Apex Legends input lag changes from day to day

[howiec]

still, why do I get better input lag, hit detection and aim assist on different resolutions on different days? Makes zero sense.
[...]
So why does Apex's servers need to know my client's resolution? I've noticed that the game's movement is controlled by the servers too.

Not sure why you guys are ignoring the underlying reasons I previously mentioned that explain the day-to-day or map-to-map or settings-to-settings impact on the end-result-effective (or perceived) aim/sensitivity feel and consistency.

Sure, the game state is managed primarily by the server and any disruption or variance in network or server performance will manifest in all kinds of things like stutter, poor hit-reg, "time warp", etc., perceived by the player (client side) but these network and/or server issues are not the fundamental cause of the issue above and can only add to the problem via a different mechanism.

To put it simply:

• I previously mentioned the major issues with the game/engine that would manifest even with perfect server & network performance
• Poor netcode, network, and/or server performance will add variance to the baseline problem above

If you want an explanation as to why the effective (or perceived) aim/sens feel is so inconsistent, ask the devs (with the requisite expertise) to explain some basic things such as:

• How exactly are the effective aim/sens + angle adjustments calculated & how are they decoupled as much as possible from inconsistent reference variables / constants / functions?
• Why does effective/perceived aim/sens differ so drastically between all maps such as firing range vs WE vs BM, etc.?
• Why does every single mouse & controller sensitivity & setting/value all affect each other (why are they not independent)?
• Why does supposedly inactive rendering functions such as DVS still drastically affect aim/sens?
• Why does even running another process with little-to-no CPU utilization drastically affect Apex's effective sens/aim even on a high-end, optimized system (13900K + 4090 FE + DDR5 7400)?

Clearly, there are fundamental flaws that are not addressed and probably never will be.

Respawn/EA has a proven expertise and history of screwing up even the simplest patch or update even when it's just for skins/events... Don't expect them to be able to handle an actual technical task.

[aress]

I am suffering from the same issue. Apex was running smoothly for two days, but then it went back to having input lag. I went from D4 to D1 in one day, but now I'm losing every single RP because I can't hit anything. Additionally, the audio has become problematic again. Some days, the game runs amazingly smooth, and other days, it lags for weeks. I have a high-end build and have had this issue for at least two years. I've written many tickets to Respawn, but unfortunately, I have never received a solution.

[howiec]

still, why do I get better input lag, hit detection and aim assist on different resolutions on different days? Makes zero sense.
[...]
So why does Apex's servers need to know my client's resolution? I've noticed that the game's movement is controlled by the servers too.

Not sure why you guys are ignoring the underlying reasons I previously mentioned that explain the day-to-day or map-to-map or settings-to-settings impact on the end-result-effective (or perceived) aim/sensitivity feel and consistency.

Sure, the game state is managed primarily by the server and any disruption or variance in network or server performance will manifest in all kinds of things like stutter, poor hit-reg, "time warp", etc., perceived by the player (client side) but these network and/or server issues are not the fundamental cause of the issue above and can only add to the problem via a different mechanism.

To put it simply:

• I previously mentioned the major issues with the game/engine that would manifest even with perfect server & network performance
• Poor netcode, network, and/or server performance will add variance to the baseline problem above

If you want an explanation as to why the effective (or perceived) aim/sens feel is so inconsistent, ask the devs (with the requisite expertise) to explain some basic things such as:

• How exactly are the effective aim/sens + angle adjustments calculated & how are they decoupled as much as possible from inconsistent reference variables / constants / functions?
• Why does effective/perceived aim/sens differ so drastically between all maps such as firing range vs WE vs BM, etc.?
• Why does every single mouse & controller sensitivity & setting/value all affect each other (why are they not independent)?
• Why does supposedly inactive rendering functions such as DVS still drastically affect aim/sens?
• Why does even running another process with little-to-no CPU utilization drastically affect Apex's effective sens/aim even on a high-end, optimized system (13900K + 4090 FE + DDR5 7400)?

Clearly, there are fundamental flaws that are not addressed and probably never will be.

Respawn/EA has a proven expertise and history of screwing up even the simplest patch or update even when it's just for skins/events... Don't expect them to be able to handle an actual technical task.

Funny, today's Apex patch notes is evidence that explains part of the problem as to why your effective/perceived aim/sens is so heavily influenced by internal rendering calculations/pipeline and hence literally everything else (due performance impact).

