- May 30, 2018 at 9:58 am #17401Super88ParticipantTopics: 4
I just mean for the majority of people who bought descent they haven’t continued playing.
Well that has alot to do with the failure of D3 and the Descent as a property basically tanking… The people at outrage messed up Descent let’s be honest. The great irony was, the person at interplay who suggested making the D1 expansion Descent 2 turned out to be correct. Descent 3 they had too much going on /w largely a divided team (volition went to THQ and outrage stayed with Interplay??) Anyway, the ultimate reason was after Freespace’s and then Descent 3’s failure most people at publishers weren’t too enthused about 360 degree space games.
Let’s remember Freespace 2 flopped and Descent 3 just didn’t do well sales wise as well.May 30, 2018 at 11:06 am #17402CDN_MerlinKickstarter BackerTopics: 15
D3 sold 50K copies.
FS2 sold about 25K copies.
Also, 6DOF games are harder to learn/master and this is why there aren’t many around. It’s the same for Flight sims, they don’t sell many copies because people don’t want to invest $$$ into joysticks/wheels/etc and the time to learn to master the game.
Most people from the 2000’s and on want instant action/gratificationMay 30, 2018 at 8:39 pm #17418DarkhorseKickstarter BackerTopics: 2
Looks like we’ve got 2 and 3. 4 is looking highly promising, and 1 is not.
Somewhat unfortunate since 1 was the most important, but may not have been trivial to implement.May 30, 2018 at 10:17 pm #17430
Hopefully there is just one reality that can be totally realized with all prediction off. With self prediction on it’s understandable to get hit even after you round a corner if you have high latency.May 31, 2018 at 2:24 am #17435SiriusKickstarter BackerTopics: 9
Nobody I’ve talked to on the Overload Discord cares about number of realities. They care about whether their skill matters. That’s been the drawback of a client/server model.
For most cases, it hasn’t been a crippling problem. Bigger games have enough going on that a little fuzzy hit detection isn’t going to make that big a difference anyway because you can’t dodge everything. The place it’s earned the most dissatisfaction is head-to-head.
From what I recall, there might be something coming at some point post-release to help with that… kind of emulate peer-to-peer I guess? Probably for smaller games though if I recall correctly. Until then, well, team games and FFA are still fairly good.May 31, 2018 at 9:09 am #17441yencilKickstarter BackerTopics: 7
Team games and FFA are what most games of this type are best suited to, anyway (and what I personally most enjoy, not only with Descent back in the day, but also games like Halo / Titanfall / etc).
The 1on1 scene with D1/D2 was certainly impressive, but it is fairly “exclusive” and not something that I think will be a draw for new players (it also takes a Long time to complete a game, another thing that has changed since the D1/D2 heyday).
On the basic premise of this thread, I think focusing on other multiplayer modes would be the best way for Overload multiplayer to truly succeed (which means, to me, drawing in and retaining a large enough playerbase).June 1, 2018 at 9:58 am #17467WhitesharkParticipantTopics: 14
First of all I agree with Sirius’ first post. I’m currently liking Overload multiplayer, but it also has some work to do (being able to choose servers should be among the first things to fix). Driller seems somewhat dominant to me because it’s not range limited, and the missile count at respawn could be reduced. I’m sure a lot will change in the future thoguh, as things become more customizable.June 1, 2018 at 10:35 am #17469yencilKickstarter BackerTopics: 7
Just wanted to add to my post above – sorry if it came off a bit “crotchety” ; ) as I do think this detailed feedback from people like Sirius is really invaluable to Revival Studios!
I also agree with his points – just wanted to express that if team / FFA modes are exciting, it’ll get people playing and then there’s room to add patches for other ways to connect / server options / etc (or things like the LAN mode).June 1, 2018 at 11:05 am #17470
There are 2 sides to the same WYSIWYG token
-I see my projectiles hit the target, but did they do damage to the target? (D1&D2 had a clanking sound)
-I see that I dodged the projectiles, but did they do damage to me? (D1&D2 had a clanking sound)
Higher the latency (ping) the more pronounced these issues become (hardly noticeable with low latency.)
Any specifics given about how exactly Overload’s architecture handles latency? Any lag compensation processing that Teamfortress 2 has, for example?June 2, 2018 at 1:57 pm #17505SiriusKickstarter BackerTopics: 9
On ship movement, they do use client-side prediction. If there is a mismatch between that and the server, the errors are corrected gradually and smoothly. I don’t know how much credit that system can take, but either way the result is currently good enough that you very seldom notice anything. You might on a really bad connection maybe, but it’s been great for me even when playing on a server in Europe.
On hit detection, there is no lag compensation as far as I recall or can notice. You can shoot someone with Driller, clearly hit on your screen, and do no damage; likewise, you can dodge the shots on your screen and take damage anyway. The effects are minor on lower pings, though apparently still big enough to matter from time to time. They’re significant on higher (around 100 ms or more) pings, to the point that a lot of people now appear to be self-segregating to their own geographical region, especially for 1v1s.
Games with more players aren’t affected as much because it’s easier to lead your shots on a target that is distracted by someone else. Not everyone likes playing FFAs with lag either, but I’ve been OK so far.June 2, 2018 at 5:49 pm #17518
Are there some network settings to control amount of prediction, or shut it off completely? In Quakeworld you could enter
cl_nopred 1 to shut off self prediction. Delayed movements feel horrible, but you know where you stand relative to enemy projectiles.
cl_predict_players 0 to shut off prediction of other players. Won’t be as smooth looking, but you don’t end up shooting at wrongly predicted enemy positions.
pushlatency -pingvalue I believe this allows you to set how far in the future to predict, ping time being a common value to choose
(QW wouldn’t let you predict your projectile shots, which I never had a problem with. Will Overload allow you to turn self projectile prediction off?)
rate lets you set bandwidth, cl_shownet and r_netgraph and showpackets showed useful networking stats
Oddly you could still movement cheat in Quakeworld by sending more than 1 second worth of commands in less than a second to the server. They could have easily programmed auto-overspeed-detection and auto kick players for doing that, but I don’t think the original servers did, probably because at first that exploitation was unheard of.
Also Quakeworld would not send other player’s position if not in potentially visible space (PVS). I assume Overload does the same. I read this on twitch
“i remember like… entering a pvp tournament once and because they didn’t want matches to last long, they enabled seeing people on the automap
: so i just watched the automap and threw missiles around blind corners
: this one poor guy kept turning a corner to eating a mega missile in the face”
You start getting paranoid that someone can see around corners if projectiles keep meeting you around corners!June 4, 2018 at 11:37 am #17543mickey1ParticipantTopics: 0
NBN chief blames online ‘gamers predominantly’ for fixed wireless congestion
Forums are currently locked.
Previous Topic: Hello Multiplayer! Next Topic: Axis binding bug with Saitek Cyborg 3D Gold