-
Posts
855 -
Joined
-
Last visited
-
Days Won
1
Content Type
Profiles
Forums
Calendar
Gallery
ARK News
Sponsored Mods
ARK Mobile News
Everything posted by arkark
-
"We’ve made the decision to postpone the Center as it wasn’t on track to launch with the level of quality we require." This statement scares me. Considering how The Island came out, I don't even want to imagine what state The Center's development must be in for them to have decided not to release it because it doesn't reach the level of quality that, considering previous releases, is already very low. Probably a flat surface with two trees or something 🙂
-
It was clear that this was going to happen. You have learned nothing.
-
ARK: Survival Ascended - Device Driver Hung - Possible Workaround
arkark replied to Hodir's topic in General Discussion
Can confirm this solution doesn't work with a 4070 Super. Been playing ASA with a 1070 (with lot of commands to reduce shadows, volum. clouds, etc) with very few crashes during this months and since I have the 4070 Super I'm getting the DEVICE HUNG crash 2 or 3 times per day. This only happen with ARK btw, I've tried other games with long gaming sessions and havent had a single crash. My crash backtraces are similar to the ones in the OP, I would also add this part that only appears inside the crashstack files: I've googled about this UE function (GetCompletedValue() from CurrentQueue.Fence.D3DFence) and the DEVICE_HUNG and this appear to be happening to more ppl when playing games with UE 5.2, like Remnant, because UE 5.2 rely on old functions when talking with DX12 API. So I dont know what to expect tbh. -
DEVS? Will performance remain the same in future?
arkark replied to Jazz86's topic in General Discussion
Unfortunately, I don't think anything will change in the future. And if it does it will be at the cost of losing functionality. Apparently someone thought it would be a good idea to reuse ASE servers that are practically 8-10 years old, and besides, instead of putting three game servers for each physical machine what they have done is to put five for each physical machine. Something inconceivable knowing the antecedents of how bad ASE was in populated maps. Obviously what they have is not enough, and I doubt very much that they are going to achieve a moderately acceptable performance during the whole life of ASA, unless they discover that there is a huge bug in the code that is weighing down the performance of the servers in a very significant way, which as I say I don't think is going to happen. I think, totally unfounded opinion, but I think one of the next steps they will take will be to eliminate the issue that the dinos move with UE5 pathing system and activate some switch for the officials to work in the old mode that is probably less expensive, since the issue that in officials the dinos remain static for many seconds, is something that has been spoiling the experience since day one, and is possibly one of the points where the hardware of the officials is not suitable. -
There are many things wrong with the new cryopods. To begin with it doesn't make any sense that they are introducing them now in PVP. They are supposed to be coming early to alleviate the serious problem in PVE with the tame cap. Tame cap problems do not exist in PVP, they have the ability to purge dinos. In PVE as they have approached it, what they are going to do is to further worsen the coexistence on the servers. They will cause a privileged tribe to obtain the areas closest to the caves to be able to introduce the dinos they want, while the rest of the players who are not their allies will have to settle indefinitely to play in worse conditions. In addition, this brings collateral effects, since there will be no more sleeping bags at the entrances of the caves. That practice so normal in ARK is over, now unfortunately you will have to go far away, until you do not find structures of other tribes to place your sleeping bag. The issue of having to release cryos near a cryofridge does not displease me, it is consistent with the idea that the important thing now is to alleviate the load on the servers and reduce the tame cap. However I would go further: - I would only allow one cryofridge per tribe. - This cryofridge would have an activation time of 24h, so that it is not mobile. - The cryofridge would be 250 slots, and the dinos in cryopod would not count for the global tame cap of the server but would count for the limit of dinos per tribe. In this way the problem is solved, without giving unfair advantages to anyone and in the most similar way that the gameplay would be in an ideal server where there would be no need to implement this emergency measure.
-
What you want to do is implemented in Ark Smart Breeding, in the tab "Stat Testing". In the case of the deino, a 0 damage points hatched deino would be 125.8% dmg, a 55 damage points hatched deino would be 449.2%. From here you can calculate how much % dmg it gains per level: 449.2 - 125.8 ~= 323.4 <- total % damage gained with levels Divide then by the number of levels applied: 323.4 / 55 ~= 5.88 <- % dmg gained per level 449.2 + 5.88 = 455.08 == 455.1% <- rounded, this is the stat that you will see ingame for 56 dmg points. Bear in mind that the decimals are rounded in ark hud, so the calcs are not perfect accurate. I would just dont re-invent the wheel and would give it a try to ASB which fits perfectly with what you trying to do.
