Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
Say for example the game decides what will spawn once per second. If you have 3 computers, low tier takes 5 seconds to calculate what spawns, mid tier takes half a second, and high tier takes 100ms, the low tier computer would probably be spawning less, and the high tier would just have more idle cpu time.
This is kind of what it seems like. Most games have the same amount of spawns, a few games have a ton (but a game later with the same host it will be normal again oftentimes), and then a few hosts you play with the enemies lag and way less spawns.
Well, I have played the game on two different systems. I have had over 400+ hours on both machines combined. And, I can tell you that on my potato PC, I NEVER had those extreme crazy insane massive hordes that I often get on my extreme-end powerful machine. However, on my beefy rig, I did STILL get many quiet boring runs as well. You see the difference?
I think it works something like this: if spawn rate RNG is on a scale of 1 to 10, the RNG on a high-end CPU will be more often between 5-10 whereas RNG on lower-end CPU will be more often between 1-5. If a guy with a 5+ ghz 8700k often gets 10, a guy with 3rd generation i5 will NEVER get above 5. But, the guy with 8700k may sometimes get 5.
I don't think a poor i5 from 6 years ago can afford to host those massive massive insane hordes.
P.S. The potato PC belongs to my brother.
Well, the difference between extreme low-end and extreme high-end is certainly huge and readily noticeable. An educated guess should lead us in the direction that between middle-end and high-end there should LIKEWISE be a difference ALBEIT a small and less-noticeable one. The difference should be proportionate to the difference in CPU powers. No? But, it becomes even more difficult to notice in the latter categories because this is also purely RNG.
Links, please!
my own testing shows that hordes are equally big(with rng variance) and I have tanked my CPU so much that I played around 10 fps and I've also fiddled with memory which caused the game to crash when it tried to spawn a horde... I have also upgraded my CPU with no spawn difference. I haven't documented or done enough testing to completely rule it out which is why I don't say it's not true.
However I do want to find these proofs that people claim exist but never link.
Well, its your experience against my experience then. I'm not here to prove anything. I am not sure myself what is causing the issue. Is it pure RNG like you said. Or, is it host CPU as well as RNG like I think is the case. Neither of us can prove either way.
https://www.youtube.com/watch?v=LAwTKyX0MnU
The low-end CPU makes the finale a ghost town accompanied by clear AI dis-functionalities whereas the high-end CPU has no problem in these regards. However, please be advised that the author is stressing the CPUs in both cases by doing a speed-run. Nevertheless, a clear relationship of the host CPU to spawn rate and AI behavior is established in this video - whether or not that also translates to other CPUs in lesser degrees and in general is fodder for thought (video does not prove that).
Except it's not. He's just saying "an old laptop" and then he throws in his theory and explanation as of WHY it happens. Nothing in this video shows anything that supports his theory tho. He doesn't even give specifications of the laptop. People asked for the specs in the reddit and he just ignored it and only replied to people that believed him.
I don't think he tries to trick or lie to people, I think he genuinely believe it's due to the CPU as he claims. You need to prove that tho, not just say it is.
Reason why I'm skeptical is that we've had weird AI behaviour when we play within the same group and same hosts and only thing that happens is that someone starts to lag. Which is why I've stated before that connection probably play a bigger part than the CPU. Reason for that is syncing issues between the computers which causes weird behaviour and maybe even halt the spawn of new enemies since it probably can't sync up the new spawns.
Considering he says "an old laptop" I think that the laptops wi-fi is causing issues... but I don't know, cause he doesn't even mention what kind of laptop or if it's wired connection or wi-fi.
If you find something in that video that can work as proof rather than showcasing a problem I'd love to hear it, because frankly I find 0 proof in that video. Only that the problem exists, but nothing on what's causing it beyond his speculation.
Unless someone looks at the actual coding in the game engine to say one way or the other, there is no such thing as "proof" in the sense that you want. All in all, I find the video more than probable, and a fairly reasonable explanation. I was under the impression that you would at least acknowledge "old laptop" CPUs failing to maintain spawns and AI, but to each their own I guess.
Anyway, I am not here to claim or prove anything. I find the CPU explanation as it circulates on reddit and these forums a more plausible "theory" than saying its just RNG and pure RNG -end of. And, the video to me is quite a bit of reasoning and demonstration to convince me. And, I agree that I am also throwing in quite a bit from my personal anecdotal experiences to come to this conclusion, but so is everyone else around here as far as this issue is concerned.
I find that Director mostly targets players/teams that are spread out or just generally out of position or in a bad place. It can tell when a player is out of line of sight from their team and generally capitalizes on it.
As host, I try to ask questions about my games like "how's the latency?" or "are you seeing pathing issues?" etc and the responses are "great!" and "nope everything looks good". It's rare that someone has latency (most players have yellow or green latency indicators) and during those times the other players experience no issues.
