Road to 2.6 | Is the Icewind Dale 2.6.3.0 Patch Ready to Release?

JuliusBorisov · February 2021

Patch 2.6 is close to release!

Steam Beta 2.6.3.0 for Icewind Dale: Enhanced Edition is available now for public testing!

Based on your feedback on Build 2.6.2.0, we've added some fixes and removed a few regressions. HUGE thanks to everyone testing these games and helping us make this patch a success.

IMPORTANT:
Patch 2.6 will NOT be the last patch for Icewind Dale: Enhanced Edition

Patch 2.6 will not fix every issue— but we hope to make further improvements in future
Here's what in the patch: 2.6 Patch Items

System Requirements have changed:
Minimum: Windows 7, 8.1, 10 64-bit, OS X 10.9, Ubuntu 18.04 or equivalent

Please check your device meets the updated System Requirements before testing
Make sure your device supports 64-bit games

TEST OBJECTIVE: Start questing! Is the patch good enough to release?
How to Participate

- Playthrough some quests & VOTE on the release
- Tell us about *new bugs (the beta shouldn't add any issues) - comment below!
- Tell us about major issues affecting saves, visuals, sound & quality - comment below!

How to opt-in & verify you're using the correct version

Please make sure you're using a clean (unmodded) install. Clear your override folder before testing the beta build
In your Steam Library, Right-Click Icewind Dale: Enhanced Edition and choose “Properties”
Click on the Beta Tab
Choose the road_to_2.6 in the drop down menu
If you don't see the road_to_2.6 in the drop down menu, please restart Steam
In-game, on the main menu, you will see that the build is 2.6.3.0

WHAT CHANGED SINCE THE LAST BETA

All Games | Green lines no longer visible in various places
All Games | Interrupting client character creation causes bad state
All Games | Debug Console should now support up/down arrow keys to select previously executed commands
All Games | Nearest Neighbor Scaling should now behave as expected
IWDEE | Polymorphing between two different forms should not create impassable blocks
IWDEE | Fixed issue occurring when casting polymorph on self
IWDEE | Fixed graphical issues when selecting Area Background in the starting area
IWDEE | Fixed issue with Brown Bear polymorph Crashing the game
IWDEE | Fixed issue with item comparison not being present
IWDEE | Fixed crash when entering the second floor of Kresselack's Tomb

We won't move forward with a 2.6 Patch release until we hear feedback from enough Windows, Linux & macOS players for each Infinity Engine Title:

- Baldur's Gate: Enhanced Edition
- Baldur's Gate: Siege of Dragonspear
- Baldur's Gate II: Enhanced Edition
- Icewind Dale: Enhanced Edition

*Note: We're focusing on Windows macOS & Linux now, but stay tuned for details on the 2.6 Patch for Android & iOS!

Join the Beta & Vote Below!

Testing Status:
6/8 Windows Players
1/3 Linux Players
0/3 macOS Players

500MetricTonnes · February 2021

Edit: I meant to post this in the BG:EE thread, sorry. I am unsure if this issue also occurs in IWD.

There's still a graphical issue that has persisted since at least 2.5, where animated background tiles do not appear to line up correctly when bilinear filtering is used.

Note how the spinning crystal in Thalantyr's tower looks seamless when nearest neighbour scaling is on:

But when nearest neighbour scaling is OFF, there is a noticeable seam:

Galactygon · February 2021

It's a known issue for large animations including large creatures which consist of multiple .bam sequences sewn together side by side. It's a legacy of the pre-EE engine which was unable to handle large .bam frames. EE can handle large frames otoh, so sewing together those multiple .bams into a single large .bam would be the best way to fix this issue. Whether or not it will be fixed is tbd for a future patch as there are quite a few animations requiring this treatment, not to mention BG2's dragon animations. It sure would be nice to see this fixed at some point in the future, I completely agree on that!

