Connect with us
Mining can kill memory chips of GeForce RTX 3080 and Mining can kill memory chips of GeForce RTX 3080 and

Components

Mining can kill memory chips of GeForce RTX 3080 and 3090 video cards, but in games, situations with heating GDDR6X looks alarming

Published

on

Information about the strong heating of the GDDR6X memory chips of the Nvidia GeForce RTX 3080 and RTX 3090 video cards during mining is not new, but now it has been confirmed once again. And, as it turned out, in games situations, if better, then not much.

This time, the temperature of the memory chips was measured using the updated HWInfo64 application (version 6.42) – it now has the GDDR6X temperature monitoring function (GDDR6X Memory Junction Temperature). Armed with the software, Metro Exodus, and a 4K monitor, the enthusiasts ran a series of tests that heated the GeForce RTX 3080 Founders Edition’s GDDR6X memory chips to 102 ° C. This is higher than the TJmax temperature – 95 ° C, at which the thermal protection is triggered and the components simply turn off for self-protection purposes. But here’s what’s interesting: when DLSS was turned on and ray tracing in real time, the temperature of the memory modules did not exceed 94 ° C. And this is a rather unexpected result of the gaming test. However, the GeForce RTX 3090 Founders Edition memory chips in the Cyberpunk 2077 game with DLSS and ray tracing still warmed up to 100 ° C.

During mining, the situation with the heating of GDDR6X looks critical – the GDDR6X chips of both video cards heated up to 110 ° C. At the same time, video cards even lowered their frequencies in order to somehow reduce heating. The situation is typical for different models of different brands, so this is a common feature of the RTX 3080 and 3090. It is also important to note that heating up to 110 ° C was recorded without overclocking the memory – typical for a 3D card used in cryptocurrency mining.

What can be done in such a situation? As noted, even lowering the frequency of the GPU and memory using the MSI Aftrburner utility by 450 and 520 MHz, respectively, did not help much: after a few minutes of operation, the video card still reduced the frequency of the GPU to 900-950 MHz. However, the efficiency of mining did not decrease. In reality, only a decrease in the Power limit threshold to 60% helped – the temperature of the memory chips dropped to a normal value of 90 ° C. However, the efficiency of mining dropped from 95 to 65 MH / s.

The conclusion of the tests is simple: mining can lead top-end Nvidia video cards to premature failure due to overheating of memory chips. Moreover, this can happen quite quickly, since such video cards operate in a heavy mode 24/7, and even with memory overclocking. In games, the chips also heat up above normal (that is, above those same 95 ° C), but, firstly, the temperature does not rise so much, and secondly, such a scenario does not imply round-the-clock work for many days or even months … And while the question remains open why Nvidia allowed the GDDR6X to work at such high temperatures.

Click to comment

Leave a Reply

Your email address will not be published.

Components

The power consumption of the Radeon RX 7000 series graphics cards will not be as huge as that of the GeForce RTX 40

Published

on

The power consumption of the Radeon RX 7000 series graphics

Rumors are actively attributing huge power consumption up to 600 W and even higher to GeForce RTX 40 video cards. Radeon RX 7000 cards will also consume more than the current generation.

The power consumption of the Radeon RX 7000 series graphics cards will not be as huge as that of the GeForce RTX 40

AMD Senior Vice President Sam Naffziger revealed some details about the RDNA 3 architecture and products based on it. He also touched upon the topic of energy consumption.

The demand for gaming and computing performance is increasing, while at the same time the development of technical processes is slowing down quite sharply. Thus, the power level will only increase. We now have a multi-year roadmap with very significant energy efficiency improvements to offset this curve, but the trend is there.

Performance is critical, but even if our designs are more energy efficient, that doesn’t mean we won’t raise power levels if competitors are doing the same. They’ll just have to raise them much higher than us.

Thus, the Radeon RX 7000 line, despite the previously stated 50 percent advantage (over the RX 6000) in power efficiency, will stand out for its high power consumption simply because of the huge performance boost. At the same time, AMD is confident that Nvidia will have to increase power consumption even higher.

For now, the RX 6000 line is more power efficient than the GeForce RTX 30, although the difference isn’t huge. Previously, there were rumors that the top-end RX 7900 XT will consume 450 watts versus 600 watts in the case of the RTX 4090, but so far these are just rumors.

Continue Reading

Components

Who will release the weakest modern video card? GeForce GTX 1630, claiming this title, will be released on June 28

Published

on

Who will release the weakest modern video card GeForce GTX

The GeForce GTX 1630 graphics card may hit the market on June 28th. This was announced by the manufacturer of 3D-maps Colorful.

Who will release the weakest modern video card?  GeForce GTX 1630, claiming this title, will be released on June 28

However, the source says that the date may change, and this is quite likely, given that Nvidia has already postponed the launch of the adapter twice.

The GTX 1630 will be the weakest video card from the current Nvidia line if we are talking about new products. Recall that it is based on the TU117 GPU with 512 CUDA cores, has 4 GB of memory and a 64-bit bus.

The performance of such an adapter will be very modest. Previously, there was evidence that the new product will be slower than the GeForce GTX 1050 Ti. This means that it will probably be weaker than the latest Intel Arc A380, which is 5-15% behind the GTX 1650.

Continue Reading

Components

Ryzen 6800U and Radeon 680M: Aya Neo is preparing another portable console – Flip clamshell

Published

on

Ryzen 6800U and Radeon 680M Aya Neo is preparing another

Chinese manufacturer AyaNeo has announced the fifth handheld game console, which will be released in the fourth quarter of this year. This is a model called Flip, which is structurally similar to the Nintendo DS.

Ryzen 6800U and Radeon 680M: Aya Neo is preparing another portable console - Flip clamshell

So far, there are no special details about the new product. It is only known that the console will receive an 8-core and 16-thread Ryzen 7 6800U (Zen 3+) with a frequency range from 2.7 to 4.7 GHz. Integrated graphics – Radeon 680M architecture RDNA 2. The device will also receive a 7-inch display. Also, judging by the published image, AyaNeo Flip will receive analog sticks and rear triggers.

Note that the company has already announced the release of AyaNeo 2, AyaNeo 2 Geek, AyaNeo Air and AyaNeo Slide. The latest version, by the way, is made in the slider form factor. There, a full-fledged keyboard is hidden under the screen.

Ryzen 6800U and Radeon 680M: Aya Neo is preparing another portable console - Flip clamshell

There is no information about the start date of sales yet, the price and configuration options are also unknown.

Continue Reading

Most Popular