-
Notifications
You must be signed in to change notification settings - Fork 12
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
v3.1 general balance (augs, weapons, skills, etc) #842
Comments
for heavy weapon skill and microfibral muscle affecting movement speed, in DeusExPlayer.uc
|
for aug level 5, we could either definte a new values array with a different name? or make the level 5 value its own standalone value? |
Going to share my reactions! I have a few proposals. Some could be used in combination, some are radical enough that they would be reworks. I am primarily concerned with speedrun balance, but I'm trying to not make proposals that would negatively affect the regular game, and will prefer proposals that can improve both speedrun and regular game. Since I'm going to be making some proposals, but don't want to waste your time responding to all of them, I'm going to color code them. A lot of these are meant to give you ideas rather than argue for a specific thing, unless I explicitly say otherwise. Legend: EMP Shield
EMP Shield has some marginal utility, but it's just not a very exciting mod. Making EMP shield consume less energy is good, so making it stronger and have less energy usage can't hurt. But I think doesn't address the fundamental design issue -- EMP damage is rare and you want to avoid it entirely. I think EMP Shield also suffers from having very weak reasons to upgrade. Reaction:
Proposals:
Energy Shield
I'm unsure whether Energy Shield is underpowered or just underutilized. Energy Shield is already significantly better than it is in vanilla just by virtue of more enemies having access to energy attacks. It's just people being too focused on other stuff to actually micromanage their defenses during speedruns. I think Energy Shield also suffers from having weak reasons to upgrade, less so than EMP Shield though. Reaction:
Aqualung
Reaction:
Proposals:
Synthetic Heart
Generally the reason Synthetic Heart is undesirable in speedruns is that you focus your upgrade canisters on Speed and Recirculator. This gives you a very lopsided distribution of upgrades, you're often maxed out, most of the time you don't have many other augs, so Synthetic Heart does little for you. Once a blue moon it might improve your targeting or combat strength to put you over the 50 damage threshold. That's basically the only time I'd ever think to use it. Reaction:
Heavy Weapon Skill
Reaction:
Microfibral Muscle
Reaction:
Proposals:
Range Mod
Reaction:
Pistols
Reaction:
Reaction:
I have a LOT more to say about pistol balancing, but I'll put that in a separate post when I have time. Sword
Reaction:
The main issue with the sword is that it's unclear what niche it covers. Most of the time you don't have inventory slots to spare. With 12 damage it's unlikely to help you break containers unless you stack both low tech and combat strength. If you have high low tech and combat strength, you'll probably prefer the DTS anyway because you can one-shot commandos. Making the sowrd any stronger in terms of damage is just going to invade the design space of DTS. The only thing that seems like could be boosted would be to give it a higher base attack speed -- then it could have comparable DPS with the DTS but inferior container breaking capabilities. This justifies why DTS occupies one more slot. So in terms of design, I am of the strong opinion that the sword should have similar DPS to the DTS. Proposal:
|
I rather like this idea, personally. Although the default max energy is 100 so 120 definitely wouldn't be the right number.
I agree that it would be exciting. It might increase the difference between a good seed and a bad seed though, which is the opposite of what we sort of want. Still, it might be a good change.
You don't think decreasing max range with the assault rifle would impact things? I've seen speedrunners practically snipe with it under the best circumstances. |
that would be 120 energy per minute, not per second |
…acy for pistols also fix Accessed None warnings for 3rd person camera
rates are still lower than vanilla
for EMP Shield aug, I made levels 1 to 3 slightly stronger (level 4 was already 100% resistance), and all levels also make you immune to scramble grenades for Energy Shield aug, I made it cost 25 energy per minute instead of 40, like multiplayer does the changes for movement speed I'm testing right now...
|
actually just the fixed gain over level 4, but it keeps compatibility with old seeds, and ensures the Synth Heart is always useful and not just wasting your energy
all the commits of allowing aug boosting past level 5 together: cea7b90...4309a3d |
Seems like a good change. I think the fundamental issue is that this aug is too situational. It's possible to go a whole run without receiving EMP damage, but this seems really hard to address, so just buffing the numbers seems like a good idea meanwhile.
