LordGopher

For any console players reading this in the future, there is now a mod available for console called prevent hibernation which does exactly as its name suggests. (Apparently the mod broke after the last update but if the mod maker keeps up with it then it should be possible for us console players to disable hibernation. )

I know I already conceded this discussion to you and I am not going back on that, but i wanted to critique this argument you made. I'm sure you are getting tired of responding to the same post. So this will be my last comment on the topic. You stated that if you compress an entire ark days worth of respawns over the course of an hour then it would be 24x the max server load. That is true if you assume that on the time based system the resource harvesting is spread out evenly throughout the day. Which it almost never is. With a time based system the server load is dependent on what the players of the server are doing and when, Which means you can have situations where players are doing a lot of things very quickly. Or basically nothing at all for that ark day. And in both of those situations the difference in server peaks is negligable. Because if a lot of resources are being destroyed very quickly, then they must also be respawned very quickly in order of thier destruction. Which is basically the same as a schedule system in that regard. Conversely. If the server is pretty chill and people aren't really doing anything then the difference is also negligible. Yes the difference is still roughly 24x. But if 120 nodes were destroyed in that time and were destroyed equally spaced through the day. (5 nodes every 2.5 mins) then the server on a schedule would have to respawn 120 nodes over 1 hour. Vs 5 nodes per hour over a full day for the time system. This is indeed a 24x increase. But these values are so small that the exponential increase should be negligible assuming your servers aren't garbage. The difference really only begins to matter in situations in between these 2 extremes. Where the rate of resource destruction and restoration gets closer to the average value for the given amount of time. (2.4k nodes over 24 in game hours would most optimally be divided proportionally to the amount being destroyed and how long they take to come back. If trees take 1 hour to respawn then that means you would need to destroy 200 per hour in order to fit 2.4k into a single day as you have to allow time for them to respawn as well in that amount of time while keeping the peak server load to a minimum. This is indeed what you have stated yourself in previous replies but I am restating it here to be thorough.) This is what wildcard is trying to do by imposing resource harvesting limits on thier official servers. As you stated above. However, the problem is that depending on how high that number is and how much time you have to wait. It can be a brutal experience for the players on that server. If they made it so you can only destroy 100 nodes per ark hour. Then that would suck because there are up to 70 people on official. That means each person per hour can only harvest 1 node if they don't want to screw anybody else over. And you are going to have situations where 1 tribe is harvesting a bunch of stuff and hogging the harvesting limit so other tribes can't collect stuff. Imagine spawning into an official server and trying to pick up a stone or punch a tree to get some wood so you can make a pickaxe but you can't because there is a person bulldozing forests with thier mammoth eatting up all 100 of the node limit. That would be infuriating to deal with. So wildcard is in a delicate situation where the limit has to be high enough to allow players to still get a reasonable amount of stuff done and not hog it all from the rest of the server, while still keeping it low enough to keep peak server loads to a minimum.

They can't change thier mind now about the dlcs being free. They already told people when they bought the game what the deal was. The moment money starts being exchanged then you are locked into agreements made when that exchange happened. Trying to change the deal at this point is just asking for a false advertisement suit. I'm not going to pay for premium mods as it would be allowing snail and nitrado to make money off of other people's work. Especially given how deplorable they are as companies. And you know damn well the majority of that money isn't going into the game to continue supporting It. It's going into snail and nitrados pockets. Which is reptared in my opinion. I am ok with paying for the new dlcs wildcard has planned such as a possible arrat prime map which takes place between extinction and gen 1. Since that is work done directly by the devs. I will not support them leeching off of mod makers though. Premium skins are ok I guess so long as they aren't to expensive. 3 bucks for a dino skin is fine. I don't think they should make skins from ase premium though. Such as the vampire eyes and the witch hat and the dino skeleton costumes. That's just lazy taking something we had for free and just slapping a price tag on it. If they want me to pay for something then at least make it be new. The entire point of a remaster is you get all the content from the previous game but revamped and for a lower price since they already made back the investment plus extra for the previous game.

