I've never noticed/felt any difference when I set it to 0, 1, 2 or anything else in either Q3/CPMA/QL and even other games using the same engine.
I've never had a bad net though, packetloss is not really something I have seen that often at home.

So, good question...

well if you have a bandwidth meter you can maybe measure the difference of packetup 0/1. I know that 1 increases the traffic but I am not sure how much. (in Q3 that was!)

I downloaded some free utility to monitor outgoing udp traffic.

This is what I observed:

cl_MaxPackets made the biggest difference, going from 63 to 125 nearly doubled the outgoing traffic. This is what you would expect.

When cl_maxPackets was 63, cl_packetdup made a big difference going from 0 to 5. Maybe 50% more traffic.

When cl_maxPackets was 125, cl_packetdup still made a difference, but it was only about 15 % at a value of 5.

So, assuming that the utility can be trusted it would seem to suggest that cl_Packetdup is effectual, but not very much at cl_MaxPackets 125.

Still I wonder if sending more packets is a good thing or not if you have no packet loss. Is the server more "certain" of your information in that case? Or is it simply bad because the extra outgoing traffic has the potential to increase latency?

maybe if you send 3 identical packets the server can process whichever one arrives first thus reducing the latency?

leave it at default 1 or do 0 it doesnt matter anymore. it was for sluggish connections back then i guess. sure it increase the traffic a tiny bit, it duplicates the pakets, is making backups.

im not saying that the uc guide is right on every topic, but still it is a good q3 guide and also for ql.

go http://wayback.archive.org/ and type http://ucguides.savagehelp.com/

sluggish connections back then eh?

still see at least 1 warper out of 5 players everytime i join a team game mode server.

QL netcode only works on lans, or with flawlessly stable connections.. if your net is shit, packetdup won't make any difference at all

edit: I just checked for you and looks like it doesn't have any effect, at least none I could capture.

if your net is shit, packetdup won't make any difference at all

its not supposed to "help" with your connection, its just to make sure your actions get actually send to the server. if thats not a problem in the first place then sure, there will be no difference

good definition

edited: You are right, indeed task manager shows some minor difference (like a few hundred bytes in each seconds, but wonder why I wasn't able to capture it with a dedicated utility), so I was wrong and indeed it could be very usefull when your first packet did not reach the server.

I'm sorry

packetdup 1 and 0 in ql is the same. only values above 1 will make any difference.

Does that mean 0 still sends a duplicate packet or 1 doesn't send any?

No backup packets are sent on 1 and 0

Are you one of those who foruming in write only mode? :) just look above a little:P

Well, first I used proffesional capture softwares to analyze what's going on, and when I saw no difference I concluded it useless because it doesn't matter if it's "1" or "0".
Nukm told me (because he knows I guess), that I'm wrong, so I fired up taskmanager and added the "bytes sent per interval" and "bytes received per interval" columns.
You can clearly see that "1" sends more data (not much, only a few hundred bytes per interval), but still measurable.
The reason why the capture software didn't show any difference is still a mistery to me.

Nukm told me (because he knows I guess)

no no
your one post led me to believe that you were expecting something weird from the cmd but i didnt really follow the discussion and i have no idea if the cmd actually does what it should at 0/1

well, #7 sounds entirely different to me tbh:)

what does it sound like to you?
i was just trying to make it clear what the cmd is supposed to do in the first place

OK,
- @#4, I stated that packetdup wont make any difference if you have a shitty connection.
- @#7 you replied that packetdup not supposed to "help" with shit net (I did't say "help" btw, but ofc I understood what you meant). That post sounded like you know if packetdup 1 works in QL or not, and it sounded like you saying that I was wrong because it will "make sure that my actions get actually send to the server", so it's working like in q3.
- @#15 I stated that it's working like how you told me, and now you are saying you don't know? o_O

btw which part of my post led you to a conclusion that I'm "expecting something weird from the cmd"?

"help with shit net"

yea, I guess the original intention why the implemented the command was to send duplicated packets on good connections just because sending things are awesome, and not to make the game more playable on shitty connections:)

It's just for users who are experiencing packetloss to the server. If you have red spikes on your lagometer, try cl_packetdup 2 or 3. 1 or 0 are for good connections with no packetloss problems.

Please tell me what's the difference between 1 and 0 if there are no backup packets sent? what are the extra bytes i see in TM?

In Q3, cl_packetdup "1" meant that the game will send the user commands to the server, and will send it again in the next packet. So with cl_packetdup "4" for example you needed a loss of 5 packets in a row to cause commands to be lost.

Could you tell me please how does it work in QL?

it is very effective, now i can warp from mh to ra :D