-
Community Crunch 389: Turkey Trial is Live!
arkark replied to StudioWildcard's topic in Announcements
It does not make sense at all to release an event with dino colors and breeding rates while most of official servers are tame capped. I think you should hurry to release cryopods at least in pve. -
Well, after searching and searching for information about this, I finally did what I should have done from the start, which is to check it out for myself. In case anyone is interested, they have replaced the Steam protocol with the EpicGames API, so to get the server status, requests have to be made to the API. The API is not public, and I haven't seen anywhere that you can request a key. However, very well known server metric services have been using it since the launch day of ASA without it seeming to be a problem for Wildcard. I would appreciate it if a moderator could ask and inform me whether I have permission to post an example of using the API that includes the key I found, or if they could shed some light on how to request a key for the API. In my case, for instance, it's to help a friend with his website for his ASE and now also ASA cluster, which shows the server status, players connected, ping, etc. I also use it for myself, to check the server status once I've disconnected from a map (very useful to know if the server crashes within 15 minutes of disconnecting, before saving).
-
Official 5193: 1 minute in dino breeding is 1 minute and 45 seconds in real time so basically the increased rates you applying are halved. Already reported, and indeed, reported also every day since days ago.
-
In ASE we could use Steam Query Protocol to get info (ping, users, name, and others..) of a server. Apparently this no longer works in ASA. What is the new method to retrieve that information and where is the documentation? Some sites like battlemetrics already have it implemented.
-
Shrinking the size of official saves (PC)
arkark replied to arkark's topic in PC Server Administration
As far as I know a tool that can do this completely offline doesn't exists (yet). -
Shrinking the size of official saves (PC)
arkark replied to arkark's topic in PC Server Administration
sure, https://github.com/alez-repos/shrink-ark-saves It can be a bit complex if you are not familiar with the subject. If you get stuck at any point, let me know and I'll give you a hand and improve the documentation. -
Shrinking the size of official saves (PC)
arkark replied to arkark's topic in PC Server Administration
Thanks. The largest one I've completed is the one at position 44 with 2.4GB. There might be others slightly larger (up to 2.8 or 3.0GB) that are feasible given my computer's specifications (32GB RAM). I tried with valguero544, and my computer crashed after 3 hours, struggling with 79GB of virtual memory used. So, given my specs, I would need to adopt a different approach or just increase RAM amount. I did another program before discovering the json server option, to parse the save file and extract all the tribe IDs. However, unlike with the JSON, I didn't studied the save file format in depth to be able to determine how many structures each tribe has. Anyway, out of almost 1900 servers only 15 would be problematic for someone with 32GB of RAM. And also most of that top 15 list are PVP servers, so I assume that the vast majority of structures and dinos will belong to the server's alpha tribe. -
Shrinking the size of official saves (PC)
arkark replied to arkark's topic in PC Server Administration
Which server number, I want to see how much time it takes -
Shrinking the size of official saves (PC)
arkark replied to arkark's topic in PC Server Administration
Yes, I've seen some json up to 0.5GB but not really getting struggled when parsing it. I extract all the info from them in about 10-15 secs. -
Shrinking the size of official saves (PC)
arkark replied to arkark's topic in PC Server Administration
Yes, I use the -parseservertojson option then either RCON or 'exec' (depending on the tribe size) command to destroy the stuff. -
I haven't been on the forums for a while but this couple of days I've been doing a little utility to clean up the maps and preserve that little piece of history that is my own bases and the ones of my friends on the official servers. Some examples of how it turned out: Server | Size in disk | Memory usage | Server save time | before after | before after after wilddinowipe | before after ------------------------------------------------------------------------------------------------------------------------------------------------------ Ragnarok 210 | 1.6GB 400MB | 18GB 10GB 2.4GB | 25s 0.7s Gen2 1008 | 2.5GB 187MB | 24GB 12GB 3.1GB | 35s 0.4s Mainly I have done this because I think that having to handle so much data that we do not need its not very environmental friendly, a waste of energy and computing resources (Wildcard, which from what I see has employees on twitter who do activism on a consistent basis, should have been more careful about this and been more responsible, facilitating it themselves), and also in terms of monthly expenses if we are going to deploy the servers at an external provider it is not optimal because, like Nitrado, they will want to charge us, rightly so, for a giant server. My utility extracts information from the save and removes everything that is not related to a list of tribes I want to preserve. It is almost fully automatic, relatively fast and secure. I don't know if there are people interested, or if there are other similar methods to clean the server in an almost unattended way and guaranteeing that everything is cleaned. At least from what I researched before I started programming my own utility is that what there was was compatible with saves prior to version 11. One of the ways to achieve something similar is to allow everything on the server to decay. But this is very inefficient, takes a lot of time (+80 ingame days due to structures like tek dedi-storage), and can also cause problems if we want to do it very fast and we have several bases to maintain and have to refresh them every few time cycles. If I see a lot of interest I might upload my utility for anyone who wants to use it, or create a discord server to support it. PS: While doing this I have seen some absolutely crazy things, for example in Ragnarok 210 there was one tribe with over 68000 structures, another with over 16000 cryoed dinos. It explains a lot why the servers were going the way they were going. I think it makes you wonder to what extent the developers shouldn't put some parameter in the official servers to limit this.