I used to think wildcard was just incompetent and didn't know what they are doing. But over time I came to realize that everything isn't thier fault. Ase was a disaster to work with since it was built on a code base made by inexperienced developers who honestly didnt think the game would get as big as it did. When the old dev team left and the new one came in, they had to work on a broken foundation which means that every change to the code could have disastrous consequences if they weren't careful. In order to solve this problem, wildcard saw ue5 and the benefits it has and decided that they could start fresh with ark and redo the code so that updating, Performance and bug fixing would be easier and more streamlined. I'm confident that's exactly what they were doing for the 1st few months of asa development. However, as I'm sure you are aware, snail games blew all of thier income on stupid tesla car rip offs and other pointless projects. They were going bankrupt and needed more money. They likely contacted wildcard to see what kind of projects they were working on and when they learned about wildcard wanting to do a free ue5 port for ark. Snail told them no, we are going bankrupt and need a new product. So asa as a stand alone title was born. I'm willing to bet the original deal with asa being bundled with ark 2 and having to buy all the dlcs again was all from snail. After the community backlash. Wildcard likely told snail that plan was terrible so snail relented and allowed wildcard to set a price point that would please the community for the most part. And so the deal we have today, $45 for asa with original dlc included was born. And for the most part everybody seems cool with that. Wildcard was probably ok at this point. Even with the changes to the plan and now charging a premium for what was supposed to be a free upgrade. They still had plenty of time to actually work on the game and fix the issues ase had. Wildcard likely told snail that it would take time for asa to release in a state they wanted. Asa in its current state feels like it could have used at least another year in development to iron out the performance issues. But snail games needing money now didn't like that prospect, and wanted them to release it now. wildcard probably resisted heavily as if they released it then, then all thier work would be for nothing because it would just be a repeat of ase. (Even though it kind of is, but at least they tried.) Snail then decided to make a deal with nitrado for funds. And that's when it all really went downhill for asa. Because now wildcard is stuck between a rock and a hard place. Thier publisher asking them to do stuff, and now the only server provider for thier game asking them to do stuff as well. I'm willing to bet as part of the contract between snail and nitrado, they gave nitrado some control over changes that are made to the game. Nitrado wanting thier money back as soon as possible. And knowing that people will buy ark regardless of its condition wanted wildcard to release it as is and just fix it later. Wildcard likely fought back against this and kept delaying it to try and fix it as much as they could before releasing it. Finally snail and nitrado told wildcard that thier time was up and so wildcard had to set a definite date and stick to it. And so the October release for pc was born. The rest is history. Wildcard has thier resources split between needing to fix the game in general as well as fix bugs. And despite the disaster, I would say I'm proud of them for doing the work they have done in these conditions. I'm confident the game will be much much better in a few months. Performance for both servers and single player for all platforms will definitely improve. Bugs will get fixed. We should never forget the role snail and nitrado played in this disaster though. Wildcard made mistakes in the past. But I don't think this one is thier fault.

I'm glad we were able to come to an agreement in the end. It's possible but not practical. Also I'm sorry your pc endeavors didn't go so well. I hope you have better fortune in the future. I know it's a few days to late but merry Christmas man.

Does that mean hypothetically your game could use a schedule respawn system. As long as you make sure your servers are beefy enough to handle the increased load during those times?

This is a prime example of why monopolies are a bad thing. It puts the consumer in a take it or leave it situation. The old don't like it don't buy it strategy doesn't work when your only option is to buy from the person you don't like. Businesses cooperating with each other to basically do the same thing is just as bad but at least that takes time for them all to agree on how to rip you off.

Your 1st point implies that nodes have thier health restored little by little over time instead of all at once after a set time. Which would mean that the resource respawn interval is really just changing how much health is restored to the node per second rather than how much time before the node returns. Which is interesting if true. Also I'm fairly certain that trees wouldn't spawn right in your face as resources don't respawn within a certain distance to a player or structure. So you shouldn't have your view abruptly blocked like that. But I can see your point. Imagine looking for a raptor to tame in a clearing then the trees respawn around it before you get there. Now you have to find it again. So that's one con. I see your point on the second paragraph. And I think I understand what you mean by server load. It's how how many computations the server has to perform over a given time. With timed based respawn the server has to perform respawn calculations as they come. Reducing the load as it only has to update a few nodes at any one time as they aren't all going to respawn together. Although the server is still having to keep track of every single node anyway. But I'll conceded this argument to you. For your final point. You are basically saying that the server can only perform so many computations. And having players with max level mammoths that were just bulldozing entire forests in minutes was stressing the server because it had to keep track of every single tree that was being destroyed. And it couldnt keep up. to that all I have to say is how bad are these servers where they can't keep up with that? These aren't indivivual rigs built by gamers. These are clusters of machines that were designed specifically to do just this and they can't even handle it while the player is performing simple tasks such as harvesting resources? One of the most fundamental things you do on ark. Either the servers are bad or the game is so horribly optimized that its asking to much from them. It is 24x more server load at once compared to the current system. My argument was that with proper optimizations it could be comparable in performance to the current system. There are things that could be done to reduce the server load such as updating a few nodes at once over time. But at that point you are basically doing what the old system does where the respawns are spread out so why even bother? I'll go ahead and conceded this argument to you my friend. Although I'm confident it could work theoretically. Practically it would not. The amount of hoops and adjustments you would need to jump through just to get performance roughly the same as the current time based system wouldn't be worth it. It would be a pyrrhic victory on the devs part.

