Skip to main content

Forums / Games / Halo 5: Guardians

Heavy Aim caused by netcode

OP IWI IUI

  1. 1
  2. ...
  3. 30
  4. 31
  5. 32
  6. 33
  7. 34
  8. ...
  9. 41
LUKEPOWA wrote:
If it's a CPU issue, would they even be able to fix it? With the game winding down with updates, it doesn't seem hopeful if it requires some type of overhaul.
An overloaded server can cause the tick rates to drop which forces the xbox to do extra prediction steps. The net code could also be forcing excessive game state checks which bogs it down. I use the word excessive here lightly though. With thrust enabling an immediate change in direction check rate might be necessary.
Aykibaby wrote:
Woooow. Guys in the Teambeyond-topic Unyshek joined in! http://teambeyond.net/forum/topic/15295-i-have-devised-a-way-to-demonstrate-heavy-aim/page-4

Finally, we made them listen... actually it was more like you, the hard working guys with all the testing, did it. Congraz to all of you @OP & @MIb2347 & to all the testers etc pp. This feels so satisfying.

Now I hope they keep informing us. Transparency would be awesome.

Thx Unyshek!
First off, nice work @OP & @MIb2347!
Maybe we can finally get this resolved, or at the very least dare hope a Halo game will NEVER again release with a broken aiming system.

Secondly, really @Unyshek ?!
Now you (as a figure of 343i) "bother" to give this topic ANY kind of response, and you won't even do it on the official Halo forum (Waypoint) ?!
This topic has been alive for months (since 1/8/2017)!!
People has REPEATEDLY asked for 343i to chime in on the topic, and you have the audacity to respond ON A DIFFERENT FORUM!!
Why does 343i even bother with having Waypoint if you won't respond to the people using it?

343i seriously needs a new PR-manager, and new PR-policies. This kind of communication with your customers/consumers (you know, the people that make sure you get payed at the end of the day!) is going to have serious backlash.
Apoll0 wrote:
IWI IUI wrote:
Apoll0 wrote:
IWI IUI wrote:
IWI IUI wrote:
MIb2347 wrote:
I have an important update.
SacUsa on TB brilliantly suggested that I turn on machinima mode as it gives numbers that show your reticle position. Here is a video of that.
http://xboxdvr.com/gamer/Mib%20HCS/video/29208763
You can see that it will land it the exact same location for a period of time and then somewhere else. It will also sometimes come back to that location as close as possible to that. I would like to do a lot more testing but I haven't got time today. I only did this because I knew how to set it up quickly and it was done in my only spare 5 minutes.
Is it the activation of the vibration that moves the aimer or the duration? If it is the duration, have you been able to verify that the weapons don't have more than one length of vibration coded into the game?
Team beyond, believes they have figured it out. Hardware issue.
I know. I was the one that discovered it was a cpu issue. I pm'd MIb about it and he made the post in beyond. Last time I tried to bring this up on the website I was immediately insulted and cast out. Probably a good thing he made the post and not me haha.
No. Your OP edits don't automatically update all the times you were quoted, just so you know. Your initial argument was that the latency to the server and -Yoink- netcode was the cause of the issue which is NOT the problem and NOT the original argument you were trying to make when people said you were wrong and didn't have legitimate evidence. You even tried to cite that this was true because it didn't happen on LAN, which everybody already knew was not true.
Dont try to make TB sound like they ran you out a couple months ago for having this same theory because that is not accurate at all. The CPU loading/latency theory is new and corroborated by the experiments here so dont try to make it sound like this was your idea all along. this is new and if you had come back with this same experiment and evidence, the support for this would have been just as strong.
TJHIGHGUYS wrote:
Couldn't there be a variable in the program/mechanism if it is used with the vibration on the controller. When taking apart many controllers I feel that the vibration may be inconsistent. Couldn't the weights that spin around the bar to create the vibration accidentally go one extra little spin? It would be equivalent to a millisecond but that would throw off the amount of time the controller spent aiming left at 100%? I feel like banking on the absolute consistency in the vibration may be an issue.
Please advice!
He is not using the actual rumble pad, just the connections to the pad. The signal that is sent to the pad should be the same every time unless they coded variability into it in the game on purpose which wouldn't make sense imo. He also ran the test with CE and it landed spot on the same place 66/67 times, so thats a good comparison.
I discovered the issue was the CPU days ago and messaged MIB about it. He made the post in team beyond. Would you like me to screen shot the pm?
And latency and server stress does have an effect. The xbox is forced to do extra client side prediction and interpolation which causes additional cpu stress.
I do believe that you found the CPU issue a few days ago and messaged mib. That's not what i was saying at all. I was saying that you tried to make it sound like you discovered it months ago and were "driven out of town" on TB which is NOT the case. You also just said before, "the problem is caused by latency" which isn't true. "latency can add to CPU stress that then exacerbates the problem" IS true, but that was not your original argument.
This new theory is totally legit and i'm behind it. My only reservation is that we need to make sure there is no designed variability with the rumble signals that get sent to the controller. We would either need a response from 343, or a way to measure the rumble signal directly and not filtered through the game.
I think we should test this with another weapon too. If we can find one where the rumble signal length is longer but a consistent length of time, it would be interesting to see if the variability stays the same to indicate variable input delay as the culprit, or if it get exacerbated over a longer length.
very interesting and damning stuff as long as we can confirm that the rumble signal is consistent.
I concur. I suggested the sentinel beam. I'm thinking having bottomless clip enabled will keep the rumble completely consistent. I can take the clip and put it in the video editor to measure the 360 time. The only problem is the test will be limited to custom games and warzone. Still workable though. I did the test before measuring terminal times but did not use the circuit technique MIB is using.

