ESP32 ee.* operations in assembly look pretty much like aliases for a VLIW bundles, on the same cycle issuing loads used in the next op while also doing multiplication on other operands. This is not a minimal Xtensa.…
Xtensa architecture is flexible and extendable by the user. Ability to define new instructions, hw features and VLIW configurations are some of the key features. You can find more details on the internet…
I wonder if ESP32 has VLIW slots and a tighter instruction packaging is possible?
I am speculating, but I'd expect that Google's reach outside of USA is much larger than the Yandex's reach outside of Russia. In 2021 Alphabet had 46% of revenue originated from USA. If applied withing the country…
It's not clear if it helps in the long run, it probably does but a lot of friendly fire should be expected. Just recent example where Ukrainian anti aircraft units were taken for Russians and killed in Kyiv (the one…
How about quantization? Does tensorflow lite perform quantization or is it tensorflow supposed to do it? Is it iterative process or straightforward? Or are you training quantized models as nn api docs say?
Yep, enjoyed this one.
They used customizable DSPs from Tensilica. I wonder if this is based on the same technology.
ESP32 ee.* operations in assembly look pretty much like aliases for a VLIW bundles, on the same cycle issuing loads used in the next op while also doing multiplication on other operands. This is not a minimal Xtensa.…
Xtensa architecture is flexible and extendable by the user. Ability to define new instructions, hw features and VLIW configurations are some of the key features. You can find more details on the internet…
I wonder if ESP32 has VLIW slots and a tighter instruction packaging is possible?
I am speculating, but I'd expect that Google's reach outside of USA is much larger than the Yandex's reach outside of Russia. In 2021 Alphabet had 46% of revenue originated from USA. If applied withing the country…
It's not clear if it helps in the long run, it probably does but a lot of friendly fire should be expected. Just recent example where Ukrainian anti aircraft units were taken for Russians and killed in Kyiv (the one…
How about quantization? Does tensorflow lite perform quantization or is it tensorflow supposed to do it? Is it iterative process or straightforward? Or are you training quantized models as nn api docs say?
Yep, enjoyed this one.
They used customizable DSPs from Tensilica. I wonder if this is based on the same technology.