07 Dec, 2007, Brinson wrote in the 1st comment:
How much bandwith does a mud use? Alot? A little?

I realized I have no idea about the answer to this question. Most hosts don't pay attention to it. Does anyone know?
07 Dec, 2007, Remcon wrote in the 2nd comment:
Well it depends on lots and lots of things on this one.
`Transfer data since Wed Nov 28  9:06:20PM 2007Received           :  11KB 627BUncompressed sent  :  272KB 104BCompressed sent    :  253KB 4BSaved by MCCP      :  1MB 418KB 323BIMC data received  :  2MB 668KB 812BIMC data sent      :  352KB 119BTotal Received     :  2MB 680KB 415BTotal Sent         :  877KB 227B`

Thats from mine. I was kind of curious about how much it would use so decided to put in something to see lol.
07 Dec, 2007, Zeno wrote in the 3rd comment:
I'm looking over my Network Probe software, and I see data from a certain port (that a MUD is running on): the Bytes column says 1.6MB (and I think that's only for today).
07 Dec, 2007, David Haley wrote in the 4th comment:
In the scheme of things, MUDs definitely don't use a lot of bandwidth. It's pretty easy to do the math and get estimates… Let's assume that you're going from room to room once a second. Assume average description length is 6 lines of 72 characters, which is 432. So, one player uses 432 bytes/sec in this scenario. That's 1,555,200 bytes per hour, i.e. 1.5 MB. So if you have, say, 20 players online, you are sending 30 MB per hour. The bandwidth in the other direction is more or less negligible by comparison.

But, that is assuming 20 players who are constantly active, moving around very quickly. More likely, you're only changing rooms once every several seconds. If you assume something more reasonable like 1 change/10s (which is still a lot of moving) you get 6 change/min so at 432 b/change it's 155,520 bytes per hour. 20 players in this scenario costs 3,110,400, i.e. 3mb per hour.

But again these are what I'd call "worst-case" scenarios. The best thing to do is get actual empirical data for your MUD. Throw in something like what Remcon did; edit your networking code to keep track of every outgoing byte. Then run your game for a few days, and see what you get.
07 Dec, 2007, Conner wrote in the 5th comment:
Remcon said:
Well it depends on lots and lots of things on this one.
`Transfer data since Wed Nov 28  9:06:20PM 2007Received           :  11KB 627BUncompressed sent  :  272KB 104BCompressed sent    :  253KB 4BSaved by MCCP      :  1MB 418KB 323BIMC data received  :  2MB 668KB 812BIMC data sent      :  352KB 119BTotal Received     :  2MB 680KB 415BTotal Sent         :  877KB 227B`

Thats from mine. I was kind of curious about how much it would use so decided to put in something to see lol.

Hmm, that output looks pretty nice. (From a numbers oriented/obsessed sort of standpoint…) Is that just an expansion of the checktransfer command you'd written me?
07 Dec, 2007, Remcon wrote in the 6th comment:
Yea, ran across it the other day and put it in and expanded it some lol.
08 Dec, 2007, Conner wrote in the 7th comment:
Looks to me like maybe a bit more than a little too.. you willing to share the changes?
08 Dec, 2007, Kayle wrote in the 8th comment:
You willing to share the whole thing? :P
08 Dec, 2007, Remcon wrote in the 9th comment:
Always willing to share it, and it will be in the next release of LOP which will likely be sometime this month lol. It's simple enough though so if you want me to I could put a snippet together for it, let me know if you would prefer it as a snippet.
08 Dec, 2007, Kayle wrote in the 10th comment:
Either way works. :P Snippet might be faster though. :P
08 Dec, 2007, Remcon wrote in the 11th comment:
Well the release is out now and has the stuff in it.
Check out transfer.c imc.c db.c comm.c (I think thats all the places that have to do with that code)
08 Dec, 2007, Conner wrote in the 12th comment:
Cool, Remy, thanks. :smile:
09 Dec, 2007, Zeno wrote in the 13th comment:
lol looks like this wasn't something I could just take transfer.c and put it in. :P After a bit of work, I got it working.
09 Dec, 2007, Conner wrote in the 14th comment:
For my mud, which already had the older version of the check transfer command he'd written, it is almost just a matter of trading transfer.c files and creating the transfer.txt file, but it's not that much additional. Nice job Remy. (And thanks for sharing.) :smile:
09 Dec, 2007, Guest wrote in the 15th comment:
AFKMud also has a version of the transfer data code in it. I think I'll be modifying ours to just give a raw byte count though. The GB/MB/KB splits look weird to me.
09 Dec, 2007, Remcon wrote in the 16th comment:
Will be a little more then that Conner, you should check all the mentioned files to make sure its all set correctly etc…

Thats actually not a bad idea Samson. Could drop it down to just two ints each and up the second one around 1billion/million and give a raw byte count.
09 Dec, 2007, Conner wrote in the 17th comment:
Oh, I saw 'em all.. and there's also something in mud.h as well if you don't already have the command, but compared to some snippets I've installed (the aquest code comes to mind…).. it's about that easy. :wink:
16 Dec, 2007, Zeno wrote in the 18th comment:
Here is an updated set of data from BIYG, in case the OP is interested how much an active MUD sends/receives:

From the 15th:
`Transfer data since Sun Dec  9 12:07:28PM 2007Received           :  3MB 51KB 840BUncompressed sent  :  400MB 560KB 700BTotal Received     :  3MB 51KB 840BTotal Sent         :  400MB 560KB 700B`

That's a LOT. My website used 939.75 MB of bandwidth during November. At this rate, my MUD will overtake my site bandwidth.
17 Dec, 2007, Conner wrote in the 19th comment:
Have no idea at this moment what my web traffic looks like, but my dev port is showing:
`Transfer data since 03:06:06 on 12/12/2007Received           :  101KB 125BUncompressed sent  :  236KB 372BCompressed sent    :  286KB 694BSaved by MCCP      :  3MB 42KB 239BIMC data received  :  840KB 35BIMC data sent      :  51KB 874BI3 data received   :  I3 data sent       :  Total received     :  941KB 160BTotal sent         :  574KB 917B`

and my live port shows:
`Transfer data since 04:12:17 on 12/12/2007Received           :  737KB 898BUncompressed sent  :  6MB 885KB 240BCompressed sent    :  2MB 227KB 2BSaved by MCCP      :  16MB 882KB 957BIMC data received  :  762KB 373BIMC data sent      :  99KB 651BI3 data received   :  I3 data sent       :  Total received     :  1MB 477KB 248BTotal sent         :  9MB 188KB 893B`

I added for it to watch i3 traffic on mine too because I still have the i3 network stuff in my codebase but I don't actually connect to i3 anymore so… also the above reflects the total sent including Remcon's latest fix for the bug I brought to his attention earlier today… also, in the time span documented above for my live port, things have been a bit slow - there have only been 77 logons since 04:12:17 on 12/12/2007.
Random Picks

0.0/19