CSGO Sticky thread

Fulcrum29 said:
The only thing your rank serves in CSGO is to get access to the limited CS2 test, that and having Prime. In CS2 you don't have any rank, you start at the bottom and have to win ten rounds to get a score (or rank). I don't know how it really works, but some call it a calibration mode which has been open to scrutiny. Some people have pointed out a problem with this calibration process. They say that weak players could have ten good games and then get a high rank (or score). This could create the situation we all know as ELO hell, which already exists with alt and boosted players.

I have seen on social media some old high ranking players complain about their low scores. Valve haven't said anything, but it seems to be calculated by win percentage rather than ELO (or whatever it is called nowadays).

I don't care. I will play my ten games and move on. I am already 3/10 down, no losses. I was hoping that you and @Neoprod would catch up.

Then, as @TribbleZA have it, once the standard competitive mode returns to active duty, then players will achieve rank, or score, per map. As it is, some players only play one or two maps at which they will excel, but won't touch any other map because they are bad at it. This is how Valve plans on balancing this issue.

The new Premier mode is made to rival other services like FACEIT, but I doubt that it will have any impact on their player base.
Click to expand...
Ah ok that makes sense.

I will try my best to catch up over the weekend... hopefully get some of my aim back.
 
I agree with this player:


I had one game that I surely had noticeable lag input, but also in another match that was not as disruptive. It is also why I quoted Valve's update addressing lag compensation to improve latency experienced with low-latency players.

I am curious whether this has to do with how Valve is approaching the tick rate, note that the tick rate is now also set by Valve. I can't see how Valve would have 'reinvented' their net code in its entirety since it worked well with CSGO. Currently, they are doing it very similarly to how Respawn is handling Apex Legends which also use Source 2.

Something is sure up with the input lag in CS2. The main problem is the delayed kills/deaths, similar to backtrack hacks. I think that the YouTuber is up to something. Valve, please Fix.
 
Apex doesn't use Source 2, that engine is modified so heavily it's basically just Source in name.
 
Soul Assassin said:
Apex doesn't use Source 2, that engine is modified so heavily it's basically just Source in name.
Click to expand...

Whether it is Source or Source (2), it is still the same base with modularity.

EDIT: But, yes, I have incorrectly used Source 2 as Apex's game engine. Though they do use Source, they have branched Source and extensively developed their own modules. I haven't seen any released datamining on their Source code so I don't know their dependencies.
 
Last edited:
Fulcrum29 said:
I agree with this player:


I had one game that I surely had noticeable lag input, but also in another match that was not as disruptive. It is also why I quoted Valve's update addressing lag compensation to improve latency experienced with low-latency players.

I am curious whether this has to do with how Valve is approaching the tick rate, note that the tick rate is now also set by Valve. I can't see how Valve would have 'reinvented' their net code in its entirety since it worked well with CSGO. Currently, they are doing it very similarly to how Respawn is handling Apex Legends which also use Source 2.

Something is sure up with the input lag in CS2. The main problem is the delayed kills/deaths, similar to backtrack hacks. I think that the YouTuber is up to something. Valve, please Fix.
Click to expand...
Also this:

https://www.reddit.com/r/GlobalOffensive/comments/16kgxg0
 
Soul Assassin said:
Also this:

https://www.reddit.com/r/GlobalOffensive/comments/16kgxg0
Click to expand...

This part here he the post author added:

EDIT: i belive i have caught a different bug here with showimpacts. it looks like the client shots always start from your uninterpolated position, which is always further ahead. funnily enough this precisely counteracts the delay created here so it is always the full 1 tick distance.
Click to expand...

That is exactly how I experienced it. I only really started to notice it in Banana, being used to player engagement positions in that area within a set time. In Nuke it becomes very notable because you play in rooms that you have to slice (peek and so on) to enter.

I was under the impression that it was due to my own latency because Cool Ideas had issues at that time, but now that I see more players bring it to attention and Valve making an change to low-latency timings it must be something considered game breaking.
 
To quote SteamDB's creator:

https://twitter.com/x/status/1702277004656050220

I guess that CSGO will remain alive, and possibly become the 'community' edition.

https://twitter.com/x/status/1702629078916887015

I will try the above. Those who have experimented with it didn't perceive any UX improvements, but alledegly it improves the 1% lows.

https://twitter.com/x/status/1702729374238167166

VAC Live is now enabled? Good.

Just one problem with that report though, it is a live ban due to report spam not because a cheat was automatically detected.
 
Fulcrum29 said:
To quote SteamDB's creator:

https://twitter.com/x/status/1702277004656050220

I guess that CSGO will remain alive, and possibly become the 'community' edition.

https://twitter.com/x/status/1702629078916887015

I will try the above. Those who have experimented with it didn't perceive any UX improvements, but alledegly it improves the 1% lows.

https://twitter.com/x/status/1702729374238167166

VAC Live is now enabled? Good.

