Larian Banner: Baldur's Gate Patch 9
Previous Thread
Next Thread
Print Thread
Page 1 of 5 1 2 3 4 5
Joined: Jun 2020
Niara Offline OP
veteran
OP Offline
veteran
Joined: Jun 2020
This is a synopsis of events and discussions from the recent stream, for those who don't have the time/energy to watch the stream itself.

The stream itself was a 2:20 minute stream. The initial official upload made it look much longer, but I think it's been trimmed down since to cut out the long intro time and the technical difficulties.

Stream begins with a six minute intro involving fan art and a new musical rendition of Down By The River. The art was nice to see.

The stream then comes in on a set of Swen in his armour at a feast table (sort of). It's a bit difficult to really focus on him, as the camera cut is a long, low angle that weaves back and forth over the table, with Swen being a tiny sometimes blurry head in the background. This is only half the stream view, because the other half is taken up by live footage of a cow. It all looks kinda tacky and over-produced. The sound quality is terrible.

Swen gets a fancy introductory nameplate across the bottom of the screen, showing his name and position and job. Neat.

For five minutes or so, he talks about how they've been listening to the community, and sites all the things they've done in response to community feedback. He mentions:

“We changed how cantrips worked” - In reality, they removed a flag from four cantrips, so they didn't spawn surfaces, and changed one damage die; an action that could be done in less than an hour or two of code tweaking.

“We revised how resting was working!” - In reality, they put an extra short rest in, and changed a UI asset to accommodate the second green dot. They didn't change how it was working at all.

“We changed approval ratings!” - I actually don't KNOW what was changed here and have never been able to get a clear answer from anyone about what is new and what is old and what is actually different. It remains, however, that this was a change to the dialogue of one character. Probably the most amount of effort and work in the list because it involved getting the VA to record some new lines (or so I'm led to believe?). The more notoriously fiddly and opaque character approval trigger sequence, Gale, has not been tuned up.

“And like this, we ah, added a... whole bunch of.. errr... things...” - Professional game developers.

“We made it so that you don't have to manually jump every single character in your party across a gap one at a time individually” - they did indeed do this, though from this viewer's perspective, fixing that (while still leaving the player-vs-UI war raging in so many other places), shouldn't be something you wave around on a banner... it's feels like trying to say you're doing good by first advertising how utterly inept at something you are/were, and saying that you got a little better at it.

“We made a UI change to shortcut some spellcasting” - In reality, they changed it to something that is still clunky, inept and archaic, just better than the original which was literally have your hotbar filled up with half a dozen variants of the exact same spell just so you can choose what level to cast at.

He also mentioned inspiration points (which were already in the game before, just without a tutorial) and cross-saving was what they defined as “A whole lot of things”

During this introduction he spends about as long talking about 'the big thing' that they 'can't talk about yet', as he does saying anythign of substance.

We move on to the first panel guest then, and spend about 15 minutes talking to David, head of production (who was not given a fancy name card on the screen like Swen got), about what was actually going to be in Patch 4. Here's what he said:

“First thing, we're introducing loaded dice!” - That thing that, as far as I'm aware, no-one ever actually asked for anywhere.
Their reasoning for this is stated as: “So, if you roll a 1 occasionally in table top, It's a bit of fun, But if you do it repeatedly in a video game, you'll get frustrated.”
It's a pretty glaring non-sequitur that they brush over - If you roll 1s repeatedly in tabletop you get frustrated as well. If you only roll 1s occasionally in a video game, it's a minor thing. If you're failing a lot more in the video game, then loaded dice aren't the answer. This is a band-aid solution to their poorly written RNG. It may solve the perceived issue, but it's not a good fix, especially when it's flavoured and worded as a flat out accusation of cheating against the player, in the name. (And yes, many have analysed it - their RNG delivers the expected spread in a long term test, but it patterns in a visible sine wave that creates issues in actual play, and results in more failures and more strings of failures closer together, and more failures-against-all-odds, than you find in tabletop, or with better RNGs)

“We've done more with Speak with Dead!” - We get some footage of what they mean here, and it looks neat. They've added sfx and overlays to the existing dialogues, and put in cutscenes for all of the existing speak with dead targets. The issue here is that they also talk about how most corpses won't have anything to say, and cite time/resource constraints. This is the problem though: the more glitz and cutscene they add to each corpse, the fewer actual corpses are going to have information, and the less useful the spell will end up being overall.

“Cinematic improvements, and the addition of some new cinematics.” - This one looks nice enough. We get some footage of before and after slides, and while prettying things up a bit is a minor comment, what we do see as a result is the strong implication is that many elements of what has been perpetual day time now will actually be taking place at night, and that's an interesting thing to see.

“We... fixed torches?” - Not addressing the limitless abusable action-free infinite candle dipping idiocy... No, instead what we see is that they're putting yet more pointless pressure on the hot bar in order to do some utterly unnecessary click-swap deal... which looks, the way they demonstrate it, as though using this click-swap is for main hand only, and you won't be able to use this new sub-system to have a weapon in your main hand and a torch in the other - you'll still have to do that manually.

