BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
buena suerte :-)
BOB Bank of Brainchip
here we know it's going to be a rocky road, but we'll walk it
View attachment 21613 LLooking forward to a nice dip at the end of the "Rocky Road"
Sirod69
bavarian girl ;-)
JK200SX
Regular
What does anyone make of this:
www.cnx-software.com
AXERA AX620A 4K AI SoC delivers up to 14.4 TOPS for computer vision applications - CNX Software
AXERA AX620A is a high-performance, low-power AI SoC with a quad-core Arm Cortex-A7 processor and a 14.4TOPs @ INT4 or 3.6TOPs @ INT8's NPU that is
www.cnx-software.com
Fact Finder
Top 20
Hi @Diogenese
Have you been doing your own research again. We will have to put a stop to that how will the WANCA’s have any chance of manipulating you.
On a thoughtful note.
Today a few things occurred that led Blind Freddie to make an observation.
The first was a post asking what happened to Renesas and pointing out that they had licensed 2 nodes of IP almost 2 years ago.
The second was an article posted here that said ARM provides 90% of the chips in all mobile phones or something like that but certainly supplies chips to Apple for the iPhone.
The third thing someone posted was that Renesas (as well as MegaChips) is an Apple supplier.
So taking these three things and adding in that different products have different development time lines and a few weeks ago Renesas stated they would be combining AKIDA IP with the ARM M33 chip Blind Freddie theorises that a possible solution as to why Renesas is still to release an MCU with AKIDA technology on board emerges.
We do not know what product Renesas is combining ARM/AKIDA for so we don’t know the development timeline.
We do know as a fact that Brainchip and Renesas continue to work together and Brainchip continues to state they are trusted by Renesas.
We know mobile phones have a three to four year development cycle.
We know from Mercedes Benz that AKIDA is 5 to 10 times more energy efficient for voice activation such as Hey Mercedes than competitor products.
We know that Mercedes’ Benz works with Apple so logic suggests what they have to offer for Hey Mercedes is five to 10 times less efficient than AKIDA.
We know from Anil Mankar’s 2021 Ai Field Day presentation that if you just want to do voice recognition you can just license 2 nodes of AKIDA IP like Renesas.
So as we do not know the product Renesas was targeting the longer it takes to appear the more likely it becomes that it is a product with a longer development timeline, that requires very low powered high performance voice recognition, that can operate without connection securely on device rather than sending to the cloud each time to verify the user.
Something like an Apple iPhone Blind Freddie suggests.
Could he be correct???
My opinion only DYOR
FF
AKIDA BALLISTA
A World Between Frames: Our own Gareth Powell provides a great overview of event-based vision and the Prophesee approach to more efficient #machinevision at VISION. He explains the concept of pixel-level intelligence that enables higher performance, lower power, better #HDR and less data processed by capturing only the changes in scenes.
Watch the video herehttps://lnkd.in/e_qpFNY3
If you like hearing the term “ Just going into mass production “ , for the Prophesee / Sony IMX 636 ES , just fast forward to the 16:16 minute mark .
More negative attention for BRN from simply wall street. How is the forecast even predicted? Who is the analyst?
|
Slymeat
Move on, nothing to see.
Looking at nothing else other than power consumption—1 Watt is NOT low-power. One Watt means it will run for less than an hour on a standard AAA battery (750mAh). Doesn‘t seem to be low enough power consumption to truly be called running at the edge, it will need power cables run to it for one thing.What does anyone make of this:
AXERA AX620A 4K AI SoC delivers up to 14.4 TOPS for computer vision applications - CNX Software
AXERA AX620A is a high-performance, low-power AI SoC with a quad-core Arm Cortex-A7 processor and a 14.4TOPs @ INT4 or 3.6TOPs @ INT8's NPU that is
View attachment 21614
Edge devices need to consume milli-Watts to be effective in a truly standalone manner.
