Beware missing specs

Twice in one week is two times too many. First the somewhat lame Kickstarter project called HERE – a supposedly active listening device that as a concept seems quite the opposite.
Developed by a company who call themselves DOPPLER LABORATORIES would make you think they might own the test gear AND the knowledge to know that stating a freqeuncy response spec this way is COMPLETE NONSENSE!

Screen Shot 2015-06-19 at 3.42.03 pm

But just incase it was a mistake I messaged them:

“A simple claim of frequency response that cites two frequency extremes unqualified by a dB specification (e.g. your spec of Frequency response: 20 Hz – 20 kHz) is meaningless and useless.

As you have demonstrated your product as working & fully functioning, no doubt you have measured the response & I question why you are not providing such important information?

Can you please provide an actual frequency response chart, showing the response from full range audio source > microphone > DSP > speaker, with dB variation across your claimed response of 20Hz-20kHZ?

Only then can anyone make an informed decision about the audio quality of this product.”

 

No response… Maybe they are too busy counting the money ($488,557) – imagine that, almost half a million dollars donated by people who do not know ANYTHING about the quality of the audio of the product they are buying. If it was some kid selling some dodgy device out of his backpack for $20 I’d forgive it, but for an experienced tech company I think there are only two possible reasons:

1. They don’t have the actual specs because they don’t have a working model (although they do say they have a working prototype)

2. They have the specs and don’t want to make them public

Either option is highly questionable, so I hope someone at Doppler Labs reads this & maybe provides some answers…

Until then I’ll just have to write off the project as a marketing exercise: style over content. Attempting to cash in on a mass market (one of their mottos is “Instagram for your ears”) doesn’t mean ommitting crucial facts.
When I heard of the project my first thought was not about frequency response so much as latency. Essentially the little ear buds are recording sound via an A-D, processing it with DSP & then converting it back to analog via a D-A and replaying it out of little speakers. To achieve that with low enough latency to be useable live takes processing power. So I was intrigued to read their stated spec of latency of 30 μs

Nick Darnell asked them on June 5:

“30 μs doesn’t seem like enough time to have accumulated enough data to do any useful sampling in the frequency domain to do any DSP. 50 μs is listed as the cycle time for highest human-audible tone (20 kHz) in wikipedia. Which begs the question, how are you doing any kind of audio analysis on frequency without at least having a delay of 50 μs? Can you please explain how you arrived at 30 μs (microseconds), is that the true wall clock time in/out for the audio?”

Their answer: “30 microseconds is the sample by sample processing latency, independent of frequency”

And another of their answers to a related question: “the current prototype already has functionally zero latency – incoming sound waves are processed at a speed that renders the latency more or less imperceptible to most humans.”

Confused much?

More or less?

And a second crowd funded audio project: the InstaMic – it seems audio developers really want to be instagram, although InstaMic then go on to express a desire to be the GoPro of audio… Basically its a very small omni mic + preamp + 24bit 48k recorder, which is not a bad idea although a directional mic may be a better option, but again my interest grinds to a halt when I read the specs:

Screen Shot 2015-06-19 at 4.01.54 pm

So it has a frequency response of

50 to 18,000Hz +/- XdB?
X = 3dB? or 10dB? or 20dB

That missing X figure gives an accurate indication as to whether this is going to be an audio device worth any further interest… You have to wonder why they choose not to test their prototypes and display the results… A part of the development cycle would be testing exactly this spec, these people are all asking for other peoples $$$$ so how about being up front with the test results?

UPDATE: and another one with inaccurate/misleading specs:

Screen Shot 2015-06-24 at 9.27.23 am

MikeMe on Indiegogo using misleading specs but they did include a frequency chart, so why not put the industry standard specs of 20-20,000 +/-5dB or whatever it is tested to be

q7pkzoyprdere1zmiroq

Leave a Reply

Please use your real name instead of you company name or keyword spam.