Your Daily Tech Dose

Quick Report

Samsung is reviving its Z-NAND high-performance flash technology, positioning it as a direct competitor to Intel's now-discontinued Optane memory in the AI-accelerated computing era. According to recent reports, the new Z-NAND implementation promises up to 15x higher peak performance than conventional NAND flash while reducing power consumption by approximately 80%. This specialized storage technology aims to bridge the gap between DRAM and mass storage, particularly for AI workloads that require faster data access than traditional SSDs can provide.

Quick Report

Intel is bringing back expanded AVX functionality to its desktop processors with the upcoming Nova Lake lineup. According to industry sources, these new CPUs will feature AVX10.2 instruction set architecture, which includes full 512-bit vector support across all cores. This marks a significant shift in Intel's desktop CPU strategy after the company previously removed AVX-512 support from Alder Lake and Raptor Lake consumer processors due to compatibility issues with E-cores.

Quick Report

Thermal Grizzly has officially launched the der8enchtable, a premium benchtable developed in collaboration with renowned overclocking experts Roman “der8auer” Hartung and Jon “elmor” Sandström. Priced at $268.98, this professional testing platform stands out from conventional benchtables by featuring an integrated active PCB with comprehensive connectivity options and fan control capabilities, making it an ideal solution for hardware enthusiasts, reviewers, and overclockers.

Quick Report

Intel has announced it will drop support for 16x MSAA (Multi-Sample Anti-Aliasing) with its upcoming Xe3 Celestial graphics architecture. While the architecture will continue to support MSAA at 2x, 4x, and 8x levels, the company cites modern upscaling technologies like XeSS and TAA as superior alternatives that make the legacy 16x MSAA obsolete for most use cases.

Quick Report

Intel has secured a significant win by landing Tesla's Dojo 3 packaging contract, signaling a new dual-supplier approach where Samsung Electronics Foundry will manufacture the custom D3 AI-training chips while Intel handles the packaging and testing. This partnership leverages Intel's Embedded Multi-Die Interconnect Bridge (EMIB) technology to connect multiple dies through silicon bridges rather than a full-wafer interposer.