SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Qualcomm Moderated Thread - please read rules before posting -- Ignore unavailable to you. Want to Upgrade?


To: pheilman_ who wrote (196170)10/26/2025 9:41:26 PM
From: waitwatchwander1 Recommendation

Recommended By
Dr. John

  Respond to of 196358
 
Yes, HBM3 will always be close to the processing. The memory pools Qualcomm could offer will be low power, more latent and used for storage of models to be fed into GPU/NPU processing memories.

With LLMs cases there are lots of options with the mgt of the necessary parameters, diversity and number of models and the bringing together of appropriate hardware components.

It will be interesting to see the niche Qualcomm has identified.



To: pheilman_ who wrote (196170)10/27/2025 1:10:45 PM
From: waitwatchwander1 Recommendation

Recommended By
Dr. John

  Read Replies (1) | Respond to of 196358
 
Near memory computing ...

Is this like treating a memory pool like a huge cache?

Beyond addressing space how does movement of bits come into play here?