“You can now target portraits.” - Swen pipes up with “Why would you want to target portraits” in response. I don't really care if it's an attempt at a fall-flat joke (and fall flat it does), it felt like it captured a lot of design philosophy.... though he then goes on to admit that it's a basic necessity of life... that they obviously didn't think about or consider to be worthwhile in their initial builds.

“Multiplayer improvements.” - They talk briefly about tightening up how multiplayer works, though actual details of what that means were scant. Can't speak to this, as my own multiplayer experiences have been stable enough, but very sketchy in their controlling and organising, and in rejoining sessions and the like. I'll take their word for it that improvements in this sector are on-going.

“ “Speaking of listening (?)” they're adding a “flee combat” function.” - Apparently in response to the community, they're also adding in the ability to flee combat with a button click. I don't recall ever seeing or hearing anyone asking for this on any of the forums or other media platforms that I've followed the game on. Not sure who they're citing listening to. This honestly just sounds like another mechanic that they're introducing that will serve the sole purpose of gaming the system and cheesing the actual content in various ways. Nice that they're so focused on putting tools into the game designed to let you subvert the intended mechanics and intended combat flow. By nice, of course, I mean, disappointing.

What we do learn during this point of discussion is that fleeing makes you teleport to your camp... so... no sign of them actually listening regarding the waves of feedback concerning how unacceptable and conceptually conflicting and contradicting the current camp system is with other elements of the game. What we can see this as is a tacit statement that they have zero plans to change the camp from functioning as an immersion-shattering teleport-from-anywhere-and-back-again mess.

“Better downloads – individual patch downloads should be smaller and more efficient now.” - They mention that they've managed to finalise their download and update system now so that you won't need to redownload a mass of repeat data every time they patch somethign, from now on (Patch 4 will be the last time). After that, patches will be smaller and more efficient in size. Hopefully it might passively encourage them to get more tweaks out more regularly so those working on feedback can respond and redirect their efforts in useful ways more often. Fingers crossed.

“UI improvements” - This sounds promising as it's first mentioned, but they talk about it briefly and we learn that he's not actually talking about actual UI improvements to the game... just re-skinning what's there and moving it around, not actually improving UI function so you don't feel like half the gameplay is your player war against it. It's worth noting, they mentioned that the options menu UI is getting significant improvements, which is great, because it was terrible for accessibility and function. Swen took time to comment that “no-one goes to the options UI, can't you talk about a different one”, and that coupled with how completely awful the options UI was/is in its current state actually tells you an unfortunate amount about how Larian are approaching the development of their game.

“Improvements to pathfinding.” - As they discuss this briefly, it turns out they don't mean general AI improvements, so much as that they've improved the pathing specifically around the Mill in the blighted village. Apparently this was a community thing? I've certainly never seen anyone, on any community location mention anything about pathing around the mill.... in other locations in the blighted village, certainly, but not the mill. Maybe I just missed those threads.

“Gith will now have Gith weaponry. Flaming fists have Flaming Fist icons on their shields.” - This is neat, and a nice aesthetic touch, for all of those five npcs that exist.

“Improvements to the event log, in terms of status conditions.” - They tell us that now you'll be able to get some condition feedback if you get a condition in combat. No word on whether this will be more or less useful than the existing ability we have already to mouse over the condition as displayed in boxes above our hot bar already, or to inspect characters and read the description there... but they do say that they're working to add more rules and information feedback to players, and light knows the event log needs to be more robust than it is currently, so this is still good to hear.

“You can disable traps by double clicking on them.” - Now, if you have a trap kit on the character, you'll be able to double click to disarm traps. Not a single click, mind you, like every other sensible game on the market or in development. there was no word on whether this includes all of the vast majority of non-disarmable traps, or if this is just literally, you can now double-click to disarm the traps that you could previously right click and select disarm on... reducing total click requirement by nothing at all. This does also reveal to us that they have no intention of leaving off with trap kits and just letting us use our theives' tools... and of course, we're still using slight of hand, rather than tool proficiency; the game still speaks about them as though they are different, when they are still functionally one and the same. No indication that that is going to change.

“Eating herbs and mushrooms is now free, no bonus action required.” - They announce that eating herbs and mushrooms will no longer require a bonus action, and those particular consumables will now be entirely free. This is puzzling, and it runs exactly opposite of what the vast majority of all voices who speak have to say about this matter.

“Devil's Sight range increased to 80 feet (24metres)” - We're getting into the minutia now; Devil's sight is getting its range extended slightly, to a point that it's still shorter than it should be.

At this point, Swen rushes David along, and gets him to hammer out a few more before cutting him off. They do a gag with an old dot-matrix printer that further adds to the white noise and poor sound quality of this part of the stream. The noise is ongoing through the next section. It only lasts a minute before Swen cuts him of and rushes on to the next section. During that minute, however, we learn these bullet points:

- Stinking cloud won't damage items
- Metal doors aren't vulnerable to slashing damage
- Hunter's Mark doesn't negate other hunter's marks
- Hellish rebuke shouldn't get blocked by terrain (no word on the legion of other targeting issues that exist with spells all over)
- Fixed multiple dashes in a turn failing to apply.
- Removed 1 turn cooldown from dash and disengage (This is really just the fix for the above point mentioned twice, also now thief rogues can bounce twice as much)
- Witch bolt will electrify surfaces, (so it's now as abusable as shocking grasp is with water. I'd prefer it if you couldn't stun a whole room for multiple turns repeatedly without a save with a cantrip, personally, but gaining the ability to do so with a leveled spell as well is not a fix for that...)
- Warlock can't use armour of shadows if wearing armour (a bug they needed to fix because they didn't implement mage armour correctly in the first place)
- Can't summon summons on top of other characters
- Sun-Walker's gift grants darkvision. (it was already working correctly, it just didn't show it)
- Wizard's arcane ward will supposedly scale properly now
- Hook horrors can be silenced (Yay!)

