August 13th, 2024 Livestream
Q&A: Can you give more detail about the CPU/GPU loads in 1.0?
This question has a related video
Suggested Wiki Reference Code
<ref name="yt-_MaS7WcgQyo">[https://youtu.be/_MaS7WcgQyo YouTube - August 13th, 2024 Livestream - Q&A: Can you give more detail about the CPU/GPU loads in 1.0?]</ref>
Topics
Transcript
in regards to optimization coming 1.0 can you, give more details about how much change we can expect in CPU and GPU loads so here's the thing about this question is that it's really hard to talk about like optimization because it's so dependent on Hardware, and like you know I think I mentioned this in that video that like you know some of you might not notice any change some of you will notice a massive change and and we have like a machine that we do testing on we have two machines that we do testing on we have like one highs spec machine and we have one really low spec machine and like the low spec machine is like I think right now barely like, the minimum minimum Target Hardware stuff so like it gets by, and we do, like automated testing on both of the machines and we check like for instance like how what is the what is the, the average Ms, on like our slow Machine versus the fast machine like compared to all these commits that are coming in and if we notice a significant drop then we you know we have data on that so we compare these two, and for 1.0 we notice like pretty significant change on the slow machine but not as much on the fast machine because I mean it makes sense so like it's really hard to talk about that stuff, I mean we could give you the numbers that we run when we run these two machines but I'm not sure if that will actually like mean anything like be meaningful to you guys because if I tell you that like oh the game runs at like 13 Ms on our machine here and has these settings like okay what does that mean for me, so, it's it's it's I mean it's possible to do it but it's just hard to to make it worth the like worth while for you guys that makes sense and and if we do these videos and say like hey because we this has happened in the past too we're like hey we've noticed on our machine that we get like a 30% increase in like performance we say that and then people get pissed off because like wait but I don't see 30% increase on my computer like you guys are just lying or bullshitting or whatever so it's like okay I guess we won't do it at all then, so yeah it's just really hard to make any like General like, any like remarks about performance and stuff like that we can only really talk about performance in general, one aspect of this as well is that like a lot of people tend to, like I'm not I don't want to throw too much shade here okay but a lot of people tend to come back with us and tell us so like they think they know more about how the game is made than than we do, and it makes also makes the conversation really really difficult, because if we try to explain something and then people have this preconceived notion that like they know more about it than what they think they do it makes it really hard to to communicate about the specific things we've done, and like you know implementations that we've done or whatever even between game devs honestly like, I've had conversations with other game Debs where they think they understood our solution to problems and then they clearly didn't, and it's just how it is sometimes with with software it's like there are so many different ways to do the same thing and there are so many like you know how the thing where you use you know one system in a certain way but if you have another system and you use them together then the tradeoffs that they combined get like are different from game to game so, so yeah it's it's a tricky thing and the only and really the only thing that's true in any of this conversation is the profiler that we use on the game, and that's the only thing we can go by so so like regardless if people say like hey I get pretty bad performance and then we perform in like I'm PR on like exact same Hardware we get like a completely different result so like it's it's really tricky it's really hard to talk about it, but we try to give you as much information we can, about this aspect we just sort of like need to find like I mentioned before like it's hard to do full deep dipes because people misunderstand and I think as to some degree that misunderstanding muddies the water in the conversation, but we also don't want to make it too vague because then it it sounds like we're just like kind of making stuff up or whatever it just sounds like yeah everything's better you know but like we notice that the game we have data to prove that the game is better but like whenever people play it might not be noticeable so like there's an as there's always an aspect of like when something is more performant whether you actually notice it is or when it actually is, you know if a game runs at 200 Ms, sorry if it runs at like 200 FPS and then the game suddenly runs at 120 FPS like that's a massive drop in in performance but not really you know what I mean like so everything's always, subjective anyways