Jump to content

stupor-mundi

Members
  • Content Count

    267
  • Joined

  • Last visited

Everything posted by stupor-mundi

  1. I've built a new pc, Ryzen 7 3700X , Sapphire RX 5700 XT Pulse. It's not in a case yet, cpu cooling is as yet temporary, i.e. there might be some cpu throttling ... This, to replace my previous setup, an ultrabook plus eGpu, which was apparently cpu bottlenecked. I've built this with a view to get a 1440p, 144hz gaming monitor, to replace my current, 1080p non-gaming monitor at 60hz. To see if I could put some load on the gpu, I set vsync off and target fps 144hz. Also, ultra settings, ultra shadows, medium mirrors x4 distant landscape detail 150km horizon draw distance landscape filter blurred ultra grass quality extreme clouds quality 4x antialiasing. Also, HDR on, but everything else on the right hand side, off. At those settings, I get >60hz, but <144hz. Often around 100hz, depending on the situation, in multiplayer. Considering I got this hardware for a 1440p 144hz, I'm a little miffed that with my current, 1080p monitor, it tops out around 100hz. Since the setting "ultra" graphics (vs "high" etc.) is independent from the detail settings for landscape et cetera, is it safe to assume it's other, independent stuff? Maybe the detailledness of planes and vehicles? the polygon count? So I'm wondering which of those settings are maybe quite luxurious, are real performance killers which don't provide all that much benefit... (I'm not dead set on having those settings. A lot of them, I just turned them on in order to load the gpu...) I'd like to get an idea at what sort of settings I'd be able to get much above 60hz at 1440p. After all it would be pointless to get a 1440p monitor with a really fast refresh if IL-2 ends up just being able to push around 60hz, or even below...
  2. Hope this isn't a terribly redundant question. I'll be building a PC (not purely gaming, but also), Ryzen (probably Ryzen 7 3700X). I'm thinking of getting a Radeon RX 5700 XT. Is Il-2 GB ok on the current generation of Radeon? ... for 1440p
  3. Hah! I should probably check again, using some more accurate monitoring software. However, assuming this is correct, everything falls into place. It's an ultrabook from end of 2017 IIRC. low wattage core i7. Certainly qualifies as a quite weak cpu in 2019. Thus, assuming the cpu is the bottleneck in multiplayer, i.e. not being able to keep up processing the incoming data from the server, makes sense. In the big thread about multiplayer stutter, Bauemer had suggested to do a fresh install to remove stutters. Since I had done a fresh nvidia driver install, and had gotten a fresh startup.cfg , I doubted doing this would have a big effect in my case. But I did it anyway--Revo uninstaller, wiping entries in the registry, everything... with very interesting results: Even though, in multiplayer, in busy areas where my rig bottlenecks in the client/server comms, my FPS were just as low as before (for instance 15fps in bad cases), they were suddenly devoid of the additional, irritating microstutter. Quite amazing. I wonder what it is that a fresh install does, or that the update process screws up over time... So, low fps don't look nearly as awful as what i had gotten used to, when the stutters are gone. I'd make it part of my routine, after one or 2 updates, if it wasn't for the extreme clunky inconvenience of the fresh install process. Also interesting is that there is the OTHER kind of stutter, which, now that one kind of stutter has disappeared, I can identify, and I believe it's not a multplayer stutter: Even flying offline, on the rhineland map, when flying towards a town, there's that moment of stutters when the buildings are loaded, which the other posters on this thread have mentioned.
