The Hidden Cost of Laptop Fans—and One Company’s Radical Alternative

Posted by

On July 6, 2026

TAIPEI—As the AI PC era becomes a runaway train, laptop makers face an old nemesis: heat generated by hot-running chips. New AI-focused processors pack ever more CPU, GPU, and NPU horsepower into increasingly thin machines. But the basic approach to cooling them hasn’t changed much in decades: The humble spinning fan remains one of the biggest constraints on laptop design.

Ventiva says it’s time to rethink how we do things.

At Computex 2026, I met with Carl Schlachte, the chairman, president, and CEO of Ventiva, a 62-person company based in Fremont, Calif., that has designed solid-state “ionic cooling” modules with no moving parts. The pitch isn’t merely quieter laptops. By eliminating fans, Ventiva says it can free motherboard space, give designers more flexibility, and help solve one of AI computing’s biggest emerging challenges: cramming massive gobs of memory close to increasingly powerful processors. And Ventiva has an automated factory in Malaysia that, Schlachte says, is ready to move millions of these modules.

What Is a Ventiva Module?

Ventiva aims to use solid-state ionic cooling to replace laptop fans, eliminating moving parts, vibration, and noise. One result? Quieter-running hardware. But the knock-on effects go beyond silence: They enable a laptop designer to reimagine the geography of a laptop motherboard.

(Credit: John Burek)

Think of it as a real estate crisis. In modern laptop design, component proximity matters more than ever. For example, Schlachte explains that running local AI inferencing requires immense memory bandwidth. To optimize that, memory must be soldered extremely close to the CPU to reduce the length of the motherboard’s traces (circuit lines).

“Father Jensen [Huang] talked about this at just the beginning of the week,” Schlachte quips. “These devices are becoming AI-centric, right? Call it inference at the edge, happening more locally than not. That is a memory-bandwidth thing.”

He continues, “Memory bandwidth means that you’re putting memory very close to the CPU. And [putting] memory very close to the CPU in this form factor is super, super hard, because if I put 128GB here, I really don’t have any space left over for anything else, right? Where Ventiva has found its stride is in devices that want to run big models locally in that kind of form factor, because we free up the space.”

I note that that sounds exactly like the kind of PCs that Nvidia outlined in its RTX Spark announcement: high-end laptops equipped with the new chip that would achieve their AI goals with large banks of unified memory. Memory proximity would be a big deal in machines like these.

Schlachte agrees. Showing a mockup laptop motherboard mounted on clear acrylic, he demonstrates how much space traditional fans waste. A typical laptop motherboard is a rectangle with two circular fan cutouts; the fans cool the CPU, GPU, and memory between them. These fans can occupy roughly 40% to 45% of the motherboard area.

(Credit: John Burek)

“In round numbers, that’s close to 8,000 square millimeters of very, very expensive real estate,” he notes.

Now imagine that space freed up for a designer to rethink the motherboard or allocate more room to other components, such as a bigger battery. Fan elimination can also allow designers to do things that were never possible inside a laptop. That’s because allocating a certain amount of fan real estate was an immutable “given.”

Indeed, Schlachte frames it as a psychological barrier in laptop design. “We call it fan blindness,” he says. “Meaning this thing has been around for so long that when you go talk to engineers who have been developing laptop motherboards forever, they cannot not see the fan!”

(Credit: John Burek)

At the core of Ventiva’s cooling tech are small, rectangular modules that can be stacked, lined up, or wedged into awkward spots. (Indeed, one of Ventiva’s side applications Schlachte describes is in rack servers, with modules addressing tricky dead-air spots between memory modules or near NICs.)

“They’re designed to be like Legos,” he notes. “You can do all sorts of weird orientations. If you want to do something at the corner of your machine, all of that stuff is possible.”

“We call it fan blindness. When you go talk to engineers who have been developing laptop motherboards forever, they cannot NOT see the fan!”

That said, Schlachte points out that in a challenging environment like a laptop or a mini-PC, the Ventiva modules are designed to push and pull air throughout the chassis, not just over specific components.

The Wire and the Wind: How Ventiva Modules Work

So how do you move air without a fan? Ventiva relies on an electrohydrodynamic (EHD) system that uses a tiny plasma field to generate airflow without any moving parts. “The physics of this is that we are charging the air,” Schlachte explains.

Schlachte walks me through the module itself. A very thin wire running the length of the module is charged, creating a plasma field that strips positive ions from the air. These ions are drawn toward a negatively charged collector, dragging nearby air molecules with them and creating airflow. “As that ion crosses the gap, it collides with air molecules, and you get billiard balls,” he explains. “Everything is cascading, everything starts moving, and you get air movement without having to move impellers.”

