 billion synapses vs VIP ones 18 million neurons... Grai only list "parameters" though and don't specify synapses
My understanding is the # of parameters that the chip can hold is calculated by doing something with the # of neurons and the # of synapses - I'll look for how this is calculated, but I remember @FrederikSchack digging in extremely deeply into this line of questioning, so maybe he could shed some light on how neurons/synapses/parameters are calculated
BrainChip has identified Grai Matter as the "nearest competitor" in their AGM presentation
There is a bunch of shared history between the two companies:
All roads lead to JAST
What is JAST? https://www.researchgate.net/profile/Amirreza_Yousefzadeh/publication/319410728_Unsupervised_learning_of_repeating_patterns_using_a_novel_STDP_based_algorithm/links/5b4932cc0f7e9b4637d5e6ba/Unsupervised-learning-of-repeating-patterns-using-a-novel-STDP-based-algorithm.pdf...
thestockexchange.com.au
Hopefully this is helpful
Last edited:
jtardif999
Regular
Synapses are collocated memory containing the synaptic weights associated with the memory of learning things. So if their Neuromorphic architecture doesnât contain Synapses how would anything be learned or indeed training be stored? Their Neuromorphic architecture is not real Kung FuMisslou
It's interesting that the Grai VIP is "a thirdâgen product" - so we are essentially comparing Akida v1 to Graimatter v3
It's on 12nm, which means they can squeeze a lot more into it
From cursory glance, Akida has 1.2 million neurons and 10 billion synapses vs VIP ones 48 million neurons... Grai only list "parameters" though and don't specify synapses
My understanding is the # of parameters that the chip can hold is calculated by doing something with the # of neurons and the # of synapses - I'll look for how this is calculated, but I remember @FrederikSchack digging in extremely deeply into this line of questioning, so maybe he could shed some light on how neurons/synapses/parameters are calculated
BrainChip has identified Grai Matter as the "nearest competitor" in their AGM presentation
There is a bunch of shared history between the two companies:
All roads lead to JAST
What is JAST? https://www.researchgate.net/profile/Amirreza_Yousefzadeh/publication/319410728_Unsupervised_learning_of_repeating_patterns_using_a_novel_STDP_based_algorithm/links/5b4932cc0f7e9b4637d5e6ba/Unsupervised-learning-of-repeating-patterns-using-a-novel-STDP-based-algorithm.pdf...
Hopefully this is helpful
uiux
Regular
Synapses are collocated memory containing the synaptic weights associated with the memory of learning things. So if their Neuromorphic architecture doesnât contain Synapses how would anything be learned or indeed training be stored? Their Neuromorphic architecture is not real Kung Fu
If it's got neurons you would assume synapses follow
My understanding is that there's a relationship between neurons and synapses, like a sliding scale. You can decrease the number of neurons for more synapses and vice versa.
Maybe Graimatter see more value in advertising # parameters vs # of synapses
I see zero mention of learning rules associated with the device. It can't perform on-chip learning or one-shot learning..
uiux
Regular
Misslou, as per the headline:
$1 million in preâorders
Versus
Megachips $2 million + royalties
equanimous
Norse clairvoyant shapeshifter goddess
The following paragraph from the GrAi article when compared with the Nviso BRN demonstration is telling as to 1,000 fps (also somewhere I saw the number 1670 fps but cannot recall where might have been said by Tim Llewellyn):
âGrAI VIP can handle MobileNetv1âSSD running at 30fps for 184 mW, around 20Ă the inferences per second per Watt compared to a comparable GPU, the company said, adding that further optimizations in sparsity and voltage scaling could improve this furtherâ
They are saying they are comparable with GPU performance per watt while AKIDA according to Nviso is 10 times better than the Jetson Nano 100 fps.
AKD1000 too cool for school. It does this at $US10 to $US15 a chip on 28nm and as you scale down to 12nm its performance will increase so GrAi are much more expensive at 12nm and well behind.
The article makes clear they do not have one shot and incremental learning.
They are using 16 bit activations and boast only a 1 percent accuracy lose when converting from 32 bit activations. AKD1000 at 4 bit activations is offering only 1 percent loss because it has on chip convolution as well as the SNN.
âGrAI VIP can handle MobileNetv1âSSD running at 30fps for 184 mW, around 20Ă the inferences per second per Watt compared to a comparable GPU, the company said, adding that further optimizations in sparsity and voltage scaling could improve this furtherâ
They are saying they are comparable with GPU performance per watt while AKIDA according to Nviso is 10 times better than the Jetson Nano 100 fps.
AKD1000 too cool for school. It does this at $US10 to $US15 a chip on 28nm and as you scale down to 12nm its performance will increase so GrAi are much more expensive at 12nm and well behind.
The article makes clear they do not have one shot and incremental learning.
They are using 16 bit activations and boast only a 1 percent accuracy lose when converting from 32 bit activations. AKD1000 at 4 bit activations is offering only 1 percent loss because it has on chip convolution as well as the SNN.
From this article, it's a whole SoC, not IP that can be applied to any new small chips coming out.
It requires dual Arm M7 ( extremely high powered CPU's) to operate it
and it uses floating point maths (ie Graphics processing which involves power and heat.)
