kinda late, but looks like no one has posted yet
___________________________________________
Hey everyone!
Thanks again for all the great questions this week; here are your answers from the team at id Software. If you didn't get your question answered this time, stay tuned for a future questions thread. (And we might pull some of the really good ones from the previous thread that were posted after the deadline!)
[Questions answered by Lead Designer Adam Pyle, Programmer Mike Rubits, and Studio Director Tim Willits]
When are you going to fix the netcode and hitreg for good? Do you intend to fix rocket launcher no-hit issues (direct and splash damage), or do you think it works fine as it is? What about rocket despawn? When will you fix the stuttering a lot of people experience?
The terms "netcode" and "hitreg" are terms that cover a wide swathe of observable issues, some that don't even involve the network at all. I'd like to take this opportunity to talk about where we're at, and where we're going, so here's a (not so!) brief summary of what's going on!
We started off the Closed Beta period with two major systems being handled on the client: movement authority, and shooting authority. This is a perfectly valid tech design decision, and plenty of high profile multiplayer games have released this way, but we felt uncomfortable with the trade-offs for a game like Quake. We made the decision to start moving systems over to server authority and client prediction, which is similar to previous Quake titles. This is a change that can occur in pieces, and we're still not completely there yet.
One area where we should be seeing more changes is literally how the server is detecting hits (so the term "hit detection" is actually appropriate here!) We currently use an animation-based system where we simulate player animations on the server, and rig up a lower-polygon, fatter version of the champion, independent from armor selection. For the pace of our game, we felt the additional server CPU usage, and the technical complexity of having to sync up the state of the animation bones and blend tree over the network was not worth having the ability to match up hits to the specific animation. Our solution is to transition to using a static mesh for each character state. (moving, strafing, crouching, jumping, etc.) This also makes supporting rollback much simpler, and should result in a more reliable experience on higher pings.
On the subject of rockets, which have been a sore spot for both us and our players, a couple of patches back we made two major changes: we fixed an issue that was causing extra latency when firing rockets, and we also started predicting rockets and local knockback on the client. This is something that's not been a part of Quake before, but we believe that this will make for a much better experience on a non-LAN environment, especially while rocket jumping or using projectile abilities. This surfaced some problems with animations simulating on the server differently than the client, and is what the initial cause of rockets not doing damage was. We've already addressed this by moving rocket collision over to the new static mesh system described above. What remains known now is strictly a client-side issue where the client incorrectly predicts a hit that the server will never detect as valid. This can even occur on LAN, where latency is not an issue. Once this is fixed, this should improve rockets, however due to the nature of client-side prediction, there may still be cases where it predicts wrongly. We'll continue to monitor the situation past this change, and see if any other issues surface.
Additionally, we have also identified a potential fix that causes other players to appear more jittery, even at high framerates. The proposed fix is currently in development, and we're waiting to evaluate the effect it has, but due to the unstable nature of the Internet, sometimes it's not possible to evaluate a fix without actually going through with the implementation and trying it in a real world environment.
Finally, we've also improved bandwidth consumption. I've attached a graph below, measured in a live 8 player TDM over a period of 60 seconds at ~160FPS.
We've reduced the amount of game logs the client sends overall, and have gated certain logs to only trigger collection when the client or the server detects specific issues within the game. Bandwidth usage should continue to improve from here, although we do not consider this issue to be resolved yet.
Hope that clears up some questions!
Will we be able to host our own servers through the server browser? Seems like people have a bad experience on your servers, with high ping. If we can host our own, we can get lower ping and good servers.
We are continually working to improve the network experience for all users, this includes server providers and the code-side, but allowing users to host their own servers is not something we’re focusing on at this time.
In the last livestream, Tim said offline mode wouldn’t be implemented because it’s not secure. Could you develop a stripped-down server binary available offline that only allows for one human local player connection? Would this be a feasible way to have offline mode with bots?
Unfortunately, this particular suggestion would not be feasible, because if you lock it down it can (and will) be hacked, hence it remains a security issue. That said, we’ll continue to look at other possibilities in the future.
What kind of feedback/bug reports are helpful to the dev team to fix netcode issues?