I haven't tested how much today's patch/change (only took the devs almost 5.5 years to do so...) improves aim/sens consistency but at least it's 1 step in the right direction:

"Fixed bug that caused high polling rate mice to occasionally encounter high frame stutters. High polling rate mice users should now see little to no frame stutters from mouse input.
- Input and window processing now occurs on its own thread instead of the render thread"

[Raphaeangelo]

still, why do I get better input lag, hit detection and aim assist on different resolutions on different days? Makes zero sense.
[...]
So why does Apex's servers need to know my client's resolution? I've noticed that the game's movement is controlled by the servers too.

Not sure why you guys are ignoring the underlying reasons I previously mentioned that explain the day-to-day or map-to-map or settings-to-settings impact on the end-result-effective (or perceived) aim/sensitivity feel and consistency.

Sure, the game state is managed primarily by the server and any disruption or variance in network or server performance will manifest in all kinds of things like stutter, poor hit-reg, "time warp", etc., perceived by the player (client side) but these network and/or server issues are not the fundamental cause of the issue above and can only add to the problem via a different mechanism.

To put it simply:

• I previously mentioned the major issues with the game/engine that would manifest even with perfect server & network performance
• Poor netcode, network, and/or server performance will add variance to the baseline problem above

If you want an explanation as to why the effective (or perceived) aim/sens feel is so inconsistent, ask the devs (with the requisite expertise) to explain some basic things such as:

• How exactly are the effective aim/sens + angle adjustments calculated & how are they decoupled as much as possible from inconsistent reference variables / constants / functions?
• Why does effective/perceived aim/sens differ so drastically between all maps such as firing range vs WE vs BM, etc.?
• Why does every single mouse & controller sensitivity & setting/value all affect each other (why are they not independent)?
• Why does supposedly inactive rendering functions such as DVS still drastically affect aim/sens?
• Why does even running another process with little-to-no CPU utilization drastically affect Apex's effective sens/aim even on a high-end, optimized system (13900K + 4090 FE + DDR5 7400)?

Clearly, there are fundamental flaws that are not addressed and probably never will be.

Respawn/EA has a proven expertise and history of screwing up even the simplest patch or update even when it's just for skins/events... Don't expect them to be able to handle an actual technical task.

When you say things like

day-to-day or map-to-map or settings-to-settings impact on the end-result-effective (or perceived) aim/sensitivity feel and consistency.

makes me think you aren't listening. Things change right in front of me in the firing range (same exact location and time in the firing range), same PC settings, close to same network (because it's within seconds of each test). Of course a game that has a server component will have inconsistencies because network and server performance that are outside our control. I'm talking about inconsistencies that normally shouldn't be there. I'm in the firing range so "map-to-map" doesn't make sense. It's the same map. Also, I test my accuracy on each resolution within minutes of each other. Each res yields wildly different accuracy (+ - 20%) and before you say "yeah each res puts different amounts of stress on your PC" it's not that easy. the res with the "best" accuracy one day will not be the "best" the next. This tells me the res-to-accuracy has nothing to do with my machine and the fact that the best accuracy/res changes day-to-day tells me I'm not just better on one specific res. Between each test (and day) my PC settings haven't changed so "settings-to-settings" doesn't apply here either.

[Sandy]

DISCLAIMER: This is only speculation based off of tests and observations but I think I have enough evidence to support these claims.

Background
I've been playing Apex Legends for about three years and I've constantly been chasing a "settings ghost" I've tried every settings guide you can image and they work...until they don't. This is how my experience goes. I play Apex for a few hours and something feels "off" (game feels heavy, aim is bad and it feels like I'm running through mud) so I Google around till I find some Apex setting I haven't tried. I give it a try and test. Wow everything feels great! I've finally solved the input lag problem in Apex!!! Till the next time I login and it's back to playing like garbage. I repeat the steps of Googling my issues and making a change only to have temp results. What gives??? This is seriously maddening. How am I supposed to improve in a game that's constantly changing? I thought it was hopeless till I came across these videos.