_Luke_ · February 2021

@Galactygon

[existed in v2.5] SPWI418D.SPL – The 'Range' field (Extended Header at 0xE) should be set to a very high value (32767 ideally) and the 'Projectile' field (Extended Header at 0x26) to a new Projectile that should be coded as suggested here:
- The same applies to all those spells such as Fireshield, to control the range of the feedback damage without generating the "One of your spells has failed" message.
[existed in v2.5] Quivering Palm – SPCL820.SPL and Stunning Blow – SPCL811.SPL should affect only fists ('parameter2' of opcode 248 should be 4 instead of 0, otherwise all melee weapons are eligible).
[existed in v2.5] CLSSPLAB.2DA is missing the SHAMAN entry (which should be `4 6` like the CLERIC/DRUID entry). Moreover, it's unclear if it's intended or not that MONKs are treated (i.e. use the same values) as CLERIC/DRUID instead of as FIGHTERs...
This 2DA is for use with the Cat's Grace spell – SPWI225.SPL (in particular, it's related to opcode #15,p2=3...)
[existed in v2.5] Otiluke's Freezing Sphere – SPWI628.SPL: all its effects should NOT bypass Magic Resistance.
[new in v2.6] SPWI626.SPL – Lich Touch: Spell description should mention the duration of the Paralysis effect (which appears to be 600 seconds – 10 in-game Turns).
[existed in v2.5] SPWI224.SPL – Glitterdust: certain effects bypass Magic Resistance, whereas other not. None of these effects should bypass Magic Resistance. Optionally, you could delete op#142 (p2=8) since the Blindness portrait icon is built-in the Blindness opcode...
[existed in v2.5] Polymorph Self, Shapechange, Druid Polymorph abilities and the like:
When you shapechange between two different non-natural forms, you will lose effects from equipped items.

I know you said this is a code issue, but for the time being, you might want to consider the idea of either using opcode #53 (p2=0) or setting op135's parameter2 to 1 (Appearance only).

Moreover, I have other things that require your attention, this time regarding spell casting opcodes and subspells, just give me the time to gather all my info...

ProfErrata · February 2021

I suggest strongly that Beamdog put a close eye on the Luke's fixes before the release of the next version of IWDEE…

JuliusBorisov · February 2021

We greatly appreciate @_Luke_ 's feedback (and you can see the list of recent fixes already incorporating some of that), but the patch 2.6 will not fix every issue. After all, the beta for 2.6 started in July, 2020 (more than 6 months ago), it can't go on forever.

We hope to make further improvements in the future.

Doubledimas · February 2021

JuliusBorisov wrote: »

After all, the beta for 2.6 started in July, 2020 (more than 6 months ago), it can't go on forever.

Although probably technically correct, in the months before november, everything was focused on just testing if saving and loading worked. In the older threads the message that was displayed:

Don't test too much. While you're free to explore the whole game, please focus on the objective. Once you hit the objective or a major bug, report your feedback and stop playing the beta for now. We'll be back with more test objectives that we hope you'll take on in the future

Only since November has there been a list of bug-fixes available and is it the explicit goal of the beta to test the whole game.

So saying it started in July 2020, while true if you look at the whole 2.6 phase, feels not like the real start of the bug-testing phase as real, solid bug-testing has, in my view, only started with the release of the bug-fixes list, somewhere in November if I'm not mistaken (time-stamping on this forum is really inadequate). Because before that Beamdog asked us not to start testing too much already (as displayed) by the message I quoted above.

Galactygon · February 2021

@_Luke_ Unfortunately we had to shift gear and prioritise other issues for this build as we're wrapping up work on 2.6. Rest assured, the fixes are already submitted internally ready to be applied; they will simply be part of 2.7 instead. Thanks for your valuable contributions.

Khyron · February 2021

Alright.. that's it. I am making a call.. and I am calling out Beamdog on this.

You're making another entry in the line of Infinity Engine games.
I'm not gonna speculate if it's BG, IWD or something completely different.

But you are making another IE game/campaign one way or another.

