[HN Gopher] Mythic Rolls Out M1000-Series Analog AI Accelerators
___________________________________________________________________
Mythic Rolls Out M1000-Series Analog AI Accelerators
Author : rbanffy
Score : 41 points
Date : 2021-08-31 14:23 UTC (19 hours ago)
(HTM) web link (fuse.wikichip.org)
(TXT) w3m dump (fuse.wikichip.org)
| PaulHoule wrote:
| That kind of thing is great for wake-word recognition where it is
| powered up and running all the time.
| datameta wrote:
| Potentially just a few of those cores alone will allow for
| wakeword recognition! This in-memory analog compute
| breakthrough is to me, in the truest sense of the word,
| mindblowing. I've been following Mythic since learning of
| memristor-based tech for TinyML applications and I'm beyond
| extremely impressed with not just the engineering but also the
| market interaction side of it.
| dogma1138 wrote:
| If these are going to be cheap it would be a game changer for the
| hobbyists/ hacker market, get the M.2 version and an SBC or a
| board for this and the Pi compute module run your main logic on
| the ARM CPU and use this the Mythic chip as an accelerator for
| object recognition or localization.
|
| Now it will all boil down to software and how easy would it be to
| use these darn things, if they are going to be cheap enough it
| doesn't even matter if they'll become obsolete quickly since
| you'll be able to just swap them.
| nharada wrote:
| So this is purely for inference right? Basically you're setting
| up a physical analog inference device by writing weights to flash
| memory and then using the device properties to do that mat-muls
| required for the forward pass?
|
| This is super cool, I'd really like to see how this shakes out in
| actual products in terms of models supported, actual costs and
| performance, etc.
| TheAdamist wrote:
| Flash has limited write cycles, I wonder what the lifetime of
| these chips are supposed to be. I would think trying to compute
| with them would hit that limit very quickly.
|
| Consumable chips I guess.
| lern_too_spel wrote:
| The weights are loaded once per model, so you would only have a
| write when you update the model, which happens once every few
| hours or days for most applications today.
| Eyght wrote:
| My mind immediately went to the artificial 4-year lifespan of
| the Nexus-6 replicants in Blade Runner.
| sdenton4 wrote:
| "The candle that burns twice as bright burns half as long...
| And you, you, Roy, have used oh so many write-cycles."
| base3 wrote:
| Positronic brains!
___________________________________________________________________
(page generated 2021-09-01 10:02 UTC)