-
Gen 2 934, 935, 936, 937, 938 and Gen 2 Legacy 946, 947. All this 7 (7!) servers are on the same physical machine. When one of this servers crash it often drags all the other servers with it. Example of this morning: At 07:33 server 938 crashes. At 07:58 while 938 is not responding (crashed, crashing, generating dump, or even booting), all the other servers on the machine start crashing one after other This is just a real example of this morning. This exact same behaviour is happening 10-15 times a day on this machine. When one server crashes it grabs all the other servers on the same machine with it. I would speculate about this is some memory leakage on the crashing server that hog the resources of the machine until every server crashes, result of limited shared resources (RAM) of the physical machine. But this is just speculation. What is clear is that until you take a look at it and move some server to another machine or increase the shared resources this situation is not gonna improve. Please do it soon so we can enjoy gen2 without being worried all the time about if we can do x thing because it may crash or not. @lilpanda @Cedric
- 6,682 replies
-
Genesis 2 performance its horrible. We are playing all the time in 938 with 200-255 ping. Dinos just teleport and dissapear all the time. Even taming a bee becomes a challenge. I think each bare metal server was hosting 3 servers before (27015, 27017 and 27019), but with genesis you seem to be running 5 servers or more in the same hardware (27015, 17, 19, 21 and 23) ? Whatever you are trying to do, probably with more powerful machines is not working at all, rendering to a very poor latency.
- 6,682 replies
-
Siempre ha sido igual, los nuevos servidores tienen 90 días en los que no se puede transferir nada más que el personaje. Cuando pase ese tiempo quitarán la limitación.
- 6,682 replies
-
This exact same crash happened to some users back in 2018 and apparently it was some DNS problem at provider (most if not all of the ppl reporting were australian telstra users): https://steamcommunity.com/app/346110/discussions/0/1694922980046496262/?ctp=5 The crash was in the same function ValidateSendBuffer() and with exact same payload size (506 bytes), so I would bet it is the same problem. Changing the DNS or connecting through VPN seemed to work, maybe you can try that.
-
Community Crunch 255: Behind the LiveOps Curtain and More!
arkark replied to StudioWildcard's topic in Announcements
Yes, the problem is that this two maps, due to how easy element farming is are saturated of giant bases with tons of cloners plus tons of dinos for selling / mutation breeding. But the long save times is not really the problem, I mean, I can live with 1 or even 2 minute save every 15 min, thats no problem, I accept it. Problem is that due to that long save times the game client (I think) believes that the server is down and disconnect with timeout message. And at the same time the server, after the save is done, still thinks we are connected so if Im doing an osd the zone where I am does not enter stasis and dinos and char get killed while we are waiting on the login lock screen. I think optimal solution would just be increase the timeout of the game client when a save its being done. At least until more appropiated solution is found it would workaround this kind of timeout disconnections. This is not only happening on the maps I play btw. Some months ago I considered relocating my bases to another servers and asked active playing friends about how they gen and ext were performing and almost all told me they are experiencing same thing on this two maps. -
Community Crunch 255: Behind the LiveOps Curtain and More!
arkark replied to StudioWildcard's topic in Announcements
Its good to see you are working on improving stability of the game but please dont forget whats going on on most (afaik) genesis and extinction official pve servers where long save time makes game clients disconnect due to timeout and then server does not allow to reconnect fast because "there is already someone connected with that account". This makes the area where the player was not entering stasis immediately, and if player was doing veins/osd/missions this much likely means player killed, dinos killed, broken gear... rendering this maps to non playable status. -
There is a maintenance atm and, at least on PC, the transfers are not working. Try to go to genesis, and in the character creation screen click on Character download button (at bottom right of screen). Your char probably will be there.
-
Hello, This server is performing very bad lately. Almost every time there is a save it kicks the players and then you cant inmediately reconnect cause "there is a player from this account already connected". I myself have reported it dozens of times thru ark.gg/outage, and right before leaving the server I asked in global and ppl told me they are reporting it a lot too. This last time we were doing a purple OSD when the server kicked us. We though, OK, all players kicked, also the zone its in stasis but nope. As soon as we were able to reconnect we were dead, the 825 imprinted gigas were dead, all our quality tek gear was destroyed and as a bonus all the dinos in the zone did not despawn so when we were able to clean the area and escape with the stuff we could pickup the server kicked us again and another dino from the osd killed us again. I think some can be learned from there: 1. Extinction 464 really needs attention. Do some clean up, move to other server, leave alone on the server it is. Something, but it is kicking everybody even with low population and I think it is because the required time to do a save is so insane and its bigger than the timeout set on the client. 2. The OSD should despawn with all the creatures if everybody around disconnect. I think it was behaving like that some time ago idk when changed. 3. The dinos should enter stasis faster, or the server detect the player been kicked and stasis the area, because I cant understand how the dinos where able to take on a 36k 124 saddle dino that fast while the area was theorically in stasis. 4. Its so pissing when the work of two weeks is destroyed that fast due to a failure on the server. Same for the repairs of all the tek we had. If the player is disconnected it shouldnt be visible for the dinos around, especially when the disconnect is not clean. Please look whats going on extinction 464.
- 6,682 replies