The MSI Claw 8 AI+ Polar Tempest model first launched in April before being removed from MSI’s website, but has now returned with a dedicated product listing and a July 15 release date for the United States. This powerful handheld leaves the Steam Deck in the dust in most regards, but has been notably hard to purchase due to high demand and a limited initial production run.
The latest run of the Polar Tempest Edition comes with 2TB of storage and is priced at $999, versus the original Sandstorm model with 1TB of storage and an $899 price tag. In addition to a US release, fans have spotted listings in Germany, which suggest Europe will also get another release this summer. The only other difference is the white front panels (hence the Polar moniker).
The MSI Claw 8 AI+ Polar Tempest sports an 8-inch IPS display with a 120Hz variable refresh rate and a maximum brightness of 500 nits. It’s powered by an Intel Core Ultra 7 and has a whopping 32GB of RAM alongside an 80 Wh battery. Interestingly, it utilizes an Intel Arc GPU rather than the more common AMD or Nvidia GPUs found in most other handhelds.
MSI
If you’re used to playing a Steam Deck, your wrists will get a workout. The MSI Claw 8 is 795 grams compared to the 669 grams of the Steam Deck; that’s roughly 1.75 pounds in total, despite being thinner. Most of that added weight likely comes from the battery, given that it’s roughly twice the capacity of the Steam Deck.
The original MSI Claw didn’t review well and was largely overlooked in the market due to its poor performance. The Claw 8 AI+ corrected many of the mistakes from the first generation and offers a smooth gaming experience, even when playing games that can be truly demanding. This latest model provides fans a chance to get their hands on one with a fresh look and double the storage.
Buying a laptop might seem easy, but if you want the best gaming laptop within your budget, you’ll need to dig beyond surface-level information. With that said, every single spec should be readily available so that you know what you’re buying, but that has not always been the case.
Fortunately, Nvidia says that it’s working with its partners to make sure that one key spec is always listed: The total graphics power (TGP). Here’s why it matters.
Why does TGP matter in laptops?
Nvidia
Unlike desktops, laptops tend to be pretty straightforward to buy. Manufacturers and retailers always supply the basic specs, such as the CPU and the GPU, and that’s enough information for most gamers to be satisfied. However, what many people don’t know is that even within the same GPU model, there is a lot of room for error when shopping, and that all stems from the TGP.
The total graphics power, or TGP, refers to how much power an Nvidia GPU can draw at maximum. In desktop graphics, you’ll see massive TGPs of up to 600 watts; laptops are locked to 150 watts and below.
Laptop GPUs, such as the RTX 5090 in its mobile iteration, are very constrained by their power limits. The thinner and more power-efficient the laptop, the lower its power requirements, and that means less juice for the GPU. As a result, even within the same model of graphics card, you might see wildly different benchmark results and performance. An RTX 5090 that’s limited to the minimum TGP of 95 watts will not put out the same performance as one that can draw up to 150 watts.
This often overlooked spec is pretty key, and yet, it doesn’t always make its way into retail listings (in my experience, it often doesn’t). However, when laptop manufacturers also omit this information on their spec sheets, that’s when it’s really a problem.
We’ve seen this time and time again in laptops: power draw is crucial. Sometimes, a GPU that’s one model down — say the RTX 3060 instead of the RTX 3070 — would actually put out more frames than the pricier, better card, all due to power constraints. This made some laptops much less desirable than others.
Nvidia’s laptop GPUs used to get labels such as Max-Q and Max-P, and that helped. Max-Q referred to power-efficient versions of the card, and Max-P covered those focused on maximum performance and full power. These labels aren’t in use anymore, which made buying a laptop a little bit confusing.
Dynamic Boost is another thing that often goes unnoticed. It’s a technology that lets your laptop dynamically optimize the power allocation between the CPU and GPU, giving your graphics card an extra boost whenever it’s needed. In the case of Nvidia’s RTX 50-series laptops, the Dynamic Boost is set to 25 watts.
