The arrival of Nvidia’s GeForce RTX 30 Series GPUs on laptops, following their smashing success on desktops, promises big performance gains and attainable ray tracing on mobile machines. Based on the same “Ampere” architecture that delivers incredible performance at a fair price in full-size desktop graphics cards, the RTX 30 Series laptops GPUs have come zooming in this week to high expectations.
Having just completed our first round of real-world testing of RTX 30 Series laptops, though, it turns out that things are not quite that straightforward. Through the process of reviewing three different laptops, each with a new GeForce RTX 3070 GPU, we have seen some impressive performance numbers—but also some variations in frame rate and performance that we could not ignore.
how we test laptops.)
You can quickly see the issue: For three GPUs with the same name, there’s quite a range of results, especially where the Asus TUF Dash laptop is concerned. It’s a multi-faceted issue, but the larger part of it ultimately lies with the wattage deployed to each GPU, Nvidia’s Max-Q design branding, and how laptop makers disclose this information.
How Max-Q Has ‘Matured’
To help understand the issue, you first need to understand Max-Q. We have a full explainer of what exactly Max-Q has meant during the past few years, but here’s the quick version.
Nvidia worked with manufacturers to develop Max-Q implementations of top-end GPUs (for example, a Max-Q RTX 2070) to fit inside thinner gaming laptops. These Max-Q editions have had down-tuned power in order to restrict their thermal output, necessitating less cooling hardware and space, thus allowing them to fit in slimmer machines. The compromise is that the limited power, naturally, led to lower frame rates versus the “full” version of any given GPU. (Incidentally, in laptop-maker lingo, the full-strength version of these GPUs deployed in larger machines were often dubbed “Max-P.”)
Hot Hardware, had a different last line:
Ultimately, like all laptop features and specs, it is up to the OEM to market what their particular laptop configuration supports.
We suspect Nvidia and its OEMs may have caught some heat for that statement, potentially leaving consumers to fend for themselves. (Indeed, we’d be the first to fire things up.)
The Range of Things to Come
Why? In essence, the Max-Q suffix is no longer a quick-glance designation of its capabilities, and you may well not see it on a given laptop’s spec sheet. Instead, Max-Q is more of a technology-platform designation for the new GPUs, encompassing a host of power-saving aspects, and you’ll instead need to look at the detailed specs of each laptop for a gauge of how the GPU is implemented, for an idea of how well it will perform.
That’s a difficult ask for laptop shoppers, who need to dive in to the specs to get a theoretical idea of a laptop’s power. If you’re an enthusiast, you may get a clue of sorts from reading specs, but the less tech-savvy will have difficulty parsing the nuances. Even GPU names alone can be confusing for many shoppers.
On top of that, with the RTX 30 Series, each GPU can be implemented at a wide range of TDP values and varying peak boost clocks, really blurring the lines between GPUs of the same name. See Nvidia’s GeForce RTX 30 Series summary chart below for what I mean…
GPU-Z to log the GPU wattage as the test ran. The Asus TUF Dash F15 peaked at 62 watts TDP and 86 watts TGP, far lower than the Alienware m15 R4’s 107 watts TDP and 141 watts TGP.
If you saw those TDP and frame-rate gaps without knowing which GPUs they were, you could be excused for assuming they’re entirely different GPUs. That illustrates the problem with the lack of clarity around this topic moving forward: Paying for an RTX 3070 set up in a lighter power-consumption state could mean you get something implemented in a way that brings it closer to a RTX 3060. In other words, some RTX 3070s might be equal in name only, and it could be on you to figure out which is which.
January 27, 2021