The sorry state of journalism these days. We previously tried to help TH out with corrections, but since they don’t care, here they are:
“Bolt does not disclose anything about its micro-benchmark or how it simulates the performance of its hardware and how it gets comparative performance for AMD, Intel, and Nvidia GPUs, so we can't really say much about these results.”
- we explicitly stated we ran a variety of real renders. Literally measuring the time it takes for various GPUs to render. (See pic below)
“Typically, the workload in synthetic tests is controlled: Rays are cast in predictable patterns against fixed triangle sets, and the acceleration structures are static and well-optimized. This test, however, produces a clean, high number that reflects the GPU's raw intersection throughput under ideal conditions, which is good enough if one only wants to learn about the theoretical limits of a GPU in this specific workload.”
- to recap, we did NOT use synthetic tests; we rendered real scenes. It’s most definitely not under ideal conditions (see pic below)
Here’s our public documentation that describes our testing methodology for path tracing:
Bolt Graphics claims its Zeus GPUs outperform Nvidia's RTX 5090 by up to 13x in internal ray-tracing tests, though real-world gains remain unknown.
tomshardware.com/pc-componen…