In my runs, I've skipped the Energy Shield often because the first two levels either can or did get randomized to pretty low values. I didn't find the cost to be the main issue. I once had an Energy Shield that started at around 50% and it was worth every bit of that 40 epm. Reducing the energy cost is also good though and it allows the randomized values to stay wide.
I didn't know Microfibral didn't help with carrying corpses in vanilla!
I'm really curious what effect this is going to have on the meta. I kept debating in my head what percentage range would be best. I also know the this percentage was picked for how it affects non-randomized effectiveness values in Zero Rando and Randomizer Lite, right? I feel like Heavy is necessary for me to go fast and I try to aim for combined Heavy/Rifle builds. The GEP gun is simply too useful, faster than demolition and more generally useful. It ensures you won't run out of explosives because grenades are still usable with low demolition, but the GEP is much less usable with low heavy. If you don't have Heavy, reloading the GEP is extremely painful unless you do it at pre-planned points (e.g. taking the elevator up to airfield). Will this change make more seeds viable for me, or will it incentivize going deeper into Heavy as a result of requiring higher effectiveness to get full speed? It could strongly buff builds that go deep into heavy, which I think is not needed in the current meta. I'm not 100% sure of my analysis here and I have my own biases. But if we conclude it does end up buffing going deep into heavy too much, one way to adjust that would be to make the spread 100% to 175% or 110% to 175%. This would make lower % more tolerable in terms of speed while still giving extra benefit for going higher. The potential downside of a wider spread is that it would make untrained heavy somewhat stronger, therefore make heavy more ubiquitous -- but heavy is already ubiquitous, so I'm not sure how much of a downside it is in practice. I also think widening the spread is compatible with Zero Rando having 100/120/150/200. It would mean that the player will receive a small speed boost while wielding Heavy even at Trained, which I actually like. I'm not sure if I'm advocating for the wider spread as of yet, it's tempting, but this stuff is hard to model! Use your judgement and later adjustments can be made after seeing us play. One thing I know this change in general, regardless of the chosen spread, is likely to do is to help incentivize presetting your weapon skills a bit less and looking at the weapon randomization before investing, especially in races. Because now 150% isn't a hard binary anymore, so it increases the number of potentially viable combinations. I think this would be a very good effect.
This is such a good aug change. Endorsed!!! I'm excited to have to make decisions between this and Combat Strength. |
I'm so excited for the Synthetic Heart change!! This is one of the coolest aug changes. It's going to take several seeds, but I can't wait to get a situation where it's actually viable to use. I think making the speed aug only improve from 170 to 180% is a bit disappointing but also very wise. Synthetic Heart is very dangerous augmentation to balance. Speed Aug already is 100% priority for upg cans, then Power Recirculator is 100% priority for upg cans. If Synthetic Heart becomes a third 100% priority, then this can have a bad effect on meaningful upg can distribution choices. |
I think EMP shield and energy shield are at least worth installing, like if they're the first augs you find. Probably not for speedruns but at least for casual runs. So slight buffs here are probably good. Energy shield's defaults are 20%, 40%, 60%, and 80%, and the new level 5 is 100%. I could probably ditch the level 5, and just make it like 30%, 60%, 80%, 100% or something. EMP Shield obviously doesn't have a level 5 since the level 4 is already 100% resistance. EMP Shield is 30%, 60%, 80%, 100% now. In vanilla EMP Shield is 25%, 50%, 75%, 100%. So I guess Energy Shield could copy one of those lineups from EMP Shield. Here's a question about aug scaling. If you want every upgrade to be just as valuable, do you...