Showcasing how Input response/aim assist doesn't act the same on resolution that you used before.

phpBB [video]

TLDR this tester is in the firing range and showing how aim and input lag change based on what resolution they are on. You might be thinking like me "ok maybe the lower resolution is easier on the CPU thus resulting in better performance" but this is where it gets interesting. He tries this on different days. One day the lower resolution provides better results (better results = better aim and movement) the next day the lower resolution results in higher input lag and worse results. The game's input lag varies based on resolutions, which change from day to day!? This seems unbelievable, so I had to test it out.

How I tested this theory.
I go into the firing range, I keep all my settings the same (low everything) and only change the resolution, making note of my accuracy percentage (you can turn this accuracy metric on in the range settings). Each resolution results in different accuracy. I choose the resolution with the best accuracy and use it for the day. When I play Apex for that day, everything feels really good - I land more shots and do more damage. But the next time I log into Apex (days later), it feels "off" again. I repeat this process, and wouldn't you know it, I get better results on a different resolution! Sometimes it's better on lower resolutions, and sometimes it's better on higher resolutions.

How can this be happening? I have a theory (this is the speculation part): EA is randomly making the game worse for retention reasons. I play longer/more when I'm doing badly in the game (gamblers addiction), than I do when I'm playing well. When I do well in the game, I play less; "That was a good session, I'm jumping off for the day." I know that sounds wild, but it's the only explanation for this strange behavior with the resolution. Please give it a try and report back your findings.

My Specs
Intel i5 13700K (water cooled not overheating issues)
MSI Tomahawk z790
DDR4 cas 13 3600 b-die
RTX 3080 ti
MSI 360hz gaming monitor
PS dual sense 250hz polling rate controller (Tried OC to 1000hz made no difference in apex inconsistency problem)

My current settings (I've tried everything. No difference in apex inconsistency problem)
Stock clocks, high performance power plan, VSYNC/GSYNC disabled, 360hz refresh rate, 120fps cap, 250hz controller polling rate, disabled hyperthreading, disabled e-cores, disabled c-states, disabled turbo boost

Disabling NVIDIA thread optimization (must choose to completely turn it off) can eliminate all input delay (global delay) and must restart the system
Refer to this
viewtopic.php?t=12580

[howiec]

When you say things like

day-to-day or map-to-map or settings-to-settings impact on the end-result-effective (or perceived) aim/sensitivity feel and consistency.

makes me think you aren't listening. Things change right in front of me in the firing range (same exact location and time in the firing range), same PC settings, close to same network (because it's within seconds of each test). Of course a game that has a server component will have inconsistencies because network and server performance that are outside our control. I'm talking about inconsistencies that normally shouldn't be there. I'm in the firing range so "map-to-map" doesn't make sense. It's the same map. Also, I test my accuracy on each resolution within minutes of each other. Each res yields wildly different accuracy (+ - 20%) and before you say "yeah each res puts different amounts of stress on your PC" it's not that easy. the res with the "best" accuracy one day will not be the "best" the next. This tells me the res-to-accuracy has nothing to do with my machine and the fact that the best accuracy/res changes day-to-day tells me I'm not just better on one specific res. Between each test (and day) my PC settings haven't changed so "settings-to-settings" doesn't apply here either.

Dude, I don't know how much clearer I can be...
I'll try to break it down as simply and concisely as possible:

• My explanations do NOT contradict what you've described. They explain the underlying issue:
  • Apex and/or the engine in general is poorly/coded or designed with regards to aim/sens consistency across basic, expected workloads.
  • The effective aim/sens is extremely sensitive to any change in system performance.
• What I mean by map-to-map, or day-to-day, or settings-to-settings changes that affect the effective aim/sens is the following:
  • If you change NOTHING, your effective aim/sens is DIFFERENT between maps such as the firing range vs WE or BM, etc.
  • If you change NOTHING, your effective aim/sens is DIFFERENT between days because if you reboot your PC or run different services/apps in the background or if even the temperature changes such that your GPU isn't boosting to the same exact core frequencies, etc., all of these changes are also slight performance changes that Apex is sensitive to.
  • If you change ONLY the render resolution, your effective aim/sens is different because your CPU & GPU loads are different, which obviously affects performance, which in turn affects effective aim/sens in Apex because of how poorly designed it is.
• You are only looking at 1 factor (resolution) which affects workload/performance.
• There are many other factors that also affect performance IN ADDITION to changing just the resolution, which explains why resolution A is better on some days, while resolution B is better on other days or even between maps.... there are multiple factors that all add up to produce the end result / effective aim/sens which is what makes this a major fundamental problem.