But then, if hooked up to a Strategic Elements self-charging printable battery, producing 1200mAh and never going flat. Maybe it could make it on the edge.
Dear Lord forgive me because I have sinned. I said the following would never happen.
1. I would buy more BRN shares.
2. I would not invest because I read something on a forum
3. I would not sell any of my RMD.
I have sinned lord, I have broken all three of the rules today. (But lord like you, I know it was likely the best decision of my life)
That Mega chips post has me very excited about the next growth spurt for BRN in the coming few years. They are shouting BRN from the roofs, and I am more than happy to invest more here. I did like looking at my previous buy in though.......
1. I would buy more BRN shares.
2. I would not invest because I read something on a forum
3. I would not sell any of my RMD.
I have sinned lord, I have broken all three of the rules today. (But lord like you, I know it was likely the best decision of my life
)That Mega chips post has me very excited about the next growth spurt for BRN in the coming few years. They are shouting BRN from the roofs, and I am more than happy to invest more here. I did like looking at my previous buy in though.......
JK200SX
Regular
So, aside from the power aspect, does it look like it may include our tech?
wilzy123
Founding Member
No it is not our tech.
Fact Finder
Top 20
He she or it seems to have taken the first half operating loss and multiplied it by two to come up with $17.2 million.More negative attention for BRN from simply wall street. How is the forecast even predicted? Who is the analyst?
- The company was expected to make a profit of US$2.89m in 2022.
- New forecast suggests the company will make a loss of US$17.2m in 2022.
That has some logic but who what or where did the expected profit come from as this would have been as a result of someone giving guidance of $20 plus million revenue for this calendar year.
The CEO Sean Hehir has put on record that the company was not able to give guidance but that the Board had agreed to consider the question again first half 2023.
Clearly this claim that Brainchip was expected to make $2.89 million profit is a complete fabrication in a blatant attempt to manipulate the market.
I would encourage you to forward the full article and link to Tony Dawe Investor Relations and request he investigate.
My opinion only DYOR
FF
AKIDA BALLISTA
Fact Finder
Top 20
What we know is that there was a slide or page or whatever it’s called on the Western Sydney University website that listed Brainchip along with one of the US Military services.
After emails were sent to Brainchip about it both Brainchip and the US Military names were removed.
As far as I know that was the original cause for speculation about Brainchip being involved with Western Sydney University.
To be clear what this page looked like. It was broken into three or four segments and each segment had the trade marks or emblems of the companies and government bodies partnering with Western Sydney Uni.
Now of course it might have been just a mistake that Brainchip was included along with the US Military and that the mistake was picked up at the exact same time after the emails were sent to Tony Dawe.
I don’t have a screen shot but someone else might.
My opinion only DYOR
FF
AKIDA BALLISTA
PS: After your comment the other day and again today if my sharing of my memory is not appreciated I am happy to ignore your posts.
WSU had an event camera (Prophesee?) launched by NASA, ostensibly to avid collisions with space junk, but also capable of monitoring earth for sudden hot spots.What we know is that there was a slide or page or whatever it’s called on the Western Sydney University website that listed Brainchip along with one of the US Military services.
After emails were sent to Brainchip about it both Brainchip and the US Military names were removed.
As far as I know that was the original cause for speculation about Brainchip being involved with Western Sydney University.
To be clear what this page looked like. It was broken into three or four segments and each segment had the trade marks or emblems of the companies and government bodies partnering with Western Sydney Uni.
Now of course it might have been just a mistake that Brainchip was included along with the US Military and that the mistake was picked up at the exact same time after the emails were sent to Tony Dawe.
I don’t have a screen shot but someone else might.
My opinion only DYOR
FF
AKIDA BALLISTA
PS: After your comment the other day and again today if my sharing of my memory is not appreciated I am happy to ignore your posts.
