youtu.be/YlZAFH4JqFE
Sjekk ut slutten av denne Intel-presentasjonen og man skjønner hvor samarbeidet med Elliptic bærer av sted
3 Likes
Tips å legge med timestamp og hva du refererer til
4 Likes
Fra 37:00
1 Like
It seems you can interpret anything in a positive manner, regardless of what the article states. Simmilar to how some poster like aght and aksjekongen can interpret everything negatively. It does not state anywhere that US sensors is more reliable. I still believe that the smartphone sensor is mostly a mid to low end product, as it is mostly implemented in that segment of phones. From what we know it is not integrated in the flagships of either Honor or Xiaomi anymore, so the most premium product uses physical sensors.
There was a bunch of complaints about Xiaomi’s reliability when it came to proximity sensors. I believe they have improved since then, but especially during 2019-2021 there was a lot of complaints. They probably got a lot of data from those years, as the rollout increased a lot. I believe that they are better than before, as I there seem to be less complaints regarding this, but it is hard to track due to the fact that a lot of the volume is sold in countries with low % of english speaking populations.
When it comes to the three different sensors you mention, the ultrasound sensor and virtual proximity sensor both go under the term virtual sensors. The complaints you see on reddit was also about Xiaomi phones, if you scroll down here. https://www.reddit.com/r/Android/comments/on7b0y/avoid_android_devices_with_virtual_proximity/
There is also this article here. Xiaomi to address the proximity sensor issue on some of its phones - GSMArena.com news
They obviously made some wrong claims about stopping using Elabs entirely, but Xiaomi shifted many high end phones to physical sensors during 2023. Including its flagship, which has not had ELABS since xiaomi 12. Xiaomi has also developed a virtual sensor, which it has in very cheap phones, including this one.
Redmi 14C Specs | Xiaomi UK - Xiaomi UK
The reason why I hope the Tap-to-share for smartphone is better than what is currently launched on Honor 400 series is mostly because it would enable Elabs to have sensors in the Mid and High-end market. The High end is an area they are currently lacking, as Honor, Vivo, Xiaomi all use non-elabs solutions.
The exception being Transsion who uses Elabs, but their flagships are priced mid-range.
When it comes to what GSMarena reports on specs, before release they report on what is rumoured and that they know. After release they update the specs. If the OEM does not specifically list that the phone has a virtual/ US sensor they can’t know before they review it themselves. Honor is the manufacturer that seems to usually have the type of prox sensor in their spec sheet on release from experience, while Vivo f.ex does not.
It is important that the discussion, both on what Elabs currently offers and how good their products are remains balanced. It is equally important that discussion about it’s future products/ implementation of them remains balanced. I don’t think its unreasonable to think there is a good possibility for the new smartphone sensor to be launched during the year, given statements about the development of it being done based on broad market demand. To speculate and say it is launched before we have confirmation seems unnecessary.
It’s important that this does not become and echo-chamber where everything is all BULL and is interpreted the best possible way.
4 Likes
I have described possibilities, scenarios, etc. If your read carefully I rarely state anything as certain. At least I dont keep things certain in my own decisions 
3 Likes
Yup, that’s how I read you Mr"could be something could be nothing" 
Jeg skrev egentlig et lengre innlegg, men fordi det var første post i denne tråden fikk jeg ikke poste lenke og ble litt demotivert av å måtte poste en gjørdetselv-url
1 Like
Regarding Hardware (HW) vs Software (SW) issues you mention above. I have mention this before (about six months ago).
I’ve been working with several software programming since 2008. There can be SW bugs in early stages, but SW will never change. As long the program gets all correct Inputs, it will give you the output it is programmed to give you. With a well tested SW it is always the HW that will have a failure if the equipment doesn’t work correctly. To be able to get correct inputs HW need to work properly.
This is why I like the strategy Elabs is doing regarding their business and deploying their product; First they deploy in small scale, then to fix SW bugs, when proper tested, release in big scale. Started with smartphones, now they’re main focus is on pc/laptops. With this kind of structure, they don’t need many to work on support on released products.
