passive, right?

Well, in this case here, this particular case I was describing is active, but the point is that you get to do it without the complexity of having an optical waveguide in your setup.

But I think that...

I walk past McGill University back and forth when I go to the IEEE session here.

McGill has been a great place for some of the more cutting-edge design principles here around transmission physics.

And whether it's waveguide-based or single-mode fiber-based or multi-mode fiber-based, right now you see optics in the hollow core space taking off.

It turns out that Similar idea to the air core dielectric.

I mean, that's where you live, right? And I always wondered, your perspective must be incredibly unique because you have to be at the forefront, right? Because everyone that's developing the next generation, they need equipment to test what they're working on, right? So how do you stay ahead of that curve and what is it like?

Mm

... a FIR filter, and a DFE and a receiver.

So the, the receive chain for the equalizers has simply gotten larger and larger.

More, more tools, more assets, more technology has been applied to compensate for the losses of these channels.

The channels get worse, the compensation gets better, and they balance each other out.

Um, right now we're at a point where, um, the, uh, in the, in the DJ specification at 200 gigabits per second, uh, we have a, a transmit FIR filter-

Mm-hmm.

You can query through CMIS and find out which one is bad, pull it out, and swap it, and you have a field technician in the back of a rack for 10 minutes and you're done.

Mm-hmm.

Uh, but the, the, the channel behavior, getting enough bandwidth in the channel may simply dictate that we have to go to a higher order modulation, and that's, that's a force of physics that people may not be able to overcome.

But that relates to the channel, and one of the challenges here is the connectors don't have enough bandwidth.

You know, if you look at most high-performance connectors nowadays, I mean, the best OSFP connector you can get your hands on, uh, ones that have one millimeter interfaces on them, uh, they resonate quite badly around 85 gigahertz.

They-

Okay.

And right now, that's a very, very narrow field of people that can do that.

But if you go to PAM, uh, PAM-6, for instance, it brings the sibling fundamental or the Nyquist frequency down to 84 gigahertz.

How is that perspective?

Well, you know, there are people who are experimenting with high-end, homogeneous board materials with high dielectrics.

It's really expensive to put a server-class system, though, on a high-end board material.

So really you're better off.

I mean, you, you've got tons of power that has to be delivered.

You've got tons, you have it physically and structurally putting stuff on a fiberglass board still makes sense because you need to hold things in space.

You need to, and these days you have to route liquid cool into everything and have to power delivery.

I

And what speeds are we, are we talking about?

about?Yeah.

So, you know, we tend to think of this from the standpoint of the single per lane speed, because that's really where the physics gets, problems get solved.

These lanes all get aggregated in typically a by eight or sometimes a by 16 configuration, or sometimes a by four configuration.

Those are the kind of IEEE conventions.

So at a single lane rate, um, we're wrapping up the 802.3dj specification, which is define the 212 gigabit per second.

So, uh, 212 billion bits per second across, uh, a PAM4 copper interface and also an optical interface as well.

... market size