4 ms·

I doubt Qualcomm will be able to increase their performance ahead of Nvidia decreasing their energy requirements.

by oneplane 3y ago

pjmlp 3y ago
Since NVidia hardly ships on Android phones, it doesn't really matter.
- oneplane 3y ago
  This isn't just about mobile telephones, or android for that matter. It's about on-device TPUs or AI accelerators or parallel compute units.
  - pjmlp 3y ago
    Which basically means Android devices, in what concerns mainstream consumers, given the state of the PC market. If on the other hand we look at embedded market, it is also not something where NVidia has any kind of strong market presence, with the exception of them trying to be in a good position on autonomous driving market.
    - oneplane 3y ago
      This isn't about mainstream consumers either. This is about a technological development where the balance between performance and energy consumption is attacked from both ends, where one vendor has the performance and the other vendor has the energy consumption but neither have both (yet).
      - pjmlp 3y ago
        NVidia isn't serving industrial embedded consumers either.
        oneplane 3y ago
        Yes they are. But it's not about that either. You are conflating end-user products with two companies developing to the same central point from opposing start positions. That is what this is about.
jayd16 3y ago
Why?
- oneplane 3y ago
  Qualcomm hasn't been able to manufacture desktop-class ARM devices yet (see SQ1 and X13 ARM edition), and the GPU IP doesn't really hold a candle to what else is out there either. They are however using significantly less power than say, x86 SoCs and GPUs from AMD, Intel or Nvidia. Nvidia on the other hand has plenty of IP for GPUs and AI workloads that is proven to have high performance, even on the Tegra scale of things. The Jetson product line as an example does still outperform whatever Qualcomm has come up so far. But they are consuming more power and don't even come close to the same thermal envelope. So Nvidia needs to reduce their energy needs, while Qualcomm needs to create actually performant devices for real-time AI workloads. That's why.

I doubt Qualcomm will be able to increase their performance ahead of Nvidia decreasing their energy requirements.

pjmlp 3y ago

Since NVidia hardly ships on Android phones, it doesn't really matter.

oneplane 3y ago
This isn't just about mobile telephones, or android for that matter. It's about on-device TPUs or AI accelerators or parallel compute units.
- pjmlp 3y ago
  Which basically means Android devices, in what concerns mainstream consumers, given the state of the PC market. If on the other hand we look at embedded market, it is also not something where NVidia has any kind of strong market presence, with the exception of them trying to be in a good position on autonomous driving market.
  - oneplane 3y ago
    This isn't about mainstream consumers either. This is about a technological development where the balance between performance and energy consumption is attacked from both ends, where one vendor has the performance and the other vendor has the energy consumption but neither have both (yet).
    - pjmlp 3y ago
      NVidia isn't serving industrial embedded consumers either.
      - oneplane 3y ago
        Yes they are. But it's not about that either. You are conflating end-user products with two companies developing to the same central point from opposing start positions. That is what this is about.

jayd16 3y ago

Why?

oneplane 3y ago
Qualcomm hasn't been able to manufacture desktop-class ARM devices yet (see SQ1 and X13 ARM edition), and the GPU IP doesn't really hold a candle to what else is out there either. They are however using significantly less power than say, x86 SoCs and GPUs from AMD, Intel or Nvidia. Nvidia on the other hand has plenty of IP for GPUs and AI workloads that is proven to have high performance, even on the Tegra scale of things. The Jetson product line as an example does still outperform whatever Qualcomm has come up so far. But they are consuming more power and don't even come close to the same thermal envelope. So Nvidia needs to reduce their energy needs, while Qualcomm needs to create actually performant devices for real-time AI workloads. That's why.