Vampire: The Masquerade - Bloodlines (DX9)

[left grove]

I see

I'd prefer going with an open source approach so that anyone could contribute

[amber atlas]

and that would literally be NONE of the remix stuff, just the actual playable they load over their VTMB

yeah im saying that as an example

you could keep the repo open, but have the OUTPUT be on moddb

[left grove]

yeah that's fair

[amber atlas]

because oh they only way to play the remaster mod is to download Remix and go on github?

[left grove]

yeah, it's more like your average person will go to the moddb and just play that

[amber atlas]

yes

[left grove]

someone else may want to look at the source and tinker with it a bit

[amber atlas]

contributors interact in the open source way

yes!

[left grove]

yeah definitely agree

the other question is moddb release cadence, do we release after every map we remaster/retexture, or only towards the end?

[amber atlas]

not sure, i did create a moddb page, right now it's set up as solo, but i'd like to change it up to be open/reference back to everyone
or, multiple mods can also be started up depending on how people want to approach the remastering etc

[neat spindle]

Okay now I'm not sure what you're talking about, this is already how things work. What we have on our repo is a raw project essentially. Remix has an export function that will package everything for an end user that we would be distributing on ModDB.

[amber atlas]

Yeah, I'm saying a moddb page hosts the mod/structure that is required to install to be a CONSUMER of the product
That product is a branch that inherits from the base, the base being something anyone can clone down and build their own project from

OR

They can contribute to the branch which is taking the remaster further

[left grove]

I believe your haven one should be a standalone completely

as a POC

and then we can make a shared one, with the same assets if you want to

and use that as a this is the mod for the full game, that we will be building over time

[left grove]

this pretty much

I just think Safemilk deserves all the credit for the safehouse one

[neat spindle]

Sorry, again I'm confused by what you're saying. I feel like you're conflating two different things here.

An end user will get an exported version. Someone who wants to make their own VTMB Remix will need raw project files. These are not the same.

[amber atlas]

im making a diagram

[left grove]

yeah but I believe safemilk also means a more raw version in case someone else wants that

just as a baseline

[neat spindle]

Yeah, I got that. But then they started talking about end-users

[amber atlas]

this feels a lot more charged than i thought it was going to be

let me plot this out and maybe it will make more sense, me typing it isn't doing me any favors

[left grove]

ok ok sure

[amber atlas]

at that middle layer, technically there is nothing different between those two branches or clones of the base

the orange one is for example, THIS group's remaster

any of these things can but don't have to produce a playable

so the community uses catto's framework on their own, or contributes to the ongoing remaster branch that's built from it

[neat spindle]

So the problem I see with this model is that the base branch isn't a playable project suitable for an end user. I think there's value in having a raw source branch that's just texture flagging and basic settings so people can make their own projects, but I don't think it's something that should be distributed to the end user on its own.

[amber atlas]

thats the bottom middle block

it can stay

or go

[left grove]

jsut a bit - phone call

[amber atlas]

you dont NEED to output something

so like, this model still works if you remove the bottom middle block and say, ok there is no base version people can download and play

[neat spindle]

But yes, that fundamental thing I do agree with.

[amber atlas]

i mean the thing i see is, person A clones SOURCE, then changes nothing, puts up moddb page of remaster

so like... why not just fill that gap?

[neat spindle]

I'm not sure how much mileage it will get, but it requires very little work to set up a base source branch people can use so we might as well.

[amber atlas]

yeah if the mileage vs the effort ratio is way off then its not worth it

but basically at that point we are not filling a gap where users will now, if they just want to see the project as close to original as is with RTX lighting and nothing else, they cannot do that easily, maybe that's ok?
but it does seem like a use case that could be up and running sooner than the potentially multi year project to remaster all assets too

if its open source and someone can do it anyway, we should just do it and represent the use cases for the community and help them do that

i think?

i dont know

just some thoughts/feelings i've been having about it

[neat spindle]

Personally I'd like to limit splintering just because of the scope of the project.

[amber atlas]

splintering in terms of end product?

or splintering in terms of branches?

[neat spindle]

Branches. At least initially. Like if we have a dozen different VTMB Remix projects with different teams that will have a lower chance of any of them making it to completion than a single centralized project.

I think remaining open source is important so people can make their own project or split off if we fail etc. But I feel that for now we would be better served trying to grow a single large team rather than enabling a bunch of a little ones to emerge.

IMO a Remix project is fundamentally different from, say, a texture pack, because it's a holistic overhaul of the game. You're not just swapping out individual models or textures, you're compiling a complete new look by adjusting lighting with asset replacement.

[amber atlas]

Yeah those are definitely valid points.

[neat spindle]

I guess to kind of summarize:

My vision is to make a cohesive overhaul that covers all of VTMB, made by a team who are coordinating to realize a collaborative aesthetic.

This is as opposed to having, say, "Safemilk's SM Haven" and "FF's Confession" etc.

