A lot of information about AMD's Radeon RX Vega is trickling to the web. Since the Radeon Vega Frontier Edition launched about a month ago, many people thought key features of the driver were missing. Frontier Edition felt like a rushed product that AMD pushed out of the door to meet the "first half of 2017" commitment it made to enthusiasts and investors.
Now AnandTech has learned that Draw Stream Binning Rasterizer (DSBR) is indeed not active in the drivers of the Vega FE. AMD claims DSBR support will be added for both RX Vega and FE when the RX Vega launches:
Speaking of Fiji, there’s been some question over whether the already shipping Vega FE cards had AMD’s Draw Steam Binning Rasterizer enabled, which is one of the Vega architecture’s new features. The short answer is that no, the DSBR is not enabled in Vega FE’s current drivers. Whereas we have been told to expect it with the RX Vega launch. AMD is being careful not to make too many promises here – the performance and power impact of the DSBR vary wildly with the software used – but it means that the RX Vega will have a bit more going on than the Vega FE at launch.
From what we've gathered, the Radeon RX Vega performance figures AMD revealed yesterday include the expected gains of DSBR.
Another question many have about Vega is why the chip is so large considering it is supposed to compete with the way smaller NVIDIA GeForce GTX 1080 GPU. AMD's engineers told AnandTech that most of the 3.9 billion additional transistors in the Vega design went to mechanisms to increase the clockspeed versus the Fiji architecture. This is quite surprising:
Talking to AMD’s engineers, what especially surprised me is where the bulk of those transistors went; the single largest consumer of the additional 3.9B transistors was spent on designing the chip to clock much higher than Fiji. Vega 10 can reach 1.7GHz, whereas Fiji couldn’t do much more than 1.05GHz. Additional transistors are needed to add pipeline stages at various points or build in latency hiding mechanisms, as electrons can only move so far on a single clock cycle; this is something we’ve seen in NVIDIA’s Pascal, not to mention countless CPU designs. Still, what it means is that those 3.9B transistors are serving a very important performance purpose: allowing AMD to clock the card high enough to see significant performance gains over Fiji.
Gamers Nexus asked AMD about the power characteristics of the RX Vega cards. AMD engineers implied the RX Vega cards have more power tuning than the FE but didn't go into the details:
As for power tuning, we asked whether this was just tightening of the power target or if actual power performance features were enabled under the hood. Turns out, RX Vega isn’t just a matter of restricting power target, they’re actually doing something for power optimization. We couldn’t get explicit examples at this time. One thing we do know is that the voltage targets change, so voltage checks are at different frequencies than FE, and voltage should be lower. We’d expect that this will align with our findings in the undervolting testing on Vega: Frontier Edition, where power consumption can equalize while improving performance. It’s still AVFS, but just a better tuning profile than FE.