All simulations are just number crunching. I never said it was just numbers sitting in a data base. Rather the numbers within the data base are used in the server computations. And yes the server itself isn't rendering anything. It has no need to. Everything happening in game can be represented by values. So no the server doesn't need to render anything visually. It does need to perform computations to determine the game state though. Which you would think for machines specifically designed to handle lots of information and number crunching would perform a lot better. This tells me that either the servers themselves are bad, or the game is so poorly optimized that it isn't leveraging the potential those servers have. Ark is a very database heavy game I will agree with that. The dinos alone have almost 10 values needed to define each one. A creature id to uniquely identify every single dino, as well as its stats and color regions. And that's not even counting needing to remember where they are or the calculations for thier ai such as when they are fighting stuff. Times that by over 1000 wild dinos on the map before even considering the tamed ones and it's easy to see why the cpu load for this game is emmense. As you stated hibernation was thier solution for this in single player as instead of needing to keep track of the entire world and what it's doing, the game only computates whats happening within render. Greatly reducing the cpu load. Lowering the render distance also reduces gpu load as well since that's less things to render. I will conced that my logic only works for aspects of games that aren't as simulation dependent. Things such as dinos for example have thousands upon thousands of entities each with an ai being calculated and recorded and wouldn't be able to handle such a system. So my apologies for my ignorance on that. However for things such as resources which are static and do not interact with the environment at all and thus require very simple calculations, that system could indeed work. It would be a simple matter of switching the numbers around when the correct time arrives. The logic could be as simple as just saying if (time) then (node health)= 100%. This would then cause the client to be prompted to render those nodes when you approach them as now the game will see thier health value is now above 0 and thus should be rendered. To be honest I never really expected to go this far down the rabbit hole when I made my original post. I just had the idea that for single player the resources not respawning when in hibernation could be fixed by just giving all resources a global timer. That way they return even if you are not there. I then thought maybe it would be a cool game mechanic to have for servers. But from what you are telling me and what I have seen myself the servers are already up in flames. Also when I said the difference is negligible I meant the difference between the two systems. I wasn't saying the calculations in general are negligible. Obviously they aren't or else they wouldn't have needed to put hibernation in the game as you mentioned.

Have you tried killing your character and respawning? Even if you can't retrieve the glitched items from your corpse that would at least empty the inventory for your respawned character.

I really don't see how simply changing resource node health values back to full is going to affect server performance. It's literally just keeping track of numbers. At the end of the day that is all a server does is just keep track of numbers. Coordinates, values such as health, xp for characters and dinos. How much ammo you have for a gun. How much gas in in your gen, ect. Ect. All of that is just numbers. The server doesn't actually render anything. It's the client that does. And it renders based on what values the server is providing. And if you aren't in render when the stuff respawns then you shouldn't feel anything. And even if you were, how many nodes are you going to be needing to render all at once? Probably a few hundred if that because of the render distance. And if you stagger the respawns so that they return in little bits instead of all at once then you have the potential of being about as performance intensive as the current system. Its not like all of the nodes on the map have to be rendered all at once. And if they did then the map maker made a huge mistake placing that many entities in such a small area. If your dedicated server can keep track of all 10k nodes on the map and thier health values as well as how much time is left to respawn for every single one of them, then it should be child's play to just set all thier health values back to 100% so the client knows to render them again. Think about it. All 10k resource nodes need to be stored. And all of those resource nodes need to have a health value so the server knows when they are destroyed, and they need a respawn timer so the server knows when to reset them. That's 30k values in total. Granted only 20k of those are permenate. The location of the node and how much health it has, the respawn timer can be applied when the node's health reaches 0. In your example you said that there are 2.4k respawns needed per ark day. That means that every real life hour the server has to keep track of approximately 22.4k values at the same time. Compared to having them respawn on a schedule in which case the server only needs approximately 20k values. Since there are only 7 main categories of resources in the game that can be grouped together. (Stones, trees, Bushes, metal, obsidian, crystal, misc such as silica pearls.) If we are talking raw numbers then my idea would actually save on server memory, approximately 11% less values to keep track of. Especially for single player where your system is both the client and the server. With the proper optimizations the difference would be marginal as both my system and the current one can be better optimized. My system is basically like having a bus take a set number of passengers down the street at set times. While the current system is like having the people walk down the street at thier own paces. Each arriving in accordance to how quickly they walk. The bus method will help logistically as you know exactly how many people will come and at what time. But that comes at the cost of having to process all those people together. The walking down the street method means less people to process at any one time because they trickle in little by little, but comes at the cost of certainty. You don't know how many people will arrive and when. Could be none. Could be a lot. Both methods have thier pros and cons. And if you really boil it down then the performance hit would also be negligible. At the end of the day it really boils down to design philosophy and how the devs want the game to work. They want a more realistic system where stuff respawns according to thier time of destruction. Meanwhile other games use a schedule for thier resources. In the end it boils down to developer preference. My preference is for them to add prevent hibernation to console so I don't have to afk on a mountain top for stuff to come back.

