Jump to content

ApexGS

Member
  • Content count

    27
  • Joined

  • Last visited

Everything posted by ApexGS

  1. The magic smoke inside your computer leaking out would be pretty bad. Speaking of power outages though, I had an extended outage a while back that cleared my CMOS battery so I had to redo all of my BIOS settings and overclocks from memory. I always write those details on a sticky note now.
  2. For the suggestions of splitting up the hitboxes, that's very doable but the problem may be in how it scales. You can use traditional hitboxes or collision volumes, but you're adding one additional component to every single player in the game world, not to mention the likely scenario of applying the same concept to vehicles. UE4 handles that sort of thing exceptionally well, but I've not done anything on UE4 with 50-100 players concurrently so I have no guess as to how an additional collision volume on 100 player entities would go for performance Edit: I forgot to mention there are excellent techniques in UE4 for using physical portions of a body in place of a traditional hitbox which might have minimal if any impact on performance. On the ballistics side you have to remember EVERYTHING in videogames is smoke and mirrors. Projectiles are not traveling their real world velocities, they're using scaled approximations of it. If a physical projectile was simulated in a game engine traveling 2900 feet per second (mapped to the game's unit scale, which is not 1:1 due to field of view warp effects) it would hit calculation precision errors, skip collision checks between frames and all sorts of other crazy things. So always keep in mind that any sort of ballistics simulation has to be approximated using clever tricks... it's less about "how do we simulate tumbling" and more about "what low-cost calculation can give the general feeling or effect of tumbling", for example. Good discussion though, even if we pretty much know what the game is aiming for already.
  3. That's because it's a 7.62x39, not the 5.45x39 AK-74 in SQUAD I could spend an entire post ranting about videos like that one; people giving guns to inexperienced shooters for entertainment value pisses me off. Meanwhile, someone who knows what they're doing running an AK-74 makes it look like the gun has zero recoil... Speaking from experience the 5.45 cartridge is a lot of fun to shoot because of the minimal impulse, it's a real shame some jackass got the surplus ammo banned from import to the US.
  4. It's not unrealistic given the variety of factors at hand in a combat scenario that two or more 5.56x45 or 5.45x39 wouldn't outright kill a target. Luckily this also helps the gameplay aspect, because one shot kills are rarely fun or fair to the majority of players. We already know from Z-man's discussions on the damage model that caliber, external ballistics, velocity over distance etc will be factored in as the core systems are fully fleshed out. Several of those are already in one way or another. Plus the incap system and medic mechanics as mentioned above!
  5. DirectX 12?

    The DX12 support in UE4.9 is a development preview, not a feature complete engine build. It would be crazy to try and jump on that during active alpha testing right now, when it could introduce tons of new issues to a renderer that already has a few kinks to work out from last weekend's build. Give it some time, folks! Looking toward Steam Early Access would be a more realistic timeline, if even then, depending on Epic's own work finalizing a stable 4.9 publish. You'll get your framerate! Patience!
  6. Type of reload

    There is currently no partial magazine retention, so you will lose the ammunition remaining in a partial mag during reload. Mag retention has been brought up a few times and discussed, but there aren't any plans for a system like that for the time being.
  7. Does lowering the graphics give you an edge?

    Resolution scale in UE4 has an enormous impact on performance, and some games use low-impact post process antialiasing and sharpening filters to offset the loss in quality moderately well. ARK for example was awful for most users in the first few weeks, even on 980ti/Titan X cards with the resolution scale maxed, but scaling to 80% or so with post process AA looked decent and ran far better. Ideally though you try to pinpoint the bottlenecks in rendering and optimize it whenever you can... but that's much easier said than done, particularly when graphics aren't the #1 priority early in development. As to why it's there... mostly a "cheap" way to get games playable on low end hardware without (or even in addition to) physically removing objects or simplifying models through aggressive culling/LOD. It's a catch-all alternative to having different resolution textures in the game, since one setting can reduce quality of all textures in one go.
  8. What Games Do You Play During The Week

    I'm heavily involved in modding 7 Days to Die and finishing up a playthrough of Crysis 2 on Twitch. I bounce around to a lot of genres and still have tons to do in Witcher 3, the latest Terraria update and have been keeping an eye on ARK to see if they sort out the issues I have with it. I keep planning to revisit Insurgency and Payday 2 but never seem to get around to it.
  9. I wont attempt any shots over 30 meters.

    There are two main ways of setting up your camera's view and your character's arms and weapon in first person. The style you see most often dating all the way back to Quake and Unreal is simply having a camera that has a pair of arms and a gun attached to it. In games where you can't see your body or feet, this is what's happening. The character model that everyone else sees in the world is completely separate from your first person view arms and gun. In a game like Counter-Strike, these are "v_model" and "w_model" for View and World respectively. The second and more recent style is a true first person controller which places the camera view where your character's head is. Your arms and gun in first person, along with your body and legs, ARE the character that everyone else sees in the world. It sounds like a no-brainer to set a game up this way, but it presents a whole host of challenges. Your character's animations need to be suitable for seeing and using your gun in first person, and since none of the animations are isolated to just your first person view they have to look right from both perspectives. Any change to the character model or animation has to gel with the first person view as well. Unlike "floaty arms" you can't just change some coordinates to shift the arms around relative to the camera! That's the gist of it anyway, I'm sure Z could go into way more detail about it. Hopefully that at least gives a decent explanation of the two!
  10. I wont attempt any shots over 30 meters.

