Tech News
Back to home
Technology

Mobile RTX 2070 with shunt mod nearly eclipses desktop performance — 60W boost provides 15% performance uplift

Source

Tom's Hardware

Published

TL;DR

AI Generated

A PC enthusiast successfully shunt modded their Nvidia GeForce RTX 2070 Mobile GPU, boosting performance by approximately 15%. By pushing the mobile GPU beyond its power limit to 175W, it nearly matches the performance of its desktop counterpart. Shunt mods involve changing a shunt resistor on the graphics card PCB to allow more current to reach the GPU without altering the vBIOS. The modder, HerrBolsch, conducted this modification as part of a project to re-house and water-cool a Zotac Mini PC. Post-modification assessments confirmed improved performance and manageable thermals, with the modded RTX 2070 Mobile keeping pace with a desktop card.

Read Full Article

Similar Articles

Seasonic's PSU calculator adds support for NVIDIA GeForce RTX 5070 Ti SUPER, RTX 5070 SUPER

Seasonic's PSU calculator adds support for NVIDIA GeForce RTX 5070 Ti SUPER, RTX 5070 SUPER

Seasonic's PSU calculator now supports NVIDIA's upcoming GeForce RTX 5070 Ti SUPER and RTX 5070 SUPER graphics cards. The calculator helps users determine the power supply requirements for their systems based on the components they plan to use. This update ensures that users can accurately assess the power needs of their PCs when considering these new high-performance GPUs.

TweakTown•
Nvidia RTX 5070 Ti Super and RTX 5070 Super TDP leaked — long-rumored RTX 50 Super series GPUs appear in power supply calculator

Nvidia RTX 5070 Ti Super and RTX 5070 Super TDP leaked — long-rumored RTX 50 Super series GPUs appear in power supply calculator

Nvidia's rumored mid-gen refresh for Blackwell is expected to launch at CES 2026, with the RTX 50 Super series generating buzz due to recent delistings of FE cards. Seasonic's PSU calculator now lists RTX 50 Super models, hinting at potential upgrades like increased VRAM capacity. The RTX 5070 Super is projected to have 18 GB of memory with a 275W TDP, while the 5070 Ti Super could feature 24 GB with a 350W TDP. These leaks align with industry rumors, suggesting the eventual release of the RTX 50 Super series. Nvidia may unveil these GPUs around CES next year, with more leaks expected closer to that time.

Tom's Hardware•
Nvidia GeForce RTX 5050 review: a necessary update, not an exciting one

Nvidia GeForce RTX 5050 review: a necessary update, not an exciting one

Nvidia's GeForce RTX 5050 is a new entry-level graphics card aimed at budget-conscious gamers, offering significant improvements over its predecessor, the RTX 3050. Powered by the Blackwell architecture, the RTX 5050 boasts higher clock speeds, more CUDA cores, and a substantial increase in L2 cache, resulting in a 44% boost in raw compute potential. Despite sticking with GDDR6 memory, the card delivers competitive performance at its price point, although it maintains a 130W power rating and retains the $250 price tag of the RTX 3050. While the RTX 5050 represents a necessary update for the entry-level market, it may not be a groundbreaking advancement.

Tom's Hardware•
The Nvidia RTX 5080 drops to new lowest-ever price — grab the PNY OC model for $929 at Walmart

The Nvidia RTX 5080 drops to new lowest-ever price — grab the PNY OC model for $929 at Walmart

The Nvidia GeForce RTX 5080 from PNY is now available at a discounted price of $929 at Walmart, $70 below the suggested MSRP of $999, making it an affordable option for those interested in 4K gaming or rendering workloads. This GPU is based on Nvidia's Blackwell architecture and features 16GB of GDDR7 memory, 10,752 CUDA cores, and advanced features like DLSS 4 and enhanced ray tracing capabilities. While it offers modest gains over the previous generation in terms of performance, it is still a solid choice for handling modern games at 4K and comes with Nvidia's latest technologies.

Tom's Hardware•

We use cookies

We use cookies to ensure you get the best experience on our website. For more information on how we use cookies, please see our cookie policy.