Nvidia is Requiring Laptop Makers To Be More Transparent About RTX 30-series Specs

Nvidia is now requiring, not just encouraging, companies selling laptops with its new RTX 30-series graphics chips to be more transparent about the kind of power people can expect. From a report: Nvidia tells The Verge these companies will have to disclose specific clock speed stats and total graphics power on online product pages -- all of which tells people everything they need to know about a laptop's graphics potential, for better or worse. However, companies won't have to mention that these chips are Max-Q variants because, according to an Nvidia spokesperson, "Max-Q is no longer part of the GPU name." Rather, Max-Q is now solely used to communicate that a laptop with an RTX 30-series graphics chip ships with efficiency features like Whisper Mode 2, Dynamic Boost 2, and Advanced Optimus. Previously, seeing Max-Q branding made it easy to determine a laptop's general performance without having to know its specific clock speeds. It's encouraging to see Nvidia no longer allows companies to hide this vital information from marketing materials. It should go far enough in helping buyers make an educated purchase without having to wait on reviewers and early adopters to report on the specs.

I doubt this will help much.

[Lordpidey]

Honestly, most people don't understand what makes a video card good.

We're nerds, so we understand rasterization vs ray tracing.

But if you ask the average person whether they will want "Bilinear" or "Arthroscopic" filtering, they won't know.

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[Lordpidey]

The problem is those benchmarks often use terms.

"Fill rate? Raycast speed? Uhh, which one does my favorite game use?"

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[DontBeAMoran]

unless you're going with a laptop with a 30 series card, 30 series cards on laptops with identical model numbers can perform vastly different, for example a 3070 could perform better than a 3080.

Which is exactly what this whole discussion is all about. nVidia made a mess out of their own product numbers and now the laptop manufacturers are supposed to be the ones to fix that? This makes no sense to me.

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[thegarbz]

All of that is completely irrelevant. Consumers can understand bigger number vs smaller number, and clock speed and core count are fundamentally determiners of generalised performance for GPUs.

At least a step towards making something understandable compared to the previous clusterfuck which directly led to people thinking the RTX2070 Super Max-Q was a faster performing GPU than a normal RTX2070 Super in a desktop, when it reality it was actually slower than previous generations GPU's of the class.

Hell when

Transparency.

[Menelkir]

Quite funny to see nvidia asking for "transparency" isn't?

Re:

[DontBeAMoran]

And it's really not that hard to do, too!

#Nvidia
{
  opacity: 0.5;
}

Not sure I understand

[robi5]

Before, Max Q had to be shown, and it's not true anymore, according to the article. How does that improve transparency?

Re:

[im_thatoneguy]

Because before a Max Q would imply that all Max Q cards were similar in performance.

Now if you thermal throttle down to 2Ghz while another card runs at 3Ghz you have to clearly label it as 33% slower.

Intel does the same thing. i7-1920u @ 3.2Ghz vs i7-1920u @ 2.4Ghz.

Re:

[thegarbz]

Before, Max Q had to be shown, and it's not true anymore, according to the article. How does that improve transparency?

If you RTFS you'd realise that Max-Q was a problem because people thought it meant faster. Removing Max-Q is only half of the change, the other half is *REQUIRING* companies to display clock speed one thing that really heavily separated the very much slower (despite what consumers thought) Max-Q from their normal brothers.

Doesn't really help.....still need to read reviews

[Computershack]

It should go far enough in helping buyers make an educated purchase without having to wait on reviewers and early adopters to report on the specs.

Unfortunately you still need to do that. Although the GPU is the same the cooling efficiency of the individual model of laptop massively affects the clock speeds and boost speeds it can achieve. So you can have the same RTX 3080 mobile GPU in two laptops and one will perform significantly worse than the other because it has a poorer cooling solution so it's base clock speed is lowered and the amount it can boost by is more restricted than the other model with better cooling.

We have a perfect CPU example of this with the new M1 Macbook Air and Pro. They both use the same CPU but different cooling solutions. The Air has no fan and a small metal heatsink for cooling the CPU, the Pro has a traditional heatsink and fan arrangement. In sustained use the Air is slower than the Pro but if you add a thermal pad to the top of the Air's CPU heatsink so it can use the base of the laptop as a heatsink then the Air's CPU will run the same as the Pro.

Laptop specs are confusing

[TJHook3r]

Shops are selling 'gaming laptops' for £500 that will definitely not play the latest games - I don't know why they are allowed to advertise as such and I know of well-meaning parents that have been caught out. At least buying a PS5 guarantees 5 years of being able to play top-tier games.