“All” graphics cards in in-depth coil whine analysis

The number of graphics cards tested over the recent period has reached a number after which we can pause a bit and specifically focus on the noise level of their coils. The latter has always been recorded in standard measurements, but in large tests with lots of other information, this unique data was getting lost. That is why it will now, within the scope of this article, be limited to these only. So which of the modern graphics cards has the quieter coils?

Conclusion

Each graphics card has a different coil whine noise level. If the noise meter registers the same value, the acoustic response will be different. This can be seen, for example, in the comparison of the AMD Radeon RX 6800 XT graphics card with the MSI RTX 4080 16GB Suprim X. In CS:GO (1080p), both come out at 40.7 dBA when measured with a noise meter, and the difference in what either card sounds like shows up only at the spectrogram level. While the specified Radeon (AMD RX 6800 XT) is noisier in the sound band around 7.5 kHz, the most pronounced tonal peaks of the GeForce (MSI RTX 4080 16GB Suprim X) are at 12 kHz. Which of these graphics cards is noisier will be different for everyone, depending on whether you’re more sensitive to higher or lower frequencies of sound. Although, of course, in both cases these are high frequencies, but still, there is a difference of some 5 kHz, and that is definitely perceptible.

Comparison of acoustic profiles of AMD RX 6800 XT and MSI RTX 4080 16GB Suprim X graphics cards in one of the test situations (CS:GO, 1080p). While the output value of the noise meter is the same, the shape of the spectrograms is different (and the color of the sound as well).

It’s also good to know that a more powerful graphics card doesn’t necessarily have noisier coils, and it can even be the other way around. Namely, that of a pair of graphics cards being compared, the “weaker” one will be significantly noisier. For example, Sapphire’s RX 7600 XT Pulse always has significantly noisier coils than Gigabyte’s RTX 4090 Gaming OC 24G. Meanwhile, the difference in gaming and computing performance is abysmal .

The RTX 4090 in the Gigabyte Gaming OC 24G design can be described as the fastest graphics card we’ve tested, yet it’s among the quietest when it comes to coils. In this respect, Gigabyte often has the edge over the competition. If we were to run a “performance per unit of coil whine noise ratio”, for Gigabyte graphics cards, it would often be one of the most favorable.

Gigabyte GeForce RTX 4090 Gaming OC 24G

In addition to the RTX 4090 Gaming OC 24G, of Gigabyte’s graphics cards that have super-quiet coils, there is the RTX 4080 Super Master 16G, RTX 4060 Windforce OC 8G or RTX 3060 Eagle OC 12G. With the RX 7900 GRE Gaming OC 16G, the coil whine noise level is already only average, but it’s still lower than many tested, even slower Sapphire graphics cards. However, it cannot be generalized that Sapphire has noisier coils. While we didn’t see a case where they were downright “quiet”, but the results of the RX 7800 XT Nitro+ or RX 7700 XT Pure are at least average. And that it can vary significantly from piece to piece even with the same model? Tests in which we took two different RX 7900 XT Pulse samples and compared them to each other don’t really agree with this. You probably wouldn’t be able to discern the sound profile of their coils, but sure, one could still argue that the variance would be greater with 5, 10 or 15 pieces. Fundamentally? We don’t suppose, as long as the same power delivery design is retained with the same component models.

But now back to the coil whine noise level of individual graphics cards. We’ve often noticed excessively quiet operation with MSI, although especially with the older GeForce 3000 models. Even the newer ones don’t exhibit downright noisy coils, though the RTX 4080 16GB Suprim X is not among the quieter graphics cards. That is, the cooler is perfectly quiet, but those noisier coils are messing it up. With the RTX 4070 Ti Suprim X 12G, the coils are noticeably quieter, similarly for RTX 4070 Ti Suprim Ventus 3X or RTX 4060 Ti Gaming X Trio 8G (also MSI cards) and then there are a few MSI models that fall into the category of cards with “silent” coils – RTX 3090 Gaming X Trio, RTX 3070 Gaming X Trio, RTX 3050 Ventus 2X 8G OC or RTX 3060 Ti Gaming X Trio. This last one (RTX 3060 Ti Gaming X Trio), at lower fps, is even the graphics card with the quietest coils we have in the results database.

