FARPY LABSPublished research, findings, and evidence.
FARPY LABS

What FARPY tested, what the evidence showed, and what you can reuse.

FARPY Labs is the public research shelf. Visitors come here to read findings, methods, and evidence from GPU tests FARPY actually ran. It is not the place to upload a .blend or track a customer render — that is Your renders.

HOW TO READ LABS

Research, graduated work, and published evidence

Start with a published record. Research holds studies and notes. Graduated holds capabilities that became public tools. Benchmark evidence lives on the GPU result pages, not on a live ticker here.

PUBLISHED NOW

Records you can open today

These are live pages, not a dashboard of unpublished work. FARPY is not inventing a feed to fill empty cards.

Compatibility study · published

Blender 4.1.1 on RTX 3090

What we learned
What qualified, what was blocked, and where the claim stops.
Evidence
Scoped GPU tests with methods, positive workloads, and negative findings.
What you can reuse
Use this record when you need Blender 4.1.1 / RTX 3090 limits, not a general GPU promise.
Read methods and findings
Verified result · published

RTX 3090 Blender evidence

What we learned
One frozen, public RTX 3090 Blender result you can inspect.
Evidence
Verified GPU result page: hardware, runtime, and score as published.
What you can reuse
Cite the published result. Do not treat it as a live leaderboard.
Open verified result
Research study · published

Blender GPU runtime reproducibility

What we learned
When a frozen Blender workload stays comparable, and when runtime variance means it does not.
Evidence
Published research study with methods, results, and limits.
What you can reuse
Read the study before treating a Blender timing as a benchmark.
Read the study
MORE PUBLISHED RECORDS

Full archive and graduated capabilities

The research archive includes additional published studies and notes. Open a record for methods, results, and negative findings. Status and dates are on each page.