Nvidia CEO proclaims ‘Moore’s law is dead’ over RTX 40-series GPU pricing

After the recent reveal of Nvidia’s 40-series graphics cards, you’re probably one of many scrunching your nose up over the pricing. Well, the company’s CEO, Jensen Huang has some words of wisdom to share with you and perhaps ease your wallet into opening just that little bit wider.

In a press conference, Gordon Ong of PC world asked, “Is there anything you would like to say to the community regarding pricing on the new generation of parts, as well as, can they expect to see better pricing at some point and basically address all the loud screams that I’m seeing everywhere?” 

After complimenting the guy on his choice of R2D2 background (seconded), Jensen had this to say:

“Moore’s law is dead. And the ability for Moore’s Law to deliver twice the performance at the same cost, or the same performance at half the cost every year and a half is over. It’s completely over.

“And so the idea that the chip is going to go down in cost over time, unfortunately, is a story of the past.”

These are bold words from the green team, reflected in the price difference we’ve been presented with between the previous generation’s RTX 3080/RTX 3090 cards and their respective successors in the RTX 40-series.

Looking at the $200 uptick from our beloved 10GB RTX 3080 to its 12GB RTX 40-series successor likely gave you a bit of a shock. There may be an increase in clock and memory speeds, but the numbers are almost negligible, especially when you spot the reduction in memory bandwidth. 

The pricing vs. specification here has had a good deal of people wondering whether the RTX 4080 12GB should go under an entirely different name. And the decision is made even more confusing when you consider just how massively cut back the newer cards are from their respective 90-tier sisters.

The difference between the RTX 3090 Ti and the RTX 4090 perhaps gives the story a little more context.

While there’s been an exceptional increase in the number of CUDA cores, as well as improvements to the base and boost clock speeds between the Hopper and Ada Lovelace flagships, Nvidia has still managed to bring the successor in at $400 less than its predecessor.

That evens everything out a little, and it’s clear there’s never an absolute correlation between specs and price. As our Jeremy points out, the numbers are not as clear-cut as they seem.

To be fair to Nvidia, we are looking at a complete switch in chip supplier—from Samsung to TSMC—between the 30-series and 40-series cards. Jensen points out that “a 12-inch wafer is a lot more expensive today than it was yesterday. And it’s not a little bit more expensive; it is a tonne more expensive.”

Since TSMC’s 4nm production capacity is highly sought after right now, we can only speculate just how much Nvidia is being charged for manufacturing the Ada Lovelace processors. Ultimately, we can’t expect Nvidia to swallow all the extra manufacturing costs if that’s been the case; some of it will inevitably trickle down to the consumer.

The question remains, are users willing to spend what Nvidia is asking for an RTX 4080 card? The recent scramble for the RTX 30-series cards would suggest it’s possible, but then we’ve got less to contend with this time around, thanks to the Ethereum merge all but killing GPU mining.

Only time will tell, but our bet is the cards will still sell out pretty darn fast. 

About Post Author

See author's posts