Swen's rushing feels a bit odd when we're only half an hour into the stream, David clearly still had a handful of other notes to share, but the stream leaves him now to return us to staring at half our screen filled with a cow, and the other half of blurry, badly shot Swen at his feast table. For the next three to four minutes, Swen takes the time to talk to us alittle more about how they can't talk about the big thing yet, but that all of that is a reflection of how much work they put in to the game. He mentioned that it's taking a lot more effort than they originally anticipated. Shocking.

We then start the next panel interview, where we get a ten minute break to talk about Larian opening their new studio on Guildford, expanding their team and coverage. Sounds like they have a fairly sizeable experience base, so that's neat. They've been working on it for a couple of years already, so they tell us. This includes a short recruitment plug, and a 4:30 minute video about Larian Guildford and their team members.

After this and another brief break, the cow finally turns into Nick (from accounting, I presume, he never gets an actual introduction, or on-screen nameplate of any kind), accompanied by a few minutes of shaky camera and cross-set calling, and eventually also Jeremy Crawford joining the stream (who also does not get any kind of on-screen nameplate like Swen got, and Swen fumbles introducing his credentials as well). Jeremy wears a very carefully constructed professional smile and reads off a sheet for a couple of minutes to give a description of what Druids in D&D are, before technical difficulties interrupt for several more minutes.

When they return, we get a 1 minute demo video of a couple of druid wild shapes and their abilities; we see a few different forms, including dire crow (flying forms at level 4), Rothe (big cow), Bear (polar), Badger, Cat and Spider. We see during these cuts, and in later ones, that wildshaping to an animal in BG3 does actually deprive you of all of your class abilities and features, and replaces them with a fixed limited set of animal abilities; we aren't able to see whether they handle proficiency gain and ability score inheritance properly, because they don't showcase it. Nick and Swen talk briefly about being initially confused at Druids wielding fire and lightning, and then learn from Jeremy that Druids aren't actually synonymous with tree-hippies, and that fire and lightning are, believe it or not, forces of nature. Jeremy talks a bit about the inspiration for some druid spells, and for the first time since joining the professionally constructed smile gives way to something that looks like genuine interest and passion. The professional smile gets reconstructed again when he finishes this description.

They then talk spells: Flaming Sphere is revealed to be one of the new spells we'll get. (Swen also says that he found a glitch with flaming sphere that he likes to exploit and doesn't want removed). Flaming sphere acts as a summon (thus displacing any other summons you have currently, such as a familiar or an animal companion). It rolls its own initiative (using what stats we don't know), and can move and jump. By virtue of acting as its own creature, it no longer takes up any part of the Druid's turn to control. Thorn Whip and Goodberry are also discussed as new spells, with some short demonstrations (such as whipping a harpy a solid twenty feet off a cliff).

For the next ten minutes, they move on to talking about wild shape directly; they reveal that they thought that changing from one wild shape to another was cool, and so they felt they had to change the rules to allow it... seemingly no-one at Larian understands that assuming a wild shape doesn't deprive you of your class abilities, and so a druid in PnP can already shift from one wild shape to another if they wish to by the core rules alone. In their implementation of wildshape we can see that it does actually deprive you of all of your class abilities and replaces them with a specific animal set... so again, we see Larian's favourite method of doing something badly, then breaking it and putting in a half-arsed fixed to make it almost as good as it would have been if they'd just stuck to the written rules in the first place.

They ask Jeremy if this is okay; his plastered-on smile has barely moved a muscle this whole time. His answer mostly deals with the supposition that by providing only a limited number of shape options, they can afford to allow it to be done more often (which, it seems, you actually can't, in BG3, judging from the videos...). They also call out to mention that there will be no dinosaur shapes, sorry Chult-fans. The reasoning seemed to be mostly down to "Swen says no.", and also resource constraints - though the latter makes no sense, since they also stress that the forms we have available now are just a selection, for now, for EA, and not the full list of options.

They also spend some time talking about using wild shapes for things other than combat, with some video cuts of different wild shapes mostly depicted fighting and killing things. At this point Larian's favourite “animals all share a universal animal language” trope is making its return. Ugh. They describe how druids in animal form will be able to talk to animals and get new dialogues, however, the demonstration that we see at this point shows us that all this really appears to be right now is a form of always-on speak with animals - the dialogue is exactly the same as you would currently get with a SwA potion.

Around this time Swen also claims that he's going to solo a dungeon that has been declared unsoloable... I'm not sure at this point what he's talking about since we've had a number of very vocal people on the Larian forum posting about soloing the entirety of act one and everything in it, quite loudly, along with threads and discussions from people about how to cheese your way through or solo various parts of that full solo run. I'm also unsure how claiming that you can solo group and full-party intended content with the new class because it's so much more OP than other things is in any way a good thing that you should be advertising on your stream. I'm just baffled by that.

