Hopefully this was the main source that cause inconsistencies from game to game because it sucks to not know what was causing it. Hopefully, this problem can be replicated with many different configurations and not just an outlier.
I really hope so too! We're super concerned with this, and finally having a way to potentially find fixes for this problem is really exciting to me. I would love to finally fix this chapter of "weird netcode."
Also in general, is there any chance yall have fixed tickrate servers as opposed to variable tick(unless the changes in tickrate im seeing are artifacts of a faulty netgraph) ?
Our servers are trying to run at 128-tick (and we have some sub-systems that are forced to run at 128-tick, like physics and movement, no matter what the actual tick rate of the server is). However, if we're unoptimized or have issues, we can't force 128 tick if we hitched and dropped lower to a lower tickrate.
Hopefully this isn't fluctuating significantly in mid-round. We're aware of end-of-round and buy phase fluctuations and we're not too worried about them right now, since there's no combat.
Would this maybe help a bit?
Windows 10 (and up): Windowed Fullscreen should provide the same (or better) performance compared to exclusive Fullscreen.
I believe we were using an incorrect legacy call for Windowed Fullscreen that introduced input latency, and I'm not exactly sure if we've addressed it or not, but it is something we will be fixing at some point.
Wow, thanks for reporting this. We're currently investigating and trying to understand the general issues everyone's been feeling around inconsistency in game feel from game to game.
Alt-tab as a trigger is something that jogs a few gears in our brains in potential causes, including a couple that we were already going to investigate, but helps in attributing a direct cause, and is a potential explanation for how netcode feel can just simply vary game to game. We're still going to investigate other potential causes, but this has me a bit more hopeful in that we're going to be able to find the crux of the issue and address it more quickly, now that we have a way to potentially trigger it.
We're getting our QA to get some reps on this and help us understand what's going on with netcode when someone alt tabs, and were already getting diagnostics for the parts of the system we think are going wrong. We don't have anything concrete to share yet, but we are definitely taki...
Read moreStill waitin on that replay system.
We're working on it, but it won't be ready any time soon. Will still take quite a while, since we're still very early on in developing it, but it's definitely something we want to explore.
Related to HRTF: I’ve noticed a couple of times while spectating, that I’ve heard loud footsteps that the person I’m spectating does not hear.
Have you heard any other reports like this?
An example: While spectating a player on Icebox: he was on A site, atop attacker-side’s “Belt” enemy Brimstone was on defenders side catwalk, but while spectating I could hear noises like brimstone moving on metal surface very near to my ally (clearly shouldn’t be the case). After the round, I asked if my teammate heard these noises, and he said no.
Hi, engineer here. We haven't heard similar reports of this bug yet, so thanks for bringing it up! Always feel free to poke us about what seems like bugs :)
We'll look into this issue. If you're able to provide us with any videos or any consistent way to trigger this, that would be super appreciated, so we can figure out what's causing this and track it down.
Footsteps should be sent by the server, and should be the same for everyone. I'd treat this as a big competitive integrity issue that we need to fix if it's inconsistent for players.
Also, did this start this most recent patch with HRTF, or did this happen before? Any additional info you can think of would be great!
We're taking a look at this right now!
We're taking a look at this right now!
Hey, this is something we're taking a look at.
We're sorry if it seems like we've been quiet but this is definitely an issue we are hard at work on. It's a really challenging problem and we can't commit to anything just yet but know that it is top of mind for us.
Thanks for these videos and analysis, it definitely does help us when y'all look at the game so closely.
I found that in comp, if I'm positive, de-escalate things, and say things like "nice play," or "nice try," I win a lot more games and that inflates my ranking. It's like a team mindset buff to compensate for my bad aim :)
First off, I really appreciate Riot's interaction with the community and the fact that you guys patrol the sub answering questions and informing us, thanks for all your hard work.
My question is, why not just disable the refunding mechanic until it's fixed? Econ is a huge aspect of both CSGO and Valorant, and it doesn't make sense to leave this feature in the game in its current state, especially in competitive.
Is it something that's become so intertwined with other code that there's no way to "switch" it off until fixed?
This has been an issue ever since the mechanic was introduced so I'm surprised it hasn't been disabled yet if there's no fix.
Truth be told, we had this on our radar for a while as well, and weren't able to make time to investigate it until recently.
As for hotfixing/switching it off before the next patch: It's a little bit nuanced; we don't want to introduce a lot of issues into a potential change in the game that we aren't able to test more thoroughly. We're trying to make our releases more stable, so we're afraid a change like that may have some other consequences. That applies to both disabling this and fixing in a hotfix now.
Will there be a hotfix or something for it or do you mean patch 2.1?
We likely won't hotfix this, so a patch like 2.1.
We are aware of this issue, and while it may not land in the current patch, we're trying our best to get it fixed in an upcoming one!
nice, grats! only one more win to lock in that purple dorito! diamond is a real big achievement!
i actually have a similar path myself, got super into competitive games in WoW arena back in the bc/wotlk days and somehow ended up working on and playing valorant :)
This is a feature we accidentally turned on, oops! We're going to be removing this in 1.11, so it will stay exposed up until then.
This is something that we are exploring to improve/adjust visuals of VALORANT. It's not properly tuned yet, so it does go closer to Borderlands that we'd like. If you do have any overall feelings about that, feel free to tell us!
It is a post-process (and a somewhat expensive one at that). Depending on your framerate and GPU, the framerate hit can be minor or significant (we haven't yet tested it internally rigorously).
hey, great news thanks for updating us
so I don't think you went through all the corners/walls of the maps and fixed them right? Something basic was broken about it and you fixed that and everything aligned ?
tldr; a bit of both
the vision cones were an old prototype feature that just ended up shipping in the same way as it was made in prototype years ago. each map has a separate vision cones mesh for perf
but the versions of the maps they were authored for are slightly older in some places and out of sync. it was a bit tedious and annoying to do these small updates, so we also made some changes to make these adjustments easier. we're also having QA help us spot issues where they dont match up with sight lines
thanks for calling this out!
Hey thank you for informing me. Since I'll be the only one seeing your reply =)
i missed the patch notes cutoff, but Ascent's vision cones should be fixed in 1.09 and Bind (and maybe more) in 1.10. the other maps shouldn't be too far behind those :)
sorry to necro an older thread but i am taking a look at this soon!
devtrackers.gg
