Just patched a long fiber link, which would've absolutely _sucked_ to get CAT.7 cable through.
The steel-armored OS2 cable is cheap and super robust/easy to work with. The LC connectors are even smaller than the hole for CAT.7 would've been.

This is nice \o/

What I want to say is: Don't be afraid of fiber for DIY networking people, it's awesome and often muuuch better than copper cable!

I also didn't have to do any annoying LSA+ terminations or crimps and the link will do 10G+ easily :)

@manawyrm I've been curious about trying this once I actually make a permanent cable run to my tinker shop...what equipment did you use for either end of the cable? :)

@ann3nova

No-name china 1270nm/1330nm SMF BiDi SFP transceivers (12$ on Amazon)

One side is my regular big Ethernet switch, which already had some SFP ports.
The other side is a TP-Link SFP media converter (also 15$).

@manawyrm @ann3nova do you need attenuators if you have an especially short run of cable between transceivers or can they just deal with ~full tx power arriving at their rx port?
Unless I fumbled 100m of an OS2 cable (assuming 0.4dB/1km) with LC/APC connectors (assuming 0.2dB insertion loss) at both ends only attenuates to about 90% of the input power

@hcsch @ann3nova
They can deal with it.

This is well specified in the datasheet of the transceivers:
https://resource.fs.com/mall/resource/sfp-1g34-bx10-datasheet-20251208090113.pdf

Average Output Power POUT: -9 to -3 dBm
Input Saturation Power P(SAT): -3 dBm
Receiver Damage Threshold: +3 dBm

So yeah, in other words: No problem, even a direct no-loss 10cm link would be (barely) in spec.

@manawyrm @ann3nova neat, I wasn't quite sure how close you can get to the saturating power, but I guess I could've just assumed that it'd be fine since they specced it to fit barely for zero-loss (which you wouldn't really get anyways).

Thanks ^^