Announcement

Collapse
No announcement yet.

Comparing UT4 to UT3 for perspective

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #91
    The flak primary is also subject to a lot of convoluted tooling which lowers the skill ceiling. The projectiles start out with a much larger collision sphere out of the muzzle, which shrinks over distance. And they are still using the flawed "use too much damage out of the gate, then use falloff to compensate" model.

    Really the flak primary should just be 10 chunks with 10 damage each and no artificial damage falloff. The damage you deal should be exactly based on how many chunks you hit, and the spread should be a non-random pattern like the CPMA shotgun, so you're not rolling dice when you shoot. That things still haven't reached this point after all this time is exemplary of how much the series has chased its own tail rather than make concerted progress in its balance.

    Comment


    • #92
      Unless you double the middle shard, you're not really going to get a symmetrical spread pattern with 10 shards. 13x9 worked well for 2k4, I wouldn't mind 10x12, but generally speaking, I agree. I'd rather see a little over 100 damage, so long as it's challenging to get all of it, and a consistent, centered spread pattern would be nice.
      Originally posted by Mysterial
      An instant hit, accurate, instant kill weapon is overpowered. There's no skill ceiling. It's limited only by the shooter's accuracy. It also severely impairs the defensive side of the game - ignoring ping, it is nearly irrelevant what your opponent does - click the right pixel and you win. Even non-instant kill instant hit weapons are often problematic - the Shock Rifle example is obvious before even getting to other games.

      Comment


      • #93
        Observation on topic:


        UT3 weapons:

        Good - solid feel of projectiles, their speed and radius, good prediction, rare and thus rewarding blind hits, last second saves, etc.

        Bad - OP combos, rockets, flak, bio rifle.

        Ugly - aim error. I swear, if I'll find this in UT4, buh bye out of my hard drive.


        UT4 weapons:

        Good - most powerful weapons have 95% efficiency.

        Bad - most powerful weapons have 95% efficiency.

        Ugly - no feel of precision on projectiles; no possibility to out-smart your opponent in a fight - same dumb pull-out-flak and ram-in-a-face gameplay; weapons are just annoying.

        Comment


        • #94
          Speaking about graphics

          UT3 had too much monotonic texturing, bloom and gimmicky DOF to compensate for lack of proper AA.

          UT4 bloom is toned down but current AA optios are hideous and eye-straining. FullHD resolution or supersampling is mandatory just to compensate atrocious view blurring. You have to choose bare minimum graphics which have less visible edges, or, if you tune settings up, between no-AA shimmering and AA smearing. Besides, we were promised visual clarity but Outpost23 thus far is the one and only map that is visually appealing.
          Last edited by Hayrack; 07-16-2017, 02:40 AM.

          Comment


          • #95
            Originally posted by -AEnubis- View Post
            Unless you double the middle shard, you're not really going to get a symmetrical spread pattern with 10 shards. 13x9 worked well for 2k4, I wouldn't mind 10x12, but generally speaking, I agree. I'd rather see a little over 100 damage, so long as it's challenging to get all of it, and a consistent, centered spread pattern would be nice.
            Sure, designing a perfectly even spread pattern with only 10 pellets is challenging, particularly with a centered shard. There's a mod for Q3 called Edawn which allows you to customize all of the weapons via a single cfg file, and using the non-random spread option sort of automatically calculates the most even spread pattern for the number of shotgun pellets you assign. Using 20x5 pellets is still far more even than 10x10, but 10x10 still gives a pretty decent, symmetrical, even pattern... the only problem being there's no pellet in the exact center.

            For hitscan shotguns, you're already guaranteed so much that I think the CPMA-style of deliberately slightly off-center non-random patterning is fine. For projectiles though, I think if you're going to have a relatively low velocity that requires precise leading, then you really should get that guaranteed center hit if you aimed correctly.

            Still, the centered shard is its own solution to all that, since you can assign it whatever additional damage is necessary to even out to exactly 100. For example instead of 12x8 for 96 damage or 12x9 for 108, you can do something like 11x8 + 1x12. Plus, IIRC every Unreal engine since at least 2k4 allows simple decimals in damage values, so you can do i.e. 12.5 x 8 and get exactly 100 damage out because the pellets aren't getting rounded up or down.

            Spawn weapons in UT have been really bad with RNG as well. The Ass Rifle of course takes the cake, but the UT3 enforcer has an unwanted bit of randomness to it, and the UT4 enforcer is much worse in this regard. And that's just talking about the primary fire. There's a lot of ways to handle this:

            1) Commit to a Halo-esque pinpoint-accurate Enforcer with no spread and a 2x or 3x scope as the altfire. Have a headshot multiplier, but lower the bodyshot damage to the point where people aren't constantly being reverse spawn fragged from across the map with just a few Enforcer bullets.