Because it is an event camera, it has sparse data, and the data was not processed onboard but transmitted to earth for processing. The project dates from before the Prophesee/Akida engagement and did not have Akida.
Whether future versions will have Prophesee + Akida is still to be seen.
The Pope
Regular
Hi FFWhat we know is that there was a slide or page or whatever it’s called on the Western Sydney University website that listed Brainchip along with one of the US Military services.
After emails were sent to Brainchip about it both Brainchip and the US Military names were removed.
As far as I know that was the original cause for speculation about Brainchip being involved with Western Sydney University.
To be clear what this page looked like. It was broken into three or four segments and each segment had the trade marks or emblems of the companies and government bodies partnering with Western Sydney Uni.
Now of course it might have been just a mistake that Brainchip was included along with the US Military and that the mistake was picked up at the exact same time after the emails were sent to Tony Dawe.
I don’t have a screen shot but someone else might.
My opinion only DYOR
FF
AKIDA BALLISTA
PS: After your comment the other day and again today if my sharing of my memory is not appreciated I am happy to ignore your posts.
thanks for the quick reply. Firstly covering the bottom of your email reply, I haven’t provided any posts for a while so you may have me mistaken for someone else. All good
Let’s just say, the conversation I have recently heard linked to WSU doing testing on AI on a few different things and I suggest it appears to confirm them involved with BRN. Sorry I can’t say any more for now. Let’s just say my ears pricked up when I heard a few things being said and appeared to strongly suggest Akida.
Last edited:
The way I see it is that the more negative press BRN receives during this present phase then the bigger the SP will pop when an announcement is released - the FOMO from blindsided investors will be huge.....
ceviche
Emerged
almost funny.
A question for those who have been in the market for a while.
How much revenue do you need to justify a market cap of $1b?
One answer, "hmmm, around $333m"
The answer for unicorns. Nothing
There is something wrong there.
At what point do you decide that a unicorn isn't worthwhile, or better what do you do with a unicorn that was looking like getting revenue but the revenue day keeps staying "over the horizon".
My answer. REDUCE.
Which i did by 40% in August at an average of around $1, then a bit more soon after.
I see some double down. GLWT.
Consider: i am still holding. BRN does have two huge advantages, high efficiency in SNN, and head start in SNN. That isn't every app.
Since Aug, did prospects of other stocks decline relative to BRN? What has BRN announced that increases its revenue stream?
Oh it got 'cheaper', compared to still not having any revenue, and for longer.
Good luck with that as a basis for investing.
I am sticking with balancing risk over a whole portfolio.
BTW.
Not saying for example, par or wbt is better long term, but i haven't reduced those. The difference is that in each case they are making progress against a schedule.
IMO> BRN need to announce something significant by soon after 1qtr 2023 at the latest.
DYOR
Fact Finder
Top 20
Sorry I laughed at yesterdays comment but thinking it was the same poster.
Anyway here is the additional unclear part of my memory which may assist.
Unclear - US Airforce
Unclear - The original discovery of the webpage was in the context of the 1,000 Eyes looking for Loyal Wingman Connections.
On the second part of your question about 90% it is not the right way to ask this question.
AKIDA achieves 100% sometimes immediately sometimes with additional training.
What you need to keep in mind is that it incrementally learns from what it is doing and improves.
The operator can also intervene and add extra shots and confirm the label each time to increase accuracy.
Take the original numbers that AKIDA produced for NaNose from the data provided from the Zayed Military Hospital it identified Covid 19 at around 87%.
The data from Wuhan it improved to around 94% and the last figures I heard it had improved to 96 to 98% accuracy.
So accuracy numbers will not be a static thing regarding AKIDA technology performance.
Also the nature and quality of the sensor acquiring the data that is being supplied to AKIDA comes into play.
If you have not listened to the Brainchip Podcast 21 with Luca Verre CEO at Prophesee it is useful on this issue as well.
Regards
FF
AKIDA BALLISTA
Similar threads
