Monday, November 14th 2022
AMD Radeon RX 7900 XTX Reference Design PCB and Cooler Detailed
AMD Radeon RX 7900 XTX reference-design isn't a first-party product with limited availability like the NVIDIA Founders Edition; but rather a classic reference-design that's sold by AMD's add-in board partners under their marquee (without sticking their own labels on the product). AMD and its partners internally refer to reference-design cards as "MBA cards" (made by AMD cards). The company gave us a technical overview of the reference-design PCB. As with every reference AMD PCB for the past several generations, the RX 7900 XTX PCB has a premium selection of components. The card uses an expensive 14-layer PCB with 4 additional layers of 2-oz copper. 14-layer PCBs are typically used with enterprise-grade products, and graphics cards typically tend to have PCB layer counts of around 10. The PCB also uses ITEQ IT-170GRA epoxy and laminate materials, which enable a glass transition temperature (Tg) of 175 °C (no, the GPU won't get anywhere near as hot).
The reference-design RX 7900 XTX PCB draws power from two 8-pin PCIe power connectors. With the typical board power of the RX 7900 XTX rated at 355 W, this falls inside the 375 total power-draw capability when you add up the 150 W input from the two connectors, and 75 W from the PCIe slot. AMD worked to minimize power-draw spikes at least from the PCIe slot. Excursions, if any, should be localized to the 8-pin power connectors. The card features 20-phase VRM solution, using "high efficiency" DrMOS power-stage phases (could be very high current). The "Navi 31" GPU is surrounded by 12 GDDR6 memory chips given the GPU's 384-bit memory interface. Two of these memory pads could end up unused on the RX 7900 XT, which has a 320-bit memory interface. Display outputs of the RX 7900 series include two standard-size DisplayPort 2.1, one USB type-C with DisplayPort passthrough; and one HDMI 2.1a.The cooling solution features an aluminium fin-stack heatsink that's ventilated axially by three fans that offer individual speed control and come with intake air temperature sensors. The cooler shroud and backplate are made of die-cast aluminium. The main heatsink uses a large aluminium fin-stack to which heat is drawn by a vapor-chamber plate. The vapor-chamber and base-plate pull heat from the GPU, memory chips, and VRM. A secondary heatspreader pulls heat from other minor heat sources and structurally reinforces the card. AMD says that the vapor chamber plate is 10% larger than the one in the RX 6950 XT RDNA2 reference-design. The company is introducing a newer thermal interface material (TIM) for the GPU and memory chips, and an "ultra soft" thermal pad for the VRM MOSFETs.
It turns out that the RX 7900 XTX reference design card is physically larger than the RX 7900 XT reference design (if it does exist in production). The reference RX 7900 XTX is 28.7 cm long and 12.3 cm tall; whereas the RX 7900 XT is 27.6 cm long, and 11.3 cm tall, making it closer to being a standard "full-height" add-in board. Both cards at 2.5-slot thick. The RX 7900 XT is compact as the cooler has to deal with a 15% lower typical board power of 300 W compared to 355 W for the RX 7900 XTX.
The reference-design RX 7900 XTX PCB draws power from two 8-pin PCIe power connectors. With the typical board power of the RX 7900 XTX rated at 355 W, this falls inside the 375 total power-draw capability when you add up the 150 W input from the two connectors, and 75 W from the PCIe slot. AMD worked to minimize power-draw spikes at least from the PCIe slot. Excursions, if any, should be localized to the 8-pin power connectors. The card features 20-phase VRM solution, using "high efficiency" DrMOS power-stage phases (could be very high current). The "Navi 31" GPU is surrounded by 12 GDDR6 memory chips given the GPU's 384-bit memory interface. Two of these memory pads could end up unused on the RX 7900 XT, which has a 320-bit memory interface. Display outputs of the RX 7900 series include two standard-size DisplayPort 2.1, one USB type-C with DisplayPort passthrough; and one HDMI 2.1a.The cooling solution features an aluminium fin-stack heatsink that's ventilated axially by three fans that offer individual speed control and come with intake air temperature sensors. The cooler shroud and backplate are made of die-cast aluminium. The main heatsink uses a large aluminium fin-stack to which heat is drawn by a vapor-chamber plate. The vapor-chamber and base-plate pull heat from the GPU, memory chips, and VRM. A secondary heatspreader pulls heat from other minor heat sources and structurally reinforces the card. AMD says that the vapor chamber plate is 10% larger than the one in the RX 6950 XT RDNA2 reference-design. The company is introducing a newer thermal interface material (TIM) for the GPU and memory chips, and an "ultra soft" thermal pad for the VRM MOSFETs.
It turns out that the RX 7900 XTX reference design card is physically larger than the RX 7900 XT reference design (if it does exist in production). The reference RX 7900 XTX is 28.7 cm long and 12.3 cm tall; whereas the RX 7900 XT is 27.6 cm long, and 11.3 cm tall, making it closer to being a standard "full-height" add-in board. Both cards at 2.5-slot thick. The RX 7900 XT is compact as the cooler has to deal with a 15% lower typical board power of 300 W compared to 355 W for the RX 7900 XTX.
21 Comments on AMD Radeon RX 7900 XTX Reference Design PCB and Cooler Detailed
Other than that, this looks like a good design - here's hoping it holds up in practice too!
I have to say it's rather interesting to not use the same PCB design for a flagship and second-tier GPU, especially when based on the same GPU die. I wonder how they worked that out economically - if I were to guess, I'd say the reference 7900 XT PCB is pin-compatible with at least one further SKU down the stack.
"6-Layer Server Grade PCB: 6-layer PCB with 2oz thickened copper provides higher performance and long-lasting system stability without any compromise." MSI
Of course, if the shroud was made of copper or another material that is more conductive than aluminium, things would be better.
Have you parked your car with the windshield oriented towards a building during colder periods when the temperature drops below 0°C?
You would see that the building heats the windshield and it remains frost free, unlike the other car screens which get frozen.
Besides, my case limits me to 30.5cm for the graphics, and I don't feel like looking for a new one. This reference design may just be the right fit at 28.7cm.
How is a 14 layer high performance pcb different from a regular 14 layer pcb? Do you mean to say you've been selling shit for the past years? (of course not, those were also high performance/premium/whatever). God damn marketing guys... Even some single use plastics are not banned because there's simply no viable alternative (medical stuff being a big example). Only common stuff like straws, forks, etc.. that are very easily replaceable even though some people like to make a big fuss about it.
But still no one who is committed to decreasing or eliminating coil whining
I honestly don't see any of this language as problematic - it's so broad that these are just general statements about the overall characteristics of the product. "High performance" - well, yes, it's a GPU meant to compete at the very peak of performance. High efficiency? Yes, they claim up to +54% perf/W - specifically. Premium means premium, as in expensive, using high grade materials, etc. If they aren't allowed to use very general descriptors like these you might as well ban the use of adjectives in advertising.