Nvidia’s plans for AI GPUs could upend PC gaming forever
Nvidia plans to significantly increase the pace at which it releases AI and heavy-duty graphics cards, and that could impact the company’s GeForce gaming GPUs.
We hear about it Semi-analysis (through Tom’s hardware) that Nvidia has big plans to dominate the AI landscape and fend off rivals by increasing the speed at which it introduces new GPUs, releasing a new range every year based on a new architecture.
So next year we’ll have Blackwell GPUs (B100), probably later in 2024, as there are also plans to debut new Hopper products (H200), so those will come first. And then Nvidia will release another new architecture in 2025, currently called ‘X’ (apparently there’s no architecture codename yet), so we’ll have the X100.
This is based on a roadmap taken from an investor presentation put together by Nvidia, although it doesn’t confirm a follow-up schedule for the annual release after that, so in that case we’ll have to take the report’s word for it.
There’s a lot more detail in all of this, but we won’t go into the details. Right now, gamers may already be gawking at their monitors and wondering why they should worry about data center GPUs. Now let’s dig into that in more detail…
Analysis: It’s all about the profit margins
Okay, so if Nvidia is going to design and implement a new architecture every year, does that mean we’ll also get new GeForce gaming GPUs every year in the future? Well no, in a word.
But isn’t Blackwell, Nvidia’s next-gen data center products, also the architecture on which the next-gen GeForce is based? Yes, it is, but that doesn’t mean the X architecture will spawn its own GeForce twist the following year (and so on).
Far from it. We’re not going to get a new crop of candidates for the best gaming GPUs from Nvidia every year, that’s just not going to happen (it’s simply not necessary).
What Nvidia’s annual AI GPUs likely herald is a further sidelining of GeForce products. Team Green is really pushing the AI side of the equation to maintain its advantage, given the large profit margins to be had there, so gaming GPUs will become a lower and lower priority.
This is partly why Nvidia has been driving prices so hard in recent generations, especially with Lovelace. Firstly, because it can, and secondly, even though in many cases it loses the goodwill of PC gamers with unreasonably high price tags – especially at the higher end of the market – it just doesn’t matter that much to the company . It’s not an area where the hefty profits are being made, so Nvidia feels free to squeeze in quite hard to get the most out of GeForce for the time being.
If gamers get really frustrated and stop buying GeForce – not that there’s any sign of that yet, not in the overall GPU picture – then Nvidia will just stop making them. Or that’s how it feels like things could play out, given the focus on AI, with an annual release frequency – if it actually goes ahead of course – further increasing the priority on these heavy graphics cards.
In short, instead of more mainstream gaming GPUs, we’ll likely see less frequent GeForce ranges, and indeed the possibility we previously raised that Nvidia could even stop making these products altogether. (The RTX 5000, and then perhaps an RTX 6000 series, may be the last efforts we see).
That’s a big call and just wild speculation at this point, but it’s not an unimaginable situation. Especially if the AI market continues to grow into a cash cow of gargantuan proportions, and Nvidia becomes consumed with milking it, which the company would be foolish not to do, let’s face it.