Something to look forward to: All indications are that Nvidia is preparing to launch the mobile RTX 30 series of GPUs early next year, which could cause a significant loss of performance for the upcoming laptops. It is confirmed that a GeForce meeting will take place on January 12 at CES, where Nvidia may unveil new GPU versions on the PC as well as its mobile counterparts to launch in 2021.
The RTX 3060 Ti, RTX 3070, RTX 3080 and the RTX 3090 graphics cards are likely to be hard to find over the coming weeks due to a combination of scalpers, a tight supply and a huge demand, but at least we know how well they perform (they are fairly fast), as well as how it compares to AMD’s Radeon 6000 series GPUs.
Not much has been made official about the mobile counterparts of the RTX 30 series, but it is confirmed that Nvidia will be hosting a GeForce event on January 12, where it could unveil the elusive RTX 3060 and RTX 3080 Ti for the desktop. The company is also likely to take the opportunity to announce the mobile RTX 30 series GPUs.
Thanks to the European laptop manufacturer SKIKK in the Netherlands, we know that a version is threatening and that we can expect them to improve just as much from the previous generation as the desktop parts.
According to complete list which was quickly removed from the manufacturer’s website, mobile Ampere will first arrive in 15-inch and 17-inch laptops. Based on what we know so far, the RTX 3080 and RTX 3070 will be in Max-P or Max-Q configurations, while the RTX 3060 will only come in a standard configuration.
Notably, the RTX 3060 will only come with 6GB of GDDR6 with a 192-bit memory bus, while the RTX 3080 with 8 or 16GB will come with a 256-bit memory bus. The RTX 3070 is the middle ground with 8GB of GDDR6 on a 256-bit memory bus.
Notebookcheck writes that the RTX 3080 mobile GPU will be based on the same GA104 GPU core as the desktop RTX 3070 GPU, albeit clocked at 6144 CUDA cores up to 1.7 GHz, depending on the model.
The mobile RTX 3070 and RTX 3060 will presumably contain 5120 and 3072 CUDA cores, respectively. The expected TGP is 115-150 watts for the Max-P variants and 80-90 watts for the Max-Q variants.