  4. I see a number of views on this, but no answers. Is it an exotic issue that others don't have? Is the topic in the wrong section, should I try to get it moved? I'll try to give some background to why I bring up this question: I'm considering: - buying a gaming monitor to get refresh > 60 hz (in this case I would go > 1080p) - building a new gaming PC Yet, I'd be quite content with the fps I get in offline/single player. My 1070 eGpu seems to be sufficient to run Il-2 GB graphics-wise, if it wasn't for the multiplayer / network induced crapness. Equally, if it were clear that my laptop's cpu is too weak, that would resolve the question in favour of building a gaming PC. But in this case I would expect to see at least 1 thread maxed out in a low fps situation, which I don't. I'm not really looking forward to the effort and expense of that, only to maybe find that the bottleneck lay elsewhere. To illustrate, here is a screenshot of resource monitor cpu cores/threads. I took it about 10, 15 seconds after being a 10fps situation on the tank server. None of the threads look maxed out. I'm actually quite surprised at the even load across all 8 threads, I had expected there would be one thread bottlenecking.
  5. I use IL-2 on a laptop with a 1070 eGpu, on a 1080p non-gaming monitor (60hz). Vsync never worked well for me, I get the (relatively) best results by setting graphics to 80 fps, combined with dynamic resolution factor (currently 0.7) Always I had problems with fps and spotting in multiplayer. After the recent update, and MS forcing the 1903 update on me, and performance further degrading, I installed the latest Nvidia driver, moved startup.cfg to another folder, and restarted IL-2, attaining a new startup.cfg, presumed default. I'm trying to understand properly what exactly happens in multiplayer that causes the performance to be so bad. On ultra settings, in single player, I can fly against 8 AIs, have them all around me in immediate vicinity, and get around 60 fps. In multiplayer, I see a strong degradation on the "big" servers, with big maps and many players. My spotting is terrible. So when I fly in multiplayer I usually go on "EU official", which has icons and smaller numbers of players. Even there the fps is much lower (than single player) with additional microstutters, but not as bad as on the 'big' servers. I find it kind of hard to grasp how and why multiplayer even impacts my GPU (i.e. fps and stutter). If it is the multiplayer data transfer ("netcode"), which causes a bottleneck, what is the nature of this bottleneck? I can only imagine it would have to be the cpu. I run on a mobile cpu (4 core, 8 threads), hence weak-ish. Yet when I fly on multiplayer with the resource monitor running, (and I get into a low fps situation), none of the cores shows above 70% load. If none of the threads is even fully loaded, how can the networking impact the FPS negatively?? Clearly it does though. The IL-2 multiplayer settings GUI used to have a field for upload and download limits, I believe. It's disappeared at some point? My new startup.cfg has got these, default, settings: client_download_traflimit = 1024 client_upload_traflimit = 1024 My old startup.cfg had: client_download_traflimit = 4096 Which I had set, based on some info on one of the threads, that you should do a speed test and divide that by roughly four, to determine the setting (or something like that). Is it still advised to edit those settings, even though they've been removed from the GUI? Does that old rule still hold?
  6. This just happened to me again, riding the 'new' T34, and it had happened a few times before, rarely, over a longer period. You ride around sort of recklessly, with the open hatch, and I have the head position adjusted to be as high as possible. Without any audible explosion, and in a location on the map where I know there are no paks etc. suddenly the commander is dead. First I think maybe someone has jumped into the comm. position and check, but no. Then I think maybe I got bombed somehow with the audio dead, so I turn 180 deg. looking for a bomb crater, but nothing. In the end I can't help wondering if maybe some physics of riding around are being modeled, where the commander is injured in the course of looking out of the hatch, while the tank is driving cross country? Did anyone else experience this? I know this sounds dumb, but I'd like to get to the bottom of it, since some possible adjustments/changes might result. For instance, if it were confirmed that commander health is being modeled with respect to the physics of cross country riding, I could adjust the head position and so on. I've often seen the health of crew members reduced when bumping into invisible trees, but usually the damage is more gradual and averaged out, with multiple crew members going yellow, instead of just the commander going red and everybody else green.
  7. I've not tried to run benchmarks. Also I don't have a 'regular' rig which would enable me to do comparisons. Guess I should mention, playing IL-2 wasn't an objective when I decided on buying the laptop/eGpu combo. From reading the various threads, and considering the issues I've had / have , around microstutters/stutters and disappointing fps in multiplayer, I get the impression that beyond the general expected performance hit compared with a normally installed 1070, the graphics issues which have (weirdly, to me) to do with multiplayer data transfer (netcode), seem to be particularly bad with eGpus. I've looked into whether the relatively weak mobile cpu might be the bottleneck but it appears not to be the case, at least according to task manager. So, for running IL-2, I'd advise strongly against getting an eGpu.