We are shown that animals, including heavy-set quadrupeds like rothe (cows) can all climb ladders.. .and oddly they've taken the time and effort to try to make sure they have properly fit climbing animations for doing so... while halflings still have completely broken climbing animations that look ridiculous. Priorities here.

They go on to talk about there being unique quests or quest chains in the and around the druid grove if you are also a druid, which they then spoil the denouement and outcome of immediately. THANKS Larian.

By this point, Jeremy looks like his face is getting tired from his enduring constructed smile, but they give him peace now and say goodbye, because talking about druid background lore was the main thing that they wanted the head of rules from WotC to guest for, apparently... oddly they don't actually get a goodbye from him, or any return gratitude... just the words from Swen that they're saying goodbye and thanking him for coming, branching straight into a 30 second discussion with Nick that doesn't give Jeremy a chance to say anything, before they then cut his camera and he is gone from the stream. Awkward. Super Awkward. I won't say that Jeremy looked uncomfortable all through this – but I will say that he looked to be in a state of diplomatically forced pleasantness, except in the couple of moments when he spoke about the background lore and seemed engaged with what he was saying. It was not a good look for Larian.

After this, they take another five minute break to briefly talk about their Divinity projects, including their progress on the Divinity Original Sin board game. I skimmed this part because it's of no interest to me, apologies.

When we come back, the stream cuts back into a full gameplay video set up, and for the next 1 hour and 10 minutes, the rest of the stream, we watch Swen play through the Auntie Ethel section in an effort to showcase some druid abilities in the field.

We see some new spells:

Enhance abilities – a second level spell with choices, like Hex. Again, we can see that they're just kicking the difficulty with the UI and choice selection vs leveled casting down the road further. That's a problem for later-Larian, which they're not addressing now.

Protection from Poison – The tooltip on this one looks as though it claims it lasts for just 1 round, but on application it does seems to last until you rest.

Moon beam – A druid favourite is making it into the game. The description onthe spell card does say that it affects a creature when it enters the beam or if it starts its turn there, as per the Phb, but when it is demonstrated, that's clearly NOT how it's working in game. Instead it deals its damage immediately upon casting. This is pretty par for the Larian course at this point though, and they show no intention of fixing it.

Flaming Sphere – As mentioned above, we get a clearer look at how the spell functions; it acts as a summon (thus supplanting any other summon you might have) and rolls its own initiative, rather than being a spell that you control with a portion of your turn.

We soon get to see that falling prone still automatically breaks concentration with no save because falling prone still inflicts unconscious (and incapacitated). Swen also showcases his terrible struggles against the UI and the targeting system, difficulties with the camera and an unclear view of what's going on because it cannot actually show both the caster and the target on the same screen at long range – leading to movement mishaps and mistakes. He also suffers from the persistent issues with downed characters at the end of combat rapidly bleeding out, none of which he seems to comment on other than occasional confusion or mild frustrations, and none of which seem to slated for any kind of fix yet. He also showcases how that things that they advertised about targeting through portraits isn't actually working, on the first thing he tries to use it on.

We then get to watch Swen return to camp and screw up their counter-intuitive and needlessly complex bartering system, accidentally giving all of his money to the skeleton instead of buying scrolls. Still no sign of a fix there. He laughs about it. So funny. He also frequently experiences the well known and much reported character selection bug where you need to click multiple times to get it to select. He blames it on his mouse, but we all know better. Why doesn't he?

We learn that while wild shaped you still gain all the benefits of your worn equipment, when we really shouldn't. He praises the cat's stealth ability, while demonstrating that literally the only thing that actually matters is avoiding the sight cones, and that he could have done exactly what he did with the cat with a t-rex just as easily. No sign of any change to this coming.

With the raven form, we get to observe that there is still no actual verticality with flight; all flying things are still permanently rooted to the ground and entirely punchable by melee, club-wielding mooks. No sign of any intention to change this either.

Swen keeps talking about the Badger's burrowing capability as a teleport, and indeed seems to think of it and treat it that way, with no concern – including highlighting the ability to 'burrow' to completely disconnected places with no land connection, or even rickety wooden structures. He then moves on to, in a positively gleeful way, exploit the locked-in-initiative vs. not-in-initiate abuse, to time-stop the entire combat while he messes around with his other party members. No indication that this is anything other than intended... Swen doesn't CARE how mechanically broken and abusable that is. He doesn't CARE how immersion-breaking it is. He LIKES cheating and exploiting. He thinks it's cool and fun/funny. Badgers 'burrowing' into hanging wicker cages doesn't create any kind of cognitive dissonance for him, because he's thinking about it as a teleport, and purely in terms of gaming a system, by whatever exploits he can find. So, it looks like we cannot really expect ANY of this to change. Apologies for the rant...

To continue, we learn that braining allies with potions is fully intended and they're even improving it by making it so that they don't deal damage when you do so. So they claim – the demonstration still did, in fact, deal damage to the target before applying the heal - you can see the damage numbers on the screen - despite both speakers claiming that it didn't.