But also, do you want every upgrade to have the same value for the player? Or should max level be extra juicy? Or should the first upgrade be the most valuable? |
To be clear, I think this design goal trades off against other design goals, so I don't think it should be followed blindly, probably not even prioritized. But I'd keep it in mind as something we can try to preserve when possible. If we were to maximize this specific design goal, then we need to consider the effect of that aug strength has for that specific aug. The answer also varies a lot with how many free aug upg cans you have available. The last part is really different between speedrunning and regular play. Not only because you can find many more aug upg cans in regular play, but also you're not obligated to sink all of them into speed and power recirculator. In regular play, before getting to Area 51 you can have as many as 15 upg cans. In speedrunning you'll have at most 10, with 7 probably being the most likely. 3 of your upg cans go directly into speed, and if you get power recirculator, another 3 go there. This just doesn't leave many cans free and it requires that subsequent levels be amazingly good to justify being used. Most of the time, it doesn't matter how good the last level of an aug is, you just won't get to it. I think we (or I in particular) might need to accept that because of how limited aug upg cans are in speedrunning, that some levels just cannot be made attractive unless something else about how speedrun mode is balanced changes. I don't actually understand how much damage a Plasma Rifle from an enemy can do and how Energy Shield interacts with that. I have only a vague understanding. The actual damage breakpoints are really important. To give a different example, you generally want enough targeting to get your Sniper Rifle over 50 damage, and any additional targeting, even if it brings quite a lot of damage, is generally not worth the upgrade. The damage scales linearly per level, but you're interested in a specific breakpoint. Similarly for Energy Shield, the breakpoints that matter are:
Besides that there's a linear health save to consider, but I'd actually say the effect of the breakpoints is bigger in speedrun because it reduces the amount of time wasted during combat. This comes in handy in chaotic situations like Cathedral and Silo. The closest I could get for speedrun mode would be something like: 30 / 50 / 75 / 90 / 100 for the old energy usage. I'm not sure how I would adjust this if the cost is reduced to 25 epm. Maybe you don't need to. One thing that might throw off my estimations of the impact of the aug in practice is how Environmental Training factors in. Since the damage reduction is multiplicative and not additive, the improvement an Energy Shield brings on top of a high Environmental Training might be less impactful than expected. For non-speedrun rando this is maybe a little strong, but then again when you're not speedrunning, a lot of augs are very powerful - we talked about Spy Drone. So I wouldn't be too worried. Also in speedrun mode, sometimes you get the second level just because it happened to be randomized high and the first level was randomized really low. Anyway, I generally think it's going to be difficult to simultaneously:
I think we should not set the first level too low just to make more room for the subsequent levels being good. If we leave room to make the max level extra juicy, it's most likely that this will just result in the aug not getting installed. Reducing the cost raises the incentive to install the aug at all, without needing to play with the per-level strength numbers. So that's neat. But also weak and cheap augs are somewhat less exciting gameplay-wise. I don't know how much of this helps. Deus Ex is a difficult design space to deal with! |
a couple more general balance changes I just did
hopefully enviro skill is still worth upgrading, and maybe we see more use of hazmats |
* add (nonfuntional) autoconsume options * combine pick up, drop and autoconsume options. use separate config values for each item class * modify `TryLootItem` check slightly * use string for loot action settings, and also keep autoconsume from crashing the entire game * ConsumableWouldHelp function * ConsumableWouldHelp checks HealingItem values * fix HX build * preproc var for bingocheat * bump version to v3.0.1.0 Alpha for devs * buff regular sword * WeaponMegaChoices now clear out any FlagRefs on the weapon choices so that they work even with Confix installed. Based on changes suggested by Wryyyong in PR #823 (real fix for #793) * Fix randomized robot names * Reorganize how pawns with different familiar and unfamiliar names are identified as having fake names and should get a random one instead * Thrown Projectiles will no longer take damage from halon * Gunther and Gilbert have ammo for the possible weapons you can give them in mission 1 and 4 #807 * If giving ammo, an add_ammo amount should apply even if the pawn doesn't have the ammo type already #807 * Throwing grenades and turning to a wall now only gives you your ammo back if it's before frame 0.7, which is when it notifies and tells the game to spawn the grenade projectile #805 * Smoke and gas leaking from barrels will now be moved when they are shuffled * Enemies in the Hong Kong Helibase robot bays can no longer open the doors themselves, and need to wait for the alarms to be set off #830 * add spaces to JacobsShadow hint (#808) * lower the alarm inside M03 boat house (#811) * sewer teleporter in 09_NYC_DOCKYARD can't be squeezed past (#816) * sewer teleporter in 09_NYC_DOCKYARD can't be squeezed past * reduce dockyard sewer teleporter radius * vans not breakable in Zero Rando * fix HX build * splits overlay is an Advanced game option (#824) * give "Trash" and "Not Trash" separate keyboard shortcuts (#829) * paul's bathroom door doesn't shut instantaneously (#817) * paul's bathroom door doesn't shut instananeously * DeusExMover -> #var(DeusExPrefix)Mover * fix HX build --------- Co-authored-by: Die4Ever <die4ever2005@gmail.com> * stand on warehouse fence without tap dancing * revert old nerf of master pistol 180% vs vanilla 200%, because we have helmets now, and rifles have been buffed a bit * EMP disables cloak * don't give "Rick" a random name when memes are disabled * Small autoturrets fire at half the rate of full sized turrets. Make sure all turrets in the game use their default firerates. * Spawned autoturrets can now randomly be small or large * Restore overheard conversation in superfreighter over "loudspeaker" in the helibay * Restore last infolink from Helios in Area 51 * Fix the "Critical Loctaion Bonus" typo in Area 51 Final... finally. * Is this a Jojo reference? * dxvk.conf max of 500 fps * More piano practice for JC * Piano will now show how many songs have been played * Piano will now make sure all songs get played an equal number of times, while also allowing certain songs to be weighted more (for bingo goals) * Each piano song can have individual weighting * Add Terran 1 to Piano * Blizzard, in it's infinite wisdom, decided it was a good idea to capitalize the C in StarCraft, but not in Warcraft * buff range mod, nerf ranges for pistols and assault rifle, buff accuracy for pistols also fix Accessed None warnings for 3rd person camera * aug tree white color for selector * slightly increase pistol ammo (#842) rates are still lower than vanilla * Oops, more bangers on the piano * I slipped and dropped these great piano tracks * weapon descriptions tell you if it's a good roll or a bad roll (#844) * weapon/aug/skill descriptions tell you if it's a good roll or a bad roll (#844) and don't show in Zero Rando mode * Item swaps fail if they move to the other side of a wall. Make sure placeholders are sufficiently high off the ground. Minor adjustment to weapon mod replacement logic so they don't even try to replace if the class matches * installer DXVK Linux, off by default * aug tree, fix draw styles * Oceanlab UC and Area 51 bunker big fans are highlightable so you can see how much damage you've done #855 * Crowd Control "give item" effects now become selectable based on the loadout in use, so you can't try to give a SWTP player a gep gun and have it rejected by the loadout #854 * Move Ocean Lab greasel lab backtrack button closer to the wall * Fan in Vandenberg Computer is now highlightable * savage doesn't give you an aug on post gas station starts * fix typo in smuggler intercom (#809) * fix typo in smuggler intercom * expand SmugglerDoorBellConvo check a bit to make sure it's in English * More extensive goal information in notes directory (#800) * add goals.json * goals with the same name are next to each other in goals.json * the exit teleporter from nsfhq can't be crawled under (#812) * the exit teleporter from nsfhq can't be crawled under * find nsfhq exit teleporter by url instead of event * don't uselessly try to change teleporter location * water cooler in versalife break room faces away from the wall (#813) * free and infinite aqualung (#845) * free and infinite aqualung * simplify DXRAugAqualung.IsTicked and DXRAugAqualung.Tick * update aqualung description * fix clothes racks with LDDP when memes are disabled (#833) * restore Everett barks (#840) * SkillComputer slightly nerf trained speed in Zero Rando so that the "defaults" display lines up better with randomized values, note this has no effect on randomized values since those only check the range of lowest to highest * teleporters block actors * DXVK 2.4 * fix "cemetery" misspelling in 09_NYC_GRAVEYARD startupMessage (#818) * fix "cemetery" misspelling in 09_NYC_GRAVEYARD