(Credit: John Burek)

I ask Schlachte about durability. Fan bearings wear out. Would the wire in the Ventiva module have a finite life, like a lightbulb filament, or is that a faulty analogy?

“No, no, it’s a perfect analogy,” he replies. “In a filament in a light bulb, you’re energizing it in order to create photons. We’re using it to create ions, so theoretically, there is a wear-out mechanism. But we haven’t found it yet! To be really honest, all our testing right now shows this exceeds the life of the machine. We’re well beyond everybody’s warranty.

(Credit: John Burek)

He says the filament itself, what Ventiva calls the emitter, is made out of a material that’s a trade secret. But the actual cooling hardware is only part of the story. Ventiva believes that replacing fans with its modules could fundamentally change how laptops are designed, too.

Without Fans, Rethinking the Laptop Motherboard

Fan removal doesn’t just affect cooling—it changes the economics and architecture of laptop design. And Ventiva chose to tackle laptops precisely because they are the hardest form factor to cool. “It’s the New York City joke: If you can make it there, you can make it anywhere!” Schlachte says.

“Laptops tend to be a very unforgiving environment for anything like thermal management, from a heat density per square millimeter standpoint,” he continues. “This is as bad as anything you see in a data center. In a lot of instances, it’s worse because you and I carry it around.” In other words, laptop cooling has to account for all the unpredictable thermal environments you might take your laptop into.

Eliminating the fans, though, can have knock-on benefits beyond reducing noise. The space factor is a big one: Ventiva modules are tiny, and removing fans could allow a laptop designer to put in a bigger battery, more than offsetting the potentially slightly higher power consumption of a Ventiva module. Or, it can allow a motherboard designer to fully rethink what goes where, and maybe even save money in the process or enable greater configurability.

On using fans in laptops: ‘That’s close to 8,000 square millimeters of very, very expensive real estate.’

Schlachte elaborates: “You put four of these things in, we’re burning a little bit more power, but we have a lot of customers that look at it and go, ‘Well, wait a minute, you just saved me all this space, I can now go with a 90-watt-hour battery, not the 65-watt-hour battery.’ Or now you can put a removable Type-2280 SSD in there, where you could never consider that before.”

(Credit: John Burek)

Another change that follows from this? The CPU can migrate around the board, opening up new possibilities for board design.

“What this does for the laptop companies is fascinating,” Schlachte says. “They’ve started to say things like: ‘Well, wait a minute. I don’t need to put the CPU in the center anymore. Before, it had to be in the center, because of the fans.

“But if I do two-piece motherboards [instead], if I take that CPU and I move it over here [he gestures to one side of the laptop], and I put all of my high-speed expensive I/O, the stuff I know I have to have, on this side….This is a 12-layer, expensive board. Then, I’m gonna make a six-layer inexpensive board over here [he indicates the other side of the laptop], and that one is going to change per SKU.”

The implication is significant: Manufacturers could isolate configuration-dependent components on cheaper second boards, reducing the cost and complexity of offering multiple laptop configurations. This could be a breakthrough in laptop economics.

Another Hidden Cost of the Fan: Pinch Points

Another breakthrough we discuss: Traditional laptop motherboards often force high-speed signals through narrow physical bottlenecks created by fan cutouts. Engineers call these bottlenecks “pinch points.”

“In all laptops, all the I/O, by tradition, is on the left or right side,” Schlachte says. “What happens? All that high-speed I/O here has to get to the CPU here. So the path it takes is this: It goes in, around, and then up.” He gestures on the motherboard model he has in hand, a path from the CPU, around the narrow part created by the fan cutout (outlined in red below), to the ports on either edge.

(Credit: John Burek)

“The trace length is actually longer than it should be, and all of the traces have to go through this point at the bottom of the motherboard….Pinch points you don’t like, because the more I/O you route through here, the more isolation is needed in the board itself. A lot of these boards are like 12-layer boards because of isolation, not because of anything else.”

“Second thing is, because of the length of that trace, somewhere in here you put a $3 repeater. The repeater is there to boost the signal. As soon as I get rid of all those, and I go straight, I don’t need the repeater. Don’t have a pinch point, so everything gets cheaper.”

Although Ventiva’s modules cost more than a conventional fan, Schlachte says several laptop makers have concluded that the potential motherboard simplifications can more than offset the difference.

“We’ve had some very large customers do a bill-of-materials analysis, and when you add in our devices, which are a little bit more expensive than a cheap $3 fan—I’ll concede that—we’re actually cheaper overall on the total cost.”

