Telemetry system


#1

After playing a fair few games of fighting OP monsters I couldn’t help remember some post from turtlerock saying how they have a new telemetry system that let’s them change stats right then and there and it’s a game changer? This isn’t a post of me being arsey, more me enquiring when we’ll get evidence of this new telemetry system.


#2

It was given a long time ago :open_mouth:
Here:

:smile:


#3

I seen that but I still see no evidence of it In action is what this post is about my friend


#4

Well apparently the changes they made were ‘larger than intended’.

Continuing the discussion from Let's talk 8.1a!:

So they couldn’t get the right numbers in for the last micropatch and that’s why things are so fucked.


#5

Oh! You want to see what it can do? Hmmm, maybe check these threads?


But who knows, maybe Mizx will give you more info :slight_smile:


#6

I’m not really bothered about heat maps. Yes it’s a cool feature and useful but more useful would be the instant “balancing” of hunters/monsters


#7

Please don’t get me wrong here buckets I’m not out to start an argument or anything. But more curious on how everyone knows gorgon is majorly OP now (and this is coming from someone who is a gorgon player) and it’s killing get the character because I don’t want to go it as it’s unfair and I don’t want to fight it because it’s a downer front the get go. I just thought the point of this system was to fix shit like this instantly.


#8

People complain about Gorgon and Elder being OP now because of personal experiences. Instead of the spider trap doing slow DOT and slow movement speed it’s now fast and does a lot of damage.

It’s how these monsters got unwarranted buffs in areas they didn’t need it.


ThunderChild is already in place as with heatmaps as proof. ThunderChild is updated telemetry so more specific data can be collected, and has no visual proof of it being there unless Devs pull the data off and put it in graphs and charts. Old telemetry system also limited weekend challenges meaning only wins or damage can be counted, but ThunderChild can now involve more advanced challenges, like dome amount.


#9

We do have a new telemetry system! It was introduced in the last title update back in January. We’ve since made it better and added a lot of data for it to collect, we’re just waiting to ship it in the next title update (TU09).

Our telemetry system only reports data to us. The system is not able to increase/decrease a characters damage. Those changes require a micropatch or title update (last micropatch was about 4 weeks ago I believe).

The balance guys use Thunderchild (the name for our new telemetry system) to identify problem areas and a starting point for what should change. If an ability is doing high damage in one set of games and low damage in another set of games, the balance bros would want to adjust that ability and bring the damage values closer together between those games. Any changes they decide to make would be shipped out in the next update or micropatch.

Hope this helps clear things up! Feel free to ask me any other questions about telemetry, I’d be happy to answer them :slight_smile:


#10

No no its all good. I just thought that the hype with thunderchild is that the changes didnt have to wait 4 or 5 months till the micro patch for changes and it could just do them as and where. E.g. oop the gorgon seems to be slaughtering at the moment, lets knock it down a few points. (Dont mistake this for me being a dick, im just worded that way)


#11

You’re right in that Thunderchild was designed for a situation in which there could be rapid iteration on game balance. The usual cycle of adjustment is something like Measure -> Propose Changes -> Implement Changes -> Test Internally -> Adjust -> Release -> Re-Measure.

This cycle works better the more rapid this loop is, ideally running multiple experiments at the same time. Thunderchild deals with the measurement part of this cycle. It uses unstructured/semi-structured data so that what is being measured can be changed with each release without breaking or delaying too many other pieces, allowing us to change what is being measured with each release in the most painless way possible.

Thunderchild is designed to support and be a part of this kind of rapid release cycle, though it does not cover that entire cycle by itself. What this implies for the future of Evolve, and why Evolve would need a system that supports this kind of iteration, is left as an exercise for the reader.


#12

Good to know such things are still in store and approaching for our beloved Evolve and for TRS.

I mean, err… Golly gee, wish I knew how to exercise my reader whatsit. :nerd:

Side note:

@junglemanlloyd

Thunderchild is a data tool, not a patching tool per se.

It will greatly assist and facilitate TRS in pushing out patches rapidly by arming them with highly detailed and useable information about how their game is performing in the hands of users in real time.

But Thinderchild =!= (does not equal) Faster Patches Now.

Just thought I’d attempt to clarify that distinction, if it helps ya.

:wink:


#13

No no that helped alot! Knowledge is power lol. Cheers dudes!


#14

Disclaimer–
I realise i keep starting with “no no” its just because
im trying not to come across as a massive bellend


#15

Good thing I like exercise!

Broad implications could be that Evolve can support a more dynamically changing META.

Faster Adjustments I think this could obviously mean what I guess a lot of us assume, more micropatches and faster. It would also help expedite the balancing of new characters and variations as they are released to prevent WAGGGIE type situations.
[SPOILER] F2P? [/SPOILER] Now this isn't a prediction, but it's rather a possible path they could take. Now look at it this way, if they were to ever adopt a model similar to where certain characters would be free and it were to rotate (possibly weekly) then we are talking large amounts of data with high variables coming in rapid fire. Having a system like ThunderChild would really help identify areas that need tweaking fast and then address them quickly.

Now I’m not saying this is proof positive of Evolve going F2P, but just a solid system that would allow that possibility in the future. Personally I put Evolve as F2P at about 25% by then end of the year and only about 15% at TU09.