Monday, June 26th 2023
More Pictures of NVIDIA's Cinder Block-sized RTX 4090 Ti Cooler Surface
Back in January, we got our first look at the cinder block-like 4-slot cooling solution of NVIDIA's upcoming flagship graphics card (called either the RTX 4090 Ti, or the TITAN (Ada). "ExperteVallah" on Twitter scored additional pictures of the cooler. Its design sees the heat dissipation surface pushed to the entire thickness of the cooler, and ventilated the entire length.
The card's PCB isn't conventional—not perpendicular to the plane of the motherboard like any other add-in card—but is rather along the plane of the motherboard, with additional breakaway daughter cards interfacing with the sole 12VHPWR power connector, and the PCIe slot. This slender, ruler-shaped PCB spans the entire length of the card, without coming in the way of its heat dissipation surfaces. The length is used for the large AD102 ASIC that's probably maxed out (with all its 144 SM enabled), twelve GDDR6X (possibly faster 23 Gbps), and a mammoth VRM that nearly maxes out the 600 W continuous power delivery design limit of the 12VHPWR.
Sources:
ExperteVallah (Twitter), Hassan Mujtaba (Twitter), VideoCardz
The card's PCB isn't conventional—not perpendicular to the plane of the motherboard like any other add-in card—but is rather along the plane of the motherboard, with additional breakaway daughter cards interfacing with the sole 12VHPWR power connector, and the PCIe slot. This slender, ruler-shaped PCB spans the entire length of the card, without coming in the way of its heat dissipation surfaces. The length is used for the large AD102 ASIC that's probably maxed out (with all its 144 SM enabled), twelve GDDR6X (possibly faster 23 Gbps), and a mammoth VRM that nearly maxes out the 600 W continuous power delivery design limit of the 12VHPWR.
145 Comments on More Pictures of NVIDIA's Cinder Block-sized RTX 4090 Ti Cooler Surface
How'd that work out with both Fury and Vega?
As I recall, the AIO cooled varieties were derided, and they had high(er) failure rates.
AIO cooled cards make a lot of sense for partners but it’s more problematic when you’re the manufacturer shipping at much higher volumes to all sorts of markets
I bet the 8060 IS this big, and half as expensive.
Those are all custom partner models. Reference/ODM-1st party designs typically do not choose water cooling. -the couple times AMD did, there were problems.
www.techpowerup.com/review/nvidia-geforce-rtx-4090-founders-edition/37.html
Nvidia makes great chips but not great coolers. The last thing they should be doing is designing water-cooling pumps and systems for their cards. Let the experts design those so Nvidia can stick to what they do best which is developing chips. You forgot the FE performs at the bottom of the product stack and they are the low end of binned chips. The best chips go to the best cards. That's how they hold the overclocks. That is why they cost more.
They are crappy and everyone knows not to buy the FE card unless that's the only way they can afford it. At least they moved past the Turbine style cards that performed even worse and sounded like an aircraft carrier.
To each their own of course, but it's true lol.
The FE is often the best way to get a good card without having to deal with retailers adding a premium on top of the premium. In France it's really a god send.
(The FE was also the only reason that I could grab the 3070 for the MSRP when everyone else was selling it for 1500€)
Go ahead and get all the FE cards your heart desires and happy gaming.
Not sure why you think I’m mad, I’m happy as a loon because it’s Friday!!!
There's no 4060 FE. there's a 4060 ti FE though.
I will say it again: the FE are good cards at a compact size. They are not the absolute best, but they are among the smallest SKU that you can get, and in that category their performance is good. If you look at that graph, you'll notice that the GPUs with the lower temperatures are much bigger GPU (in length and height) with triple fans. The 4060ti FE beats the other dual fan. The TUF is 1% faster. wow. I'm impressed. Meanwhile Palit run hotter and louder without even getting a 1% advantage.
Have you even considered that the FE running at Nvidia specs is the reason as to why they are not going to run faster than a strix or a suprim ? The AiB are not running at stocks specifications, they are using more power by default, therefore they run faster. 2 or 3% faster. You can get an additional 5% more speed with an OC. But the FE also get a 5% improvement from an OC. But you talk as if you are getting a 15% speed increase over the FE. Many people even favor undervolt to get performance close to default, but at a much lower power consumption. Increasing the power limits give you little return. Even
der8auer think that the 4090 FE is already using more power than necessary. Yes, an overclocker tuned the 4090 to get 5% less performance, but a 33% reduction in power draw. In the current industry, "stock" is already pushing the silicon close to the limits, there's not much more performance to gain. That's valid for GPUs and CPUs.
In which world a cooler cooling at 64°C for 29dBA in a small size is bad? You are too self-centered mate. The beauty of DIY is that there's many people building computers for different reason. Not everyone is looking to squeeze the most out their chips while having to deal with a massive cooler. The people who want to build a small PC are not beneath you. And the FE are filling a market that most AiB are not covering adequately.