Motherboard Hell
Nothing is easy when everything is new
DON’T GET ME WRONG, I have a cushy job—I’m fully aware of that. But that doesn’t detract from the fact that motherboard group tests are some of the most infuriating roundups you can ever work on. Soul destroying.
Take Z170, for example. You pick five to ten motherboards, and choose a whole suite of benchmarks to check performance across multiple factors: rendering, gaming, storage support, overclocking. The thing is, you’re not looking for variance, but for them all to be identical. Reason being, a good motherboard shouldn’t affect performance outside of overclocking. If it does, something’s going wrong somewhere, and you can bet your last dollar that power draw, clock speeds, or any number of other factors will be far higher than they should be.
Take Ryzen, on the other hand, and you’re welcomed into a whole new world of hurt. New architectures and chip designs are fraught with problems. Realistically speaking, Intel’s Core architecture design is coming up to 11 years old. It may receive tweaks here and there, added I/O support, and drops in transistor size, but that’s about it. In short, it has essentially been the same design layout since the days of Sandy Bridge. If you compare Ryzen to Piledriver, however, it’s entirely different. A completely new design from the ground up. And because of that, it’s loaded with problems, particularly on the BIOS front, as AIB partners race to release a new and improved BIOSes with better support for memory, and general overclocking stability for a chip that isn’t just a rehash of a design that’s been gradually refined for the last 11 years.
To cut a long story short, I killed two motherboards via BIOS updates, another was dead on arrival, and I managed to nuke one of our Ryzen 1800X samples in the process of this group test. To say it was hell would be an understatement.