  8. Happened to me again, tried the position cycling thing, which fixed it. That's definitely the thing to do! Of course, in many situations, the time lost on this will at least have cost you the opportunity, or you might be dead.
  9. It seems the new system is pretty much like what they had introduced with the 'new' T34's periscope. I guess now all the tanks, and all the sights, get a chance to 'enjoy' that 'feature'. In normal, it seems they've added some useful yellow arrows, which unfortunately we don't get on the tank server. Also it seems they's added more instability, when driving, and after coming to a stop. The guns swings stupidly. I think it's this new feature which Java Keben has described as 'drunk'. It's extremely irritating. I don't know if it's a mechanical gun imbalance which is being simulated here, or the gunner's physiology, of not being able to hold the gun steady after suddenly stopping. At any rate, it's awful. It makes it a lot harder to shoot down the bleeding attack aircraft as well. I wish they'd fix the bloody invisible trees instead of screwing up everybody's aim. Increasingly it feels to me as if I was unknowingly taking part in a psychiatriatric experiment, to add frustration at every update, and determine the player's breaking point. Mine is up soon.
  10. Not yet. Strictly speaking, have not seen it in a KV either, and possibly not in the 'old' T34, not sure. I've only spent a few hours in online missons since the update, and those gun failures have been only a small percentage of that total time. I'm hoping that people who spend more time in German tanks can shed some light on this. It might well be a bug that's confined to the 'new' T34.
  11. On a few missions since the update, I've had stretches of time where I couldn't shoot the main gun. Tried various things like changing ammo which didn't fix it. On some occasions it mysteriously started working again at some point. Obvs I checked whether the gun was considered damaged but everything was green. I hope it's not some new physiology feature where the gunner got seasick from driving around or something.
  12. That, in a nutshell, is the problem. =D
  13. I wonder whether something can be done, in the misson design, regarding winnability, purely by aircraft? I was just on the Seversk (I think) mission, the one with the town bridge and the railroad bridge. I was a very well populated mission, plenty of tanks on both sides. The mission felt like it had just gotten underway--red had got across the bridge initially, then blue got across, but neither of the town bases had been taken by the opposing side yet. The mission seemed to have at least another hour to go until one side would take the opposing town base, and then the back base. Then all of a sudden it was over, because apparently one aircraft (apparently Zommer) had destroyed lots of back base targets. From a tanker's point of view, it felt like a waste of time. When the next mission (a winter mission) came up, most of the tankers (certainly all the red ones) didn't join, apparently just as annoyed as I was. I don't know if the game logic (or mission design logic) makes it possible to avoid such an outcome? Would it be possible to apply a factor (for instance 0.5) to the points caused by aircraft? Or to the points caused by destroying stats? Something like that?
  14. Can player kicking be caused by client software? We all know the different variants of players being kicked from a server due to bad ping. The most common one, one player kicked. Another one, all players kicked. Usually but not always in that case you find that, probably, the server crashed, and a fresh map is up. But it also happens sometimes that everybody got kicked and the map is still there. And you can find, after everybody got kicked, sometimes, the server doesn't show up in the list, which players generally take to mean that the server crashed and is still down. We just had a mission going on with more or less balanced blue/red, and the tide was just turning, with red taking an important base. Then it looked as if everybody got kicked with bad ping, but afterwards I saw the server on the list, re-joined, saw the map was still up, still in the same state, and spawned. I was IMMEDIATELY attacked by a blue plane at spawn which made me deduce that at least some blue players hadn't been kicked. I looked at stats and found I was the only red player. 6 blue players. No red players re-joined, which led me to guess that potentially the server wasn't visible to them, or didn't let them re-join. I continued for a while but no red players joined. Then I quit. This all leads me to wonder--are players using client software that can sabotage the server? Is this technically possible?
