AMD Radeon RX 7900 XTX Temperature Problems Are Getting Worse, and PowerColor Intervenes to Assist Customers

Recently, a guy on Reddit stated that the reference AMD Radeon RX 7900 XTX graphics card he was using was hitting temperatures of 110 degrees Celsius in the Junction while he was playing the latest version of Modern Warfare II. However, when he reached out to AMD with an RMA request, he was informed that his request would not be granted since the business stated, “The temperatures are at their typical levels.

Please get in touch with us if there is any problem.” Another person was successful in making their request. However, they were not granted a refund for the package that had been opened before. Since then, the corporation has been the target of some criticism for the matter, particularly in light of the increasing number of incidents that have surfaced.

There have been increasing reports of excessive temperatures on graphics cards manufactured by AMD Radeon RX 7900 XTX, and firms like PowerColor are stepping up to assist customers.

Before this new invention from AMD, NVIDIA was criticized for the overheating and melting 12VHPWR connections on the flagship NVIDIA GeForce RTX 4090 GPU. More than thirty people reported having trouble with the graphics card after it was released.

The issue is now impacting both reference and AMD-manufactured Radeon RX 7900 series GPUs, which is one of the AMD problem’s distinguishing characteristics. It applies to any cards that are sold by AMD partners or the firm itself, but it does not include GPUs that have been custom-created and use the same graphics architecture. As De8auer explains in the video that can be found below, there is a widespread belief that several coolers and mounting designs have a fault.

It’s been three years since AMD published an item on the corporate blog with the title,

More Control Over GPU Power and Performance, Enhanced Thermal Monitoring, and Maximized Performance are Among the Highlights of the Latest AMD Radeon Community Update”

In the section on the increased thermal monitoring of the AMD Radeon VII GPU devoted to discussing junction temperatures, they wrote the following statement:

… Instead of setting a conservative, ‘worst case’ throttling temperature for the entire die, the RadeonTM RX 5700 series GPUs will continue to opportunistically and aggressively ramp clocks until any one of the many available sensors hits the ‘hotspot’ or ‘Junction’ temperature of 110 degrees Celsius. Operating at up to 110C Junction Temperature during typical gaming usage is expected and within spec. This enables the RadeonTM RX 5700 series GPUs to offer much higher performance and clocks out of the box, while maintaining acoustic and reliability targets.

“Kevin,” one of the technical engineering leads working on the AMD Radeon RX 7900 XTX, has stated that the company is aware of the hotspot issue as of the previous day. Furthermore, the company is actively investigating the matter and putting together a plan of action “to repro and collect serial numbers,” according to Kevin. He clarifies that the present COVID scenario in other countries is causing the process to move at a slower pace. Still, he acknowledges that producing and satisfying an RMA for customers would include addressing mechanical issues.

However, they are also investigating the possibility of updating the firmware to resolve specific occurrences in which a software patch rather than a hardware replacement would be necessary. Additionally, “Kevin” covers edge heating, which refers to the temperature along the edge of the silicon on a GPU. He states that edge temperatures exceeding 90 degrees are considered abnormal, but edge temperatures of 70 degrees are considered usual.

I’m a user on Reddit and the ambassador for PowerColor on the platform.

“PowerColorSteven” is doing everything possible to help customers with the present problem concerning the AMD Radeon RX 7900 XTX graphics cards. For the benefit of readers who might not remember Steven, we mentioned in this article that the company had a previous problem in which the AMD RX 6700 XT series GPUs were shipped with the thermal pad covers still attached, which, at the time, would cause high temperatures to be produced by the graphics card. For readers who might not remember Steven, this problem was brought up in this article. Since a couple of days ago, the sales representative has been paying particular attention to this issue, and only today, they submitted the following as a pinned topic on Reddit:

uh… just adding this in here in case anybody has not seen my other posts.

if you are dealing with the 110c issue and nobody else is helping you out. doenst matter which (AIB you bought this from) you bought it from. send me an email or direct message (since my emails are not working right today) (also, please dont use chat, that shit dont work right). i am putting together a list for my AMD contacts of how many issues we are seeing. they asked if it was just a few users. i said its def more than a handful, so now i need to show some sort of backing for my fat mouth. send me serial numbers. i am sending an email to the AMD guys with how many units are affect. powercolor. sapphire. gigabyte. amd direct. whatever. send me serial numbers. hopefully amd will get some news for all of us, but regardless, when i hear something of substance, i’ll make a post to provide some level of guidance on how to go about getting this sorted out.

— User “PowerColorSteven,” aka “sales guy or something” on Reddit

However, it is difficult to determine which businesses accept RMAs and fulfill them because several recorded cases have stated that some are permitted. In contrast, others are denied, even when it is the same manufacturer. This makes it difficult to determine which businesses accept RMAs and fulfill them. Several other hypotheses are being proposed about the issue, including using an uneven cold plate or a graphite thermal pad as a potential solution.

It is not suggested to ease the problem because the number of cases grew during the early reports. Instead, it is recommended to hang on until AMD makes an official announcement or accepts a series-wide RMA. Until then, the problem should be alleviated.