As for a shorter burst that is not user controlled. I was thinking the h2 br. It's possible they kept the original rumble signals.
IWI IUI wrote:
LUKEPOWA wrote:
If it's a CPU issue, would they even be able to fix it? With the game winding down with updates, it doesn't seem hopeful if it requires some type of overhaul.
An overloaded server can cause the tick rates to drop which forces the xbox to do extra prediction steps. The net code could also be forcing excessive game state checks which bogs it down. I use the word excessive here lightly though. With thrust enabling an immediate change in direction check rate might be necessary.
Computational tasks can also be prioritized and scheduled in the game engine. So for example if the controller input calculation needs to "wait its turn" behind a physics calculation, it could introduce lag to the input. There are a million ways all this stuff can be handled and im sure the Xbox one's anemic hardware doesn't help.

No excuses though, Halo seems to be the only game that really has this problem on the platform
IWI IUI wrote:
Apoll0 wrote:
IWI IUI wrote:
Apoll0 wrote:
IWI IUI wrote:
IWI IUI wrote:
MIb2347 wrote:
I have an important update.
SacUsa on TB brilliantly suggested that I turn on machinima mode as it gives numbers that show your reticle position. Here is a video of that.
http://xboxdvr.com/gamer/Mib%20HCS/video/29208763
You can see that it will land it the exact same location for a period of time and then somewhere else. It will also sometimes come back to that location as close as possible to that. I would like to do a lot more testing but I haven't got time today. I only did this because I knew how to set it up quickly and it was done in my only spare 5 minutes.
Is it the activation of the vibration that moves the aimer or the duration? If it is the duration, have you been able to verify that the weapons don't have more than one length of vibration coded into the game?
Team beyond, believes they have figured it out. Hardware issue.
I know. I was the one that discovered it was a cpu issue. I pm'd MIb about it and he made the post in beyond. Last time I tried to bring this up on the website I was immediately insulted and cast out. Probably a good thing he made the post and not me haha.
No. Your OP edits don't automatically update all the times you were quoted, just so you know. Your initial argument was that the latency to the server and -Yoink- netcode was the cause of the issue which is NOT the problem and NOT the original argument you were trying to make when people said you were wrong and didn't have legitimate evidence. You even tried to cite that this was true because it didn't happen on LAN, which everybody already knew was not true.
Dont try to make TB sound like they ran you out a couple months ago for having this same theory because that is not accurate at all. The CPU loading/latency theory is new and corroborated by the experiments here so dont try to make it sound like this was your idea all along. this is new and if you had come back with this same experiment and evidence, the support for this would have been just as strong.
TJHIGHGUYS wrote:
Couldn't there be a variable in the program/mechanism if it is used with the vibration on the controller. When taking apart many controllers I feel that the vibration may be inconsistent. Couldn't the weights that spin around the bar to create the vibration accidentally go one extra little spin? It would be equivalent to a millisecond but that would throw off the amount of time the controller spent aiming left at 100%? I feel like banking on the absolute consistency in the vibration may be an issue.
Please advice!
He is not using the actual rumble pad, just the connections to the pad. The signal that is sent to the pad should be the same every time unless they coded variability into it in the game on purpose which wouldn't make sense imo. He also ran the test with CE and it landed spot on the same place 66/67 times, so thats a good comparison.
I discovered the issue was the CPU days ago and messaged MIB about it. He made the post in team beyond. Would you like me to screen shot the pm?
And latency and server stress does have an effect. The xbox is forced to do extra client side prediction and interpolation which causes additional cpu stress.
I do believe that you found the CPU issue a few days ago and messaged mib. That's not what i was saying at all. I was saying that you tried to make it sound like you discovered it months ago and were "driven out of town" on TB which is NOT the case. You also just said before, "the problem is caused by latency" which isn't true. "latency can add to CPU stress that then exacerbates the problem" IS true, but that was not your original argument.
This new theory is totally legit and i'm behind it. My only reservation is that we need to make sure there is no designed variability with the rumble signals that get sent to the controller. We would either need a response from 343, or a way to measure the rumble signal directly and not filtered through the game.
I think we should test this with another weapon too. If we can find one where the rumble signal length is longer but a consistent length of time, it would be interesting to see if the variability stays the same to indicate variable input delay as the culprit, or if it get exacerbated over a longer length.
very interesting and damning stuff as long as we can confirm that the rumble signal is consistent.
I concur. I suggested the sentinel beam. I'm thinking having bottomless clip enabled will keep the rumble completely consistent. I can take the clip and put it in the video editor to measure the 360 time. The only problem is the test will be limited to custom games and warzone. Still workable though. I did the test before measuring terminal times but did not use the circuit technique MIB is using.