  15. I think/hope this is a joke rather than a serious suggestion. There must be tens of thousands of invisible trees in the various maps. A task for a purpose built detection tool...
  16. I never quite understood the rationale here, since the mission builder is also part of the whole IL-2 BoX package, thus this would be a bug for the devs either way. At any rate, to put this question to rest, I created an offline single player mission based on the virgin Kuban map, untainted by any previous editing. I know because one of the missions on the EFront server is in Seversk, that the East waterfront of the river there is full of invisible trees. I placed a tank next to the waterfront, started the mission, drove along the river for a couple hundred meters and, bonk, bumped into an invisible tree. Case closed, the invible trees are part of the base-map. The devs should look into this.
  17. yes, to different degrees depending on the tank. However, if you try it out, you'll see that you often have to aim faaaar next to a trunk to get past it.
  18. This keeps happening to when flying. However I don't record tracks or take videos, and there's always the possibility that I didn't see the plane, it was on my low six and thus hidden, and so on. So I usually don't press the point. However when tanking on a few occasions I have seen * tanks initially incomplete, with a missing (or transparent turret), which appeared later * tanks initially entirely invisible which then appeared, part by part. I want to stress that there is also a completely different thing going on with 'furniture' tanks (or other objects) , where the object is rendered differently and completely disappears as you zoom out. Then you zoom in and it's back. I've seen this effect with an AI stuka which kept disappearing every time I zoomed out and reappeared as I zoomed in. There must be a setting in the mission editor which allows for this, maybe to reduce load? But regarding the missing parts of player controlled vehicles. Would this be an indication that the online multiplayer is implemented in such a way that the actual 3d models of vehicles are sent? This seems hugely inefficient to me (because the client can have all the 3d models) but would maybe explain the struggles with the netcode on today's fast broadband connections when, already at the end of the 90s, I was playing warbirds and there were no such issues. If parts of vehicles are actually transmitted and rendered this would seem rather exploitable to me. In such a case, in the interest of fairness, I would suggest to have an implementation where, if missing parts can somehow be detected, a cube or something, the size of the largest vehicle, is rendered instead of an incomplete vehicle. Or something like that, which would make it easier in an MMO for the players to detect that something is amiss.
  19. Regarding the drawing: When you try to fire around a single tree in a probing manner, you don't always find the collision model of the trunk to be bigger than the trunk, to the left AND right. I can often shoot by the trunk, quite close to it, on one side, whereas on the other side I have to give it a lot of extra space. I.e. the model seems somewhat shifted relative to the trunk. I'm sure the way it's been implemented is NOT that way that each tree has a 3d collision model in that space which is checked. Rather there are optimisations going on which mean that if you're farther away your chances of shooting around the trunk are worse. When two tanks are facing off, and a tree is much closer to one of them, i.e. that tank is sort of half-hiding behind the tree, but able to shoot by the trunk, often the further-way tank is unable to do the same. As soon as multiple trees are in play, I believe there is some collision model simplification going on, which usually boils down to not at all being able to shoot in between the tree trunks. When you're fighting a Pak or an AI tank which is able to freely shoot through entire forests, and you can't reach it although you have a clear line of sight, this is particularly annoying.
