AMD May Be Prepping New, Higher-Power Ryzen Mobile APUs

When AMD launched its Ryzen Mobile chips late last year, the company chose to focus on the 15W power envelope. To-date, all four of the company’s Ryzen Mobile parts target this space. This made good sense for a company with limited resources and a gradual ramp — the 15W space has been extremely popular the last few years, and Intel has focused for years on improving its own performance at the 15W point or below as opposed to emphasizing the 25W – 45W mobile power band. If AMD wanted to retake space in mobile and challenge Intel’s laptop dominance, competing at 15W was an absolute necessity.

Now, AMD may be preparing to launch new parts with a bit more headroom and TDP to breathe. We don’t know details yet, but Videocardz suggests the Ryzen 5 2600H has a stock clock of 3.3GHz, while the Ryzen 7 2800H has a 3.4GHz base clock. The rest of the report refers to a 3DMark score, but the listed results don’t make much sense. Given that 3DMark isn’t particularly great at identifying hardware anyway until properly updated, we’re going to leave the actual test alone.

The idea that AMD would launch a new suite of higher-TDP Ryzen Mobile chips easily passes the sniff test. When a semiconductor company builds CPUs, it checks their power consumption, clocks, and voltages, then bins them accordingly. In other words, AMD and Intel don’t have a factory line dedicated solely to stamping out CPUs that are all guaranteed to run at 3GHz in a 15W TDP while requiring just 1.05v (all numbers are for demonstration purposes only). Instead, AMD and Intel build a CPU wafer, test the final product, and then assign CPUs to the product bins they need to fill. If all your SKUs are 15W, then a CPU that draws 20W under those same test conditions is a garbage part. If, on the other hand, you have a separate TDP bracket designed for 25W – 45W products, a CPU that draws 20W is no trouble at all.

In a way, it’s actually a pretty good sign that AMD felt confident enough in its low-power 15W CPUs to lead with them out of the gate. The alternative — launching a handful of 15W chips while leading with a 25W to 35W mainstream part — would imply that the company was struggling to bring its power consumption low enough to field these chips in volume, and was instead limited to the best of its parts. Pushing out higher-performance cores later doesn’t tell us as much about AMD’s overall production volume, but it does suggest that the company is making a pretty standard move to broaden its portfolio and address the needs of customers who want a bit more CPU horsepower and are willing to accept higher TDPs to get it.

The fact that these are still quad-core CPUs with more-or-less the same GPU (though it’s still possible that we might see tweaks on that front) also makes sense. Ryzen’s GPU performance is memory bandwidth bound more than anything; squeezing more frames out of a laptop with an APU is a question of RAM frequency and dual-channel configuration more than it relies on higher GPU core counts. Higher base and boost frequencies, on the other hand, should translate directly into higher overall performance.

As for whether we’ll see AMD challenge Intel with its own lineup of six-core mobile chips, I’m inclined to think we won’t — at least not any time soon. When AMD built Ryzen, it decided to save money by using as few die designs as possible. AMD’s mainstream desktop chips share a single Ryzen core and structure as a result, as do its APUs — and its APUs top out at 4C / 8T configurations. AMD surely wants more mobile space as well, but the company chose to initially prioritize the two markets where it historically held its strongest positions: consumer desktop and servers.

As such, I’d expect AMD’s explicit laptop play to continue evolving over time, but to also follow a conservative path through 2019. Any radical departures from the status quo — like, say, an HBM-equipped mobile APU, or six-core mobile APU — probably won’t tip until 2019, when Ryzen 2 hits the market. If AMD is going to redesign its APUs for faster memory, new GPU technology, or higher CPU core counts, it makes sense to do the work when it’ll already be adapting those designs to a new CPU architecture.

Continue reading

Rumor Mill: Nvidia Prepping Flagship RTX 3090 Ti GPU

Nvidia is reportedly ready to launch the swan song of its Ampere lineup, the RTX 3090 Ti.

Nvidia Reportedly Prepping Two New RTX 3050s, One with Just 4GB of VRAM

The company seems intent on giving gamers, and miners, some options in the sub-$300 space.

Nvidia May Be Prepping a New GTX 1050 Ti Max-Q to Counter Intel, AMD

Nvidia is reportedly readying a new version of the GTX 1050 Ti with a Max-Q spin on the silicon. Is it a move against AMD's just-announced Vega Mobile or a shot across the bow of Intel's Vega-equipped CPUs?

Apple Prepping New Coprocessors, Macs, iPad for Later This Year

Apple is already hard at work on a new set of chips and hardware for this year, but Meltdown and Spectre could have ramifications well beyond Intel.