Recommended by Our Editors

At Computex 2026, a Partnership With Asus, and Future Designs

At Computex 2026, Ventiva announced a strategic partnership with Asus. Part of that was on display at Computex at Ventiva’s suite: Ventiva modules deployed in an Asus NUC Pro 16, one of the company’s stock compact business/AI mini-PCs.

(Credit: John Burek)

In a demo room, one of Schlachte’s employees demonstrates the NUC Pro 16 in action. In it, Ventiva stacked its modular ionic coolers to handle a 45-watt-TDP prototype unit.

(Credit: John Burek)

Schlachte also shows off a prototype AMD Ryzen-based laptop. The unit is configured for a 28-watt TDP, and the Ventiva modules are housed in a strip across the back edge of the laptop, venting out the rear.

(Credit: John Burek)

He notes that Ventiva also has a laptop prototype running Intel “Panther Lake” Core 3 Series chips. (As for Snapdragon X: “I think we’ve looked at some of them with the Qualcomm architecture, but I don’t think those are as far along.”) Also on display was a sample Dell unit using Ventiva, dubbed on the bottom “Silent Thermal Solution.”

(Credit: John Burek)

Now, Ventiva is not the only company pursuing this goal. I ask Schlachte about another buzzy approach to fanless cooling, AirJet, introduced a few years back by Frore Systems. AirJet uses piezoelectric modules with a vibrating element that forces air through the module’s body. As he explains, a bank of Ventiva modules is intended to cool the whole laptop, whereas an AirJet cooler is designed to sit atop a hot chip or other element. Schlachte concedes that the AirJet architecture is “super-innovative,” but that Ventiva is a different animal.

“For an AirJet Mini trying to cool, let’s say, 15 to 20 watts, it has to sit directly on the heat source. Take that to a laptop and your grizzled thermal engineer, the very first question they ask is: Why? You’re going up in Z-height, and the difference…is that you’re gonna get a lot of back pressure off of theirs, but not as much flow.

“Our belief is directed air movement in tight spaces is exactly the way to go, and that’s what laptop companies are ultimately going to want.” That said, we later learned that at Computex, Tom’s Hardware did get wind of (though did not actually see) an Intel “Wildcat Lake” 15-watt reference system running AirJet.

What’s Next for Ventiva?

I ask Schlachte where else we might see Ventiva modules. There are limitations to the modules’ airflow due to size, of course. “We’re not your 1,400-watt GPU cooler,” he notes. “That’s not our gig. So in servers and data centers, for example, we typically would say we don’t have an application. But a very interesting use case for us, in data centers, is the stuff at the back: the NIC cards, the MOSFETs, all the rest of it. Those things overheat.”

(Credit: John Burek)

“We actually learned this with Dell…in the transition from coolers to cold plates, they pulled out fans, so there’s less air movement at the back of the server now. Trying to get air in tight spaces is almost impossible. If you shut down a NIC, it doesn’t matter how well you’re cooling the GPU up front. The server’s still dead.

“Anything we can do to give them a little bit of flow—2cfm here, 1cfm there—improves the whole thing. Teeny little tiny spaces, put us in there…all of a sudden, everything’s good. So, we’ve got a very, very good and growing business there.”

I posit that 2026 may be a right-place/right-time situation for Ventiva, with the brand well positioned to address thermal needs that are ramping up in the era of the all-AI, all-the-time PC.

“We’ve got designs with four of the top five laptop companies in various stages,” Schlachte reveals. “You should see shipping laptops here in the next 18 months.”

(Credit: John Burek)

Following from that, I reinforce that the Ventiva model seems perfectly timed for Nvidia’s upcoming RTX Spark and similar platforms, like AMD’s Ryzen “Strix Halo,” relying on big banks of unified memory near the CPU. Schlachte gets animated.

“The RTX Spark machine they’re talking about is 300 megabits per second,” he notes, gesturing at an Apple laptop in front of him. “This machine, this is an M5 MacBook Pro. This is capable of around 600 megabits per second. It’s like double, right? How Apple did it is that in the M-series processor, it has a substrate, and they put the memory die right around it, very close. The Spark machine, at least the one that I’m aware of, has LPDDR soldered down.” I mention that soldered-down memory seemed to be what I’d seen in the RTX Spark motherboard samples shown by Microsoft for its Surface Laptop Ultra the day before, during PCMag’s briefings.

“If you’re RTX Spark, you have a real big problem, because even with a 15-inch laptop and two fans here, what you’re going to lose are these two edges,” he says. “All the memory has to be here and here, because of the fans.”

A pause. “And so you know: We’ve looked. We’ve looked at this design over and over and over, and we’re just waiting. If you want that edge, you want to get to 600 megabits per second, you have to get rid of the fans, or you have to go with a much bigger laptop. Somebody’s going to give us a call, right?”