            2) Replace the Enforcer with the Dispersion Pistol. The velocity on primary fire can still be high enough to be decently accurate, but the inherent bit of uncertainty to it will be based on leading skill instead of RNG spread. The charge-up alt-fire allows for vastly superior decision-making than the "all-in diceroll" of all the stupid Enforcer alt fires.

            3) If you must have a hitscan spawn weapon with imperfect accuracy, a Q1-style shotgun with a non-random pattern is the only way to get the job done without resorting to RNG.
            Last edited by Gnalvl; 07-17-2017, 01:11 PM.

            Comment


            • #96
              I'm not saying there has to be a middle shard, but 10 will mostly leave you elliptical.

              The Halo Pistol was grossly OP.

              Whether Dispersion pistol, or a fast projectile Enforcer, going projectile is the best way to limit the range of the starting weapon, and you can still keep it pixel accurate, with damage and rate adjustments to make sure it needs no RNG. I also wouldn't be opposed to bringing the original Stinger tooling back, and just doing without a "Minigun" at all, as it's current uses are either redundant, or OP. Then you'd have a MG/SG starting weapon, which was all Tyridium projectile, and relatively low rates and damages, but significantly more versatile and range limited.
              Originally posted by Mysterial
              An instant hit, accurate, instant kill weapon is overpowered. There's no skill ceiling. It's limited only by the shooter's accuracy. It also severely impairs the defensive side of the game - ignoring ping, it is nearly irrelevant what your opponent does - click the right pixel and you win. Even non-instant kill instant hit weapons are often problematic - the Shock Rifle example is obvious before even getting to other games.

              Comment


              • #97
                The spread on the UT3 Enforce prevented it from outclassing the stinger minigun at longer ranges. The Enforcer in UT3 does pretty consistent damage when used within it's intended firing range. Also if you slowly fire off single shots you can use the Enforcer to defend yourself from long range assault, that is each first shot from the Ennforcer's primary fire was spot on. I think the UT3 Enforcer was a great starting weapon because it could be used at any range but excelled at nothing until you had two. The alt fire while nothing special did have it's place. The alt fire increased the damage per second of the weapon a bit and could be used to quickly finish a weak target. The alt fire also made the Enforcer OP in UT3 because with it you had just enough time to fire off five shots while your spawn protection was active.

                Originally posted by Hayrack View Post
                Bad - OP combos, rockets, flak, bio rifle.

                Ugly - aim error. I swear, if I'll find this in UT4, buh bye out of my hard drive..
                What is this "aim error"?

                I think combos were fine in UT3. Flak primary and bio full charged alt I agree did too much damage. Rockets were great in UT3, by far the most well balanced of all the Unreal games, dodging them is like playing frogger at times. It feels great when you squeeze between rockets fired in a spread.


                About the graphics, I think UT4 is coming along nicely visually. Outpost 23 looks great but it is not my favorite looking map. I like maps with more plants, detail, a sense of life like Titan pass and that large community map with the large bridges and birds flying around, that map is gorgeous. Facing worlds looks great too. The out door area of Outpost 23 is great having those water falls.


                How about some thing like this for weapon and movement tweaks.

                The hit box of all shots matches their size. If you evaded it, you evaded it. If you did not land it you did not land it.
                Flak primary max damage some where from 115 to 125
                Slightly increase rocket speed
                Decrease rocket blast radius
                Remove 20 minimum rocket damage
                Shock beam some where from 35 to 40 damage
                Shock ball some where from 55 to 60 damage
                Enforcer some where from 20 to 22 damage, I think 21 is worth a try.
                Sniper 70 to 80 damage
                probably slower fire rate on sniper, scale according to damage change.

                To clarify I don't mean to give weapons a damage range.

                Slightly increase run speed like say 10%
                Increase dodge speed like 20 to 30 percent
                Decrease run deceleration and or acceleration time
                Last edited by Unreal Gladiator; 07-17-2017, 12:50 AM.

                Comment


                • #98
                  Originally posted by Gnalvl View Post
                  Community-wise, I feel like someone just needed to implement 2-3 versions of the nerfs in a mutator and let people try it out, then let them debate which version they prefer, and model a final version based on a compromise of various feedback. Without a playable demo, all people really have as a source for their arguments is the subjective picture in their head which is mostly based on arbitrary aesthetics and doesn't necessarily line-up with how the changes might play out in reality.
                  Unfortunately, we've already tried this approach multiple times with little to no yield. Maybe the movement is as "not bad" as it is because of months and months of prototyping, videos, having the only populated servers in the game, etc. But even if so, it wasn't worth the effort and led to no actual discussion.
                  HABOUJI! Ouboudah! Batai d'va!
                  BeyondUnreal - Liandri Archives [An extensive repository of Unreal lore.] - Join us on IRC [irc.utchat.com - #beyondunreal]

                  Comment


                  • #99
                    Originally posted by Unreal Gladiator View Post
                    What is this "aim error"?
                    Mouse feel in UT3 was very controversial, many people complained, tried various tweaks but nothing removed the strange constant acceleration/smoothing/lag/drugs, whatever the hell that was.