As for a shorter burst that is not user controlled. I was thinking the h2 br. It's possible they kept the original rumble signals.
yeah, hoping that Unyshek replies to that TB thread. He should be able to have someone look at the code and see if the game ever sends variable rumble signals to the controller.
Glad to see that all the hard work you guys are doing is being noticed and getting traction accros communities, hopefully you can pinpoint the source of all this and 343 can do something about it.
Apoll0 wrote:
IWI IUI wrote:
LUKEPOWA wrote:
If it's a CPU issue, would they even be able to fix it? With the game winding down with updates, it doesn't seem hopeful if it requires some type of overhaul.
An overloaded server can cause the tick rates to drop which forces the xbox to do extra prediction steps. The net code could also be forcing excessive game state checks which bogs it down. I use the word excessive here lightly though. With thrust enabling an immediate change in direction check rate might be necessary.
Computational tasks can also be prioritized and scheduled in the game engine. So for example if the controller input calculation needs to "wait its turn" behind a physics calculation, it could introduce lag to the input. There are a million ways all this stuff can be handled and im sure the Xbox one's anemic hardware doesn't help.
No excuses though, Halo seems to be the only game that really has this problem on the platform
Apoll0 wrote:
No doubt. On an authoritative server physics tasks are run by the server. A lowered tick rate will force the xbox to do physics calculations for other players through interpolation. It bogs down the CPU and causes varying frames. What you just mentioned is exactly what happens.
Here is a link to a dev talking about it. Scroll down to Matso's post.

link
Aykibaby wrote:
Woooow. Guys in the Teambeyond-topic Unyshek joined in! http://teambeyond.net/forum/topic/15295-i-have-devised-a-way-to-demonstrate-heavy-aim/page-4

Finally, we made them listen... actually it was more like you, the hard working guys with all the testing, did it. Congraz to all of you @OP & @MIb2347 & to all the testers etc pp. This feels so satisfying.

Now I hope they keep informing us. Transparency would be awesome.

Thx Unyshek!
First off, nice work @OP & @MIb2347!
Maybe we can finally get this resolved, or at the very least dare hope a Halo game will NEVER again release with a broken aiming system.

Secondly, really @Unyshek ?!
Now you (as a figure of 343i) "bother" to give this topic ANY kind of response, and you won't even do it on the official Halo forum (Waypoint) ?!
This topic has been alive for months (since 1/8/2017)!!
People has REPEATEDLY asked for 343i to chime in on the topic, and you have the audacity to respond ON A DIFFERENT FORUM!!
Why does 343i even bother with having Waypoint if you won't respond to the people using it?

