PUBG vs Fortnite Netcode Analysis - General

Interesting look at these two similar multiplayer games. Explains a lot about why PUBG hit register can be very strange sometimes. [hd]https://www.youtube.com/watch?v=SwZ_NUruGTM[/hd]

Fortnite is way laggier than PUBG is for me

PUBG is way laggier than Fortnite is for me

[QUOTE=BANNED USER;52820235]PUBG is way laggier than Fortnite is for me[/QUOTE] I can go onto pretty much any PUBG stream at a given time and see how bad the lag is when they try and loot something, the same can't be said for Fortnite.

PUBG has so much room for improvement in just about every category, really disappointing that something this fundamental is pretty garbage in both games.

[QUOTE=Socram;52820265]PUBG has so much room for improvement in just about every category, really disappointing that something this fundamental is pretty garbage in both games.[/QUOTE] The only reason I'm so excited this game is so popular is that it'll push other actually competent developers to make a better one.

I thought Fortnite was fine the first few times I played it, then yesterday while someone was shooting at me and at least 2-3 seconds AFTER when I had a chance to run into full cover behind a wall, turn around and run back to the corner and look at him with 3rd person, he damaged me while I was running away behind the wall shooting thin air then while I was sitting watching him shot the air way off in the open again and killed me. My best guess is he was lag switching or playing with some insane delay of like 2-3 seconds for him to still be able to see me that long after I was in cover. Had one or two other incidents after that where someone would rush me with a shotgun while spamming jump and their character would occasionally lag out and jerk a bit.

is there even any excuse for a game like PUBG to have such glaring issues with lag and stuff? they're pulling in such a gargantuan amount of money. i'm tired of playing games where i'll go to pick something up and there's a stupidly obvious delay in picking it up, which in a fast paced shooter can mean the difference between winning and losing. i've also fallen victim to the "my screen shows me in cover, but bullets still hit" shit.

[QUOTE=Hogie bear;52820295]The only reason I'm so excited this game is so popular is that it'll push other actually competent developers to make a better one.[/QUOTE] I'm sure I heard that exact sentence when talking about DayZ, and all of it's similarish battle-royal-esque clones.

[QUOTE=Cpt.Funkymonk;52820692]I'm sure I heard that exact sentence when talking about DayZ, and all of it's similarish battle-royal-esque clones.[/QUOTE] Day Z wasn't as successful as battlegrounds though. This game broke records even though it had glaring issues.

[QUOTE=loopoo;52820620]is there even any excuse for a game like PUBG to have such glaring issues with lag and stuff? they're pulling in such a gargantuan amount of money. i'm tired of playing games where i'll go to pick something up and there's a stupidly obvious delay in picking it up, which in a fast paced shooter can mean the difference between winning and losing. i've also fallen victim to the "my screen shows me in cover, but bullets still hit" shit.[/QUOTE] As player counts increase you either need to crank up the servers processing capacity, or drop the update rate. Due to the nature of hosting PUBG (and I assume Fortnite) on cloud services there's only so much you can do in terms of increasing server processing capacity. And for every extra player added the server has to perform a handful of reasonably intensive calculations (physics, input registration, etc.) against nearby players (so at the start of every server update there'll be a broad "cluster players nearby" phase to minimise work). But this all needs to be done within a minuscule number of frames to ensure that clients have something to work with. Lowering the tick rate of the server is significantly easier and gives you more time to work with players, obviously at the expense of fine-grain accuracy between client frames. Obviously a higher tickrate would be much nicer, but from a business perspective it's probably not worth having to request Amazon host some specific, super large instances of their Gamelift service just for you. That stuff costs enough as it is. A potential solution could be to make the tick rate dynamic, start of with slower, less frequent ticks as we have now and slowly ramp them up as the player count drops. Spectators cost very little to keep in the game, just a tiny bit more networking. But this also increases the cost of running the server, as cloud services tend to charge by compute time, not server lifetime.

[QUOTE=Hogie bear;52820730]Day Z wasn't as successful as battlegrounds though. This game broke records even though it had glaring issues.[/QUOTE] I agree that DayZ Mod never had the popularity that PUBG has right now, that being said, it's hard to ignore the fact that DayZ Mod was one of the most talked about games in 2012. Spawning multiple copies such as: H1Z1, Unturned, and Infestation Z. Late 2012/Early 2013 it felt like all we heard coming out was survival based, resource gathering, perma-death based games.

[QUOTE=Cpt.Funkymonk;52820770]I agree that DayZ Mod never had the popularity that PUBG has right now, that being said, it's hard to ignore the fact that DayZ Mod was one of the most talked about games in 2012. Spawning multiple copies such as: H1Z1, Unturned, and Infestation Z. Late 2012/Early 2013 it felt like all we heard coming out was survival based, resource gathering, perma-death based games.[/QUOTE] Yeah you have a point. Guess there's still the chance the game will die within a month.

loot-based survival multiplayer games seem to be demanding, netcode wise: you've got the usual requirements of a mulitplayer shooter (which can be demanding enough), coupled with synchronizing a good quantity of items/loot, vehicles and vehicle physics, large maps, and complicated environments. even DayZ, after tons and tons of improvements, still has lag issues on high-pop servers and in detail-heavy towns/areas

[QUOTE=paindoc;52821020]loot-based survival multiplayer games seem to be demanding, netcode wise: you've got the usual requirements of a mulitplayer shooter (which can be demanding enough), coupled with synchronizing a good quantity of items/loot, vehicles and vehicle physics, large maps, and complicated environments. even DayZ, after tons and tons of improvements, still has lag issues on high-pop servers and in detail-heavy towns/areas[/QUOTE] Planetside 2 has giant maps, tons of players and vehicles too, and while the netcode there was bad at times, it was surprisingly steady for a game of that immense scale.

[QUOTE=Recurracy;52821118]Planetside 2 has giant maps, tons of players and vehicles too, and while the netcode there was bad at times, it was surprisingly steady for a game of that immense scale.[/QUOTE] Except for massive portions of the playerbase.

[QUOTE=Doctor Zedacon;52821660]Except for massive portions of the playerbase.[/QUOTE] Well, that's out of the scope of netcode :v:

[QUOTE=Socram;52820265]PUBG has so much room for improvement in just about every category, really disappointing that something this fundamental is pretty garbage in both games.[/QUOTE] PUBG is essentially the best shitty game of all time. At least they've improved the overall performance tenfold compared to how it was in the beginning. It ran so poorly it was almost impressive considering it's on one of the most polished engines available

The fact how PUBG STILL has absolutely wonky scale on the objects in the menu screen baffles me. Like the AK for example is almost longer than the player character.