Asus? The Dual RTX 4070 12G fits in the quieter half. But the TUF RTX 3080 O10G Gaming is already on the edge (of the quieter and noisier categories). This graphics card is also notable because its PCB is the basis for the “Noctua Edition” models. Within these, it has a great, extremely quiet cooler, but you won’t miss the coil whine. Therefore, in our opinion, this is not a suitable foundation for graphics cards with vastly oversized coolers, as ultimately the coils will be what the target audience of very demanding users will complain about. And even Noctua sees some room for improvement in these designs (Asus graphics card PCBs). But that’s already stuff from internal communications that we shouldn’t elaborate too much on publicly, and it’s also not necessary for this article.

We’re at the end, if there’s anything else you’d like to know and you didn’t find it in the text, bring it up. You can go through everything in the discussion below.

English translation and edit by Jozef Dudáš


Alienware vendor locks Ryzen X3D CPUs, effectively bricking chips

AMD processors have an unfortunate feature: the PSB security technology that makes it possible to lock a CPU so that it cannot be used in any motherboard other than the one that locked it with its key. This typically makes a CPU unable to boot anywhere else and this lock is irreversible. It used to affect just server processors, Threadripper and Ryzen Pro chips, but beware: Dell is apparently locking CPUs in gaming PCs as well. Read more “Alienware vendor locks Ryzen X3D CPUs, effectively bricking chips” »

New Radeon GPUs Likely Won’t Arrive Until Second Half of 2027

We recently reported the new generations of GPUs from Nvidia (GeForce RTX 6000 / Rubin) and AMD (RDNA 5) will not arrive until 2028. There was supposed to be one exception, however—AMD could reportedly release one of the new-generation cards in 2027. The information at the time did not say when exactly in that year this might happen. Now we have clarification, and it seems it will not be anytime soon either—as might have been expected. Read more “New Radeon GPUs Likely Won’t Arrive Until Second Half of 2027” »

AMD Reportedly Raising GPU Prices. Apart From One Radeon card?

Radeon GPUs have been significantly better value than similarly fast Nvidia cards lately. These lower prices (by today’s dismal standards) may not last much longer, however. According to news from China, AMD and AIB makers are now preparing a substantial price increase, despite similar price move happening recently. The reason is the AI mania again, which has massively driven up the cost of memory, as well as manufacturing itself. Read more “AMD Reportedly Raising GPU Prices. Apart From One Radeon card?” »

Comments (5) Add comment

  1. So, there are now additionally spectrograms with the scale and the Y axis label (with the noise level) in the article. Thanks to “the patient” for adding it quickly. 🙂

  2. Pretty incomplete when PASSIVE (KalmX) GPU versions from Palit not included in such test..

    1. Yes, a Palit KalmX graphics card would have been great for a test, I agree. Unfortunately, we couldn’t get our hands on one. It’s similar to when people point out the absence of Thermalright products in our tests. Their availability in our CZ/SK market is quite limited, which makes them harder to source. Even though our magazine is localized into English, many companies still consider HWCooling as a “Slovak” medium with a smaller reach. That’s the reality. 🙂

  3. What I’m wondering is why the PWM fan peak doesn’t show up at all in the graphs? I’ve had a Zotac 5070 Ti, a KFA 5070 Ti, and an ASUS 5070 Ti, and all of them produced a strong peak between 13 kHz and 15 kHz as soon as the fans started spinning — sometimes even without any GPU load, just because the fans were running. On my current KFA 5070 Ti OneClick OC, that’s even around 35 dB near the card. Unfortunately, I have tinnitus, and frequencies like that trigger it badly, especially when they’re that “loud.” But in your charts, I don’t see those fan peaks at all — did you measure differently, or somehow filter them out?

    1. — „What I’m wondering is why the PWM fan peak doesn’t show up at all in the graphs?“

      Yeah, I reckon what you are missing is the results is not there due to the diversity of testing methodologies. Other people will test fans differently than we do. 🙂

Leave a Reply

Your email address will not be published. Required fields are marked *