As I mentioned in a previous post, additional evidence of poor Apex code can be found directly from their own patch notes where they moved input processing to its own thread vs having it on the same thread as rendering.
This glaring design flaw should've been obvious since the very beginning...

[howiec]

Disabling NVIDIA thread optimization (must choose to completely turn it off) can eliminate all input delay (global delay) and must restart the system
Refer to this
viewtopic.php?t=12580

It certainly makes a difference but does not resolve the effective aim/sens consistency when zero settings have changed.
I.e., if you don't change any setting, your effective aim/sens is different between Firing Range vs World's Edge vs Broken Moon.

[Raphaeangelo]

Disabling NVIDIA thread optimization (must choose to completely turn it off) can eliminate all input delay (global delay) and must restart the system
Refer to this
viewtopic.php?t=12580

I have NVIDIA thread optimization set to disabled and I agree it helps a lot! Thanks.

It certainly makes a difference but does not resolve the effective aim/sens consistency when zero settings have changed.
I.e., if you don't change any setting, your effective aim/sens is different between Firing Range vs World's Edge vs Broken Moon.

100% agree

[Sandy]

Disabling NVIDIA thread optimization (must choose to completely turn it off) can eliminate all input delay (global delay) and must restart the system
Refer to this
viewtopic.php?t=12580

It certainly makes a difference but does not resolve the effective aim/sens consistency when zero settings have changed.
I.e., if you don't change any setting, your effective aim/sens is different between Firing Range vs World's Edge vs Broken Moon.

Yes, I also found the hit registration problem! As far as simply solving input delay is concerned, many people don’t know that disabling NVIDIA thread optimization (globally disabling) and restarting the computer can eliminate input delay. Now my input delay problem is solved. Next, I will continue to study the hit registration problem. I suspect the problem is still in the NVIDIA control panel. I am currently using Windows 7. I plan to switch to Win10 to test it.

[Raphaeangelo]

My explanations do NOT contradict what you've described. They explain the underlying issue:

Apex and/or the engine in general is poorly/coded or designed with regards to aim/sens consistency across basic, expected workloads.

Agreed.

The effective aim/sen is extremely sensitive to any change in system performance.

Agreed but I'm not changing anything. In fact I set all the bios options individually to static amounts rather than "auto" for this very reason.

What I mean by map-to-map, or day-to-day, or settings-to-settings changes that affect the effective aim/sens is the following:
If you change NOTHING, your effective aim/sens is DIFFERENT between maps such as the firing range vs WE or BM, etc.

All my tests are in the same firing range at the same exact spot.

If you change NOTHING, your effective aim/sens is DIFFERENT between days because if you reboot your PC or run different services/apps in the background or if even the temperature changes such that your GPU isn't boosting to the same exact core frequencies, etc., all of these changes are also slight performance changes that Apex is sensitive to.

Agreed but like I said I meticulously hard set these settings to avoid fluctuations. CPU and GPU power plus timings.

If you change ONLY the render resolution, your effective aim/sens is different because your CPU & GPU loads are different, which obviously affects performance, which in turn affects effective aim/sens in Apex because of how poorly designed it is.

Lets say the res with the best aim/sen is 1280 x 720 This makes sense because the lower res means less work load for both the CPU and GPU. You would think "Great set it and forget it" but the interesting thing is the very next day 1280 x 720 has horrible aim/sens and 1920 x 1080 has good res/sen. This doesn't make sense. I need to find a different res EVERY DAY because EVERY DAY the best res changes. This make zero sense because NOTHING on my PC is changing.

You are only looking at 1 factor (resolution) which affects workload/performance.

Read above.

There are many other factors that also affect performance IN ADDITION to changing just the resolution, which explains why resolution A is better on some days, while resolution B is better on other days.... there are multiple factors that all add up to produce the end result / effective aim/sens.

Im sure there are but what I'm saying is those factors are not on my end. Like I said before there are two factors outside of my control, network and servers. Since res has NOTHING to do with network lag (if I had network lag or high ping it would be present regardless of res) so it HAS to be something on the server side. My question is WHY? WHY does the server need to know my res? Why is the server responsible for all of my movement? Why does the server handle all of these things that the client should? Because the server handles so much it makes complete sense that the game feels super inconsistent because the servers are super inconsistent and outside my control.