Next-gen GPU rumors are coming, and fast. The latest is that Nvidia’s upcoming RTX 40 series of graphics cards will be built with TSMC’s 4nm process node, giving the series a slight edge over the 5nm process AMD is expected to use for its RDNA3 GPUs. next generation.
Moore’s law is dead (by wccftech) believes that Ada Lovelace will be built using TSMC’s 4nm node. Talking about us can be confusing. It was thought that Nvidia would use a node called 4N which is a 5nm custom class process, however the MLID specifically states ‘4nm’. He knows what he’s talking about.
Nvidia Workstation ‘Hopper’ architecture it’s a 4nm product, but it’s a very high-margin product, and Nvidia can easily recoup the sales costs to ensure high-end wafer starts. We know that Nvidia is paying a lot of money for access to TSMC production, but is the asking price of 4nm too high for a lower-margin GeForce card?
Confused? So do we. If Ada Lovelace turns out to be a 5nm or 4nm or 4N or whatever, the vast majority of consumers won’t pay attention. As long as the chosen card performs as expected, the node in use will not matter.
Notice I didn’t mention PCIe 5.0 in my Lovelace slides: https://t.co/n7sdIL1UkuSpeaking of “4”, #NVIDIA Lovelace is indeed 4nm!🤠🎉😋 More to say about the upcoming Broken Silicon… https : //t.co/jjJ1GSod0jApril 24, 2022
Whether that means Nvidia will actually have a single-nanometer advantage over competing AMD products remains to be seen, but in the end, it might not make much of a difference. Rumors point to Ada Lovelace being very hungry for powerwhich means Nvidia isn’t using the small power advantage it might have enjoyed as it seeks maximum performance.
There is also the possibility that AMD’s chiplet design may utilize different nodes, making node comparisons a largely pointless endeavor. Your cards should also increase power consumption.
Everything will become clear with time. The RTX-40 series cards are expected to launch sometime in the second half of 2022.