Nvidia is also launching the first Ada Lovelace graphics cards for notebooks at CES 2023: the GeForce RTX 4000 laptop GPUs. With RTX 4090, 4080, 4070, 4060 and 4050 laptop GPUs, the range right at launch is bigger than ever. OEMs will have even more leeway with the TGP in the future.
Die GeForce RTX 4000 Laptop GPUs im Detail
Compared to Ampere two years ago, Nvidia brings more variants right from the start, including the 90 series for the first time in the notebook, until now it was the end of a GeForce RTX 3080 Ti laptop GPU. Thus, the "biggest" GeForce was reserved for the desktop PC to date, which in principle was consistent with the fact that notebooks generally only used the second-biggest desktop GPU (GA103) instead of the biggest (GA102).
Again, given that Nvidia (and AMD) consider the nomenclature in desktop and notebook to be separate, there was no argument why the fastest mobile graphics unit shouldn't also have the same name as the fastest model in desktop. With the new generation of laptop GPUs, Nvidia is taking exactly this.
Nvidia promises the biggest leap in performance in generations, which is largely due to the biggest leap in efficiency Nvidia has ever realized in a notebook. Based on Ada, Nvidia promises the same performance in the notebook as with Ampere with only a third of the power consumption - even if Ada has already proven to be extremely efficient, DLSS 3 should have been used for this comparison.
A look at the key data compared to the old generation reveals: The two largest models and the RTX 4050 laptop GPU are better than the old generation in terms of shaders, while the 4070 and the 4060 are weaker. Thanks to the new, much higher clocked architecture, both will still be faster, but the premium should be slightly smaller compared to the 4050, 4080 and 4090.
The TGP bandwidth will be even larger
What is also striking is that all known series have even wider TGP bands than they have had up to now. It is also interesting that the RTX 4070, 4060 and 4050 for notebooks operate in exactly the same TGP range and therefore the corresponding notebooks from Nvidia are also mentioned in the same breath. This means that the issue of transparency in the communication of the TGP actually used in a notebook should be even more important than with RTX 3000 laptop GPUs. Nvidia is again completely silent on the Dynamic Boost surcharge that is to be added to the TGP depending on the model.
How transparent is the TDP handled?
As a reminder: With the GeForce RTX 3000 laptop GPUs, Nvidia generally wrote “laptop GPU” in the designation and thus basically made it clear that it is a mobile variant. In return, "Max Q" as a label for the particularly efficient, but also particularly slow versions within a variant (e.g. RTX 3080 laptop GPU) was deleted. How fast a specific GeForce RTX 3080 laptop GPU was was not clear from the designation and the technical data for the notebook.
Available from February 2023
Notebooks with the RTX 4000 laptop GPU are expected to go on sale in February. The top models with a mobile RTX 4090 or 4080 will be available from February 8 from $1,999 before taxes, the variants with a mobile RTX 4070, 4060 or 4050 from February 22 from $999 before taxes.
AMD is expected to introduce new competitor models to the GeForce RTX 4000 laptop GPUs with AMD Radeon RX 7000 Mobile. In particular, the issue of power consumption and efficiency should be critical, because RDNA 3 has not been able to convince in this respect compared to Ada Lovelace on the desktop.
