Thursday, January 9th 2025
NVIDIA GeForce RTX 5090 Founders Edition Hands On, Taken Apart
At the 2025 International CES, we went hands on with the new NVIDIA GeForce RTX 5090 Founders Edition "Blackwell" graphics card. This thing is huge—longer and taller than the RTX 4090 FE, and yet just 2 slots thick. This is because NVIDIA's designers have figured out that the heat dissipation area of the heatsink lost to thinning the card can be recovered by stretching it in other directions. The card retains the essential aesthetic of Founders Edition cards from the past two generations going back to the RTX 30-series, but changes the concept of the dual-axial flow-through.
While past generations used an intake fan on one side, blowing onto the PCB, and another fan at the tail end of the backplate pull air through the heatsink and out the back, the RTX 5090 FE has two large fans, both of which blow cold air through the heatsink, and out the back of the card. The PCB is located in the center of the card, and relies on a set of breakout PCBs for host interface and display outputs.
The biggest component on the PCB, which takes up nearly 1/3 of its board area, is the "GB202" GPU on which the RTX 5090 is based. The GPU has a gigantic pin-count not just for its power needs, but its 512-bit GDDR7 memory interface. At one corner of the PCB, which sticks out from the top of the cooler, is the card's 12V2x6 power input, which is rated for 600 W (we don't know the TGP of the RTX 5090 yet).
The card uses a VRM solution with 19 phases for the VGPU, and 8 phases for the memory. Much like an AI GPU board, NVIDIA resorted to high-density PCB engineering, not wasting any space on either sides of the PCB. The chokes and DRMOS (made by MPS) are on the obverse side, surrounding the GPU on three sides; and the capacitors are on the reverse side.
The reverse side of the PCB has connectors that lead to its two breakout components. The first one connects to a PCB with the PCI-Express 5.0 x16 gold fingers. The other connector leads to the display I/O breakout. Both connections are made using thin ribbon cables like the ones you find in laptops, and routed along the edges of the cooler, so as not to impede airflow.
While past generations used an intake fan on one side, blowing onto the PCB, and another fan at the tail end of the backplate pull air through the heatsink and out the back, the RTX 5090 FE has two large fans, both of which blow cold air through the heatsink, and out the back of the card. The PCB is located in the center of the card, and relies on a set of breakout PCBs for host interface and display outputs.
The card uses a VRM solution with 19 phases for the VGPU, and 8 phases for the memory. Much like an AI GPU board, NVIDIA resorted to high-density PCB engineering, not wasting any space on either sides of the PCB. The chokes and DRMOS (made by MPS) are on the obverse side, surrounding the GPU on three sides; and the capacitors are on the reverse side.
The reverse side of the PCB has connectors that lead to its two breakout components. The first one connects to a PCB with the PCI-Express 5.0 x16 gold fingers. The other connector leads to the display I/O breakout. Both connections are made using thin ribbon cables like the ones you find in laptops, and routed along the edges of the cooler, so as not to impede airflow.
122 Comments on NVIDIA GeForce RTX 5090 Founders Edition Hands On, Taken Apart
But I agree the GPU needs the wc the most. I have a 7800X3D CPU that seems to have a terrible heat conduction problem (seems like mine even more than those from reviews). I tripled check the thermal paste and cooler attachment so it doesn't come from that. At some point I think I will pop out the IHS and see what I can do with liquid metal because I have a Noctua D15 from my previous PC, and it doesn't ever gets hot, I think I could run it fanless with 0 problem (I recently limited its max fan speed to 60% because more didn't change anything), but in the meantime I am temperature limited in Cinebench, and pretty easily reach 85℃ in demanding apps (in most games around 65-70). The 7800X3D uses only 80W max so there is 0 need for a wc for it, and my D15 is way oversized as well.
A GPU on the other hand has a huge power consumption and a huge dissipation surface and since it's a PCI card there is also little space above it, so it is basically the perfect case of use of a wc.
With water, there are so many more points of potential failure...
Heat pipes are obviously not enough, why not invent something new ?
This is a very cool design but we're gonna learn real fast how many people manage to ruin it during maintenance.This is where we're at. The co sees a problem and runs it right into the ground.
PCB too heavy? "MaKe A BeTtEr sLoT!!!1"
A 128LB GPU is heavier than my grandmother and it still doesn't do anything about the real problem.
Gigglebyte.
New generation Fermi incoming ?
80c is still okay temp for GPUs
And lets hope Amd can release 5090 perf card before 2030
Nvidia's Blackwell flagship GPU uses liquid metal instead of thermal paste to reign in the 575W TGP | Tom's Hardware
I wish an AIB purposefully made a variant with crap cooling just so I can throw it away and slap on that WB. My 3090 FE's cooler didn't even spin it's fan (in my hands) - I dismantled it before powering it on. Not the best feeliing during bootup though..
Watch and see, I was actually hoping to see one of the AIB's follow suit, by they didn't, but hey, they just slap a triple/quad slot cooler on such a tiny thing and then complain when customers send in cards with cracked PCB's or lifted components.
5000 Serie generates 3 more frames
6000 Serie may generate 7 more frames
7000 Serie may generate 15 more frames....
Oh they angled the connector again, this is a step forward then I think, albeit with needing a ton of room in case with the length.I expect some AIB's to copy, either in revised models for the gen or next gen, seems like Nvidia doing more innovation than AIB partners now for cooling.I expect it will cool fine being honest the 4080 FE cooler is overbuilt, the 3080 FE cooler did fine for same TDP. This thing with massive extra airflow so I think will be good enough. On legacy designs the air would just hit a PCB and go splat outwards with nowhere to go, vacuum to try and help it, this just goes straight through.
I am just extremely annoyed to find out that reviews (with the exception of the 5090's) will only be made available on the 30th... the day the cards launch, so I can't make an informative purchase beforehand, I am not a brand fan-boy, I want the best, reliable product for my buck.