I've been working with Ethernet for the better part of 30 years and I just now learned what a "runt Ethernet frame" is.

An Ethernet runt frame is a frame who's signal has degraded to the point that the start of frame bit isn't recognized and that a subsequent bit is incorrectly recognized as the start of frame bit. Thus yielding a recognized frame that is shorter than was originally transmitted.

My adaptation / interpretation of what William James Yeager wrote towards the end of § 3 ¶ 3 on page 7 of The Origins of the Stanford University SUMEX-AIM Router Technology - An Historical and Technical Overview.

This same paper confirms a suspicion as to why the Ethernet that we know and love is referred to as Etherne-2.

Ethernet-1, now often called research Ethernet, ran at 3 Mbps.

Ethernet-2, often stylized as Ethernet-II, ran at 10 Mbps.

Hence why we see Ethernet II in places like Wireshark.

#learningIsFun #Ethernet

@drscriptt there were actual frame format differences IIRC, so you can't just bridge them.

@LapTop006 That doesn't surprise me at all.

I had to re-read part of § 3 ¶ 4 multiple times.

"Moreover, each interface's MAC address was two bytes. The first byte is the unique host part of the destination host's PUP address on a subnet, and the second is the host address of the transmitting host. Consequently, the MAC address of a packet's encapsulation is 16 bits."

I was struggling with "MAC address" seeming to be singular while seeming to refer to two hosts. Now what we'd refer to as the destination MAC address and source MAC address, or two things.

I wonder if this is possibly lost in editing and was more mean to be "MAC addressing" as in "the addressing used by the media access control (sub-layer)".

So, yes, an Ethernet frame using a 1-byte destination and 1-byte source MAC addresses would very much so be a different frame format.

It's always interesting to learn where things came from.

Though I feel like I should read O'Reilly's "Ethernet: The Definitive Guide" by Charles E. Spurgeon.