startupMessage * add additional check in "Cemetary"->"Cemetery" fix * CleanerBots don't move when disabled (#847) * Fix the "You can dance if you want to" goal not being marked if the ScriptedPawn being watched isn't in the Dancing state (#835) * fix "dance if you want to" goal for NPCs in Conversation state * add PawnAnim_* event to ScopeView * restore commando barks in 06_HONGKONG_STORAGE (#814) * restore commando barks in 06_HONGKONG_STORAGE * Update DXRFixupM06.uc --------- Co-authored-by: Die4Ever <30947252+Die4Ever@users.noreply.github.com> * change btnLabelResetToDefaults string so it doesn't conflict with ButtonUpgradeLabel (#827) * change btnLabelResetToDefaults default * old old "Restore Defaults" shortcut; disable it in DXRMenuScreenNewGame * DXRAnimTracker work in progress * Singular bingo goal descriptions (#803) * add singular bingo descriptions * Update DXREvents.uc * Update DXRModules/DeusEx/Classes/DXREvents.uc --------- Co-authored-by: Die4Ever <30947252+Die4Ever@users.noreply.github.com> * Chateau DuClare with fixed max value (tour 5 parts, and collect 3 keys) (related to #803) * open vandenberg comm building doors immediately on 129+ starts (#853) * open vandenberg comm building doors immediately on 129+ starts * Update DXRModules/DeusEx/Classes/DXRStartMap.uc Co-authored-by: Die4Ever <30947252+Die4Ever@users.noreply.github.com> --------- Co-authored-by: Die4Ever <30947252+Die4Ever@users.noreply.github.com> * Improvements and fixes for adding goals and DXRStartMap (#797) * move StartMapSpecificFlags call from PreFirstEntry to PostFirstEntry * improvements to GetGoalText and AddGoalFromConv. use latter in most places goals are added * MarkConvPlayed adds "_Played" automatically * MarkConvPlayed takes a string instead of a name * restore "NSF Defection (Streets)" start * break StartMapSpecificFlags into PreFirstEntryStartMapFixes and PostFirstEntryStartMapFixes * AddGoalFromConv uses ConEventAddGoal.bPrimaryGoal rather than a parameter * restore modified DefeatNSFCommandCenter goal text * restore riot cop barks in 08_NYC_HOTEL (#815) Co-authored-by: Die4Ever <30947252+Die4Ever@users.noreply.github.com> * change "woman" unfamiliarName to "Woman" (#819) * change "woman" unfamiliarName to "Woman" * move "woman"->"Woman name fix to DXRFixupParis * Adjust size of teleporters in Cathedral, Ship, and Ocean Lab so you can't get around them #804 * Make sure you can't frob the ramp in HK Level 2 labs * fix HX build * preproc var for bingocheat * bump version to v3.0.1.0 Alpha for devs * buff regular sword * WeaponMegaChoices now clear out any FlagRefs on the weapon choices so that they work even with Confix installed. Based on changes suggested by Wryyyong in PR #823 (real fix for #793) * Fix randomized robot names * Reorganize how pawns with different familiar and unfamiliar names are identified as having fake names and should get a random one instead * Thrown Projectiles will no longer take damage from halon * Gunther and Gilbert have ammo for the possible weapons you can give them in mission 1 and 4 #807 * If giving ammo, an add_ammo amount should apply even if the pawn doesn't have the ammo type already #807 * Throwing grenades and turning to a wall now only gives you your ammo back if it's before frame 0.7, which is when it notifies and tells the game to spawn the grenade projectile #805 * Smoke and gas leaking from barrels will now be moved when they are shuffled * Enemies in the Hong Kong Helibase robot bays can no longer open the doors themselves, and need to wait for the alarms to be set off #830 * add spaces to JacobsShadow hint (#808) * lower the alarm inside M03 boat house (#811) * sewer teleporter in 09_NYC_DOCKYARD can't be squeezed past (#816) * sewer teleporter in 09_NYC_DOCKYARD can't be squeezed past * reduce dockyard sewer teleporter radius * vans not breakable in Zero Rando * fix HX build * splits overlay is an Advanced game option (#824) * give "Trash" and "Not Trash" separate keyboard shortcuts (#829) * paul's bathroom door doesn't shut instantaneously (#817) * paul's bathroom door doesn't shut instananeously * DeusExMover -> #var(DeusExPrefix)Mover * fix HX build --------- Co-authored-by: Die4Ever <die4ever2005@gmail.com> --------- Co-authored-by: theastropath@gmail.com <theastropath@gmail.com> Co-authored-by: Die4Ever <die4ever2005@gmail.com> * slightly move M01 boat Top of Statue so it's easier to see from UNATCO * M06 rename DTS boatperson location to canal waterside apartment * m08 free clinic location closer to door * slightly faster sword animations (#842) * move ConsumableWouldHelp to DXRPlayer * autoconsume some cleanup * disable refuse button for items set to autoconsume * Update DXRVanilla/DeusEx/Classes/Player.uc Co-authored-by: Die4Ever <30947252+Die4Ever@users.noreply.github.com> * tweak defaults for melee weapons --------- Co-authored-by: Die4Ever <die4ever2005@gmail.com> Co-authored-by: theastropath@gmail.com <theastropath@gmail.com> Co-authored-by: Die4Ever <30947252+Die4Ever@users.noreply.github.com>