    Hey this isn't my first alpha, pre-alpha or first time working with UE4 I like the discussion threads but people always have to remember not to veer off of discussing possibilities and into insisting or demanding things, you know? This is the sketch that becomes a painting, or the primer-covered body that becomes a show car. I'm damned impressed with it already, which says a hell of a lot from a guy who bitches unrelentingly about FPS games.
  11. I wont attempt any shots over 30 meters.

    It's all good man! I'm definitely not pushing, just throwing out some discussion in between playtest weekends I'm definitely looking forward to the "arms length" tweaking in the future and testing out any experimental setups you guys come up with. I'm 110% in favor of seeing some new ways to go about it, the tired old fixed-crosshair FOV-solves-everything model has worn out its welcome many times over. I personally don't mind FOV shift in moderation when used properly, which is why I brought it up. There's just no easy way to simulate eyeballs and peripheral vision on a flat screen, eh?
  12. I wont attempt any shots over 30 meters.

    Definitely not a simple solution, no doubt about that. Juggling the camera controller to handle both real worldmodel animations and adjusting it to fit the first person viewport is tricky to say the least. I like the current focus system a lot, it can only get better with fine tuning later on. I'm sitting here in the shop sighting down a couple of AR builds with various optics and trying to visualize the best way to handle it to give reasonable visual acuity on a flat screen. My biggest concern is the "pixel hunting" aspect of hardly being able to see, none the less sight up and put rounds onto a target more than 40-50 meters out. In the real world this isn't a problem, irons or optics alike, since we're not limited to the pixel density of a flat display. I know there's significant and justified concern about making your average rifleman too effective at extended ranges, but I think the sway mechanic and stamina already take care of much of that. Sorry if I'm getting ramble-y and not being clear, it's been a long day working in the northeast heat wave
  13. Good deal, I think that alone will help some of the "floaty" feel with aiming, especially when using focus mode. I found myself not bothering to use my DPI switching because frankly I have enough going on with my comms and weapons being on the mouse (Naga 2014) that there's no comfortable place to throw a sensitivity clutch. I'm already probably crouching, shifting around, talking on comms, aiming and focusing down a target, my thumb can only hold so many keys
  14. I wont attempt any shots over 30 meters.

    I believe that a small amount of FOV shift for basic ADS use is a vital component in a shooter, for the sole reason that it's the only reasonable mechanic we have on a flat screen to simulate hard focus tunnel vision on a target/sights. You need to be able to see the sights, and having an 80 pixel wide Aimpoint lens with a small dot isn't helping you see the target with the limited amount of pixels we have to work with in the viewport. Conversely you can pull the gun in closer to the screen, but I don't know how your camera controller works I'm definitely looking forward to the next weapon pass and tweaking the ADS. The focus mechanic is really solid right now, though I would prefer it still be usable while moving (limited to a walk, and sway is still obviously a factor) since it likes to knock you in and out of focus while shifting positions slightly since it conflicts with sprint. I'd just make focus take precedence over sprint and require the player to let off of it and press the key again to sprint. If you guys can hammer out the "holding rifle with arms extended" feel, using Spectator6's apt description above, it will vastly improve the feel of the game. All in good time!
  15. I was recently dragged back into Planetside 2, after many months of playing and getting fed up around BR96 with the aforementioned issues. I spent plenty of time on the subreddit (which is a joke these days) bumping into Roy that I can absolutely say he and team have a firm grasp of what's wrong with other combined arms games and their shady business models I only had one single map round over the weekend in which no one on my team used comms, and it was probably after a server reboot where everyone was taking a break and re-forming elsewhere. Every time I joined one of the NY or Virginia servers (I'm east coast and dragged my friend over despite the ping) they were full of people using comms, coordinating, calling fire with the compass heading and just plain doing the things you'd expect of a squad. It was awesome, even in this early state!
  16. A discussion on Sights

    I've been doing a Crysis 2 livestream playthrough on Twitch, but I don't keep VODs saved typically. I'll record a video with some light commentary on the mechanics if I get a chance this week and link it here. It's a real shame the third game dropped most of the things that made the weapons feel awesome to use, but then the third Crysis kind of stopped making sense anyway.
  17. I wont attempt any shots over 30 meters.