                    Meanwhile, I was fed up with game's consolization and was developing a total re-balancing mutator - going across the settings of all weapons and vehicles in attempt to bring back some play fun, especially because vehicle handling logic in the game was WUT? (Goku-Manta, inflatable Leviathan, Ghetto-bender b-b-b-bouncing around da hood, Raptor with no rear thrust, La piñata Cicada, tanks that can't push jeeps out of the way, etc.). Playing with default weapon settings in the editor, I found a variable "Aim Error". If I recall correctly, it was set at "600", can't remember if the same for all weapons and vehicle weapons. Could this have been the cause of unresponsive mouse in UT3? I changed it to "0", compiled paks, mutators and launched the game,

                    O-M-G,

                    it was like playing UT2004 again, with drug-free, robust mouse and positive results all around. Ultimately, I concluded, that implementing "Aim Error", which intentionally ***** up aiming, was the thing they had in mind when "fixing hitscan domination", along with other things that chopped wings of UT3 and never let it truly rise.

                    In short, from technical perspective, in order to fix hitscan domination, they intentionally programmed aiming to be unresponsive. From player's perspective, people couldn't handle your skill, winning and awesomeness, so they broke your wrists to "bring back the balance".
                    Last edited by Hayrack; 07-17-2017, 04:33 AM.

                    Comment


                    • Originally posted by Hayrack View Post

                      Mouse feel in UT3 was very controversial, many people complained, tried various tweaks but nothing removed the strange constant acceleration/smoothing/lag/drugs, whatever the hell that was.

                      Meanwhile, I was fed up with game's consolization and was developing a total re-balancing mutator - going across the settings of all weapons and vehicles in attempt to bring back some play fun, especially because vehicle handling logic in the game was WUT? (Goku-Manta, inflatable Leviathan, Ghetto-bender b-b-b-bouncing around da hood, Raptor with no rear thrust, La piñata Cicada, tanks that can't push jeeps out of the way, etc.). Playing with default weapon settings in the editor, I found a variable "Aim Error". If I recall correctly, it was set at "600", can't remember if the same for all weapons and vehicle weapons. Could this have been the cause of unresponsive mouse in UT3? I changed it to "0", compiled paks, mutators and launched the game,

                      O-M-G,

                      it was like playing UT2004 again, with drug-free, robust mouse and positive results all around. Ultimately, I concluded, that implementing "Aim Error", which intentionally ***** up aiming, was the thing they had in mind when "fixing hitscan domination", along with other things that chopped wings of UT3 and never let it truly rise.

                      In short, from technical perspective, in order to fix hitscan domination, they intentionally programmed aiming to be unresponsive. From player's perspective, people couldn't handle your skill, winning and awesomeness, so they broke your wrists to "bring back the balance".
                      That's ****ing disgusting. No wonder the input is broken. **** those bastards at Epic, I bet they hardcoded it for this game. The smoothing is so damned strong.
                      Last edited by Calypto; 07-17-2017, 06:50 AM.
                      UT2004 Movement Mutator

                      Comment


                      • @Calypto

                        Yeah. I feel very uneasy when I see posts that state "omg hitscan is too stronk" (it is, lol, but NOT in UT4, at least not in current public matches) because they may push slipping "Aim Error" back in or coming up with some other ridiculous "balancing" idea.

                        I don't feel any kind of artifical aiming interference in UT4 but precision on fluctuating and lower framerates feels really floaty.

                        Last edited by Hayrack; 07-17-2017, 08:50 AM.

                        Comment


                        • AimError is a weapon property that is only called on bot classes. Without it bots would have 100% hitscan aim.
                          It has no way of affecting where a real player's shot lands, let alone their mouse input in general.

