![\"\"](\"https:\/\/www.hwcooling.net\/wp-content\/uploads\/2025\/10\/aorus-geforce-rtx-5080-master-16g-gv-n5080aorus-m16gd-test-recenze-review-b04.jpg\")
<\/a><\/p>\nGigabyte refers to this solution as “Screen Cooling Plus.” We have encountered Screen Cooling technology on Gigabyte products for nearly twenty years. The term “screen” does not refer to a display but is used more in the sense of a grille or mesh. It originally appeared on passively cooled graphics cards, where the open-pass-through fin stack allowed the card to be cooled by air flowing inside the case.<\/p>\n
Modern high-performance cards can no longer be cooled passively due to their high power consumption. However, the shorter PCB of graphics cards allows for cooler designs where a large portion of the heatsink consists of an open-pass-through fin stack. Similar to CPU coolers, graphics card manufacturers began experimenting with mounting fans on both sides of the heatsink in a push-pull configuration. This increases both airflow and static pressure, which is intended to improve cooling efficiency.<\/p>\n
In today’s article, we will focus on the practical impact that adding a second fan to the opposite side of the heatsink has on the card’s cooling. It won’t be entirely straightforward, because the fourth fan is not a standard part of the card but is supplied as an accessory. You will search in vain for a header on the card to plug it into, and the automatic speed regulation is similarly unresolved. Gigabyte avoided the problems associated with properly synchronizing and regulating two different fans in a stacked arrangement by leaving it up to the owner to figure out how to manage the fan speeds.<\/p>\n A silicone frame is included in the accessories for mounting the fan. The fan is attached to this frame.<\/p>\n I’ll admit, I didn’t know where to start. Finding the optimum by trial and error would have been tedious, so I opted for measurements in larger steps, combining measurements across the entire operating speed range of the cooler’s fans with the full operating speed range of the supplemental fan.<\/p>\n We will look at how different speed combinations affect the card’s overall noise levels and what influence this has on other operating characteristics. Before that, let’s just recap the card’s basic parameters, and then we’ll get started.<\/p>\n For this card, Gigabyte significantly increased the clock speed in the BIOS from the reference 2617\u00a0MHz to 2800 MHz. The power limit on both BIOS versions is set to the reference 360 W; it can be reduced by 31% to 250 W or increased by up to 25% to 450 W. The upper limit corresponds to the limit of the power adapter with three 8-pin connectors (but an additional up to 66 W on the +12V rail can be delivered to the card via the slot).<\/p>\n The card features a dual BIOS with Silent and Performance modes. The core settings of the card do not differ between them; the GPU and memory clocks, as well as the default power limits, are the same. The difference lies in the cooling curve settings. In Silent mode, the fans run at lower speeds at the same temperatures than in Performance mode. The GPU can run at slightly higher clock speeds at lower temperatures because clock speeds are usually limited by the chip’s power draw, which is, to some extent, dependent on its temperature.<\/p>\n GeForce cards typically have two headers reserved for fans, with separate regulation and RPM monitoring. The Gigabyte card has all three fan headers monitored and controlled independently in GPU-Z. Their speeds are synchronized under load, running at the same RPMs.<\/p>\n The parameters stored in the Silent BIOS are shown below.<\/p>\n We will record data using HWiNFO, which can extract the most information from the monitoring systems of most cards. You can see the list of sensors available for the tested card in the latest version of the application in the image below.<\/p>\n I will measure the operational properties on a different setup than in standard card tests. It is assembled from identical components, and the BIOS and Windows settings are also the same, but the measured values (e.g., total system power consumption) may not be directly comparable due to component variances; they might differ slightly.<\/p>\n<\/a><\/p>\n<\/a><\/p>\nAORUS GeForce RTX 5080 MASTER 16G (GV-N5080AORUS M-16GD)<\/h2>\n
<\/a><\/p>\n<\/a> <\/a> <\/a> <\/a> <\/a><\/p>\n