Please include your KSIVA in any and all bug reports. You can locate your KSIVA on the watermark at the bottom of your screen. This is the most useful thing, as this code allows us to look up your game sessions, which provide a wealth of information.
What do you think about QWC pro players almost exclusively using Sorlag, Nyx, Slash, Anarki, and sometimes Ranger and Galena? Would you consider buffing other champions to also be useful in team/duel conditions?
For our very first world championship, we are encouraged with the diversity of Champion picks. That said, we do recognize that we had a few Champions that weren't selected often and we have a lot of small balance changes coming that we hope will bring even more diversity into player's lineups. We also understand that there will always be some Champions that remain more popular with pro players and others that are popular among more casual players.
Slash’s ability automatically moves her forward; will this be changed to allow her to use the ability in the direction she’s already moving? (Backwards, strafing.)
Yes, we plan to keep the speed boost, but not force the direction.
Are you working on a fix to Slash’s crouch slide suddenly stopping if you touch a small bump, e.g. mega health on Ruins?
Yes, when we find areas in levels we make adjustments to the collision to address the issue. We are also looking into her slide code to help resolve a lot of these issues programmatical, but it is difficult to address it in code without also altering how her slide feels, so we have to be cautious in making these changes, so we will continue to address it via map collisions first when possible.
Will we get a champion with a grappling hook?
We all have enjoyed playing CTF from the very beginning, and we are currently making a Champion with a Grappling Hook ability. We look forward to sharing her with all of you.
Will we see any maps with flesh textures? (Like q3dm13.)
Not yet at least. We tried incorporating some into the Dimension of the Doomed themed Goroth maps, but in the end, it felt out of place and was removed. If we decided to add fleshy bits, we would introduce a new environment similar to Q3's Gothic Hell style used on Lost World.
When will the next maps be introduced?
We have an Ithagnal Duel map coming out this fall.
Any plans to rebalance item locations on some maps, like Lockbox?
After we introduced the Tri-Bolt we found some of the more powerful weapons were too centrally located. We plan on adjusting the item layout to distribute the power throughout and encourage players to make use of more of the map.
Are you planning to incorporate more elements of Lovecraftian lore into future maps? A deathmatch in the ice-choked ruins beneath the Mountains of Madness would be a dream come true!
Absolutely. Tim loves the lore.
Why is there a longer weapon respawn time on Church of Azaloth?
This was just a bug that has since been internally fixed, ready for a future update.
What are your plans for the Tri-Bolt? It’s still useless…
We do feel it needs some adjustments. We are currently testing changes to it that we could hopefully roll out soon. Stay tuned for more information.
Will you allow us to change our name through Steam/Bnet?
Name changes will occur in-game, agnostic of platform. This will ideally become an automated service players can access. We’ll share more about this as the feature is closer to release.
Do you have any plans on adding zoom sensitivity settings?
Yes. Zoom sensitivity is a feature that we need to get in quickly. We are speaking with the team and should be rolling this out soon.
Why can’t we queue for FFA with our friends?
Playing against people in your party is open to abuse, where those players can assist you in intentionally raising or lowering your skill rating. As a rule of thumb, you play with people in your party, not against them. You can currently play FFA with your friends in Custom Game.
Are there plans to give players more feedback about their stats in comparison to other players, e.g. leaderboards, divisions, public MMR display or something of the sort?
Yes, yes, yes, and yes.
Is crosshair customization coming?
Yes, this will be coming sooner than later.
Can we get options to resize/customize HUD elements?
We do not currently have plans for resizing HUD elements, but we do have upcoming plans that would allow players to change HUD styles.
The general feedback I've had from getting newb friends to play is that they are almost always confused by how they died, who killed them and where they were shot from. Are there any plans for a killcam feature or anything that better illustrates how you died?
We understand their struggles, we've seen similar feedback. We have some plans on the backlog that we hope will address these concerns.
For the advanced movement tutorial, would you consider having an indicator similar to the DeFrag cgaz hud or Reflex Arena's accelmeter in order to elucidate to players where they need to move their crosshair?
Maybe, but we haven't started on one yet.
Will we be able to join games as a spectator in the future?
Yes.
