It’s the quintessential problem within the PC industry: how do you jam the maximum amount of horsepower into the smallest space possible? That’s a question that has been plaguing the notebook market since its inception, particularly when manufacturers went to build systems meant for gaming. To a certain extent, NVIDIA’s Pascal architecture brought the desktop / notebook divide closer than it’s ever been, bringing at least a partial answer that that question. But now with their MAX-Q design, high-end compact gaming laptops are one more tentative step closer to reality.
The goal behind MAX-Q was a relatively straightforward one: harness the existing higher-end Pascal notebook GPUs – which already matched the performance specifications of their desktop brethren – and cram them into a space-defying form factor. Unfortunately that’s easier said than done, since a notebook has several other components that needed to be brought into alignment if there was any hope of delivering the optimal blend of size and performance that NVIDIA was looking for.
Just look at how things have progressed in a relatively short amount of time. About 3 years ago NVIDIA introduced the GTX 880M, a graphics processor that was – for the time at least – considered an extremely potent gaming companion. However, it was basically a gussied up rebrand of the GTX 780M, which itself was based upon a lower-end desktop card (the GTX 770). Even with those lowered specifications taken into account, the GTX 880M-based notebooks were heavy, thick, ungainly, loud beasts. On average they were over 5cm thick and weighed in at a portly 10lbs or more.
Compare this situation to the GTX 1080 with MAX-Q design. A notebook based off of it will conceivably offer more than three times the performance while utilizing a chassis that is 18mm thick or less. Think of this as a full gaming computer being shoehorned into a structure that’s no thicker than an early 2016 Macbook Pro. More importantly, these notebooks may actually end up being less expensive than the behemoths of yesteryear since OEMs don’t need to use such extensive cooling assemblies or a massive chassis to insure proper functionality.
One of the tenets of modern transistor design is that any piece of silicon will reach optimal performance output at a relatively low power input level. Anything above that point and all that happens is throwing power into the wind without resulting in all that much when it comes to extra performance. However, many graphics processors are made to take advantage of this relatively small gain with features such as NVIDIA’s GPU Boost and AMD’s PowerTune algorithms.
Another good example of this is how actual in-game performance doesn’t scale in a linear fashion when overclocking. We’ve constantly seen that overclocking a graphics card 10% results in significant thermal and power input increases, but the associated frame rate boost isn’t anywhere close to that 10%.
What MAX-Q endeavors to accomplish is find that “sweet spot” of efficiency and performance by sacrificing the minimal gains that are typically netted through higher, more power hungry core frequencies. Actually, this is just NVIDIA’s fancy way to state that the architecture is being underclocked without offering up framerates like a sacrificial lamb.
As you can see by the chart above, any NVIDIA mobile graphics card with the MAX-Q designation will indeed feature lower clock speeds than its closest sibling, but there are still some important nuances to take note of. First and foremost, only the GTX 1080, GTX 1070 and GTX 1060 will undergo the MAX-Q treatment. The mobile versions of NVIDIA’s GTX 1050 Ti and GTX 1050 were designed from the ground up with efficiency in mind and have indeed already proven their efficacy in the thin & light segment.
Another thing to look closely at is the actual frequencies these graphics cards reach since they vary to an almost wild extent. Whereas memory frequencies and capacities remain unchanged, the Base and Boost clocks have a range of almost 200MHz in some instances. The reason for this is straightforward though: notebook manufacturers are able to work within that clock speed range to optimize these cards for optimal performance within a given design. For example, an ultrathin gaming notebook may require a GTX 1080 to run at 1101MHz / 1278MHz rather than a higher spec due to various thermal and spatial limitations. Meanwhile, a slightly thicker design may be able to push the GTX 1080 MAX-Q’s performance towards the upper echelon of those numbers.
Regardless of the somewhat murky process of choosing the right MAX-Q GPU for your purposes, there’s no denying NVIDIA’s tight grip upon clock speeds has done wonders for power requirements. That once-ravenous GTX 1080 gets pushed down below what a mobile GTX 1070 once needed. That in itself speaks volumes.
With that being said, MAX-Q is more of a design philosophy than a technology, since achieving peak efficiency through optimal clock speeds is just one of its four principle cornerstones. It is also achieved by leveraging GeForce Experience’s vast array of automatically applied game optimization presets to deliver playable framerates at lower TDP levels.
