Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Is the heart rate sensor in the Series 2 the exact same one as the on in the 1st Apple watch?
- Thread starter Benz63amg
- Start date
- Sort by reaction score
It is physically the same. But, it is key to know that the effectiveness of an HR sensor is as much (or more) the software than it is the hardware. So, the HR sensor software should improve with every OS release.
I understand that watchOS improves the sensor's effectiveness and performance in every new WatchOS update but what I was wondering is hardware wise, physically. If it's exactly identical to the heart rate sensor from the very first Apple Watch, is it the same part # in the latest teardown ifixit did?
Yep... I was just adding that comment, because some folks think of the physical sensors alone relating to performance. I did not know where your question was coming from. IMHO, it is highly likely any AW issues with HRM are the firmware, not the hardware. The hardware are pretty typical off-the-shelf bits that are probably very good by themselves. It seems like everyone who attempts to go their own with optical HRM sucks, versus licensing a system from a vendor who specializes in awesome. If Apple had Valencell's software engineers, then likely the AW HRM would be awesome using the very same hardware it has now. Theoretically, the AW's LED and sensor package is far more sophisticated than the totally awesome package in the Scosche Rhythm+.
iFixit did not note different part numbers. And, they did not respond to any questions on their blog asking if the PNs are different for the LEDs or sensors.
iFixit did not note different part numbers. And, they did not respond to any questions on their blog asking if the PNs are different for the LEDs or sensors.
The sensor is not very accurate, I have series 1 but as far as I know it is not improved in series 2.
This morning I climbed 8 floors pretty fast because the elevator wasn't working and obviously I was out of breath when I reached the top floor. I activated the heart monitor and the beat was normal, just the same value that I get while taking a stroll.
Sometimes it is quite precise, but sometimes it is completely wrong. I wonder if they can improved it via software
This morning I climbed 8 floors pretty fast because the elevator wasn't working and obviously I was out of breath when I reached the top floor. I activated the heart monitor and the beat was normal, just the same value that I get while taking a stroll.
Sometimes it is quite precise, but sometimes it is completely wrong. I wonder if they can improved it via software
I think you mean you have an S0 and not an S1 since if it had been changed (which it wasn't) the S1 would likely have the same sensor as the S2 unless Apple noted a difference.
From the stuff I have read, the problems are nearly always software algorithms. The process is quite simple-- LEDs flash, and sensors measure reflected light after each pulse. Then, the software applies various DSP transforms to what it sees in the reflected light intensity to try to determine if the measured value is due to watch movement, outside light, actual blood flow, etc. The more pulses it takes per second, the more data it has to try this detection. The algorithms in the software are the secret sauce that control and then make sense of the light pulses.
I have series 1, but it is basically the same product as series 0 with the new chip, so the sensor on the back has to be the same as the one found in series 0
[doublepost=1475856088][/doublepost]
Yes the algorithm can be changed via software, but what if the sensor is not very accurate?
I'm not 100% sure about that, I think I read something about the sensor not changed in series 2 but I may be wrong
Until The internal part numbers are confirmed by ifixit tear down or someone else there's no way to tell
It Could have been silently improved and updated internally even though it's identical in physical appearance
I get where you are coming from. But, the quick answer is almost "the software is the sensor."
If Apple had licensed a complete HRM from from Valencell, then the question about sensor accuracy applies, because the the watch just asks the Valencell sensor (a complete HRM subsystem) what is the HR.
However, Apple elected to assemble an HRM from the individual parts-- they selected the emitters, optical sensors, and DSP, and Apple wrote their own algorithms to control and interpret it all. So, the software is the thing that actually does everything. The software is the thing that is either accurate or inaccurate, because the physical optical sensor does not present an HR, the optical sensor simply states that "at this instant, I am detecting x intensity of light." The software is the thing that determines that the optical sensor reading is a peak value, and that it last responded with a peak value 923 ms ago; therefore the HR must be 65 BPM.
I get your point, but what if the sensor cannot detect light properly? I'm not saying the sensor itself is faulty, but maybe they placed it in a position were it cannot get the data in an optimal way.
Software is almost everything today, but if you have an hardware problem it isn't always possibile to find a sw workaround.
Anyway, Watch is not a professional heart monitoring device. It sometimes get wrong data, but isn't that bad. I wonder if a band could monitor the heartbeat more accurately, and I'd like to see bands with additional health functionalities in the future, so the Watch can be primarily focused on the software and a band can be optimised just for a particular task
The dark magic of wrist-based HRM subsystems for sport use has largely been solved by companies like Mio, Valencell, LiveQ, etc. And these companies license their tech to other device makers. I think that Valencell's is the best, and the funny thing is that Valencell is suing Apple for stealing its HRM IP. Apple has a great defense-- they could argue that they couldn't possibly be using Valencell's IP in the AW, because the AW HRM subsystem sucks, and it would clearly be awesome if it had a Valencell HRM.