How Nvidia is addressing the problem of TGP in laptops
Nvidia
As reported by ComputerBase, Nvidia is aware that TGP, as well as other laptop GPU specs, often go overlooked. This is still true for Nvidia’s RTX 50-series.
Although the issue was much more prevalent in previous generations, you’ll still find laptops that don’t list the TGP as part of the laptop specs. VideoCardz spotted that some of the recent Razer gaming laptops, which have been reviewed upon the launch of the RTX 5090 mobile GPU, don’t tell you just how much power the GPU will have.
I took a look myself and found that indeed, Razer’s upcoming laptops — which are currently available to pre-order — do not disclose the TGP. The company mentions Dynamic Boost, but not the TGP, which could drastically alter the performance of the RTX 5090 housed within the thin laptop chassis. One would need to read the product description to find the TGP, which is still better than it not being there at all, but ideally, it should also be listed as part of the full specifications.
Mind you, these new laptops are pricey. The Razer Blade 16 that I was looking at here costs a whopping $4,499.99, although the price can be altered to be lower or higher depending on your choice of SSD, RAM, and CPU.
The good news is that spec sheets such as these might soon become a thing of the past.
In a statement made to ComputerBase, Nvidia said: “We work with laptop manufacturers to ensure they list graphics power on their product websites.” It’s a small declaration, but hopefully, this will be just the thing that eradicates the last of these cases of missing TGP.
Do you need to worry about the TGP in laptop GPUs?
Jacob Roach / Digital Trends
I know that this article might make it seem like buying a laptop for gaming is some complicated thing, but it isn’t — not really.
If you just want a laptop that works and you don’t feel the need to squeeze out every last bit of power it can provide, then it’s a much easier task; for help, check out our laptop buying guide.
On the other hand, if you want to make extra sure that you’re getting what you’re paying for, and high frame rates are your goal, the situation is flipped on its head. You should worry about the TGP, the Dynamic Boost, clock speeds, and video memory (VRAM). All of those things are deciding factors in the performance of your laptop graphics, and thus will have a massive impact on overall performance in games (and even during many types of work-related tasks).
Let’s hope that all of Nvidia’s laptop partners will soon make it very easy to spot all these key specs. At that point, all you might need to do is go to the manufacturer’s website to find them if the retailer fails to provide them. That would be the next step for Nvidia, although very hard, if not impossible, to enforce all around the globe.
One leak could be a fluke, but two leaks? MSI might be cooking. As per (now) two leaked images, MSI might have an exciting new GPU in the works, and it’d be one that could rival some of the best graphics cards. The GPU in question is another version of the RTX 5080, but this time, it’s said to come with 24GB memory — a major upgrade over the base version. This could finally push it past the RTX 4090, but will it really happen?
Both leaks were spotted by VideoCardz, but surprisingly, MSI itself is the original source for both stories. First, the company posted a promotional video showcasing the RTX 5080 Vanguard GPU, and on the box, it’s advertised as “24GB GDDR7.” This is an extra 8GB VRAM over the RTX 5080 that’s currently available. Coincidentally, it’s also the exact same memory capacity as the RTX 4090.
This could’ve been a mistake, but MSI then went on to post that exact same card on its website, listing it as being compatible with the MSI X870 motherboard. That’s twice now that MSI alluded to the existence of a 24GB VRAM RTX 5080.
The RTX 5080, despite many hopes and expectations, fails to beat the last-gen flagship RTX 4090. Bumping up the VRAM in such a major way, as well as overclocking the card (which MSI would undoubtedly do), would push the performance higher. Would it be high enough to beat the RTX 4090? Maybe not, as there’d still be a major difference in the CUDA core count — but it’d be a lot closer than it is now, that’s for sure.
The question is: Can we truly expect such a GPU to be revealed? Right now, it certainly doesn’t seem real, but it’s not impossible. More likely than not, we’ll see a better memory configuration in the RTX 5080 Ti or Super, whenever that makes an appearance. Unfortunately, I don’t expect it to be cheap, given that some RTX 5080 models already sell for up to $600 more than the MSRP.
