AMD's newly-released Radeon PRO W7600 professional GPU faces issues within its thermal design, causing random system blackouts due to overheating.
AMD Radeon Pro W7600 GPU Gets Very Hot Out of The Box, Replacing Thermal Pads Solves The Problem
In extensive testing of the AMD Radeon Pro W7600 GPU by Igor's Lab, it was revealed that applying load on the GPU would result in black screens at different intervals, ultimately leading to a system crash. Like every consumer, Igor initially believed the problem was present in the DisplayPort, but things got more complicated upon diving into the issue.
Igor implemented the usual SOPs of the diagnostic process, and the first thing he did was connect the monitor to the integrated graphics on the CPU, which eliminated the problem. Hence it was concluded that the "blackouts" were caused by the AMD Radeon PRO W7600 GPU. Upon expanding on the diagnostics, it was revealed that the GPU was facing an overheating problem, with the VRM section particularly being affected the most, which explains the blackouts.
So the reason behind the issue, as Igor's Lab explained, was the inaccuracies in applying thermal pads on the VRMs of the AMD Radeon Pro W7600 GPU. The thermal pads were placed in a way that prevented thermal dissipation and compromised the vapor chamber's performance, resulting in significantly high temperatures. The thickness of the pads was also above the usual standards, which restricted thermal conductivity.
Gladly, resolving the problem didn't require extensive steps; Igor replaced the thermal paste of the GPU and utilized thermal pads with optimal thickness. Moreover, he also applied thermal paste between the vapor chamber and the cooler as well, which solved the issue as expected.
Thermals were also lowered to a great extent, as revealed in testing by Igor's Lab, which you can look at below:
We haven't seen the thermal issue being reported on other units of the Radeon PRO W7600 but it is a relatively new GPU so if the problem is more extensive, we should know in a few weeks time. However, it is disappointing to see the problem exist in the first place since this is mainly due to mismanagement in the production and QC stages. This would also be the second case of RDNA 3 GPUs facing overheating issues with the first one being the 7900 XTX which had to be recalled due to a flaw within the vapor chamber.
News Source: Igor's Lab
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
