SFP? You mean the every device has slots to plug in different transceiver modules? I guess that would make it more future proof, but I think that will raise the cost, and might confuse ordinary people.
You have to think about the slot-transceiver compatibility and transceiver-medium compatibility then. Hmm… but I guess that would make it more transparent what is going on than having those chips embedded inside the cables, but not sure if we can leave them out, and require the end users to take care of thinking of all these compatibilities themselves or risk fire hazards.
yeah, I guess tvs and receivers would come with active optical cables to make it simpler, but the main thing is that optical is much cheaper and faster than copper once you get the economies of scale down on the transceivers. 1 terabit over 100km, down a cable thinner than a USB cable, is no problem with the right lasers. Meanwhile, I have interference and patent issues at 0.02tbps on hdmi cables less than a meter long.
Plenty of cheap optical HDMI cables out there, but they have compatibility issues. It would be so much easier with standard mmf mpo or SMF lc cables.
apalrd did review a unique product recently that embeds a mmf transceiver into the existing HDMI for factor, though.
SFP is the modern standard for pluggable laser modules. RJ45 sfp modules exist, but only for 1G and 10G. There’s also DAC cables for sfp, but those are limited to 2-3m, and the point was to focus on the benefits of fiber. Maybe the economies of scale necessitate some modern silicon photonics like a fiber on package option, but then you have repairability issues.
The minimum bend radius is mostly because of complete internal reflection, fiber is very flexible, and it’s not really possible to break an armored fiber cable by hand. You do have to worry about dust on the ends, though.
Toslink is cool, but it’s a very low bandwidth standard, less than 1gbit. You need proper glass fiber and lasers for high bandwidth.
Imagine putting out a new high bandwidth cable standard in 2025 based on copper.
The sooner display and networking move to SFP, the better.
SFP? You mean the every device has slots to plug in different transceiver modules? I guess that would make it more future proof, but I think that will raise the cost, and might confuse ordinary people.
You have to think about the slot-transceiver compatibility and transceiver-medium compatibility then. Hmm… but I guess that would make it more transparent what is going on than having those chips embedded inside the cables, but not sure if we can leave them out, and require the end users to take care of thinking of all these compatibilities themselves or risk fire hazards.
yeah, I guess tvs and receivers would come with active optical cables to make it simpler, but the main thing is that optical is much cheaper and faster than copper once you get the economies of scale down on the transceivers. 1 terabit over 100km, down a cable thinner than a USB cable, is no problem with the right lasers. Meanwhile, I have interference and patent issues at 0.02tbps on hdmi cables less than a meter long.
Plenty of cheap optical HDMI cables out there, but they have compatibility issues. It would be so much easier with standard mmf mpo or SMF lc cables.
apalrd did review a unique product recently that embeds a mmf transceiver into the existing HDMI for factor, though.
https://youtu.be/1aIK01S5qa4
I think you are mixing up SFP and optical fiber. SFP modules with copper wires exist, and are common. (e.g. SFP 1000base-T modules)
You can also use optical fiber without SFP, like the toslink connection.
Optical fiber also has issues with requiring a larger minimal bend radius, because they easily break. So you have to handle them more carefully.
SFP is the modern standard for pluggable laser modules. RJ45 sfp modules exist, but only for 1G and 10G. There’s also DAC cables for sfp, but those are limited to 2-3m, and the point was to focus on the benefits of fiber. Maybe the economies of scale necessitate some modern silicon photonics like a fiber on package option, but then you have repairability issues.
The minimum bend radius is mostly because of complete internal reflection, fiber is very flexible, and it’s not really possible to break an armored fiber cable by hand. You do have to worry about dust on the ends, though.
Toslink is cool, but it’s a very low bandwidth standard, less than 1gbit. You need proper glass fiber and lasers for high bandwidth.