So not competition, except possibly for a small number of specialised applications. Certainly not at the edge.
uiux
Regular
It's not a fair comparison looking at NVISOs reported FPS and Grais - they are running two different neural network architectures
On Akida hardware, Nviso were only running the gaze tracking in that graph which from memory was running on 5 nodes compared to mobilenets ~30-123 - which is a significant footprint increase
Model zoo performances â Akida Examples documentation
uiux
Regular
The article says they are aiming for near sensor applications. Though the wording of the article makes appear a little cagey, eg.
"GrAI Matter sees its offering in between edge server chips and tinyML, though its device is intended to sit next to sensors in the system. An ideal use case would be GrAI VIP next to a camera in a compact camera module, he added."
D
Deleted member 118
Guest
The shorter is giving it a good go this morning and there is only 1 thing I need to say.
uiux
Regular
It's not a fair comparison looking at NVISOs reported FPS and Grais - they are running two different neural network architectures
On Akida hardware, Nviso were only running the gaze tracking in that graph which from memory was running on 5 nodes compared to mobilenets ~30-123 - which is a significant footprint increase
Model zoo performances â Akida Examples documentation
doc.brainchipinc.com
Actually, nviso did port all their models over.... So not entirely sure about # of required nodes of all models..... Either way, it's not a fair comparison with Mobilenet as apples vs oranges
D
Deleted member 118
Guest
Accumulator is back and picking everything up at $1.13
Bravo
If ARM was an arm, BRN would be its bicepsđŞ!
Xhosa12345
Regular
Bravo sneaking back in this morning..... love it
Slymeat
Move on, nothing to see.
âBetterâ is open for interpretation. Iâd say different and limited. It seems to be a chip that can be trained to do a specific task in a way that is generally acceptable to call it AI. Not in my opinion, but seemingly accepted by many others. But then I could do the same in a sequential programmimg language also. No AI need be involved.
For each pixel
if( pixel changes from previous state) then
do something
end if
Store current pixel state for next iteration
end loop
A couple of things that stood out for me were:
1) âGrAI VIP can handle MobileNetv1âSSD running at 30fps for 184 mW, around 20Ă the inferences per second per Watt compared to a comparable GPUâ
Comparing it to a power-hungry GPU is a bit naughty. Everyone knows they are power hungry and anyway, GPUs donât do inferences per seâjust sledgehammer, power hungry, high level maths. Well considering multiplication to be high level that is.
Akida has helped achieve 1000fps and uses ÂľW
2) it uses 16 but floating point in calcs. That would be compute intensive.
3) the system can be trained, but I saw nothing about it learning.
and
4) it seems very specific to processing images only. Although they do also mention audio, bit their example is only for video.
IMHO it seems like they are closer to a normal, and single tasked, CNN and are using the word neuromorphic in a very loose manner. Pretty much just as a buzz wordâprobably to get search engines to find the article. Sure they call things neurons, but so to do many other implementations call memory cells neurons, and call what they have neuromorphic.
As @jtardif999 stated, they donât mention synapses, and I donât accept that if you have neurons, then synapses naturally follow. They should, in a true neuromorphic implementation, but so many are using that term for things that are very loosely modelled on only part of the brain.
As an example I refer to ReRAM implementations of âneuromorphicâ systems. They store both state and weight in memory cells, and use the resistive state of memory cells to perform analogue addition and multiplication. But I think all such âneuromorphicâ implementations suffer the same limitation of not being able to learn, they can only be trained. And once trained for a task, that is the only task they do until re-trained. And if that is all you want, then is your definition of âbetterâ.
This raises a VERY relevant question, is Akida too good. The world has time-and-time again gone with simple to understand, and simple to use solutions, over complex multi-faceted solutions. The world especially likes mass-produced widgets that do a required task well-enough. Some of these other âneuromorphicâ solutions may prove to be just that. People seem happy to throw money multiple times at an inferior product rather than pay extra for the product they really need.
Thereâs enough room in the TAM for multiple players. Iâm happy for Akida to occupy the top spot, solving the more difficult problems, and leave the more mundane to others.
Dang Son
Regular
"Don't miss the bus" , they said over and over and over.
Well, wot the FK?
<iframe src="https://giphy.com/embed/3o6ozsc6CnNmjHfBeM" width="480" height="239" frameBorder="0" class="giphy-embed" allowFullScreen></iframe><p><a href="">via GIPHY</a></p>
Fullmoonfever
Top 20
@uiux @misslou
Not sure if this helps but does give some insight into Grai and their original AI Neuronflow Architecture on the GrAI One and they have now moved to the GrAI VIP as an evolution so I could be wrong but, expect the underlying architecture of the VIP is just some enhancements and not something entirely new or ground breaking.
Whilst obviously suitable for certain cases and close to Akida in particular areas eg digital, event based, would appear no on device learning and indicate to me still not near Akida overall.
Original May 22 paper attached.
Not sure if this helps but does give some insight into Grai and their original AI Neuronflow Architecture on the GrAI One and they have now moved to the GrAI VIP as an evolution so I could be wrong but, expect the underlying architecture of the VIP is just some enhancements and not something entirely new or ground breaking.
Whilst obviously suitable for certain cases and close to Akida in particular areas eg digital, event based, would appear no on device learning and indicate to me still not near Akida overall.
Original May 22 paper attached.
uiux
Regular
This is an awesome find, thank you!
I am looking for the slide from a BrainChip presentation that shows Graimatter next to BrainChip with a few of the other chips to the right
Anyone know which one it's from? I thought it was 2022 AGM preso
Similar threads