                          Comment


                          • There were a bunch of undocumented mouse settings in the INI that made UT3 feel that way. There are a few guides on how to remove them. One of the issues is that, by default, the mouse has desynced scaling on each axis of mouse movement. I can't remember what the default was, but if you set the X and Y scaling down to 1, the mouse felt fine. UT4 also had this problem early on until enough people complained about it.
                            HABOUJI! Ouboudah! Batai d'va!
                            BeyondUnreal - Liandri Archives [An extensive repository of Unreal lore.] - Join us on IRC [irc.utchat.com - #beyondunreal]

                            Comment


                            • No, it's still there. Even at 300+ frames, it's not as responsive as UT2004. There's some kind of smoothing that's not there if you turn on mouse acceleration ingame.
                              UT2004 Movement Mutator

                              Comment


                              • Originally posted by -AEnubis- View Post
                                The Halo Pistol was grossly OP.
                                The factors which made the Halo pistol OP have little bearing on UT... the biggest one being that 90% of the other weapons in the game couldn't hit the broad side of a barn. Then there's Halo's botched locational damage system where shot placement only matters for 1/3 of the hits in a "perfect" headshot kill. The least discussed factor though is the fact that killing in 5 bodyshots (the equivalent of the Enforcer's usual 20 dmg) is extremely brutal in a game with such slow movement as Halo. A good deal of the time the average player was mowing people down with the Pistol, they weren't even getting headshots and their victims were dieing so fast regardless that no one really noticed.

                                Overall the idea of starting with an accurate semi-automatic, high velocity weapon with a scope and headshot multiplier is sound. I've tried 17 dmg for bodyshots with 2x headshot multiplier in UT3 and it's not remotely as powerful as in Halo, because targets in UT are much harder to hit at all - let alone headshot. Even so, what'd be better, particularly with zero spread, is more like 11-13 bodyshot damage, with 34 on headshot to start...possibly reduced to 25 if anyone is really scoring such a high headshot rate that it seems OP to anyone.

                                I still prefer something like the dispersion pistol, but I still think a sensibly-improved Halo approach is valid.

                                Originally posted by -AEnubis- View Post
                                I also wouldn't be opposed to bringing the original Stinger tooling back, and just doing without a "Minigun" at all, as it's current uses are either redundant, or OP. Then you'd have a MG/SG starting weapon, which was all Tyridium projectile, and relatively low rates and damages, but significantly more versatile and range limited.
                                I've always thought the idea of choosing from a selection of spawn weapons would be nice for stuff like Quake and UT, and Quake Champions finally did it (badly). Having a choice between something like a traditional pistol, a blaster/dispersion pistol, and a projectile nailgun/stinger with a rapid primary fire and charge-up spread blast secondary, would be pretty nice.

                                I've always wished TDM would gain more traction as a competitive mode, in which case 30 second weapon respawns are crucial, but the resulting potential to be stuck with a "crappy" spawn weapon for large portions of the game isn't appealing to many people. I think having a choice of spawn weapons which all have the potential to be decently damaging without being overpowering or relying on luck, provides a solution to all that. Even in the unlikely event that you never get your hands on a pickup, you would have some variety and strategy in your weapon choice, and be able to contribute decently with any of said options.

                                Originally posted by Unreal Gladiator View Post
                                The spread on the UT3 Enforce prevented it from outclassing the stinger minigun at longer ranges...
                                I fully understand the purpose of the spread, and I agree that the UT3 enforcer was a relatively reliable sidearm compared to i.e. the 2Kx ass rifle or the Quake MG. However if we're talking about actually advancing the design of the game, relying on RNG to balance a weapon is a bad practice which should be phased out.

                                I mean we could say the same things about the Q3A shotgun. Even with the random spread, it fills its intended role and kills reliably within its intended range; but if we're talking about improving it there's just no reason to leave the luck component there.

                                I would also add that by UT3 and UT4 the fact that they didn't at least add crosshair bloom to show exactly how the spread is increasing and decreasing with time, is just ****ing lazy. I don't care for CS, but if you must have a spread-based recoil system, the basic solutions they've employed to avoid making things a total crapshoot are pretty important to pay attention to.

                                Originally posted by Hayrack View Post
                                Playing with default weapon settings in the editor, I found a variable "Aim Error". If I recall correctly, it was set at "600", can't remember if the same for all weapons and vehicle weapons. Could this have been the cause of unresponsive mouse in UT3?
                                Originally posted by Calypto View Post
                                That's ****ing disgusting. No wonder the input is broken. **** those bastards at Epic, I bet they hardcoded it for this game. The smoothing is so damned strong.
                                lol, you guys... Stolid is correct that "aim error" is a bot-exclusive setting to prevent them from having a 100% hit-rate with every weapon. These exact same aim error settings were in UT1 and UT2k4 as well, which you can see for yourself if you look at the source code.

                                I would agree that mouse input feels a bit smoother and more precise in UT2k4, but the benefit of that feeling is pretty heavily negated by factors like this and this.
                                Last edited by Gnalvl; 07-17-2017, 02:31 PM.

                                Comment

                                Working...
                                X