They then spend a few minutes displaying that they don't know how their own game works; Swen asks for strategies on beating the hag after it all goes badly, Nick suggests a surprise attack with the bow – Swen wants to know if using his action to do that, it will still let him then jump behind her and shove her immediately, nick reassures that it will (it won't, for two distinct and independent reasons; the first being that jump and shove are both bonus actions and he only has one, the second being that when you initiate combat and surprise creatures, you don't actually get the rest of your turn there and then - you all roll initiative). Reassured, he picks an attack and asks if Rain of Thorns is a good choice for this plan; he can see on the tooltip for the spell that it takes both his action and his bonus action. Again, Nick reassures that it is. While this is going on, we're treated to the display that the enemy AIs have access to infinite healing abilities, healing themselves repeatedly while they stand in gale's necrotic aura and not moving out of it. Aside from a giggle bout Gale's necrosis working in their favour by undermining the AI, no other mention of this is made.

After struggling with the UI and the movement and targeting systems for a few more minutes, Swen comments that he's probably boring people. This viewer was feeling more disappointment than boredom, really... but it was not a particularly fulfilling of edifying use of stream time, especially when they had rushed the other stuff so much to get here.

We do get a peek at another spell; Spike Growth. Swen doesn't read the tooltip on the spell card before casting it and then asks Nick why it didn't deal damage to the hags when he cast it. (This viewer is going to lay a wager that when patch 4 drops, or within a couple of hotfixes thereafter, Spike Growth absolutely will deal damage when you first cast it, because Swen didn't think that it was fun enough like this)
We see that Frightened still acts functionally as a turn skip, no sign of any change to that. The party wipes shortly after, and that is the end of the gameplay section, and the stream, more or less.

They round off the stream by telling us that we can expect the patch sometime in the next week, though they aren't sure when, and then show a neat little 2 minute video that does a better job of showcasing all of the things the new druid class can do, than the previous hour managed, with a little teaser/spoiler on what the tadpole power for druids will be.

Frankly, maybe I'm just not the target audience for these kind of panel shows – but I feel like I've seen satisfying reveal and information panels before, and they were fine. This one felt like a LOT of fairly tacky showmanship, smoke and mirrors and attempts at 'wow-factor' to disguise the fact that they don't actually have a lot to talk about at all.

Last edited by Niara; 19/02/21 03:38 AM.
Joined: Oct 2020
old hand
Offline
old hand
Joined: Oct 2020
Don't worry Niara, It Just Works


Optimistically Apocalyptic
Joined: Apr 2020
Location: Boston , MA
addict
Offline
addict
Joined: Apr 2020
Location: Boston , MA
A sloppy Pannel that matches a sloppy game.

It lacks direction and vision, therefore is hard to be hopeful.

Last edited by IrenicusBG3; 19/02/21 03:58 AM.
Joined: Jan 2021
S
addict
Offline
addict
S
Joined: Jan 2021
Don't you know they are designing for millions though?! We just dont get it.

Joined: Feb 2020
Location: Belgium
veteran
Offline
veteran
Joined: Feb 2020
Location: Belgium
Originally Posted by IrenicusBG3
It lacks direction and vision.

100% this.

It looks like every Larian's developper add things they find cool in the game without thinking about it as a whole.

WTF about mushrooms ?
WTF about torches ?
WTF about flee that TP you to the camp ?
WTF about sorting the very limited number of items in containers ?

Listening to him, it really looks like Swen doesn't know the game and its issues at all.

Last edited by Maximuuus; 19/02/21 04:44 AM.

French Speaking Youtube Channel with a lot of BG3 videos : https://www.youtube.com/c/maximuuus
Joined: Jul 2014
Location: Italy
veteran
Offline
veteran
Joined: Jul 2014
Location: Italy
Well, that was a great summary and avividreminder of a lot of awkward moments and complaints going through my head during the vision. Congratulations for the solid work here.

On a side note reading it made me alternately burst out laughing and feeling more and more depressed and pessimistic about the game.

Someone else pointed it out in another thread, but just to reiterate: a big part of the issue here is that Larian not only is purposefully ignoring a lot of feedback, but it PRIDES itself about a large part of the junk they are pushing into the system.

Last edited by Tuco; 19/02/21 09:24 AM.

Party control in Baldur's Gate 3 is a complete mess that begs to be addressed. SAY NO TO THE TOILET CHAIN
Joined: Nov 2020
O
OcO Offline
enthusiast
Offline
enthusiast
O
Joined: Nov 2020
Originally Posted by Dexai
It Just Works

Oh snap, you went there. hehe

Joined: Oct 2020
L
addict
Offline
addict
L
Joined: Oct 2020
I’m with Tuco… this was hilarious and depressing at the same time. Superbly written, if only we could be sure someone at Larian actually bothers to read it.

Joined: Oct 2020
journeyman
Offline
journeyman
Joined: Oct 2020
It was so boring, watch to the end only in the hope of a patch release, when it is possible read patch notes at last...without all this bla-bla. Even this did not happen.

Joined: Oct 2020
addict
Offline
addict
Joined: Oct 2020
I am still convinced that NOBODY at Larian played and finished BG2. NO WAY.

