Search results

NVIDIA development directly on Tensor cores - specifically the GB10 on the DGX Spark

arm64

dgx-spark

gpu

multithreaded

nvidia

tensor

Status: Open.

#47 In ObrienlabsDev/performance;

· obriensystems opened on Dec 24, 2025

OSX 26.2 RDMA low latency across Thunderbolt 5

apple

gpu

rdma

Status: Open.

#46 In ObrienlabsDev/performance;

· obriensystems opened on Dec 21, 2025

Performance baseline for NVIDIA DGX Spark GB10 128G vram

arm64

dgx-spark

gpu

multithreaded

nvidia

Status: Open.

#45 In ObrienlabsDev/performance;

· obriensystems opened on Nov 9, 2025

Propeller 64 bit 8 core microcontroller 20MHz assembly for collatz

Status: Open.

#44 In ObrienlabsDev/performance;

· obriensystems opened on Oct 13, 2025

iperf for tripple local 1, 6 and 9 gbps fiber

Status: Open.

#43 In ObrienlabsDev/performance;

· obriensystems opened on Aug 20, 2025

Java Vector API 10th incubation in Java 25 LTS

avx

cpu

Java

Status: Open.

#42 In ObrienlabsDev/performance;

· obriensystems opened on Jul 28, 2025

64 bit collatz single threaded CPU implementation in C - Apple Silicon

64

apple

cpu

singlethreaded

Status: Open.

#41 In ObrienlabsDev/performance;

· obriensystems opened on Jun 1, 2025

64 bit collatz single threaded CPU implementation in Objective C - Apple Silicon

64

apple

cpu

singlethreaded

Status: Open.

#40 In ObrienlabsDev/performance;

· obriensystems opened on May 31, 2025

Java periodically drops core count to 6.5-9 from 20+ on an M3 Ultra 28 core studio - has happened before on an M2 Ultra - add restart check

128

apple

cpu

Java

multithreaded

Status: Open.

#39 In ObrienlabsDev/performance;

· obriensystems opened on May 26, 2025

Collatz deep research report from OpenAI o1 pro

Status: Open.

#38 In ObrienlabsDev/performance;

· obriensystems opened on Feb 5, 2025

Java: AVX capability - check vector API JEP 414/417/426/438/448/460/469/489 in Java 16/17/18/19/20/21/22/23/24 - out of incubator status in Java 25 LTS

128

avx

Java

Status: Open.

#37 In ObrienlabsDev/performance;

· obriensystems opened on Jan 18, 2025

Increase performance of CUDA code past 0.9 TeraOPS

128

gpu

Status: Open.

#36 In ObrienlabsDev/performance;

· obriensystems opened on Jan 13, 2025