Meanwhile, every notebook with a MAX-Q graphics card will also have an advanced multi-stage thermal solution that is tailor-made for each individual chassis design. According to NVIDIA, by teaming up closely with laptop manufacturers at the design stage, they are better able to guide best practices and ultimately achieve a more holistic end result.
Moving beyond the notebooks themselves, NVIDIA has implemented a new regulator design on all MAX-Q cards. While we have very little in the way of concrete details about what that entails, it looks like quicker switching frequencies, optimized PCB trace routing and upgraded components factor heavily into its implementation.
Another key addition is something NVIDIA is calling Whisper Mode, and in my opinion this is arguably the most notable addition to MAX-Q’s stable of features. At one point in time we’ve all seen claims of fast, thin gaming notebooks only to be met with fans that wail like banshees as they try to expel the system’s rampant heat buildup. Now to be honest with you, MAX-Q doesn’t guarantee silent running despite NVIDIA’s promises of operational acoustics of 40 decibels. If necessary, a notebook’s system fans can go above that if the situation warrants, thus pushing noise to unacceptable levels.
To combat this, Whisper Mode is selectable within GeForce Experience and can be completely customized by the end user. With this feature you are able to set noise, in-game setting and framerate targets in an effort to balance out acoustics with performance. Think of this as NVIDIA’s Battery Boost and Frame Rate Target technologies further stepping into their next logical evolution. With it we can also now see how GeForce Experience itself is evolving into a common launching platform with a myriad of different features for desktop and notebook users alike.
One of the best examples of the possibilities with MAX-Q is the new ASUS ROG Zephryus GX501, which we saw at Computex. By just looking at the Zephryus, you wouldn’t be mistaken to think it resembles an ultrabook from last year. It’s razor thin at just 17.9mm thick and it weighs just 2.24 kilograms. This is actually the thinnest ROG notebook ASUS has ever created, but it is also one of the most powerful.
While there will be many differet configurations available, in its highest-end form there will be an Intel Core i7-7700HQ processor and a GTX 1080 graphics card housed within. There’s also up to 32GB of DDR4-2400 memory and an insanely fast NVMe SSD capable of pushing speeds up to 3500 MB/s that will be available in capacities up to 1TB.
Those components power a 15.6” 1080P display with an optional G-SYNC module and refresh rates of up to 120Hz. Supposedly ASUS will also be offering the Zephryus with an IPS display, but there’s no information yet about whether or not that will be the 120Hz G-SYNC option. We should know more in a few weeks.
I talked about advanced thermal dissipation systems a bit earlier, and ASUS engineers created something called the Active Aerodynamic System or AAS to keep all the components cool. Basically, when you lift the lid a small bottom panel opens up which allows the fans to breathe in fresh air. It is a brilliant design that works alongside interior air ramps that insure cool air is always circulating around the heatsinks. Meanwhile, the CPU has a dedicated blower fan, radiator, and heatpipe array while the GPU uses the same design but adds a second heatsink for good measure.
Supposedly all of these innovations allow the Zephryus GX501 to achieve a noise level of just 40 decibels alongside a complete lack of throttling when gaming. Then again, there’s no word on exactly which specification of GTX 1080 MAX-Q this notebook will use. ASUS is also promising lower surface temperatures than competing solutions, and thank goodness for that since the last thing anyone wants is to burn their lap while gaming.
OK, about that odd keyboard positioning that I am sure you were all asking yourselves about…..it is weird to say the least, and it won’t be a trend among all MAX-Q designs. Supposedly, the keys are moved closer to the front edge to attain a more desktop-like hand position, and we can believe that since your wrists won’t have to sit atop the notebook’s edge when typing. It is also pushed forward since ASUS needed to maximize the chassis depth above critical components to maximize heatsink size and airflow potential. Meanwhile, the pinholes above the area where a normal keyboard would be located are used for additional fan intakes.
While the Zephyrus gave us a tantalizing look at what the future will hold for gaming notebook design, there were several key elements that ASUS couldn’t discuss. How much does it cost and how does it compare to other comparable and potentially less expensive alternatives? Both of those are first and foremost in my mind, since I highly doubt the novelty of immensely powerful thin and light notebooks will go unrealized. The Zephyrus and its ilk will be treated as premium products with an associated pricing structure to match. How much that will be remains to be seen.