Is flamethrower ammo too plentiful? Or flamethrower enemies too common? |
I'm not sure. Why do you ask? Sometimes they are extremely common per run. I run into 3 of them in the subway. Other times I don't find one for a while. Subjectively it feels like the flamethrower is more common than the Plasma Rifle. I have had situations where I don't have enough flamethrower ammo. If you think the weapon is too good, it might be worth slightly increasing the number of shielded enemies among non-MJ12 factions (MJ12 already has the Commando). Though one beef I have with the shielded enemies is that I do not find them visually distinct enough from regular enemies. Also some update on my experiences with the pistol. I've had numerous runs now where the pistol is an amazing addition to my heavy-centric run. This happens when the pistol rolls 14-18 damage. I'm not actually sure what the breakpoints are in terms of number of headshots required to eliminate an MJ12 commando. But even when the damage is a bit lower, the pistol still doubles as mini-sniper rifle. This is all made possible by the fact that pistol is extremely fertile grounds for a laser mod. The crossbow cannot benefit from the laser mod, so it will always go unused until pistol training receives a buff rather than pistols. Crossbow is difficult to buff because this indirectly buffs all crossbow enemies which are sufficiently dangerous at the moment. When I used the pistol in a true sidearm role, I tended to not run out of ammo. I've never had a single run where the stealth pistol was an attractive option. Its damage is lower, so it will inevitably eat up more ammo, making its higher base fire rate a very minor benefit. I have no idea what one could do to make it more attractive in speedrun mode. I've always been a stealth pistol hater for normal mode, I think the regular pistol eclipses it there too. But I know there are differing opinions. I'd consider raising its base damage by 1 maybe. Overall, my opinion is that the regular pistol is actually sufficiently strong in this sidearm role. It's the crossbow, pistol training and to a lesser extent the stealth pistol that need to be stronger. I don't have any good ideas currently about how these could be made stronger without inadvertently affecting normal mode. Pistol training is still very strong in normal mode, as it gives you access to a weapon with plentiful ammunition and high ammunition efficiency if you can do headshots. One maybe-decent idea is to make the crossbow be affected by the laser mod. Increasing the base accuracy of the pistol is a good idea. It does not buff its lasered sidearm role at all, but makes it otherwise more viable. If you buff it for the regular pistol, you should buff it for the stealth pistol, as has an even more abysmal starting accuracy. We also have to keep in mind it's incredibly easy to accidentally make pistols too good in speedrun mode simply because their inventory space is so small, so their opportunity cost is also very small. If an item is on average pretty good and has very small opportunity cost, you will start seeing it in every run, which is probably not what we want. Then again, the fact that the pistols randomize with crap stats half of the time should take care of that. |
Small change I made. Robots now have 25% damage resistance to plasma, we had it at 0% before. Vanilla has it at 75% but I think that's because the intended damage of the plasma rifle seems to be 40 (but it is bugged and only does 8 damage lol), so they would've intended it to do 10 damage to robots. Now we have it at 18 damage as a base, and robots are now reducing it down to 13. This will maybe be compensated by people having motivation to further upgrade heavy skill anyways. I need to play with pistols more, I have already slightly buffed their accuracy to match the assault rifle, but they could maybe use a little more. +1 damage for stealth pistol probably makes sense, so it will more frequently be better than the regular pistol. I'll look into what it would take to make the laser work with the crossbow, this would probably also fix laser on plasma. Another option could be making accuracy mods help more? I think they're maybe slightly underwhelming right now. |
some ideas to MAYBE do before closing this
|
I quite like making the heavy speed improvement start at 100%. Both for the reasons you said, and because it's a lot more intuitive in my opinion. |
slightly buffed from 125% to 175%, also cleaned up the code for it slightly
also slightly buffed level 1 and 2 of AugVision |
merges
in the compiler to help?)The text was updated successfully, but these errors were encountered: