Subnet Mask Query - Why 255?

[Scott]

This is probably a stupid question, in fact I'm pretty certain it is, I'm considering expanding my network past 255 as I have a ridiculous amount of smarthome stuff. I know that I can easily achieve this by changing the subnet mask.

But..... it got me thinking. What are the different mask calculations for? Is 255 actually faster than 254? I understand that it can impact stability and what-not, but I think most networks would be stable with 500 IPs?

Does setting the mask to 254 and then not using the extra scope still impact performance?

 

[RichLan564]

It won't change the speed of anything (that you would notice) it just allows for more addressable IP's so if you need more than 254 addresses then just widen the subnet mask as you suggest.

But it may depend on what IP class you are working with if you can actually use a 255.255.254.0 subnet, the default for most 192.0.0.0 addresses is 255.255.255.0 so 254 allowable addresses in each network, some devices won't allow that IP class to have anything other than that subnet...

 

[Scott]

It won't change the speed of anything (that you would notice) it just allows for more addressable IP's so if you need more than 254 addresses then just widen the subnet mask as you suggest.

But it may depend on what IP class you are working with if you can actually use a 255.255.254.0 subnet, the default for most 192.0.0.0 addresses is 255.255.255.0 so 254 allowable addresses in each network, some devices won't allow that IP class to have anything other than that subnet...

This is what I was finding. But, if that's the case why is a larger pool of IPs not the default?

I was wondering what the down side of 254 was really, otherwise why wouldn't it be the default over 255 (or 252 etc).

 

[Scott]

I had it setup with 254, 192.168.0.1 and 192.168.1.1 for the routers, worked great but the limit I used was 127 so I'm still well within the 255 that I was worried about. Went back to 255 and 192.168.0.1/2 for the routers.

 

[RichLan564]

This is what I was finding. But, if that's the case why is a larger pool of IPs not the default?

I was wondering what the down side of 254 was really, otherwise why wouldn't it be the default over 255 (or 252 etc).

No downside to using 254 in the correct IP range, but as i say some devices may see it is illegal and refuse to take it if the first octet is 192

Class C IP's are designed for use in small LAN's so limited to 254 addresses by subnet mask, if you want to go wider and have no issues with devices not accepting an effectively illegal subnet mask you should use a Class B, if most kit is DHCP its an easy change, bit more of a pain if you have a load of static kit.