How do performance tiers work and what do the advertised speeds actually mean? 

Unlike other providers who base pricing on still scene file image resolution, CPU GHzh ratings, or other delightful hardware specifications, rPinin takes the headache out of production budgeting. CBRF offers a work-based pricing model, while still being grounded by industry standard benchmarking platforms such as Blender, CineBench, and OctaneBench, 

CBRF uses reference benchmarks and throughput measurements (in Frames Rendered Per Minute) for each DCC supported. 

The throughput measurements for the reference benchmark are then used as the basis for the targeted throughput of each DCC's performance tier group. This allows you to forget the hardware and focus on performance. An example helps make sense of this. 

Test rigs used to establish our reference benchmark for supported digital content creation software tools, evaluated for globally recognized benchmarking platforms including:

- Blender Benchmark () 

- Octane ()

- CineBench ()

While these are great for measuring the capability and efficiency of the test rig and its hardware using standardized inputs, they don't directly tell what us how much work that rig will do for for us. For this reason we do an additional real-world benchmark to measure the time taken to render the frames in the benchmarked payloads (files) tested. Measured in frames rendered per minute, this rendering throughput becomes the basis for performance tier speed targets. CBRF queue specifications are configured and tailored to achieve throughput equivalents to the reference benchmark system specifications. Generally, slower, throttled speeds equate to cheaper costs, while higher speed, nominally throttled queue configurations command a premium.

tune and benchmark by deadline cloud fleet configurations to achieve throughput that's equivalent or relative to that the test rig's performance