Can we buy shaders with shards?
There may be new premium shaders in the future.
The sell price for vanity items is higher than the number of shards you get for duplicate vanity items. Will this be fixed?
Yes, this will be addressed ASAP.
Will you rework the rune system? I don’t think runes should drop randomly from backpacks. You should be able to play the champions you like to play and work on their runes first.
Runes are intended as a long-term goal for players, but we will continually monitor how players are receiving and completing runes, and look for ways to improve challenges and rewards.
Will there be cosmetic effects? (Abilities, weapon effects, etc.)
The QuakeCon 2017 Lightning Gun is our first example of a weapon with cosmetic effects, and we plan to continue adding custom effects to vanity weapons in addition to their custom sounds.
Can we have per-part shaders for headgear, torso, and legs?
Different shaders for cosmetics would create some technical and performance issues for players, so this is unlikely to be something we’ll enable any time soon.
Will you adjust the shard price for cosmetics now that there are so many more things you can get in loot chests that don’t have a shard value? (Runes, weapon shaders, etc.)
We are continuously evaluating the economy, it is our recommendation that if you see something you want to buy it immediately.
Can we get taunts like in DOOM multiplayer?
Taunts have been in Quake games since Quake 2. We are working on Quake Champion taunts and look forward to rolling them out and getting feedback from players.
Any hints about upcoming modes? (Not CTF.)
We're exploring options for additional modes. Some examples could be Instagib and Clan Arena, but we're evaluating how well these modes play in Quake Champions. We’re always interested in hearing your suggestions for new modes!
When will we have ranked mode?
End of the year.
When will you start sharing more news on CTF?
We just announced that we will be adding CTF. We are currently working on the designs to ensure that it plays the best possible way it can in Quake Champions.
When can we expect the gibbing system we had a glimpse of in the Quakecon video?
Hopefully by the end of the year, and it'll be worth the wait.
ETA on Vulkan support?
Vulkan support is coming along nicely. We're making progress and have the game up and running on it, but we're not ready to share it just yet.
https://bethesda.net/community/topic/90348/as...9-1?page=1
___________________________________________
Hey everyone!
Thanks again for all the great questions this week; here are your answers from the team at id Software. If you didn't get your question answered this time, stay tuned for a future questions thread. (And we might pull some of the really good ones from the previous thread that were posted after the deadline!)
[Questions answered by Lead Designer Adam Pyle, Programmer Mike Rubits, and Studio Director Tim Willits]
When are you going to fix the netcode and hitreg for good? Do you intend to fix rocket launcher no-hit issues (direct and splash damage), or do you think it works fine as it is? What about rocket despawn? When will you fix the stuttering a lot of people experience?
The terms "netcode" and "hitreg" are terms that cover a wide swathe of observable issues, some that don't even involve the network at all. I'd like to take this opportunity to talk about where we're at, and where we're going, so here's a (not so!) brief summary of what's going on!
We started off the Closed Beta period with two major systems being handled on the client: movement authority, and shooting authority. This is a perfectly valid tech design decision, and plenty of high profile multiplayer games have released this way, but we felt uncomfortable with the trade-offs for a game like Quake. We made the decision to start moving systems over to server authority and client prediction, which is similar to previous Quake titles. This is a change that can occur in pieces, and we're still not completely there yet.
One area where we should be seeing more changes is literally how the server is detecting hits (so the term "hit detection" is actually appropriate here!) We currently use an animation-based system where we simulate player animations on the server, and rig up a lower-polygon, fatter version of the champion, independent from armor selection. For the pace of our game, we felt the additional server CPU usage, and the technical complexity of having to sync up the state of the animation bones and blend tree over the network was not worth having the ability to match up hits to the specific animation. Our solution is to transition to using a static mesh for each character state. (moving, strafing, crouching, jumping, etc.) This also makes supporting rollback much simpler, and should result in a more reliable experience on higher pings.