Last edited by mr_planescapist; 19/02/21 08:49 AM.
Joined: Sep 2017
Location: Norway
S
addict
Offline
addict
S
Joined: Sep 2017
Location: Norway
Originally Posted by Niara
“First thing, we're introducing loaded dice!” - That thing that, as far as I'm aware, no-one ever actually asked for anywhere.
Their reasoning for this is stated as: “So, if you roll a 1 occasionally in table top, It's a bit of fun, But if you do it repeatedly in a video game, you'll get frustrated.”
It's a pretty glaring non-sequitur that they brush over - If you roll 1s repeatedly in tabletop you get frustrated as well. If you only roll 1s occasionally in a video game, it's a minor thing. If you're failing a lot more in the video game, then loaded dice aren't the answer. This is a band-aid solution to their poorly written RNG. It may solve the perceived issue, but it's not a good fix, especially when it's flavoured and worded as a flat out accusation of cheating against the player, in the name. (And yes, many have analysed it - their RNG delivers the expected spread in a long term test, but it patterns in a visible sine wave that creates issues in actual play, and results in more failures and more strings of failures closer together, and more failures-against-all-odds, than you find in tabletop, or with better RNGs)

Disagree vehemently with your portrayal of this aspect. You even contradict yourself, pointing at sine waves (clusters) of bad RNG as the issue - while dogmatically brushing off the optional loaded-dice (who breaks clusters) as a solution.

People may not have asked specifically for a loaded dice, but you must have been living under a rock on these forums if you failed to notice (and correlate this implementation) to the long vocal opposition to just about any kind of RNG. I suspect, largely by people with no measurable appreciation for neither D&D, nor for Larian's many efforts to combat binary RNG, nor for the imbalances and need for evermore homebrew these efforts likely will entail (ie. skills and skill-focused classes are nerfed by this, staple abilities like the Barbarian's staple Reckless Attack is worse than useless, martial classes are generally strongly buffed, spells like Sleep or Fireball are nerfed).

Let me count some of the ways for you:
1. Larian increased health and lowered AC as part of this, because missing was never fun.
2. Larian virtually guarantees players advantage on attack rolls (two attack rolls instead of one), because missing was never fun/to incentivize tactical movement. AI does use height for their advantage, but the evil enemies are obviously too gallant to exploit flanking like players easily can -- which gives a lopsided/gimmicky feel to combat.
3. Larian implemented "inspiration points", which is opportunities to re-roll.
4. Larian implemented "illithid wisdom" that can guarantee success.
5. Larian buffed cantrip Friends giving advantage on charisma checks against non-hostiles (D&D RAW makes them hostile which severely limits the use).
6. Larian has implemented a layered skill system where the player most of the time has multiple chances to avoid failure. Even failure is a chance to explore fun roleplaying avenues and not some binary game over catastrophic failure.
7. Save-scumming is made a feature by even allowing quick load/save in mid-conversation.
8. Additionally, the soon-to-be implemented feat Lucky, will provide 3 re-rolls per long-rest. Which in the game is really not much of a limitation due to resting being virtually unlimited.

There is a psychological difference between tabletop and digital gaming: People are more untrusting of processes they can't verify with their own eyes, and as a result oftentimes blame AI/devs for cheating when they simply have real or imagined bad luck. People tend to notice failures/misses a lot more than they notice successes/hits (particularly when the failures comes at critical moments) and this causes confirmation bias. This is amplified by social media/internet where like-minded people subconsciously seek out this bias.
"If you FEEL like failing a lot more in the video game", then this is likely due to this bias or because in the video game you play a more condensed version where more RNG-moments inevitably will be compressed over a short amount of time.

Since we're dealing with a largely psychological phenomenon, there are other ways to mitigate this issue. An old suggestion on these boards is to differentiate between failures in attack rolls. Instead of purely portraying failed attack rolls as misses, we could have the enemy dodge (with some slight alterations to animation) when for instance the dexterity-modifier is the difference between a hit and a miss. The same thing can be done for shields by making block animations when the shield AC is enough to make a hit a miss. The player will feel less failure by not actually hitting the enemy, while also providing more visual variety and feedback that the enemy we fight is quick or skilled/well-equipped.

Slamming Larian for not doing enough to combat RNG-issues, while also slamming them for the unintended consequences and inevitable departure from D&D rules is grossly unfair. Damned if you do, damned if you don't, I guess.

Last edited by Seraphael; 19/02/21 11:23 AM.
Joined: Nov 2020
B
Banned
Offline
Banned
B
Joined: Nov 2020
Well nice summary Niara. It makes it’s very obvious that Marian doesn’t read any forums (not even their own), or that the moderators are tasked to read and report (and those are doing a bad job) or that they fully aware and just give shit on their community.

They don’t even comment on the issues like there aren’t any megathreads about it.

Larian lost touch with its comunity. And instead trying paint their game golden by asking wotc to give them an approval seal.

Joined: Jul 2014
Location: Italy
veteran
Offline
veteran
Joined: Jul 2014
Location: Italy
Originally Posted by Seraphael
Disagree vehemently with your portrayal of this aspect. You even contradict yourself, pointing at sine waves (clusters) of bad RNG as the issue - while dogmatically brushing off the optional loaded-dice (who breaks clusters) as a solution..
Because it IS a poor solution.
The ideal design around a system that is perceived as "way to reliant on RNG and too unpredictable on its outcomes" is to tweak rules or the scenarios offered to give better predictability. Not to cheat behind the curtain lying to the player about the dice rolls.
A classic example is a situation where a failure doesn't force a re-roll or reload of a previous save game, but creates an alternate and equally interesting outcome.