Just one problem with that report though, it is a live ban due to report spam not because a cheat was automatically detected.
Click to expand...
But at least we now know that reporting is effective and it doesn't have to wait for overwatch.
 
TribbleZA said:
But at least we now know that reporting is effective and it doesn't have to wait for overwatch.
Click to expand...

From how it looks now VAC Live can be abused, but yes, it does seem to work at least. Overwatch had its own problems and was also abused, but Overwatch haven't been all too reliably used by Valve either. They disabled it at times as a VAC mechanic and since the CS2 Beta released, completely disabled it in CSGO.

It's plausible to speculate that Valve utilised Overwatch as a VACNet training tool. The question is, are their data any good. Models need to be trained, and that data needs to be unblemished.

Now with lag issues I can only see such data collected in CS2 being discarded. It is not known whether there is any active VACNet training with CS2. Generally, I think this is why Valve has been hardcoding conditional values into CS2, in order to have control over their telemetry, to make their game and UX more reliable.

I want to see Valve produce positive things. Their GCSD, or VACNet as known to us, is an additional anti-cheat measure, and the aim is not to eliminate cheating, but to reduce cheating. The last thing I want is VAC Live used to harm another player’s reputation because some players on the team tilted. It is a daring decision, kick or report.
 
3kliksphilip's attempt to put the matter to rest...


Placebo? I have never experienced this 'backtrack' in CSGO. Anyhow, it is best then to watch Valve's update notes. I haven't played the game since the lag compensation patch.
 
MrMaxim has now also released a video:


In my opinion, sub-tick is not optimised. Simply, sub-tick does not include all variables in its calculation. I won't know, but this is what I am deducing. The question I have is why it is only picked up now and not with the earlier limited test release? Either something changed, or Valve is aware. They have been making updates to address this issue, but are not engaging with the community on this issue.

I am sure that the issue will be ironed out in time. I have experienced this issue, but not in all matches.

I see the pros are making moves to call on 128 tick servers, but is avoiding that it has to do with this 'sub-tick' issue. This being said, there is a chance that this issue may have nothing to do with sub-tick, but sub-tick is what is new to the game.

128 tick is the way to go. Valve is cashing it with their marketplace, and could put that money to good 128 tick public servers.
 
https://twitter.com/x/status/1703538164105658635

Tick-rate rant

Disclaimer: This is about tick-rate, not sub-tick. If Valve can make sub-tick better than 128t, that would make everyone happy. I hope they do.

I've joined surf servers many times which are using 64t, then go into an aim map forgetting that it loads up the same 64t from the previous server. That is offline, however in a few moments I see - oh damn, time to restart the game. Need to load it up in 128t. I literally get bad hit reg and my strafing/counter-strafing doesn't match up to my shots like they do in pro games. That is in offline playing in a map where you get 1000FPS, no performance issues.

In the end it's easy to notice you are playing 64 tick when you are strafing, shooting, bhopping, throwing nades, getting bad hitreg. Stuff like that simply doesn't happen as much as in 128t, it's more consistent and fits the gameplay better. The argument, oh well why stop at 128t, lets go for 256t is just a bad take. Of course it would be an improvement, but an unnecessary one. I believe 128t is already at such low latency, that it's extremely rare to have an impact on gameplay events. The advantage of also having a tick-rate is consistency. That's what makes 128t as good as it is, because the delay is small enough and so consistent, that getting bad hitreg would be very rare, the game doesn't feel off, strafing and shooting feels better - because that delay is small enough for a person, even a skilled one, to have no effect.

We have to realize that we are human, not robots. If you push the tick-rate high enough it might be possible to notice a difference, but the impact is minimal. My point is that 128t is already high enough to not have issues in your gameplay. 64t on the other hand, especially for high level gameplay is not good enough. The feedback you get from the game especially when you are good at it - it's trash. You feel it. As launders pointed out, CS is all about getting a grip of your game and playing on feeling. That's why we are using 128t today in every server.

I think it's a similar issue to monitors, there's some which have 500Hz, but people are still using 240/360Hz or even 144Hz. While I am switching between 240/360Hz depending on the tournament, I would never go back to 60Hz. No one would. If you go high enough, eventually it doesn't matter. You just need to find the sweet spot.

This is definitely about finding the balance and seeing how far it is necessary to push the limit. I think by now we have enough data that 64t is bad and 128t is good enough.

I hope Valve can make sub-tick as consistent or even more consistent than nowadays 128t servers.
Click to expand...

Lol,

https://twitter.com/x/status/1703542650261885341

easily said without considering this game or any games netcode. There are many online shooters which have less than 64 ticks, but they are by design made that way.
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X