It makes absolutely no sense to keep patching this game if your goal is to make money.
Which I assume it is.
You won't be selling much more of the already released EE games, they've been out for several years already.. patching games this old makes no sense at all, unless your patches are Engine level changes and improvements which swipes across all EE games, paving the way for a new title to be released on a tried and tested engine.

Confirming that 2.7 is a thing makes me convinced there's something in the pipelines.

Let the speculation commence..

Continuin speculation topic here, to avoid derailing:

https://forums.beamdog.com/discussion/81528/speculation-beamdog-is-making-another-ie-game-campaign/p1?new=1

Cahir · February 2021

There is nothing from IE engine in the pipeline. Beamdog just cares for these games and want to leave them in the best state possible. Patching takes time because BD is a small studio and this task is far from number one priority at this point. And patching these games absolutely MAKES SENSE.

What I'd personally see in 2.7 (since there were several confirmations that it WILL come eventually) are UI fixes (reported already by the community) and more engine externalizations and modders' quality of life features. Maybe there is something from EEex that can be adopted as well. There are tons of tons of things modders could do with these games yet, if they have the right tools for it.

_Luke_ · February 2021

@Galactygon

Even if you are basically done with v2.6, you might want to add the following to your to-do list (many thanks to @kjeron who helped me finding the following issues/quirks!)

[existing in v2.5] Mold Touch spread – "SPPR326B.SPL": The caster should be able to contract the mold if he/she gets too close to someone that is infected, so you might want to remove the "Ignore Center" bit (@ 0xC) from "idmold.PRO".
[new in v2.6] Vitriolic Sphere – "SPWI432.SPL" is partially broken.
- The number of delayed op146 effects (i.e. the number of times the initial target is given a chance to Save vs. Spell to avoid further damage) should be on par with the number of delayed op12 effects.
  
  As a result, the first 3 abilities (MinLev = 1,8,9) need more. The delayed subspell and delayed damage should be in equal quantities, with the subspell 5 seconds before the damage. The last 3 abilities (MinLev = 10,11,12) are correct.
- Description states that this spell splashes acid in a 5-ft. radius around the primary target ("SPWI432Y.SPL").
  
  Once again, the caster should probably take this "splashed" damage if he/she is too close to the primary target. As a result, you might want to remove the "Ignore Center" bit (@ 0xC) from "VSPHERX.PRO"...
  
  BUT, as you surely know, that will introduce another issue, namely the following one:
  The "Ignore Center" bit enables subspells to check Magic Resistance when they are applied through op326/333/146*p2=2 ("SPWI432Y.SPL" is indeed applied through op326).
  Without that bit, the secondary targets would not get a Magic Resistance check, which would be incorrect.
  
  As a result, if you want to have both a Magic Resistance check and affect anyone (including the caster!), you could use a two-stage subspell, something like:
  1. Main (single target) => op326 => subspellA to main target (no resist)
  2. subspellA (area-effect) => op326 => subspellB to each target (no resist)
  3. subspellB (single-target, ignore center): applies effects to each target (resist)
  That should also fix/bypass another known issue involving op326/333/146*p2=2, that is:
  those 3 opcodes break when they interact with Deflection/Turning/Trap opcodes. For Vitriolic Sphere, creature(s) near the primary target will still take splash damage, the primary target will be offered the saving throws to end nothing, while the original caster suffers the full main damage without any saving throws.
  
  However, I'm not 100% sure how exactly this spell should be changed, so if @kjeron could provide a full example, that would be great...
[new in v2.6] Shroud of Flame – "SPWI524.SPL": this spell also needs a two-stage subspell (since it is using op333), similar to Vitriolic Sphere, with the difference that all spell casting opcodes should NOT bypass resistance (since this time the spell is not meant to burst upon impact, so if the primary target resists, then nobody will be affected...)
[existing in v2.5] Mordenkainen's Force Missiles – SPWI430.SPL: this spell also needs a two-stage subspell (since it is using op333), similar to Vitriolic Sphere.