Party control in Baldur's Gate 3 is a complete mess that begs to be addressed. SAY NO TO THE TOILET CHAIN
Joined: Sep 2017
Location: Norway
S
addict
Offline
addict
S
Joined: Sep 2017
Location: Norway
Originally Posted by Tuco
Originally Posted by Seraphael
Disagree vehemently with your portrayal of this aspect. You even contradict yourself, pointing at sine waves (clusters) of bad RNG as the issue - while dogmatically brushing off the optional loaded-dice (who breaks clusters) as a solution..
Because it IS a poor solution.
The ideal design around a system that is perceived as "way to reliant on RNG and too unpredictable on its outcomes" is to tweak rules or the scenarios offered to give better predictability. Not to cheat behind the curtain lying to the player about the dice rolls.
A classic example is a situation where a failure doesn't force a re-roll or reload of a previous save game, but creates an alternate and equally interesting outcome.
Re-read the entirety of my post please, because you obviously didn't read more than the opening paragraph. This is one of MANY "solutions", one of MANY "tweaks".

Past these many efforts, more than dealing with "way too unpredictable RNG and outcomes", we are dealing with the catering to the lowest common denominator of obsessive compulsive behavior coupled with a strong intolerance of any deviation from perceived perfection. "Snowflaky entitlement" issues if you will.

Let's take the situation with the dying illithid at the crash site near the Ravaged Beach to illustrate my point on their "layered RNG-system" alone:

You can kill the illithid from afar (knowing its mental powers and being informed the people are thralls). Opting to not do so and approaching, you have a DC 15 persuasion check to convince thralls to stand down. Fight vs. thrall commences with failure. Mid-fight, you get another DC 10 wisdom check vs. mind influences of the illithid. Success leads to thralls breaking free. You can now again kill the illithid (definitely knowing how dangerous it can be), or you can approach it. You get a DC 10 intelligence to (foolishly) lock minds with the illithid, if you "succeed" you get another DC 10 wisdom check to avoid being killed. Even when killed, you are rewarded with an exciting battle (and, should you win, more XP than otherwise I believe), at the cost of using one scroll of revivify.

Last edited by Seraphael; 19/02/21 11:37 AM.
Joined: Feb 2021
M
stranger
Offline
stranger
M
Joined: Feb 2021
Originally Posted by Niara
This is a synopsis of events and discussions from the recent stream, for those who don't have the time/energy to watch the stream itself.
[snip]

Frankly, maybe I'm just not the target audience for these kind of panel shows – but I feel like I've seen satisfying reveal and information panels before, and they were fine. This one felt like a LOT of fairly tacky showmanship, smoke and mirrors and attempts at 'wow-factor' to disguise the fact that they don't actually have a lot to talk about at all.


Thank you, Niara! I watched the panel and read your summary. Yor comments are very fitting.

Joined: Nov 2020
E
addict
Offline
addict
E
Joined: Nov 2020
Good synopsis. The whole affair left me a bit deflated and seemed to suggest they’re forging ahead with things in much the same chaotic vein. Whilst it would be impossible and impractical to act upon all the variant strains of feedback I do think some of the more prevalent ones could have been acknowledged at least.

For my part I wish they would just come out and categorically state what will be amended and what won’t, if only to put an end to posters going over the same things time and time again.

Joined: Oct 2020
T
stranger
Offline
stranger
T
Joined: Oct 2020
I'll just wait the video made from Sven, a couple months after the release, that will remind me of this one:

[Linked Image from i.imgur.com]

Hubris is one hell of a drug.

This project brings with it a lot of terrible design from Divinity, because this company is just a plaything for Sven.
And no amount of community of colleagues' advice will save it until that man get the **** out.

Joined: Oct 2020
B
enthusiast
Offline
enthusiast
B
Joined: Oct 2020
First of all - thanks Niara for your post. Great summary as always.


Yeah, this patch was the one I was waiting for to see what direction Larian has decided to go with the game and by now I know that BG3 is not the game for me. Watching Sven play was a show of all the things I despise about BG3 and where many people usually come up with 'they are listening, its just EA' - but no. Its their philosphy when it comes to create games.

BG3 is the first game I fully regret buying in a long time. I'm usually forgiving that some game might not be my style of play or whatever, but this one is a design mess by a well founded and large enough studio that could and should do better, while sadly failing at absolute basics. Its not a matter of bugs, but tons of questionable desicions and no coherent vision beyond 'I want chessy ways to handle things, I don't care about the core game'.