    Sometimes it's fun trying to explain to people the difference between a controlled pair and a "double tap" The things holding back controlled pairs and accurate shooting in general are the sights and camera controller setup right now, which seem to be high priority along with prone. Hopefully we'll see a lot of improvements sooner rather than later! Even with the sway I enjoyed the weekend playtime tremendously and can't wait to see things brought in and quickly iterated through UE4.
  18. Focus

    I'm on the fence about it, and probably won't get too involved in debate over it until the next weapons pass cleans up the first person camera controller a bit. I posted some thoughts in the optics/sights thread regarding how games attempt to work around how the human eye works (e.g. tunnel vision effect when hard focusing) and there's not too much you can do outside of VR headsets to mitigate the FOV zooming. Without it, you can't see a whole lot beyond vague blobs outside of say 50 meters, and the game becomes very pixel-hunty so to speak. A big part of gameplay in the future will be supported firing positions and cover system, which also dictate much of how the first person camera will work in regards to sights and long range accuracy. Hitting a man sized target at 100 meters with a supported position isn't tough with irons in the real world, and we already have the sway mechanic and stamina to cover battlefield stress. Optics should absolutely be vastly superior to irons at ranges beyond 50-ish meters, but they should also not be categorically worthless if authenticity is what we're aiming for!
  19. Weapon Jam With Tactical Reload.

    To further support the well-maintained firearm point, the reliability standards of most firearms are stringent. Here's the actual mil-spec for the M16A2 and subsequent variations taken from Black Rifle II: Nine failures in six thousand rounds is the maximum allowable for acceptance into service, that's pretty excellent reliability if you think about it! If you consider your round count in any given round, and that dying and respawning effectively resets that... there's an incredibly slim chance for a decently maintained rifle or pistol to fail under those circumstances. It's a good discussion to have though, and I WOULD like to see a game fully handle malfunctions with proper animations and mechanics that didn't hurt the game. I just don't think Squad fits that, and that it would be much better suited to something open world and survival-esque personally.
  20. A discussion on Sights

    The biggest problem with sights in games compared to years of real world experience is trying to deal with how the human eye works and applying it to a flat display. The tunnel vision effect of focusing hard on something like a target you're trying to shoot is challenging to simulate, to say the least. The primary way most games do it is by over-zooming the FOV and pulling the gun and sights much closer to the screen, which in some cases isn't as bad a solution as it sounds on the surface. Right now the pistols are great but the rifles are held out and away from your body in a similar fashion, which I suspect is one of the first things being changed on the weapons pass. That's a huge part of the reason the sights are so tough to use. I'd take a look at the second Crysis for how to best pull off depth of field on weapons and hands. The game had its share of flaws, but its gun mechanics and visuals are not among them. They had an excellent animation setup with "loose" ADS / sight misalignment while moving or snapping between targets which felt excellent compared to most center-fixed crosshair games.
  21. UE4 has the same capabilities for interactive foliage sim that CryEngine has, if not better. It's all up to the implementation and the assets utilized, so there's tons of ways the team can go forward in the future with improving the foliage. ARK for example has physic sim foliage which is fairly CPU heavy, but it was introduced right off the bat and remains part of the reason the game runs so poorly. All in good time, eh? I can imagine seeing a squad advancing through the poppy fields with the plants bending and swaying out of the way though. We'll have to see what the future holds! Edit: Also back to the original topic, try ReShade/SweetFX with that one guy's tweaked Squad config. I just set it up myself and should be testing it out this afternoon, the screenshots look great though. The thread was linked from the post-process sharpen filter thread.
  22. Weapon Jam With Tactical Reload.

    I'll be blunt, America's Army did a TERRIBLE job of doing malfunctions. It's been a very long time since I played it so things may have improved, but as I recall they had basic RNG chances, "click click click" (you know what I mean there) and a single vague fumbling-with-gun animation to clear it. The only games that have done worse jobs are STALKER ("press reload to clear!") and Far Cry 2. I'm of the opinion that if you're going to shoot for the kind of realism that requires handling malfunctions, you should have type 1, 2 and 3 and animations to go with them. There are reasonable ways to simulate malfunctions without being pure RNG, but the question then becomes dev time cost versus gameplay gain, which tends to be a very poor ratio when so many other features take precedence. I also don't really feel like it's especially useful to add in a respawn based team game like Squad, though it could potentially add a lot to a persistent open world shooter e.g. Fallout, STALKER (if it were done well). Just two cents from someone who's pondered the concept for many years and tinkered with it in UE4.
  23. I have to second the quality of the sounds, and I'm a stickler for them being in the gun industry. If you ever have need of a source for various handling sounds, I've been thinking of picking up a quality studio mic and recording various guns that come through the shop. Always been a big fan of your sound packs for various games, Anders. Keep up the good work man.
  24. I support the muffled audio in some form, however it's decided works best overall. Perhaps the muffled/shifted game sounds combined with a dying heartbeat to really drive home the death screen. Comms open is fine to me personally.
  25. The sounds...they are so good...

    I think the reload sounds could use some work and a volume bump (not too loud, we're comparing them to gunfire after all) but I'm really pleased with the vast majority of sounds. That's something very few games even try to get right, leaving aside the difficulties in recording gunfire and explosions. Equipment rustle and general handling sounds are tough to get right as well. I swear one day I'm going to pick up a studio mic and start recording the guns that come through the shop or that I own personally. The sounds in Squad are just fueling my OCD with other games doing it wrong.
×