Improbable Icon

Laggy Wizards Demo


#1

Hello everyone,

SpatialOS sounds super interesting and a really cool piece of tech! So I got some time to play with it today, and I fired up the wizards demo by myself. This is between 9:30 - 10:30 pm PST on Monday, and I’ve noticed quite obvious lag, sometimes taking multiple seconds where NPCs just start marching on the same spot, and then suddenly fast forward to catch up. Also experienced rubber banding a few times with my player. I have to say I was surprised to have a laggy experience in a demo where I’d expect you guys to put your best foot forward, not to mention having only a single client active in the game is hardly pushing the limits of what you advertise to be possible.

So my question is: was my experience just a fluke due to maintenance, or geographical location, or something else? Or is it as good as it gets in the current state of the technology?


#3

Hello @taylank,

Welcome to the forums and to SpatialOS. The experience you are describing does sound a bit odd as it should not be that laggy.

Could you potentially provide the following details so we can narrow down the source of the issue?

  • Did you run the cloud version of the demo as per the hands-on tour or did you build the project locally to run it on your own machine?

If you used the cloud version:

  • Where are you geographically located (EU, US West / East, …)?
  • Did you use the exact command-line provided in the documentation or did you potentially modify it?

Best regards,
Duco


#4

I used the cloud demo provided without changing anything, copy pasted the command line. I am located in Vancouver, BC (west coast). Later on I tried deploying to us clusters with the --cluster argument. I got a permissions error for us1 and us2. Us3 finally let me in and it was somewhat better, although it felt like it was having occasional hiccups still. I’ll try again at some point, to see if I was just unlucky with the timing.


#5

Hello again @taylank,

Thank you for the feedback. The permission denied for US1 and US2 are normal as only US3 and EU4 are available for alpha / beta users. The best way to circumvent the try-out of every cluster is to use the --region flag instead. It will start your deployment in the specified region (US / EU…) in a cluster to which you have access.

By going to your console you can check the history of the deployments of your project and by viewing the one where you experienced the serious lag you can see on which cluster it has run (lower-right corner, see screenshot).

If it ran on eu4-prod then that could potentially be the source of lag.

Don’t hesitate to ask any further questions you may have. We are happy to talk about any concerns you might have.

Best regards,
Duco