On the subject of rockets, which have been a sore spot for both us and our players, a couple of patches back we made two major changes: we fixed an issue that was causing extra latency when firing rockets, and we also started predicting rockets and local knockback on the client. This is something that's not been a part of Quake before, but we believe that this will make for a much better experience on a non-LAN environment, especially while rocket jumping or using projectile abilities. This surfaced some problems with animations simulating on the server differently than the client, and is what the initial cause of rockets not doing damage was. We've already addressed this by moving rocket collision over to the new static mesh system described above. What remains known now is strictly a client-side issue where the client incorrectly predicts a hit that the server will never detect as valid. This can even occur on LAN, where latency is not an issue. Once this is fixed, this should improve rockets, however due to the nature of client-side prediction, there may still be cases where it predicts wrongly. We'll continue to monitor the situation past this change, and see if any other issues surface.
Additionally, we have also identified a potential fix that causes other players to appear more jittery, even at high framerates. The proposed fix is currently in development, and we're waiting to evaluate the effect it has, but due to the unstable nature of the Internet, sometimes it's not possible to evaluate a fix without actually going through with the implementation and trying it in a real world environment.
Finally, we've also improved bandwidth consumption. I've attached a graph below, measured in a live 8 player TDM over a period of 60 seconds at ~160FPS.
We've reduced the amount of game logs the client sends overall, and have gated certain logs to only trigger collection when the client or the server detects specific issues within the game. Bandwidth usage should continue to improve from here, although we do not consider this issue to be resolved yet.
Hope that clears up some questions!
Will we be able to host our own servers through the server browser? Seems like people have a bad experience on your servers, with high ping. If we can host our own, we can get lower ping and good servers.
We are continually working to improve the network experience for all users, this includes server providers and the code-side, but allowing users to host their own servers is not something we’re focusing on at this time.
In the last livestream, Tim said offline mode wouldn’t be implemented because it’s not secure. Could you develop a stripped-down server binary available offline that only allows for one human local player connection? Would this be a feasible way to have offline mode with bots?
Unfortunately, this particular suggestion would not be feasible, because if you lock it down it can (and will) be hacked, hence it remains a security issue. That said, we’ll continue to look at other possibilities in the future.
What kind of feedback/bug reports are helpful to the dev team to fix netcode issues?
Please include your KSIVA in any and all bug reports. You can locate your KSIVA on the watermark at the bottom of your screen. This is the most useful thing, as this code allows us to look up your game sessions, which provide a wealth of information.
What do you think about QWC pro players almost exclusively using Sorlag, Nyx, Slash, Anarki, and sometimes Ranger and Galena? Would you consider buffing other champions to also be useful in team/duel conditions?
For our very first world championship, we are encouraged with the diversity of Champion picks. That said, we do recognize that we had a few Champions that weren't selected often and we have a lot of small balance changes coming that we hope will bring even more diversity into player's lineups. We also understand that there will always be some Champions that remain more popular with pro players and others that are popular among more casual players.
Slash’s ability automatically moves her forward; will this be changed to allow her to use the ability in the direction she’s already moving? (Backwards, strafing.)
Yes, we plan to keep the speed boost, but not force the direction.
Are you working on a fix to Slash’s crouch slide suddenly stopping if you touch a small bump, e.g. mega health on Ruins?
Yes, when we find areas in levels we make adjustments to the collision to address the issue. We are also looking into her slide code to help resolve a lot of these issues programmatical, but it is difficult to address it in code without also altering how her slide feels, so we have to be cautious in making these changes, so we will continue to address it via map collisions first when possible.
Will we get a champion with a grappling hook?
We all have enjoyed playing CTF from the very beginning, and we are currently making a Champion with a Grappling Hook ability. We look forward to sharing her with all of you.
Will we see any maps with flesh textures? (Like q3dm13.)
Not yet at least. We tried incorporating some into the Dimension of the Doomed themed Goroth maps, but in the end, it felt out of place and was removed. If we decided to add fleshy bits, we would introduce a new environment similar to Q3's Gothic Hell style used on Lost World.
When will the next maps be introduced?
We have an Ithagnal Duel map coming out this fall.
Any plans to rebalance item locations on some maps, like Lockbox?
After we introduced the Tri-Bolt we found some of the more powerful weapons were too centrally located. We plan on adjusting the item layout to distribute the power throughout and encourage players to make use of more of the map.
