Jump to content


  • Content Count

  • Joined

  • Last visited

  • Medals

Everything posted by dnk

  1. [TABLE=width: 700] [TR] [TD]Weapon[/TD] [TD]Helmet Head[/TD] [TD]Plate Carrier [/TD] [TD]Tac Vest[/TD] [TD]Unarmored Chest[/TD] [TD]Arm[/TD] [TD]Leg[/TD] [/TR] [TR] [TD]7.62 ABR[/TD] [TD]17/45/100% (2)[/TD] [TD]29-35% (3-4)[/TD] [TD]35-37% (3)[/TD] [TD]100%[/TD] [TD]25%[/TD] [TD]44-50%[/TD] [/TR] [TR] [TD]6.5 MX[/TD] [TD]32/100% (2)[/TD] [TD]10-25% (5-6)[/TD] [TD]23-30% (4)[/TD] [TD]100%[/TD] [TD]17-22%[/TD] [TD]31-36%[/TD] [/TR] [TR] [TD]5.56 SOAR[/TD] [TD]17/100% (2)[/TD] [TD]7-15% (5-7)[/TD] [TD]7-17% (5-7)[/TD] [TD]100%[/TD] [TD]5-10%[/TD] [TD]16-19%[/TD] [/TR] [TR] [TD].45 Vermin[/TD] [TD]22/100% (2)[/TD] [TD]10% (7)[/TD] [TD]12% (5)[/TD] [TD]100%[/TD] [TD]15-25%[/TD] [TD]25%[/TD] [/TR] [TR] [TD]9mm Pistol[/TD] [TD]10/100% (2)[/TD] [TD]5% (9)[/TD] [TD]7% (6)[/TD] [TD]100%[/TD] [TD]10%[/TD] [TD]5-13%[/TD] [/TR] [/TABLE] UPDATED WITH 556 STANAG Here are the rough damage ranges for each weapon at close range. Generally, a second head shot always kills, as does a 4th or 5th chest shot. It seems scripted in that way, since even the 9mm can kill in 6 shots. The SOAR is probably bugged, since EVERY shot has the almost exact same percentage (usually there's a random factor), and it's always too low. After extensive testing in editor, these are the values I came up with. You can set it up in the editor easily. Run: [this] execVM "hurt.sqf" for each soldier you place, then have a "hurt.sqf" in the folder with this: Note that I said it was scripted (senso lato), in that after X shots given a certain armor, there is a very high likelihood that the next shot will be fatal, regardless of current damage. So after 4 .45 shots the health is still ~50%, but the 5th shot gets a big bump up to 52% to finish him off almost every time. I'm not sure if this works for arms and legs, though. Not sure I like this, since I'd rather the values were higher overall without some scripted "X shot death" thing.And in general I find the values a bit low for my taste, particularly for PvP. The table is for close range, and the lower caliber bullets lose a lot of their killing power at range. Yeah, 6+ shots even from 6.5 at 300m for a kill probably is common now. I guess that's somewhat realistic with 2035 body armor... not a huge fan for gameplay, and I don't like the non-1 shot helmet, even for 7.62. I get the feeling a helmet isn't stopping such a round from killing or severely incapacitating you even at a distance. I want to echo others who have said that if this is the way BI wants to go with the damage system, we NEED some sort of effect for being hit, like falling down or something more than a small jerk/nothing as it currently is. Turning everyone into bullet sponges makes for tedious gameplay, even if it means you live longer (which is always nice, except when "you" is the guy I'm shooting at :P).
  2. dnk

