By advantage of its relentless pursuit of ever quicker, ever extra highly effective GPUs, Jen-Hsun Huang has claimed that Nvidia, over the previous 20 years, has pushed the “price of computing down by a million occasions”.
While you have a look at the rising prices of recent graphics playing cards in contrast with their forebears, that is possibly arduous to fathom. It certain appears like the price of a GPU has simply been steadily rising to most of us after we have a look at the objects of our silicon wishes. However if you have a look at simply what the graphics chips of at present are able to, the extent of uncooked computational energy on the disposal of even a lowly RTX 4060 would have appeared borderline legendary 20 years again.
A GeForce 6800 Extremely from 2005 delivered a whopping 6.4 GFLOPS, whereas the underside of the Ada Lovelace era comes with 15,100 GFLOPS of processing grunt. That is an entire world of distinction from a $499 card of 20 years in the past versus a $299 GPU of at present.
And that is not even a card wherever close to the highest of the stack, nor near what you will get from Nvidia’s strongest enterprise GPUs.
Jen-Hsun, at at present’s morning-after-keynote Q&A session, in contrast what Nvidia has performed in creating extra highly effective graphics silicon, pushing down the relative value of GPU computational energy, to the affect of Moore’s Regulation.
“The explanation Moore’s Regulation was so necessary within the historical past of the chip is that it drove down computing prices,” Huang remarks. “In the middle of the final 20 years we have pushed the marginal price of computing down by a million occasions.
RelatedPosts
“A lot that machine studying grew to become logical: ‘simply have the pc go determine it out.'”
Mainly, there’s a lot computational energy obtainable for such comparatively little money that you simply would possibly as effectively simply begin throwing it at AI to unravel all our issues. Or, you recognize, draw us an image of a gold fish if you completely, positively simply have to have a freshly generated image of a fish.
There is no getting away from it, the graphics card, and its constituent part, the GPU, have grow to be a very powerful items of silicon in our fashionable time. There’s additionally no getting away from the truth that Nvidia is liable for a few of the most necessary silicon of our time, nonetheless you are feeling concerning the rise and rise of synthetic intelligence and its potential affect on the world and humanity.
Does Jen-Hsun’s maths add up? I do not know, he did not present his workings, however what’s true is that for the reason that delivery of the GPU as we all know it, the cost-to-performance ratio has solely been going in a single route.
