NIC is usually the networking device ... it may have multiple chips. It
used to be a separate card ("network interface card"), but these days it
is often part of the mainboard. Sometimes it's even embedded in the
chipset. (For some embedded systems, it might even be part of a single
chip -- SoC -- system-on-a-chip types of designs.)
MAC is the media access controller. Its sort of a logical networking
layer. For example, all ethernet devices use the same MAC protocols, so
its common to use a single MAC device. On a NIC, this is usually the
main brains of the device.
The PHY is the physical layer part -- the part that does the actual
electrical or optical signaling, such as the transceiver.
It is not uncommon to have a single MAC controller that can have
alternate PHYs, e.g. variants for optical or copper signaling might use
the same MAC, but different PHYs.
-- Garrett
Lars Tunkrans wrote:
> Hi,
>
> I am probably a novice at this subject.
>
> I have been looking at the new Motherboards with the Nvidia Nforce
> MCP72/73/78 chipsets .
>
> Apperantly the Network part of this MCP78 chipset is only the MAC
> and the motherboard vendor needs to provide a PHY ?
>
> so from what I can read some motherboards with the MCP78 chipset
> comes with a
> REALTEK PHY, others comes with an Atheros PHY.
>
>
> So, Now that we have the nvidia Gigabit driver ( nge ) is the
> driver dependent on the PHY ?
>
> and can the tasks and the responsibilities of the MAC and the
> PHY be breifely
> summarised for us novices ?
>
> Regards
>
> //Lars
>
>
> This message posted from opensolaris.org
> _______________________________________________
> networking-discuss mailing list
> [email protected]
>
_______________________________________________
networking-discuss mailing list
[email protected]