    Damage system sucks - fix needed

    UPDATED Here are the rough damage ranges for each weapon at close range. Generally, a second head shot always kills, as does a 4th or 5th chest shot. It seems scripted in that way, since even the 9mm can kill in 6 shots. [TABLE] [TR] [TD]Weapon [/TD] [TD]Helmet Head[/TD] [TD]Plate Carrier [/TD] [TD]Tac Vest[/TD] [TD]Unarmored Chest[/TD] [TD]Arm[/TD] [TD]Leg[/TD] [/TR] [TR] [TD]7.62 ABR[/TD] [TD]17/45/100% (2)[/TD] [TD]29-35% (3-4)[/TD] [TD]35-37% (3)[/TD] [TD]100%[/TD] [TD]25%[/TD] [TD]44-50%[/TD] [/TR] [TR] [TD]6.5 MX[/TD] [TD]32/100% (2)[/TD] [TD]10-25% (5-6)[/TD] [TD]23-30% (4)[/TD] [TD]100%[/TD] [TD]17-22%[/TD] [TD]31-36%[/TD] [/TR] [TR] [TD]5.56 SOAR[/TD] [TD]17/100% (2)[/TD] [TD]7-15% (5-7)[/TD] [TD]7-17% (5-7)[/TD] [TD]100%[/TD] [TD]5-10%[/TD] [TD]16-19%[/TD] [/TR] [TR] [TD].45 Vermin[/TD] [TD]22/100% (2)[/TD] [TD]10% (7)[/TD] [TD]12% (5)[/TD] [TD]100%[/TD] [TD]15-25%[/TD] [TD]25%[/TD] [/TR] [TR] [TD]9mm Pistol[/TD] [TD]10/100% (2)[/TD] [TD]5% (9)[/TD] [TD]7% (6)[/TD] [TD]100%[/TD] [TD]10%[/TD] [TD]5-13% [/TD] [/TR] [/TABLE]
  3. This isn't devbranch specific, but I'm pretty sure it's a mistake from the current main build patch: 556 SOAR is the weakest gun in the game. Weaker than the 9mm pistol even. It takes 20 shots to an armored chest to kill, whereas even the 9mm can do it in 5-6 shots every time. Clearly this is wrong. Unarmored shots still only do 10% damage to the chest. Basically, don't use 556 until this gets fixed.
  4. dnk

    Daylight fog too blue?

    I'd also like to see realistic (gray) fog. The blue looks SO gamey.
  5. Yeah, it would be nice to choose an armor loadout, especially once BI gets the weight/fatigue system implemented (where there are real tradeoffs to weight).
  6. wait, so these are tweakable in a file somewhere then?
  7. It was not clear, no. Good luck with that then.
  8. First off, people should note that this was originally started in the recoil thread, taken to the dev fatigue thread, and then appropriately moved to its own thread when it overgrew that thread. This was all suggested as an alteranative to the current recoil mechanism of "shooting to the sky" if you didn't constantly adjust, which many players have complained about (I'm less concerned about that aspect of it). One alternate was random direction for recoil, but in the original thread it was complained that this felt like "fighting against the avatar", which I tend to agree with. The purpose here is basically to simulate the soldier and negate the gamey tactics that have come to dominate play, particularly online. Regarding inaccuracy/weapon sway. I came up with this idea (well, copied a bit from another game) for changing the current recoil/inaccuracy system. It's related to fatigue, and since you're working on this already, felt I might mention it here as well. I'd just settle for real inaccuracy [rather than recoil], and have that increase substantially for X milliseconds after a shot is fired, cumulative, up to some hard upper limit. So, each weapon has a set base inaccuracy (much higher than current, since it's not only weapon-specific MOA but also the shooter's non-super-delta-force-ability/stress), which is adjusted for stance, health and fatigue. Then, each weapon also has an amount of inaccuracy increase per shot. Then each weapon also has a set reduction of inaccuracy per second, as well as a maximum inaccuracy (you're never going to accidentally shoot 90-degrees to your intended target due to recoil). Example: That makes way more sense, I think. You could add a camera-only shake/movement (ppeffect) for each shot (that doesn't move the cursor, and which will automatically return the camera to be centered on the cursor after X amount of time). That would add to immersion and at least be slightly disorienting vis-a-vis the sight picture, without requiring constant micro-adjustments of the mouse.
  9. This would be pretty cool.The community really does need like a "new features" addon collection at least. Something that takes in a lot of community wishes and creates a standardized package. Something that has: RWS Weapon resting scripts ACSE / JSRS ASR AI TPW mods Something like that, I'm sure I missed a few. Aside from the sheer boredom of playstyles/missions, I can't play big COOPs without these things since otherwise it just lacks the necessary immersion/challenge to make it worthwhile. For me it's generally SP+mods or PVP.
  10. ^I guess setting up the recoil based on "Regular" values, since that seems to be the server default as well. Of course, many players might not want ridiculous recoil AND no peripheral dots AND super AI, but the options to fully customize the playing experience are lacking quite a bit still...
  11. It pains me because this looks awesome, yet almost no public servers will use it or allow it. Anyway, as I said, looks awesome! Will this have any effects like limping (speed limitations at least), increased weapon sway, increased fatigue (fatigue should increase faster and have a higher minimum value with bloodloss), etc?
  12. dnk