I think part of this that people are forgetting is that good latency to the server greatly improves this game, even with high spawn rates. You can timed parry attacks against you, push against charged attacks, etc because you're closer to the real time processing of the server.
There are bugs that happen in most of my games, like monsters starting a charged attack against a player in front of them then turning 180 degrees in the middle of the action and targetting someone behind them instead. This usually results in someone getting downed/grabbed. This even happens to me as the host so I know it isn't latency causing it, but it appears to be a bug in the targetting system that allows changing targets mid-action.
Personally, I love playing Legend even when Director is being a ♥♥♥♥ about it. The next time this happens in your games, type something in chat to Director. This usually gives my team a morale boost, because it changes the focus for them from the skill of their teammates to the Director being a ♥♥♥♥ and it typically brings the team together and makes it more fun. The Director wants you to have a good time and it will try to kill you again, and again, and again. It's a challenge and each game is different, meet that challenge as a team.
No, you can definitely prove this by doing more controlled tests instead of just using "old laptop" as a metric. You can first of all say which hardware it runs on. You can show how much the CPU load is. You can throttle the CPU and show that, you can also test various connection methods and you can use net limiter to throttle your connection to see if it things changes. Plenty of ways.
I plan on doing a thorough test myself I just need to figure how cause I only got one computer.
However In this case I'm simply asking for one... which that video doesn't provide. Old laptop and showcasing a problem doesn't give one hint of what's causing it except hardware or connection is causing issues.
Reason why I asked is that your first long post you said how much proof there was. I mean in the end we all beileve what we will believe considering the lack of testing on this matter exists. So i'm not saying you are wrong or that he is wrong cause I can't know. I was just curious cause you said there was proof, but alas there isn't any.
I never said in the first post that there was "proof" of different spawn rates between CPUs in general, i.e., the ones used by gamers by and large. However, I did say that there is proof of host CPU anomalies if tested between extreme-low and and high-end CPU as hosts (e.g. old laptop CPUs?). I said that the proof exists for very low-end CPUs such as the one shown in the video linked. I thought I made it very clear when I said "most people are not using such low-end CPUs anyway". I find it weird that you say I said there is proof when my long post ends with "why this is is anybody's guess". Doesn't sound like an ending to a post coming from somebody who says he has proof, does it?
With that said, I STILL stick to what I said. Yes, there is proof of low-end laptop CPUs creating ghostwed towns and AI overcloads in a scenario where a high-end CPU does just fine. There is NO proof of difference between more wide-spread middle-end to high-end i5/i7/ryzens that most gamers use for this game, only guesses. Do I make it clear? Thank you. You can sit here and argue all day about what proof means to you and what it means to me when it comes to extreme-end low-power laptop CPU demonstrating AI overload and spawn rate decrements (not general CPUs).
Also, why does he have to do all that you ask? Is he on a trial in the court to prove a case? Remember, he is DEMONSTRATING something, not PROVING for you. Only because he is not showing afterburner OSD, making detailed mention of "old laptop" specs and all that does not mean he is mistaken and/or lying in what he is demonstrating. What he has showcases enough. There is no need for unreasonable skepticism.
Except there is no proof that it's tied to CPU... only that there are differences between hosts.
He doesn't... I'm simply saying why his video can't be used as PROOF. Because it includes 0 proof. It's showcasing a problem and speculation.
There is proof that low-end LAPTOPS are creating it... not necessarily the CPU. How can you even say it's the cpu when he doesn't even mention what cpu it is? That's just ridiculous.
"Extraordinary claims require extraordinary evidence." Those making the claims are responsible for backing it up with evidence, not anecdotes. There have been legend runs on high-end hosts that resemble ghost towns, that's the nature of RNG based level fill generation. Sometimes the level is filled with ambients, other times it's a "ghost town". Sometimes finales are hard as hell, other times they're a walk in the park.
What has been shown here is known as a "correlation", but as we know, that does not imply causation. Empirically showing a cause/effect relationship between host specs and outcomes is something that can only be shown with statistical analysis of a very very very large set of recorded match histories linked to hardware specs and/or an analysis of the code, both of which none of us are equipped to gather accurate evidence about given that match histories are not provided via an sdk and gathering the amount of data to show any statistical significance would require an immense level of effort and rigor from the community as well as a waay to control for people just straight up lying about the specs to intentionally skew the results (see: trolls).
So really it's a waste of time to even be discussing this unless it is just blatantly obvious which it very obviously is not.
And of course we can just say logically if you're computer is not fast enough to spawn enemies, it won't, although this would lead to a ton of other problems people don't commonly mention in the whole cpu spawning more thing.
I think its very likely cpu can affect spawn rates, and probably does in speedruns. But I think the percentage of games where it has any noticeable affect is likely very low, and does not account for all the people claiming legend is hard because of their cpu. The people who have a system that is actually spawning less are probably the same people who the game keeps crashing for.