However, in this particular case, there's no reason for Mordenkainen's Force Missiles to not use op146*p2=1 instead of op333 (since `# hits` is always equal to 1). As long as the subspell (namely "SPWI430B.SPL") is of type (@ 0x1C) Wizard, it will be cast at the same level (i.e., Casting Level is preserved).
About subspells in general (all games):

Primary (@ 0x25)/Secondary (@ 0x27) type should always match parent spell, both for interaction with opcodes 220/221/229/230, and with the specialists save bonus/penalty (and op346). Also for Opcodes 200-205, 223, 226-228, but those should already catch the main spell.

For instance, "SPWI432Y.SPL" is flagged as `NONE`/`NONE` instead of `CONJURER`/`OFFENSIVEDAMAGE` like its parent "SPWI432.SPL"...

This issue seems to be common (see Shroud of Flame – "SPWI524C", "#SOFLAMC", Mold Touch – "SPPR326B", Otiluke's Resilient Sphere – "SPWI413A", Wall of Moonlight – "SPPR428A/B", Spirit Fire – "SPPR450B", Animal Rage – "SPPR522B", Soul Eater – "SPWI631B.SPL", IWD:EE version of Death Fog – "SPWI614B/C", Bigby's Clenched Fist – "SPWI818B", others...?)

Also, unrelated to subspells and spell casting opcodes, but still relevant:

[new in v2.6] Blood Rage – "SPPR422.SPL" is broken.
- There are about a dozen extra op321 effects removing ITSELF, scattered throughout its Extended Header. Only about 25 effects (of 90) actually stay on the creature.
- The op318 effect towards the end, should be the absolute last effect, and is there to prevent you from recasting the spell before its fatigue period ends, but it also has the wrong duration (should be 240 since the fatigue period kicks in after 120 seconds and lasts for 120 seconds).
[existed since forever...?] You should properly distinguish Hold (opcode #175) from Paralyze (opcode #109) when it comes to opcode #101 (all games).

"Hold" is a mental effect, whereas "Paralyze" is a physical effect (typically chemical, intended to disrupt the nervous system to prevent the body from receiving signals from the brain).

As a result, all Berserk/Rage spells (e.g. Blood Rage – "SPPR422.SPL") and class abilities (e.g. Barbarian Rage – "SPCL152.SPL" and Berserker Rage – "SPCL321.SPL"), Chaotic Commands – "SPPR508.SPL", Mind Blank – "SPWI802.SPL", Impervious Sanctity of Mind – "SPPR733.SPL" and the like, along with items such as the Greenstone Amulet ("IPSION.ITM" in IWD:EE, "AMUL17.ITM" in BG2:EE) should NOT grant immunity to opcode #109 (i.e., they should NOT grant immunity to the bite of a Carrion Crawler or Ghoul/Lich Touch).
Only Free Action – "SPPR403.SPL" and similar spells/abilities should protect against both forms as per P&P...

This tweak is similar to the one about properly distinguishing "Sleep" from "Unconsciousness/Knockdown/Nausea", except that it should be much more easier to implement/code since we already have two separate opcodes (so no need for external EFFs or a new opcode...)
[existed since forever...?] Hakeashars and Nishruus are partially broken (all games).

As you surely know, the script trigger "SpellCastOnMe()" does NOT work for SPL files whose Ability target (@ 0xC) is `4|Any point within range`. As a result, the following block (from "NISHRUU.BCS")
```
IF
	OR(3)
		SpellCastOnMe([ANYONE],CLERIC_DISPEL_MAGIC)  // SPPR303.SPL (Dispel Magic)
		SpellCastOnMe([ANYONE],WIZARD_DISPEL_MAGIC)  // SPWI302.SPL (Remove Magic)
		SpellCastOnMe([ANYONE],2326)  // SPWI326.SPL (Dispel Magic)
THEN
	RESPONSE #100
		Kill(Myself)
END
```
will NEVER return true (since all those spells have an Ability target of 4). And even when you directly click/tap on the in-game sprite, you still target its location, not the creature...