Are you planning to incorporate more elements of Lovecraftian lore into future maps? A deathmatch in the ice-choked ruins beneath the Mountains of Madness would be a dream come true!
Absolutely. Tim loves the lore.
Why is there a longer weapon respawn time on Church of Azaloth?
This was just a bug that has since been internally fixed, ready for a future update.
What are your plans for the Tri-Bolt? It’s still useless…
We do feel it needs some adjustments. We are currently testing changes to it that we could hopefully roll out soon. Stay tuned for more information.
Will you allow us to change our name through Steam/Bnet?
Name changes will occur in-game, agnostic of platform. This will ideally become an automated service players can access. We’ll share more about this as the feature is closer to release.
Do you have any plans on adding zoom sensitivity settings?
Yes. Zoom sensitivity is a feature that we need to get in quickly. We are speaking with the team and should be rolling this out soon.
Why can’t we queue for FFA with our friends?
Playing against people in your party is open to abuse, where those players can assist you in intentionally raising or lowering your skill rating. As a rule of thumb, you play with people in your party, not against them. You can currently play FFA with your friends in Custom Game.
Are there plans to give players more feedback about their stats in comparison to other players, e.g. leaderboards, divisions, public MMR display or something of the sort?
Yes, yes, yes, and yes.
Is crosshair customization coming?
Yes, this will be coming sooner than later.
Can we get options to resize/customize HUD elements?
We do not currently have plans for resizing HUD elements, but we do have upcoming plans that would allow players to change HUD styles.
The general feedback I've had from getting newb friends to play is that they are almost always confused by how they died, who killed them and where they were shot from. Are there any plans for a killcam feature or anything that better illustrates how you died?
We understand their struggles, we've seen similar feedback. We have some plans on the backlog that we hope will address these concerns.
For the advanced movement tutorial, would you consider having an indicator similar to the DeFrag cgaz hud or Reflex Arena's accelmeter in order to elucidate to players where they need to move their crosshair?
Maybe, but we haven't started on one yet.
Will we be able to join games as a spectator in the future?
Yes.
Can we buy shaders with shards?
There may be new premium shaders in the future.
The sell price for vanity items is higher than the number of shards you get for duplicate vanity items. Will this be fixed?
Yes, this will be addressed ASAP.
Will you rework the rune system? I don’t think runes should drop randomly from backpacks. You should be able to play the champions you like to play and work on their runes first.
Runes are intended as a long-term goal for players, but we will continually monitor how players are receiving and completing runes, and look for ways to improve challenges and rewards.
Will there be cosmetic effects? (Abilities, weapon effects, etc.)
The QuakeCon 2017 Lightning Gun is our first example of a weapon with cosmetic effects, and we plan to continue adding custom effects to vanity weapons in addition to their custom sounds.
Can we have per-part shaders for headgear, torso, and legs?
Different shaders for cosmetics would create some technical and performance issues for players, so this is unlikely to be something we’ll enable any time soon.
Will you adjust the shard price for cosmetics now that there are so many more things you can get in loot chests that don’t have a shard value? (Runes, weapon shaders, etc.)
We are continuously evaluating the economy, it is our recommendation that if you see something you want to buy it immediately.
Can we get taunts like in DOOM multiplayer?
Taunts have been in Quake games since Quake 2. We are working on Quake Champion taunts and look forward to rolling them out and getting feedback from players.
Any hints about upcoming modes? (Not CTF.)
We're exploring options for additional modes. Some examples could be Instagib and Clan Arena, but we're evaluating how well these modes play in Quake Champions. We’re always interested in hearing your suggestions for new modes!
When will we have ranked mode?
End of the year.
When will you start sharing more news on CTF?
We just announced that we will be adding CTF. We are currently working on the designs to ensure that it plays the best possible way it can in Quake Champions.
When can we expect the gibbing system we had a glimpse of in the Quakecon video?
Hopefully by the end of the year, and it'll be worth the wait.
ETA on Vulkan support?
Vulkan support is coming along nicely. We're making progress and have the game up and running on it, but we're not ready to share it just yet.
https://bethesda.net/community/topic/90348/as...9-1?page=1
18476 Hits