343i seriously needs a new PR-manager, and new PR-policies. This kind of communication with your customers/consumers (you know, the people that make sure you get payed at the end of the day!) is going to have serious backlash.
Since Mib made the thread with the amazing post on Beyond I find it most fitting for Unyshek to respond there. I am just glad we finally got a real response to this matter. Wherever it is :)
Just fix the freakin aiming 343i.
IWI IUI wrote:
LUKEPOWA wrote:
If it's a CPU issue, would they even be able to fix it? With the game winding down with updates, it doesn't seem hopeful if it requires some type of overhaul.
An overloaded server can cause the tick rates to drop which forces the xbox to do extra prediction steps. The net code could also be forcing excessive game state checks which bogs it down. I use the word excessive here lightly though. With thrust enabling an immediate change in direction check rate might be necessary.
Something they could do easily and quickly is lower the overall game resolution, or max from the dynamic res feature. Lower resolution results in less polygons being rendered by the CPU. That obviously wouldn't be "pretty" but I'd bet you a million to one we'd all rather have consistent aiming rather than extra polygons at this point.
Is it possible 343 could implement DX12 support to the console version of the game, and would that help reduce the cpu stress? I know some xbox one games like SW:Battlefront use DX12.
Jabbo wrote:
IWI IUI wrote:
LUKEPOWA wrote:
If it's a CPU issue, would they even be able to fix it? With the game winding down with updates, it doesn't seem hopeful if it requires some type of overhaul.
An overloaded server can cause the tick rates to drop which forces the xbox to do extra prediction steps. The net code could also be forcing excessive game state checks which bogs it down. I use the word excessive here lightly though. With thrust enabling an immediate change in direction check rate might be necessary.
Something they could do easily and quickly is lower the overall game resolution, or max from the dynamic res feature. Lower resolution results in less polygons being rendered by the CPU. That obviously wouldn't be "pretty" but I'd bet you a million to one we'd all rather have consistent aiming rather than extra polygons at this point.
I could be wrong here, but according to my knowledge, resolution has no impact on the utilization of the CPU, since the CPU is tasked with telling the GPU what objects it has to render, and the GPU is responsible for determining the amount of polygons in which to render each object. The CPU would only be under less stress if the actual number of objects on the maps were decreased, or the framerate cap was lowered from 60. If I am wrong though don't hesistate to let me know haha
Jabbo wrote:
IWI IUI wrote:
LUKEPOWA wrote:
If it's a CPU issue, would they even be able to fix it? With the game winding down with updates, it doesn't seem hopeful if it requires some type of overhaul.
An overloaded server can cause the tick rates to drop which forces the xbox to do extra prediction steps. The net code could also be forcing excessive game state checks which bogs it down. I use the word excessive here lightly though. With thrust enabling an immediate change in direction check rate might be necessary.
Something they could do easily and quickly is lower the overall game resolution, or max from the dynamic res feature. Lower resolution results in less polygons being rendered by the CPU. That obviously wouldn't be "pretty" but I'd bet you a million to one we'd all rather have consistent aiming rather than extra polygons at this point.
I could be wrong here, but according to my knowledge, resolution has no impact on the utilization of the CPU, since the CPU is tasked with telling the GPU what objects it has to render, and the GPU is responsible for determining the amount of polygons in which to render each object. The CPU would only be under less stress if the actual number of objects on the maps were decreased. Thats why forge maps with tons of objects feel especially crappy to aim on. If I am wrong though don't hesistate to let me know haha
No you aren't wrong but resolution does play a little part of cpu processing as well. I guess a less appealing alternative would be to simply reduce some non-essential objects in the game scenery per maps or reduce the FOV (field of view) so less objects are on the screen at once. Or maybe have a "barebones" option where no custom skins/armor variants are loaded and everyone is forced to default. Not sure if that impacts this but I'd certainly like for 343 to try all possible avenues.
FINALLY !!!

