Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
Nvidia now requiring laptop vendors to list GPU power
https://www.theverge.com/2021/2/5/22266921/nvidia-requiring-companies-share-clock-speed-power-gaming-laptop-rtx-30-series-gpu-specs
In some previous generations, Nvidia introduced their Max-Q laptop GPUs. The idea is that if you take a high end laptop GPU and clock it a lot lower, then it will burn less power, allowing it to fit into a thinner and lighter laptop. Of course, it will also offer a lot less performance. Even so, if it was branded Max-Q, you knew that, while the GPU silicon was the same as the normal version, the performance would be much less.
Then Nvidia decided to drop the Max-Q branding for the new RTX 3000 series laptop parts. This made some amount of sense, as instead of having only two possible power levels, they could allow laptop vendors to pick from a lot more intermediate options. The problem is that it meant that two different laptops that both had an RTX 3080 might offer very different levels of performance. For that matter, a laptop with an RTX 3060 and a 115 W TGP will often feature more GPU performance than a laptop with an RTX 3080 and an 80 W TGP.
Nvidia "encouraged" laptop vendors to list the TGP of the GPUs that they used. Most didn't. So now Nvidia is changing it to require laptop vendors to list the TGP in the laptop specs. This is a good thing, as it means that now instead of having to take a wild guess as to how fast an Nvidia laptop GPU will be, it will be right there in the specs. You just have to remember that performance scales both with higher model number (within a generation) and also with higher TGP.
That said, it isn't automatic that you want the highest possible TGP, nor that the higher end cards like the RTX 3080 are a waste. Higher power usage means more heat, which requires a more robust cooler, and that makes the laptop thicker and heavier. Higher end cards can scale well to higher power usage than lower end cards. And more shaders clocked lower can beat fewer clocked higher in the same TGP, so an RTX 3080 with a TGP of 80 W will tend to be meaningfully faster than an RTX 3060 with the same TGP of 80 W. Giving consumers information about what they're buying so that you can make an intelligent decision is a good thing.
In some previous generations, Nvidia introduced their Max-Q laptop GPUs. The idea is that if you take a high end laptop GPU and clock it a lot lower, then it will burn less power, allowing it to fit into a thinner and lighter laptop. Of course, it will also offer a lot less performance. Even so, if it was branded Max-Q, you knew that, while the GPU silicon was the same as the normal version, the performance would be much less.
Then Nvidia decided to drop the Max-Q branding for the new RTX 3000 series laptop parts. This made some amount of sense, as instead of having only two possible power levels, they could allow laptop vendors to pick from a lot more intermediate options. The problem is that it meant that two different laptops that both had an RTX 3080 might offer very different levels of performance. For that matter, a laptop with an RTX 3060 and a 115 W TGP will often feature more GPU performance than a laptop with an RTX 3080 and an 80 W TGP.
Nvidia "encouraged" laptop vendors to list the TGP of the GPUs that they used. Most didn't. So now Nvidia is changing it to require laptop vendors to list the TGP in the laptop specs. This is a good thing, as it means that now instead of having to take a wild guess as to how fast an Nvidia laptop GPU will be, it will be right there in the specs. You just have to remember that performance scales both with higher model number (within a generation) and also with higher TGP.
That said, it isn't automatic that you want the highest possible TGP, nor that the higher end cards like the RTX 3080 are a waste. Higher power usage means more heat, which requires a more robust cooler, and that makes the laptop thicker and heavier. Higher end cards can scale well to higher power usage than lower end cards. And more shaders clocked lower can beat fewer clocked higher in the same TGP, so an RTX 3080 with a TGP of 80 W will tend to be meaningfully faster than an RTX 3060 with the same TGP of 80 W. Giving consumers information about what they're buying so that you can make an intelligent decision is a good thing.
Comments
Most big vendors do not usually worry about the consumers getting shafted by these type tricks, of selling stuff with misleading info.
Nice job Nvidia.
If Nvidia is trying to make sure the consumer is not mislead by lame marketing tactics then kudos to them but I somehow doubt that is their true agenda.
Never forget 3 mile Island and never trust a government official or company spokesman.
I would say, outside of forums like these (and here, I use that loosely) - most people only know that a bigger number is probably better. And they are going to buy whatever either the salesperson at Best Buy pushes on them, what their favorite Twitch streamer uses, or what some good review on a random internet site says is good.
And then they will complain the laptop is hot, doesn't last long on battery, etc.
It will be just another random number on the side of the box that they don't really know what it means.
I don't dispute that more information is better than less. But this doesn't really give the consumer useful information in a way most people will understand.
https://videocardz.com/newz/chinese-gpu-miners-are-now-bulk-buying-geforce-rtx-30-laptops-to-mine-ethereum