devtrackers.gg
It's okay everyone, I remembered what I googled and found the Stack Overflow page again.
It's okay everyone, I remembered what I googled and found the Stack Overflow page again.
I'm pretty sure the memory leak has been addressed. Last Saturday I had the client up for 5+ hours and was at 1.2gb of memory. In ye olden day I would have been at 2+ by then.
There are many sources of lag, if it's varying from day to day the most likely source is internet lag. Sadly there isn't a ton we can do around this one as the game is server authoritative.
From my experience, I think a likely culprit is the log file being an issue when it gets really large, but we need to investigate the reports more to know for sure. There can also be pauses when a particularly complex game state comes down from the server, both from server processing, internet lag, and client processing.
If you are having issues please submit reports with your log file and dxdiag is a bonus.
It's hard to say without seeing the board, but my guess is at some point your client entered into a bad state for some reason. If you're not getting response windows the best course of action is to exit the client and re-join. This will often fix these issues.
Hey while I have you here, why does Bo3 only pay out double? That means that bo1 is still in aggregate a better deal for your time.
EDIT: I was wrong about this.
It's funny, we're actually pretty sure that at +2/-2 Bo3 is better for climbing than Bo1. It's something we'll watch closely.
So you actually didn't need 2 weeks to fix this huh?
Ya, most of it was written earlier in the week. With everything else, I missed updating the date to be a shorter timeline. I've already upgraded my software to avoid including dates in the future. Beep Boop.
What you see post-match is accurate for your ranking at that time. Milliseconds later someone could, in theory, knock you down if their match finished right after. The profile page updates on a regular cadence, but on the minute's scale.
This is weird though, I've forwarded it for investigation.
I missed top 8.
I'm not actually a WotC employee, but I'm curious if my name tricks it.
Also, WotC, since I know you'll read this...
My name is for sale.
$1
Don't tell me what to do!
I'm the guy from the "Magic Arena saved my marriage" post. Do you have data that could be used to determine normal and abnormal amounts of Nexus of Fate casts per game? Is there a threshold that could be reached that would flag the player for a looping warning, perhaps in the future when the game can be analyzed? This could be a deterrant to ensure that the (evil) deck is played with the goal of reaching a win condition (Teferi mill, Chromium damage, etc).
Plausible for sure.
Test post please ignore
Is it true the guy who looped Nexus infinitely against Shahar Shenhar was only banned for 2 hours?
Correct, it was an initial ban with an e-mail sent to them explaining why. Repeat offenses would escalate the time, with the potential of the account being permabanned. The first goal is to educate the player to stop the bad behavior. If they don't respond to that, then we escalate.
I bow before our new digital overlords.
Dublin, Ireland here.
I've noticed that the in-game mouse cursor stays responsive while the rest of the game freezes up. Maybe that'll help to identify the problem?
My GPU is a GTX 1070.
With everyone so far being over the pond, and knowing that the cursor stays responsive my experience say this is an internet issue. We're investigating.
If its happening when you play a card (or when the opponent does) I'd expect it to be an issue with communicating with the servers, or an issue processing the game state from the server and not a video card issue.
Can people having the issue share rough geo-locations?
It's annoying that you start to time out when your opponent combos off. Timed out because a life gain deck was comboing off, and had about 20 triggers to all manually click resolve on.
Definitely not a feature I like.
Just wanted to pop in to say this is something we're actively working on. Specifically around how to address instances when there's a large stack that needs resolving.
Oh man, what a read to start my Monday morning. So many mixed emotions.
11/10
At least you get a nice strong hug before you're smashed into the ground...
We couldn't reproduce this in house. Do you have a twitch clip of the issue or a more complete reproduction of the steps to see the issue?