I hope that 343 will fix heavy movements and hit registration too.....
If only 343 cared as much/was as competent as the best of the community.
I don't know if this correlates with heavy aiming but I will post it here and not at a new thread:

Some time ago I got the feeling (and today it happened more often) that sometimes I don't get aim-assist/ red reticle in gunfights.
I don't have anything to prove it, but I will try to explain it.
For example: Today in a 1v1 I couldn't get my reticle on the enemy and it was flying over him from left to right and back. The distance was maybe like 10 meters (not close enough for melee but definitely close enough to not use zoom). While I was trying to shoot him, I noticed my reticle wouldn't turn red, when it was on his body.
I'm not a pro player, but I think I'm not too bad and can say that I'm able to aim properly. Also I'm using a quite low sensitivity of 4.

If this has nothing to do with heavy aiming, I will maybe open a new thread.
Collet005 wrote:
FINALLY !!!

I hope that 343 will fix heavy movements and hit registration too.....
I think the heavy movement and heavy aim is of the same or caused by the same factor. IMO once or if they figure what might be causing this there will be much better over all performance and things in general will be more spot on.
Starboyyyy wrote:
I don't know if this correlates with heavy aiming but I will post it here and not at a new thread:

Some time ago I got the feeling (and today it happened more often) that sometimes I don't get aim-assist/ red reticle in gunfights.
I don't have anything to prove it, but I will try to explain it.
For example: Today in a 1v1 I couldn't get my reticle on the enemy and it was flying over him from left to right and back. The distance was maybe like 10 meters (not close enough for melee but definitely close enough to not use zoom). While I was trying to shoot him, I noticed my reticle wouldn't turn red, when it was on his body.
I'm not a pro player, but I think I'm not too bad and can say that I'm able to aim properly. Also I'm using a quite low sensitivity of 4.

If this has nothing to do with heavy aiming, I will maybe open a new thread.
Starboy I personally think your experiencing heavy aim or what is being discussed here. Heavy aim seems to happen in stages of severity. Your 1v1 sounds like you were experiencing the windmill effect when your aim becomes uncontrollable. That is part of the heavy aim effect. I think many players if not all players experience some sort of aim issues.. Its just that many cant explain it or don't think there is anything wrong and just right it off as being a bad game. Hope that helps!
So what happens now? We wait for 343 to come up with a solution?
The Limpo wrote:
Aykibaby wrote:
Woooow. Guys in the Teambeyond-topic Unyshek joined in! http://teambeyond.net/forum/topic/15295-i-have-devised-a-way-to-demonstrate-heavy-aim/page-4

Finally, we made them listen... actually it was more like you, the hard working guys with all the testing, did it. Congraz to all of you @OP & @MIb2347 & to all the testers etc pp. This feels so satisfying.

Now I hope they keep informing us. Transparency would be awesome.

Thx Unyshek!
First off, nice work @OP & @MIb2347!
Maybe we can finally get this resolved, or at the very least dare hope a Halo game will NEVER again release with a broken aiming system.

Secondly, really @Unyshek ?!
Now you (as a figure of 343i) "bother" to give this topic ANY kind of response, and you won't even do it on the official Halo forum (Waypoint) ?!
This topic has been alive for months (since 1/8/2017)!!
People has REPEATEDLY asked for 343i to chime in on the topic, and you have the audacity to respond ON A DIFFERENT FORUM!!
Why does 343i even bother with having Waypoint if you won't respond to the people using it?

343i seriously needs a new PR-manager, and new PR-policies. This kind of communication with your customers/consumers (you know, the people that make sure you get payed at the end of the day!) is going to have serious backlash.
Since Mib made the thread with the amazing post on Beyond I find it most fitting for Unyshek to respond there. I am just glad we finally got a real response to this matter. Wherever it is :)
That is true, Mib did make the topic there and I understand why Unyshek replied there instead of here, it made sense this time but I think LovelyJoey21605 has a point that it seems like more times then not 343I either 1) just generally always replies on other sites to topics in general way more then topics on Waypoint​, thy OFFICIAL Halo site and/or 2) News, press releases and so forth seem to always be announced everywhere else first (Team Beyond, Reddit, Twitter etc) over Waypoint, again the OFFICIAL Halo site.

I sometimes wonder why I even bother coming on Waypoint for anything when I always and I mean ALWAYS hear things first somewhere else and questions/topics seem to get a lot more "love" per say everywhere else aswell. It's sad really...
  1. 1
  2. ...
  3. 30
  4. 31
  5. 32
  6. 33
  7. 34
  8. ...
  9. 41