The most straightforward fix would be that of deleting that block from "NISHRUU.BCS", providing the relevant creatures (e.g. "NISHRUSU.CRE", "HAKSU.CRE") with a unique identifier (f.i. via the Specifics field @ 0x274), and finally adding an op177 effect (p1=%new_specific_value%,p2=6,resource=DESTSELF) to all Dispel/Remove magic spells/abilities.

EugVVl · February 2021

I found a small issue. I am playing with Russian translation and the title of the local map screen is in English. In L_ru_RU.lua

AREA_MAP_TITLE = "Карта местности",

but title is "Area Map".

Update: The same issue exists in BG2:EE.

Baptor · February 2021

Cahir wrote: »

There is nothing from IE engine in the pipeline.

That's a shame, really. I felt like BD was finally ready to make a full length game. It seemed like a natural progression: 1) upgrade originals to EE, 2) add content to existing originals, 3) create an expansion for an original game...it just seemed like 4 would be create a full length game.

I think BD could do it, and I would certainly buy it if they did. Shame they won't.

Hafir · February 2021

Can you sort one bug as well? I mean on HoF when you leave your companion for a while then you let them join again they have extra hp sometimes 3-4 times or more more.

JuliusBorisov · February 2021

Hafir wrote: »

Can you sort one bug as well? I mean on HoF when you leave your companion for a while then you let them join again they have extra hp sometimes 3-4 times or more more.

Can you please provide a save?

Hafir · February 2021

Done

_Luke_ · February 2021

@Galactygon

Since a new beta build is coming, I'd like to report another issue in spell implementation (affects all games):

Let us have a look at "SPWI818A" – Bigby's Clenched Fist and "SPWI918A" – Bigby's Crushing Hand.

It seems you have already bumped into the op39 issue I told you some time ago, and this is exactly the case.
In these two SPL files you apply an op337 (p1=39|Sleep,p2=101|Immunity to effect) effect right before applying op39, most likely because that op39 effect is not a "real" sleep effect, but something like "unconsciousness / knockdown", so you want to be sure the spell can affect creatures such as the Undead.

However, there are 2 issues:

op337 always acts as permanent (even if applied with a 0 second duration)
as already said, op39 is used for everything (Sleep/Unconsciousness/Knockdown/Nausea)

As a result, you're basically preventing Undead, Golems, Dragons, Slimes and the like from being immune to Sleep effects, which would be incorrect. So for instance it could happen the following:

Cast Bigby's Crushing Hand – "SPWI918" at a Skeleton.
If the Skeleton fails its Save vs. Paralysis, the subspell "SPWI918A" will kick in (along with op337).
From now on, that Skeleton will be vulnerable to the Sleep spell "SPWI116.SPL" (incorrect).

As already said, we either need an alternative Sleep opcode or you need to rely upon op318 + EFF files...

Galactygon · February 2021

Yes, unfortunately that is true, even equipped effects aren't refreshed as I thought they would be. That effect has already been removed in all internal builds @_Luke_ Chill Touch in IWD:EE will need to be addressed as well, most likely in 2.7.

Many of the other reports submitted by various users has made it in the current build. Hence the extra delay for the final release.

_Luke_ · February 2021

Galactygon wrote: »

Yes, unfortunately that is true, even equipped effects aren't refreshed as I thought they would be.

Yeah, equipped effects should refresh only upon reloading a game...

Chill Touch in IWD:EE will need to be addressed as well, most likely in 2.7.

Right.
However, this should be easier to fix. If you set `parameter2` of op24 to `1` (or any non-zero value), it should bypass the immunity granted by op101 (see the IESDP...)

Also, did you perhaps manage to have a look at my other reports (including those related to the BG games – in particular Hold Person/Creature spells and "BDSUM00.BCS")? You know, so as to know if those spells/things are indeed broken or not...

Balquo · March 2021

@_Luke_ BDSUM00.BCS should be fixed in the latest beta. Not sure about the spells though. Those may have to wait until 2.7, if they are not included now.

Road to 2.6 | Is the Icewind Dale 2.6.3.0 Patch Ready to Release?

Comments