  20. Some thoughts regarding the criticism of players using the 'old' T34. First I think it's silly if someone who always tanks blue, tanks red for 1 day or so, and then declares the periscope sight of the 'new' T34 to be perfectly fine. ... and thus implies that those who tank red all the time, and who have been complaining about this periscope, must be making things up or having ulterior motives. The problem with the mouse and the periscope sight is not, that it's impossible to aim or that, when *aiming*, the turret behaves in an erratic way. The problem is that : You have to zoom out quite far to see the black outside the view circle, which gives you visual feedback regarding whether you are moving the mouse too far, left or right. OTOH, when being zoomed out, the mouse response is hypersensitive. I.e. you might want to zoom in just to calm the mouse response down, but then you lose the black rim which tells you if you're pulling too hard into one direction. As Jordan (I think) pointed out, if you do pull too far into one direction, you've now pulled the 'desired' view direction by more than 180deg and you're pulling into the other direction. This is the view behaviour of looking around in space, for instance, looking around inside the tank when not using a sight, and it's inappropriate for sights in my opinion. I think there's no good reason why anyone would desire this directional reversal behaviour, and I regard it as a bug. In contrast with this, the mouse response of the 'old' T34 is smooth, in no way weird, and feels completely natural. Until recently, I was using a razer gaming mouse and now have a steelseries, but the problems I have with new T34 were the same with both mice. An advantage my new mouse has is a button at the top to change DPI. This makes using the new tank's persicope a little less awful. A thing that really annoys me is the assumption that players use the old T34 because of an assumed deficient DM. In most missions there are plenty of tigers. I find it pathetic that blue are trying to cast themselves in the role of the underdog here. Once we have mostly missions without tigers that would be different, but I can already see that blue players just leave when there are no tigers on offer. So as long as I'm fighting tigers, I will take the best tank available to me, and that's the 'old' T34 because a) I can turn the engine off and listen, and still turn my turret with reasonable speed, and b) I can look around with the hatch open, have the turret pointed in the right direction, close the hatch and the turret will still be pointing in the right direction. That the devs chose to make the turrets of the new tanks point to 12:00 whenever you open the hatch is something I complained from day 1 and I think it's a huge mistake.
  21. I'm sure the devs know the details, since the sound bug is so extremely prevalent, especially when tanking. Nonetheless, just in case, some details: The likelyhood of sound bug occurring within a mission goes up just through the passing of time. However, what REALLY increases the chances, is getting hit. So, what you see in multiplayer on the tank server is that a lot of players, every time they got hit, died, and would normally just start a new mission, leave server, restart the game, and then start a new mission. That's how extreme it is. Just now a rare variant of this bug occurred to me, maybe that sheds some light. I was tanking in a summer mission, dry weather, had got hit but survived, drove on for a bit, opened the hatch and shut down the engine. Normally in this situation, when having sound bug, you hear maybe nothing, or a reduced set of the sounds that should be there. I heard rain. Quite clearly. I even took off my headphones to check if it's IRL, but it wasn't.
  22. I'll try this for a bit. I think there a several variants of this UI. The one I have doesn't have a place to enter such values.
  23. I only get the "master server" dialog very rarely. But when I got the "game server" message, when I was in the list, it must have been a misleading dialog, since in that situation there is no connection to individual game servers. It must have been the central server. At any rate, that dialog appearing while being in the server list was unusual, whereas the in-mission discos are extremely frequent now. I have been (in the last weeks) kicked far more frequently than ever before, also when flying, so this appears not to be tank specific. What I've forgotten to mention in the first post is that this occurs on servers where I don't in any way have a slow ping. I checked my multiplayer settings because I wanted to make sure a setting I had once set was still there ... there was a rule you should check your transfer speed on one of the test sites, and then divide that by something, and enter it into the setting for multiplayer transfer, something Mbit... I now saw in the settings that this config has disappeared. Apparently the game attempts to calibrate itself in this regard? Could it be that this change is connected to the issues? Two observations which might have something to with it (I'm suspecting that some bulk data transfer happens which screws up the ping): When tanking, in recent weeks, let's say I drive from A to B at top speed, and suddenly the game freezes for half a second or so. Sometimes, at this point, disco happens, other times, it recovers. This type of freeze is completely new. I used to have multiplayer stutter or microstutter which mostly went away by setting no vsync and 80fps (I set that some months ago), but this is entirely different. When my tank is destroyed, involving fire, it now often happens that I'm already in 'spectate' for a few seconds, i.e. nothing much graphics wise is happening, but then the ping message appears and i get kicked. I suspect that also in this situation a bulk transfer kills the ping.
×
×
  • Create New...