Is anyone experiencing counter window problem in legends? Like the counter window is closing up fast or times when counter window takes awhile to close after finishing a move
Probably lag. Which we all face. What you see on your screen may be a half or full second slower than what your opponent sees. Check your ping. Low ping = goodness.
This. When did you start noticing it? I wasn't able to log on last night but the night prior everything seemed normal to me. Iirc you're also on playstation correct? EDIT: Also keep in mind you can't close a counter window sooner (unless it's something channeled), but you can in fact extend them. So that may have something to do with it.
Could be a lag, its just that today pvp legends seems kind of off for me. I might need to check connection first. I did see a superman flying towards me while getting up from a ko animation .
When GU47 hits, what you see from another player will be at least twice as responsive. The time it takes for you to see someones actions or movement is, the other player's latency to the server (typically 150ms or less), the time to apply the data in the server (typically 25-75ms), then the latency to yout client (150ms or less). The client does buffer a few frames of data from the server to allow for smooth playback of other players or AI. For the full list of Combat Responsiveness speedups, check the Testing Feedback thread. -dirt
It will probably help US server though I have my doubts regarding EU when we're playing with players from Australia or South America which basically makes EU server something more like RESTOFTHEWORLD server. Nonetheless every bit helps, thanks for these new changes.
The only thing I can't help is the Internet. All my changes of course were not the Internet. So let me know what you think once its Live. Unless your ping is wildly bad (in which case there is no hope), you will notice improvements. Some of the players on PCTest are from EU and they have noticed a difference that is worthy of note (at least I believe so). -dirt
No, no, you're correct there. There's quite a few players who normally have wildly bad connections and have noticed rather large improvements.
Just for clarification, you said counters on Live will also be 50 to 100 ms slower than it is on Test correct?
PCTest runs a lighter load than Live typically, per server process. It can take longer to move the data around the server on Live because of increased player counts. I think it will be marginally worse on Live (it won't be better than PCTest). Let's see what you think when it's out to Live. Before GU47 changes, it was still mentioned that PCTest PvP was so-so. So now that PCTest feels responsive with GU47, it will only translate to Live in a positive way. Basically I am saying combat will be the most responsive it has ever been on Live since launch. -dirt
The counter window timings are unchanged, as far as I know. You will be able to more easily catch someone who opened a window because you will see them do it closer to when the server actually knows the vulnerability is available. With GU47, there are less delays in getting your client critical PvP (combat) information (about other players) so that you can react faster. If someone lets off block, there is still a 300ms hangover for trying to block in the first place, same as on Live. It is somewhat hard to explain other than it's going to feel much better (according to the PCTest feedback). I tried to outline the changes in the Testing Feedback thread, calling out specific changes that will be more responsive. I guess if you have not been a witness of the PCTest changes, then you can't see the fruit of the labor, yet... I have asked some of the best PvPers in the game how it feels on PCTest and they seem to be giving me the thumbs up. -dirt
Yes it is, that's intended. Dirt has been informed of the block cancelling bug that is used after the roll is cancelled however, which is definitely not intended. Late block breaks will still happen, but due to the changes that dirt has made they should hopefully seem far less drastic. The window itself is around 300-350ms long but people were experiencing over one second late delays (that's anecdotal evidence though) which should be lessened due to the increased responsiveness in the build.