Can confirm this is still happening for xbox series x single player even after the latest patch.

I'm fine with the way dino spawns work now. My suggestion only pertained to resources. Either way until they add prevent hibernation to the xbox version I'll just have to go afk and wait by the metal spawns.

I can see your point. The comparison to destroy wild dinos isn't exactly 1 to 1 though. As not every resource on the map is going to be destroyed. Some will still remain so not every node is being respawned at the same time. Only the ones that were destroyed. They could also stagger the respawns. When the time reaches the appropriate o'clock instead of everything coming back all at once it could repopulate the world over the course of an in game hour or so. This gets rid of the all at once lag spike you mentioned. And also can provide that natural look as stuff comes back little by little instead of all at once. Which would help maintain the emmersion. This idea was my way of fixing the hibernation issue in single player. As it wouldn't matter if you are in render or not; when the time strikes the appropriate o'clock then stuff will respawn regardless of player presence. I really only posted the official server schedules as a proof of concept. And really my intention was for this to be added as an option you can check such as the use single player options. Having a schedule on official would make resource runs more interesting though as knowing when stuff will respawn can allow tribes to better plan thier farm runs. And if you time your run correctly you can even get a double round of resources as stuff respawns as soon as your done harvesting what was there before. Let's just see what they do with the game. There is still plenty of time for them to fix and change stuff. Just thought I would throw out an idea. Edit. I had a feeling the devs don't look at these. Oh well, thanks for taking the time to give your feedback on my idea though.

I have no idea if the devs even read these but I thought I would try my hand at posting an idea. Currently resources in ark ascended work by having individual timers. This puts more stress on the servers and on your system in single player since they have to keep track of every individual node and its timer. This performance hit is probably negligible but it's still there. This also means that on single player resources in hibernation won't respawn unless you are in render. My proposal? Have resources respawn based on a schedule instead of a timer. For example at a certain time or day in game, all resources of that type on the map will respawn. As long as they meet the requirements to respawn (not to close to structures or players.) The schedule can be similar to how aberration works where certain resources will be slated to respawn on certain days for servers or certain times for single player. The exact values can be adjusted as needed, but the general consensus I came to was having the values for single player be these. Single player: Bushes: 00:00, 06:00, 12:00, 18:00 in game everyday. Trees, rocks, silica pearls: 12:00 and 00:00 in game everyday. All endgame resources such as metal, obsidian, crystal, black pearls, and element: 00:00 in game everyday. Cave loot drops and deep ocean crates: 00:00 in game everyday. For official servers the values could be these. Bushes: 12:00 in game everyday. Trees, rocks, silica pearls: 00:00 in game every day. Endgame resources: 00:00 on days ending in 5 and 0. (5 in game days between respawns) Cave drops and deep sea crates: 00:00 on days ending in 5 and 0. How would this work for single player and private servers? In the main menu where you customize your server settings you will see resource respawn schedule instead of respawn interval. Underneath this will be a list of every resource in the game as well as 2 numbers, time in game written in military time. And days, which is how many in game days you want it to take between respawns. The 2 numbers together will decide on what day and at what time that resource will respawn. For example if you set metal to have the values of 00:00 and 5. Then you will have metal respawn at midnight in game every 5 in game days. (Days are counted from 0) I know this is a lot to read so thank you to whoever took the time to read this and please give me your feedback in the comments.