Joined: Oct 2020
old hand
Offline
old hand
Joined: Oct 2020
Originally Posted by Seraphael
Originally Posted by Niara
“First thing, we're introducing loaded dice!” - That thing that, as far as I'm aware, no-one ever actually asked for anywhere.
Their reasoning for this is stated as: “So, if you roll a 1 occasionally in table top, It's a bit of fun, But if you do it repeatedly in a video game, you'll get frustrated.”
It's a pretty glaring non-sequitur that they brush over - If you roll 1s repeatedly in tabletop you get frustrated as well. If you only roll 1s occasionally in a video game, it's a minor thing. If you're failing a lot more in the video game, then loaded dice aren't the answer. This is a band-aid solution to their poorly written RNG. It may solve the perceived issue, but it's not a good fix, especially when it's flavoured and worded as a flat out accusation of cheating against the player, in the name. (And yes, many have analysed it - their RNG delivers the expected spread in a long term test, but it patterns in a visible sine wave that creates issues in actual play, and results in more failures and more strings of failures closer together, and more failures-against-all-odds, than you find in tabletop, or with better RNGs)

Disagree vehemently with your portrayal of this aspect. You even contradict yourself, pointing at sine waves (clusters) of bad RNG as the issue - while dogmatically brushing off the optional loaded-dice (who breaks clusters) as a solution.

People may not have asked specifically for a loaded dice, but you must have been living under a rock on these forums if you failed to notice (and correlate this implementation) to the long vocal opposition to just about any kind of RNG. I suspect, largely by people with no measurable appreciation for neither D&D, nor for Larian's many efforts to combat binary RNG, nor for the imbalances and need for evermore homebrew these efforts likely will entail (ie. skills and skill-focused classes are nerfed by this, staple abilities like the Barbarian's staple Reckless Attack is worse than useless, martial classes are generally strongly buffed, spells like Sleep or Fireball are nerfed).

Let me count some of the ways for you:
1. Larian increased health and lowered AC as part of this, because missing was never fun.
2. Larian virtually guarantees players advantage on attack rolls (two attack rolls instead of one), because missing was never fun/to incentivize tactical movement. AI does use height for their advantage, but the evil enemies are obviously too gallant to exploit flanking like players easily can -- which gives a lopsided/gimmicky feel to combat.
3. Larian implemented "inspiration points", which is opportunities to re-roll.
4. Larian implemented "illithid wisdom" that can guarantee success.
5. Larian buffed cantrip Friends giving advantage on charisma checks against non-hostiles (D&D RAW makes them hostile which severely limits the use).
6. Larian has implemented a layered skill system where the player most of the time has multiple chances to avoid failure. Even failure is a chance to explore fun roleplaying avenues and not some binary game over catastrophic failure.
7. Save-scumming is made a feature by even allowing quick load/save in mid-conversation.
8. Additionally, the soon-to-be implemented feat Lucky, will provide 3 re-rolls per long-rest. Which in the game is really not much of a limitation due to resting being virtually unlimited.

There is a psychological difference between tabletop and digital gaming: People are more untrusting of processes they can't verify with their own eyes, and as a result oftentimes blame AI/devs for cheating when they simply have real or imagined bad luck. People tend to notice failures/misses a lot more than they notice successes/hits (particularly when the failures comes at critical moments) and this causes confirmation bias. This is amplified by social media/internet where like-minded people subconsciously seek out this bias.
"If you FEEL like failing a lot more in the video game", then this is likely due to this bias or because in the video game you play a more condensed version where more RNG-moments inevitably will be compressed over a short amount of time.

Since we're dealing with a largely psychological phenomenon, there are other ways to mitigate this issue. An old suggestion on these boards is to differentiate between failures in attack rolls. Instead of purely portraying failed attack rolls as misses, we could have the enemy dodge (with some slight alterations to animation) when for instance the dexterity-modifier is the difference between a hit and a miss. The same thing can be done for shields by making block animations when the shield AC is enough to make a hit a miss. The player will feel less failure by not actually hitting the enemy, while also providing more visual variety and feedback that the enemy we fight is quick or skilled/well-equipped.

Slamming Larian for not doing enough to combat RNG-issues, while also slamming them for the unintended consequences and inevitable departure from D&D rules is grossly unfair. Damned if you do, damned if you don't, I guess.

I think you're misunderstanding the situation. The problem with the RNG isn't just or mainly psychological. The problem is that the RNG is badly coded and gives clumped results. This is a common issue for badly written RNGs.

The problem with the loaded dice solution is then that it isn't a solution -- it is a crutch. The difference between a solution and a crutch is that with a solution the problem is solved, but with a crutch, the problem is worked around but remains. The RNG is like a leaking pipe and the loaded die crutch is a bucket placed under it instead of fixing the pipe. If they said that hey this is a short time measure because rewriting the RNG is a big procedure and will take time, that would be fine. But if this is their long time "solution"... well, it doesn't solve anything.

So no, Niara is not contradicting herself. And the people who complain about misses aren't just silly people with negative confirmation bias. This is Larian making a RNG-based game with a mediocre RNG code. And that might have been okay if this was a small indie game but it's far below AAA quality.


Optimistically Apocalyptic
Joined: Nov 2020
Location: Silverymoon
old hand
Offline
old hand
Joined: Nov 2020
Location: Silverymoon
Thank you for the summary; I much prefer five minutes of reading over two hours of marketing fluff.

The Speak with Dead changes sounded neat at first, but after thinking about it for a minute, they might actually stop me from using the spell altogether. Cutscene closeups of the dead bodies is not something I want at all.

I'll have fun playing a druid. The other changes don't look like they'll affect my game very much.

Page 1 of 5 1 2 3 4 5

Moderated by  Dom_Larian, Freddo, vometia 

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.5