Disclaimer: I own Elabs shares 
11 Likes
Stein på stein mot mid-term target 
1 Like
Kan bli stort dette, hentet fra nettsiden;
Ceva is Building the Future of the Smart Edge
Ceva architects, develops, and licenses the industry’s only comprehensive communications and scalable edge AI IP portfolio. What makes our technology special is our ability to deliver outstanding performance at ultra-low power within a very small space, enabling our customers to turn great ideas into extraordinary products. We proudly partner with the industry’s leading semiconductor companies and original equipment manufacturers with one simple goal in mind – to deliver the technology that enables a smarter, safer, and more interconnected world. And our goal is becoming a reality. Ceva powers more than 19 billion of the world’s most innovative smart edge products from phones to drones to cellular base stations, and more, with an additional 60 devices added every second. (Ticker: CEVA).
3 Likes
Trenger ikke mange ørene per device her for å få fart på omsetningen med 60 devicer added per second 
1 Like
Men det var et selskap med bare 24 millioner USD i inntekter i 1Q, så ikke akkurat så mye penger der om jeg leste riktig.
Men selvfølgelig positivt med nye bruksområder 
1 Like
Hjemmesiden Deres:
19 Milliarder produkter Globalt til nå.
2 Milliarder pr. år.
Vil nok ta litt tid før det blir penger av dette, men så kan dette gi et bra bidrag i omsetningen.
1 Like
Dette forklarer nok hintene fra Elabs siste halvår om aktivitet innen IoT.
5 Likes
De har jo hatt samarbeid med Syntiant, lanserte ett produkt som ikke ble til noe mer. Så får håpe dette samarbeidet er basert på forespørsel om funksjonalitet fra en kunde. Kanskje samsung ønsker seameless funksjoner mellom earbuds og mobil feks, hadde vært en bra case
1 Like
Ja lige præcis. Nu må vi se om nogle af disse samarbejder fører til noget godt. Intel og Syntiant har ikke blevet til det helt store indtil videre
Det er tydelig at forenklet oppkobling mellom enheter er viktig for mobilprodusentene for tiden. Nå jobber Vivo med en funksjon som : “ability to connect with iPhones for a seamless cross-platform experience.”
Vivo er som kjent en stor kunde av Elabs der proximity sensor er i de fleste nye modeller.
https://www.gizmochina.com/2025/06/10/vivo-x-fold-5-iphone-sync-details/
7 Likes
Kan dette være relatert til enten samarbeidet med Intel eller den kontrakten som inkluderer accessories, eller er det helt fjernt?
Slik jeg forstår hva Ceva gjør så designet de NPU’en, så de eier designet på den. Denne er designet for å kunne kjøre på lite strøm og uten en CPU hvis det trengs, spesielt for å kjøre AI-modeller. Tror CEVA kun selger IP, så de produserer ikke selve NPU’en, men lisensierer ut designet slik at f.ex lenovo/intel kan integrere denne i sine chipper/produkter.
Vil dette si at en kunde som f.ex ønsker å bruke en Elabs løsning da kan komme kjøpe denne ferdig designa NPU’en som de så produserer selv i egen fabrikk? Eller er det slik at kundene går til CEVA, også tilbyr de produktet til kunden basert på hva de trenger? Er Elabs kunde av Ceva, omvendt eller selger de det samlet?
Gjerne noen rett meg om jeg tar feil eller ELI5, for her er jeg utenfor egen kunnskap.
Er dette en NPU som kun er designet for iot som wearables eller kan den integreres i en laptop/smartphone også? Med tanke på at for å kjøre en presence detection osv på Windows må kjøres uten tilgang til Os, eller for at smartphones kan kjøre det uten å bruke mye strøm osv.
" enabling next-generation, context-aware sensing on ultra-low power edge devices. " Svarer mitt eget spørsmål, men kan sikkert være nyttig for noen andre.
Hvis det har noe med andre kontrakter/sammarbeid er det kanskje nærmere accessories kontrakten eller så står det på egne ben. Uansett positivt. Virker som om også denne spesifikke NPU’en har vunnet priser og har god market traction, hvilket sikkert ikke er negativt. Ceva Embedded AI NPUs Gain Traction in AIoT and MCU Markets - Ceva
Yeah, I agree with you on that point. I also believe it could be the best solution at some point. Maybe it already is less flawed than physical sensors at this point, but the bugs and issues from the first “waves” of smartphones makes it so it is less used in high-end smartphones to this day.
Like most things that are made for the first time there were issues, but unlike most other things software can be patched and improved upon. I believe they did that during those years, as from what I have in my notes Xiaomi pushed out several software updates for proximity after launch
1 Like