    Realistic damage

    Easy to script. Run it in a loop somewhere: {_dam = getdammage _x; if (_dam > _x getvariable "currentdam" + 0.2) then {_x setdammage 1}; _x setvariable ["currentdam",_dam];}foreach allunits; Change 0.2 to whatever you like, but that would be about an arm hit. Alternatively, just to increase damage (3x modifier): {_dam = getdammage _x; _diff = _x getvariable "currentdam" - _dam; if (_diff > 0) then {_x setdammage (_diff * 3)}; _x setvariable ["currentdam",_dam];}foreach allunits; That may or may not work... If the mission lacks a loop, a crude way is to just add while{true}do{ xxxxx}; where xxxxx is all the code above. Add that to each unit's initialization, or add it to the init.sqf. For a quick 3min script, it may cause some performance issues, and for more complex missions might be an issue. all_units = allunits - player; {_dam = getdammage _x; _diff = _x getvariable "currentdam" - _dam; if (_diff > 0) then {_x setdammage (_diff * 3)}; _x setvariable ["currentdam",_dam]; all_units = allunits - player;}foreach allunits; That may exclude the player from this effect. Or cause your computer to explode. One of the two.
  13. My heaviest core is around 50-65% I think. Total is around 40-50%. Hell if I know, but apparently something. I don't engineer the things, and any attempt to explain a possible reason would be the most half-assed of technical speculation: Could be more complex or shared cache management. The AMDs in particular have shared caches and front-end components, which may be where the performance is being hit most (especially for AI). Lots of little overhead and latencies resulting in large waits per frame, resulting in lower performance in FPS. You'd need far better observational tools than the freely available MS technet stuff to get at that, though.Additionally, Arma might have lower performance for these processors since it's all still mostly limited by the first core, which might be less of an issue with other games, especially ones without the complex AI that Arma has (which all seems to run off that 1 thread on that 1 core and holds the rest of the simulation up). Other games wouldn't have such an effect, and all the cores would be used more, making it seem comparatively that these 8c processors were underperforming considerably in Arma. Let's keep in mind that a lot of low-performance issues are subjective as much as objective - players expecting X FPS at Y settings based on experiences with other games, then complaining when they get 1/2 X or have to use 1/2 Y or something of that nature. And then there are clear cases where something's just wrong and it's not just subjective... but a lot of complaints are seemingly subjectively based on expectations, and as such this lack of real support for more than fully utilizing 2 or 4 cores can create a "performance issue" for the user, even if objectively their system is running as would be expected for the engine and quality settings.
  14. AND WITH LOWER QUALITY SETTINGS WHAT VD CAN YOU GET?Seriously, I've spent more time responding to you than it takes to change your ULTRA ULTRA settings down one or two notches to test this quickly. You especially set the object quality to high and you have a monstrous geometry to calculate and monstrous amounts of data to reference at each increase of VD because the models drawn out into the distance remain pretty high detail (with high detail textures to boot). On the lower settings you get LOD switches pretty quickly with distance, plus a lower %age of models shown, which makes a HUGE difference when you increase the area of the scene from 2Mm2 (1500VD) to 11Mm2 (3500VD). I have no issues with a 3500VD because my object/tex quality is on low/standard always, and the models are mostly not displayed beyond 2000m and those that are have very low poly counts. In fact, there's no appreciable difference in FPS between 1000m and 3500m VD for me, save when I turn quickly and need to suddenly load a bunch of new models/textures/terrain up (and on the lowest quality settings, this is not much of an issue). Nor do mine, and I have an i5 3350P. Some sort of latency has to be the issue. Whether it's RAM or CPU architecture related or graphics memory bus related I don't know. Clearly, the simulation has a lot it needs to do, but something is holding it up from fully using resources. What is it?You can explain further how the filemapping API gets around RAM latency. I would be interested to hear about it.
  15. Yes, this stuff really bothers me. I get that "the dots" would not be possible on higher difficulties, but why in the world aren't they toggleable in the lower settings? I usually run lower settings because they give me all the options, but then I'm stuck with immersion- and challenge-breaking dots and now I don't get to choose my preferred recoil either... I have to choose to have either super-accurate AI without dots and high recoil OR realistic-accuracy AI with dots and low recoil. I have no other options.This sort of stuff is a lot less excusable. Just add a bloody check box or slider for these "features"...
  16. Have you tried lowering the settings other than resolution and a few minor settings? That's all I noticed you mention... The reason your hardware might not be doing anything is because you have it loading up a TON of data every frame. I specifically mentioned texture and object quality settings, but VD is important also. Have you tried even turning all your settings to the lowest option yet, then moving each up to see where it starts to eat into your performance too much, then balancing your quality/performance to taste? It seems like you haven't from what I'm reading...
  17. Right, and the CPU runs at 40-50% usually for me. Why is that? fanboy fanboy fanboy!Fact is like 2% of gamers have more than 4 cores. Anything more complex and far rarer than a standard (2 and 4 cores) is "exotic" in my book. What's your non"fanboy!" definition of "exotic" when referencing computer hardware? I'm not saying all players with these have problems, but certainly when I see these performance threads, usually there's some oddity in the rig, something that you'll never see outside of the ~3% of hardcore computer users. They're "oddities" specifically because they're extremely rare in the general gaming population. Certainly, they may seem more common because the enthusiast custom builders are far more likely to take part in forums, and far far more likely to complain if they think the game isn't living up to very high standards of quality settings (not actual quality, mind you) and FPS.
  18. Where did I say 4c was exotic? SLI/Xfire is exotic however, simply because it is more complicated than a normal rig. I can find no stats, but I assume that it's also far less common than a single card.
  19. dnk

