Today we have planned to start our retail testing of high tickrate for PC. The end goal is to get enough data to make confirm our findings when it comes to server and client performance, and in the end allow Rental Server Providers to rent out high tickrate servers to players.
We have a special server that we are going to use for the occasion (it's an exact mimic of a regular high end 64 player server). The testing is going to gradually increase the tickrate from the basic 30hz up until we see it has reached it's limit on the current hardware (issues start happening). Once we reach the limit, we will go on to the next set of tests.
First out will be 64 player servers. And since this is a "regular server", we will run 4 instances (IE 64 * 4 player slots). For comparative results, we need to
fill all these slots for about 2 rounds per test. Note: The quicker we can reach full 64x4 for two rounds, the quicker we will get to the next set, finish this testing and be able to release this in the wild!
We will start each set with a 30hz setting to have a baseline as well. The plan for Wednesday (9/16) starts like this:
SET 1 (64x4 players) - Wednesday 9/16:
- 30hz - 64x4 Conquest Large (base game maps) [Base line]
- 45hz - 64x4 Conquest Large (base game maps) [Works fine]
- 60hz - 64x4 Conquest Large (base game maps) [Our top limit, red icons across the board]
- 60hz - 64x2 Conquest Large (base game maps) [left overnight]
SET 2 (32x4 players) - Thursday 9/17?:
- 30hz - 32x4 Conquest Small (base game maps) [Base line]
- 45hz - 32x4 Conquest Small (base game maps) [Probably will work fine]
- 60hz - 32x4 Conquest Small (base game maps) [Probably will work fine]
- 75hz - 32x4 Conquest Small (base game maps) [Probably will work fine]
- 90hz - 32x4 Conquest Small (base game maps) [might reach the limit for 32x4]
- 120hz - 32x4 Conquest Small (base game maps) [improbable for 32x4 players]
- 144hz - 32x4 Conquest Small (base game maps) [improbable for 32x4 players]
SET 3 (20x4 players) - Friday 9/18?:
- 30hz - 20x4 Team Death Match (base game maps) [Base line]
- 45hz - 20x4 Team Death Match (base game maps) [Probably will work fine]
- 60hz - 20x4 Team Death Match (base game maps) [Probably will work fine]
- 75hz - 20x4 Team Death Match (base game maps) [Probably will work fine]
- 90hz - 20x4 Team Death Match (base game maps) [Probably will work fine]
- 120hz - 20x4 Team Death Match (base game maps) [might reach the limit for 20x4]
- 144hz - 20x4 Team Death Match (base game maps) [improbable for 20x4 players]
Once this testing is over, we might do a quick run using less instances (IE instead of 4 servers we run 2) - to make sure our assumptions about max player counts and the upper limits of tickrate for the server hardware are correct.
I'll update this thread as soon as we have more information (like the server names etc). Also watch my twitter if you are interested in taking part - and help spread the word so we can fill these servers up for 2 rounds per!
Twitter:
https://twitter.com/tiggr_
/David & Tyler