Yeah, in the end that Eloq 4.7 test was not good and a success, but it got me to realize: The odd artifacting does not happen at speech rates below 60! Very baffling, as you can set it to 55 even with the older one and you won't hear it. Then crank it to 60 and press the number 3, and boom, it's there. I also realized why it worked for some and not others! The copy I had used for myself made one additional change to pipe the audio not through NVWave but direct MME, but the one I published I only made one single change to load the other library to quickly swap it in to the add-on. That's what I get for doing it at a time when I'm already exhausted :) Anyone who has the broken driver can revert it to the latest copy and it'll go back to the one you had before all that mess. (eurpod.com/Eloq47-2024.nvda-ad…)
![antirez 3 hours ago | next [–]
About "people still thinking LLMs are quite useless", I still believe that the problem is that most people are exposed to ChatGPT 4o that at this point for my use case (programming / design partner) is basically a useless toy. And I guess that in tech many folks try LLMs for the same use cases. Try Claude Sonnet 3.5 (not Haiku!) and tell me if, while still flawed, is not helpful.
But there is more: a key thing with LLMs is that their ability to help, as a tool, changes vastly based on your communication ability. The prompt is the king to make those models 10x better than they are with the lazy one-liner question. Drop your files in the context window; ask very precise questions explaining the background. They work great to explore what is at the borders of your knowledge. They are also great at doing boring tasks for which you can provide perfect guidance (but that still would take you hours). The best LLMs (in my case just Claude Sonnet 3.5, I must admit) out there are able to accelerate you.
duped 2 hours ago | parent | prev | next [–]
> Try Claude Sonnet 3.5 (not Haiku!) and tell me if, while still flawed, is not helpful.
It's not as helpful as Google was ten years ago. It's more helpful than Google today, because Google search has slowly been corrupted by garbage SEO and other LLM spam, including their own suggestions.
mvkel 1 hour ago | parent | next [–]
I'm surprised at the description that it's "useless" as a programming /...](https://fedi.ml/photo/preview/640/625051 "antirez 3 hours ago | next [–]
About \"people still thinking LLMs are quite useless\", I still believe that the problem is that most people are exposed to ChatGPT 4o that at this point for my use case (programming / design partner) is basically a useless toy. And I guess that in tech many folks try LLMs for the same use cases. Try Claude Sonnet 3.5 (not Haiku!) and tell me if, while still flawed, is not helpful.
But there is more: a key thing with LLMs is that their ability to help, as a tool, changes vastly based on your communication ability. The prompt is the king to make those models 10x better than they are with the lazy one-liner question. Drop your files in the context window; ask very precise questions explaining the background. They work great to explore what is at the borders of your knowledge. They are also great at doing boring tasks for which you can provide perfect guidance (but that still would take you hours). The best LLMs (in my case just Claude Sonnet 3.5, I must admit) out there are able to accelerate you.
duped 2 hours ago | parent | prev | next [–]
> Try Claude Sonnet 3.5 (not Haiku!) and tell me if, while still flawed, is not helpful.
It's not as helpful as Google was ten years ago. It's more helpful than Google today, because Google search has slowly been corrupted by garbage SEO and other LLM spam, including their own suggestions.
mvkel 1 hour ago | parent | next [–]
I'm surprised at the description that it's \"useless\" as a programming /...")
JamminJerry
in reply to Tamas G • • •Tamas G
in reply to JamminJerry • • •JamminJerry
in reply to Tamas G • • •Tamas G
in reply to JamminJerry • • •JamminJerry
in reply to Tamas G • • •