    We need 64 bit support !

    It really is necessary for the next installation. Probably necessary for this current one also. I don't think I could ever run certain settings on their highest levels because 4GB would be totally insufficient to handle all the data without lots of hiccups. The time has passed where the 32-bit addressing limitations are no longer acceptable. I can understand why the switch wasn't made for A3, given how stretched their resources were, but it's a major issue now that we're still stuck in 32-bit space, and I think quite a few performance complaints boil down to RAM usage (or lack thereof). Point is, performance may already be heavily constrained by RAM limits, and any future releases are going to see that constraint only tighten much harder as quality expectations (and RAM requirements) rise. They need to spend a lot of resources now on making the transition for the future.
  20. No. I'm just claiming they didn't take into account exotic hardware when designing things and/or that exotic hardware may have issues with the bottlenecks of the game (latency/bandwidth/memory, I think). SLI/X'fire further complicate this particular issue, which may just lead to drastically lowered FPS. Likewise, 8-core CPUs won't see any major performance boost over an equivalent 4- or even sometimes 2-core since the game is mostly limited by just 1 of the CPU cores for the main thread. Given that things like SLI, >4-cores, and such are very uncommon setups, it's understandable that BI didn't spend a lot of resources on designing for them.I don't use the term "optimize" since it's a very vague and nearly meaningless buzzword, so don't ever read that into my statements.
  21. UltraUltra Ultra Ultra Ultra Ultra Ultra Ultra 3800VD Particularly texture and object quality should be tried on Standard or High at most. How many GBs in that GPU? Less than 1.5GB and I wouldn't go higher than High for texture quality. It gets eaten up quickly. This game is horribly capped by bandwidth/latency issues. That's what I've gathered from months on these forums and personal testing. The core/CPU don't get used so much because of all the waiting around for RAM and data to get transferred from it to the processing centers, and then handled by those. When you set everything to "Ultra", you're creating massive loads of data. Also since it's all one one thread basically, any little wait caused by the engine can hold everything else up, leading to this underutilization. If the GPU core isn't being fully utilized, why is that? I can think of one obvious thing, the most simple answer: it's having to wait too much for a bottleneck somewhere in the data stream. Might just be all that ULTRA QUALITY data you're sending at it. How big are all those files...? Also make sure your graphics driver is up to date. It made a considerable impact in some cases for me. Certain performance issues I was having that weren't clearly caused by the GPU were fixed all the same. PS This is exactly what some posters have been saying: your quality expectations are way too high. You can't just set everything to "ultra" and a big VD and expect it to run at 50FPS on any hardware currently available. If this isn't the case, someone show evidence to me.
  22. How much of a difference is there IRL between the US and Russian armies? I'm not aware of 556 being a whole lot different from 545 or the 30 rounds in the AKs being much different than the 30 in the Ms or the hand grenades in the Russian pouch being different from the ones in the US.
  23. dnk

    Multiplayer dying fast??

    @Bvrettski @ Varanon/NeuroFunker Or because the COOP community hasn't moved over 95% of their missions OR ACE and wants to keep on with A2/OA. That's what the A2/OA numbers show, mostly COOP players. That's not BI's issue or whatever, it's just understandable since many of the communities and missions would have a hard time moving over to a new game with or without a flawed MP (and it's not totally flawed by any means). They will move slowly, and much faster when an ACE-like mod comes about, but that can take literal years.
  24. dnk

    Development Blog & Reveals

    Well, the fictional universe is one where the Iranians have continued to advance while the West has stagnated and both sides have reached a rough parity. Having a rough balance therefore is expected. You have to look at this game as more of a Russia-USA matchup as in A2, than Insurgent-USA as in OA.They'd need to have more assets in the game to start imbalancing things, otherwise players would complain that one side is missing an X.