Switch 2 motherboard maybe leaked

[numberwang]

sc94597 said:

numberwang said:

I am skeptical of claims that put Switch 2 performance above the Steam Deck in handheld mode. The Switch 2 SOC is slightly larger with ca. 210mm² compared to Steam Deck 163mm² at the original TSMC at 7N process. With a Samsung 8N node that would mean fewer transistors for the Switch 2. Samsung 5N should give Switch 2 more transistors. Clock speed on Switch 2 is much lower to accommodate 10W power and a longer battery life.

Switch 2 die size: 210mm²

Steam Deck die size: 163mm²

Switch 2 GPU Hz handheld: 561MHz

Steam Deck GPU Hz: 1600Mhz

Switch 2 TFlops: ?

Steam Deck Tflops: 1.6 TFlops

The [original Steam Deck] Van Gogh graphics processor is an average sized chip with a die area of 163 mm² and 2,400 million transistors.

https://www.techpowerup.com/gpu-specs/steam-deck-gpu.c3897

TFLOPs are a function of clock-frequency and core-count. Both of those are knowns now. The node isn't important anymore. So yes, handheld Switch 2 is 1.72 TFLOPs. For raw-rasterization, that's not enough to say it is better than the Steam Deck 2 though, because 1 Ampere TFLOP ~ .7 - .75 RDNA2 TFLOPs when it comes to inferring rasterization performance. On paper, in a pure rasterized workload a max-TDP Steam Deck would outperform a Switch 2 handheld, all else kept equal. But with DLSS and in mixed ray-tracing/rasterized workloads (which are increasingly more common) the Switch 2 handheld should make up the gap. 

Again, why can the Switch 2 pull this off at a lower wattage and frequencies than the Steam Deck? Because the Switch 2 has three times the shading units/cores (1536 for Switch 2 vs. 512 for Steam Deck.)

The Steam Deck starts to collapse in terms of power-efficiency at about 1200Mhz or more, and the voltage has to rapidly increase to increase frequency beyond this, and therefore the power-consumption increase quadratically with voltage. So much so that in order to get that last 400 Mhz the power-consumption has to double on the Steam Deck. You can get 75% of the Steam Deck's performance when running at half its max TDP. 

Edit: Also that is without considering that most Steam Deck games run on a compatibility layer with performance loss + x86 (even AMD x86) is less efficient than ARM at sub-15W TDPs unless you use actual x86 efficiency cores (that cut out some of the instruction set), which nobody does because of compatibility issues. 

We don't know the shader count of the Switch 2 or really anything about its internal soc structure. The leaked image does not confirm anything that was speculated before with that T239. Switch 2 having 3x the "shading units/cores" to a Steam Deck is implausible on a comparable die size / transistor count without sacrificing something else on the die.

Look at it this way: Switch 1 had a die 118 mm² (TSMC 20nm) with 307MHz in handheld mode with 0.157 TFlops. A 5x increase to 0.8 Tflops on Switch 2 handheld through a bigger die, a better fabrication node and higher clocks is reasonable.

https://www.techpowerup.com/gpu-specs/switch-gpu-20nm.c3104