[amber atlas]

The only real inflection point in the end I guess is if that middle bottom box exists or not which is where I think yours and my thinking diverges a bit.

Also yeah I don't envision a world in which there's a mod that's like... Safemilks haven etc

But I also didn't think anything I was saying was advocating for that amount of granularity in active projects

More advocating for community flexibility

[neat spindle]

Prior to the recent DXVK developments I would have been more open to the bottom middle existing, but since the recent updates have made every texture with an emissive map becomes primarily transparent by default I don't see a value in distributing that.

The point where we agree is on the root of the tree existing: a raw source project repo with textures flagged and basic settings for anyone who wants to start their own project from scratch.

[amber atlas]

I think that's another super valid point it definitely creates a lot of work. I guess the question is will that work need to get done regardless? Not if we assume we are remastering every texture. Which is the plan of course.

But I guess if we provide a cloneable framework where 90 percent of textures are broken as is then is it viable for other hobbyists? Definitely a lot of moving parts to the decision...

[left grove]

haven't fully read everything, but I do agree we shouldn't really be saying X's this map or Y's this map. I do believe we should have a credits bit tho, thanking everyone who did help in any way

[neat spindle]

Yeah, the sheer amount of work to make a Remix project is kinda why I just am not sure anyone other than our team (however many people it eventually expands to) will be making one.

I'm thinking of our project's scale on the megamod/TC side. Black Mesa/Skyblivion/Fallout: New California.

[left grove]

it is a joint effort to get the E2E game remix remastered

[amber atlas]

i guess there is the point you made earlier about mileage, vampire is more of a niche game, so the subsection of contributors is an even smaller slice of that smaller slice, so finding quality contribution for the project may be difficult to do at meaningful scale at any consistent level

[neat spindle]

I think the one advantage we have is that people who will be into VTM are more likely to be artsy types, haha. Similar to how many physics/math/computer nerds HL attracts in its modding community.

[neat spindle]

Oh, something else I remembered! In the Remix Toolkit you can set existing mods as dependencies if you want to make an override. So someone could just do that with our mod.

Like if they wanted to make Vampire: Toon Masquerade where everything has cel-shaded textures they could easily do that.

[left grove]

oh god

made a start on that

ignore the hideous texture it was my first attempt

[bleak barn]

Can someone pin that video so we don't lose it?

[left grove]

anyone know what sort of noise patterns this would be? (using InstaMat)

I will make that ceiling

[left grove]

In my defense it did look a lot better on InstaMat

[neat spindle]

You know that part in the tutorial where he makes the weird brain texture on accident? Copy that and then invert the mask.

[amber atlas]

https://www.youtube.com/watch?v=fvcCamGa88Q

this is for substance designer

i dont know what nodes you have access to in there

but i bet you could use some of the same principles

and get some sharper peaks etc

though this does rely on a node i haven't seen in previous versions of substance d, but maybe theres a noise generator in there than can get similar results?

[left grove]

I think so the tutorial guy was mentioning it a lot

and the process boils down to blending a bunch of stuff together (for my simple brain )

[broken scaffold]

Not to butt in but is there any additional issues you guys are seeing as a general source engine thing other than the sudden jitter when the sky disappars / reappars on the vis?

[neat spindle]

I think the closest starting point is Voronoi Noise, maybe Slope Blur it or blend it with a cloud noise texture.

[broken scaffold]

Just want to see wher to focus the last of my efforts

[neat spindle]

All of the door frames don't render properly, can't remember the term for what those are.

Brushes, that's the term

[amber atlas]

the disappearing camera, and some of the doorframes conistently rendering poorly are the biggest open item that seems to be plaguing our version of source, other than studiomdl compiler just not working anymore which is a WHOLE different issue outside of remix issues

[broken scaffold]

Get me a pic of the doorframe issue

as for the disappearing camera and the vis

I can look into it

but this is more likely a surce engine patch needed

fingers crossed tho

[neat spindle]

Oh, and the portal thing for the TVs.

[left grove]

TV portals work, right?

[neat spindle]

Oh, is that fixed now?

[left grove]

I had footage of them working did it get broken?

I can check quickly

the TV works

but I loaded a save while next to it and it has no audio 😂

works fine

[neat spindle]

Oh yeah, we still need to figure whatever that bug is that causes assets to not load properly when you load a save in a map.

[left grove]

is that a remix thing?

[neat spindle]

@broken scaffold Brush issue, plus a new oddity.

[broken scaffold]

Are those things in the skybox?

check the scale when you arei nthose positions

[neat spindle]

They definitely shouldn't be skybox objects, one of them is a character model

[broken scaffold]

Yeah and the skybox looks corre4ct

[neat spindle]

Also it's an interior location

[broken scaffold]

I was just thinking maybe the skybox like

mesh is seeping in

and it's causing a Z issue

but that doesnt' make sense here

especially not with a body like that

any cutscenes

is there a TV screen that leads here?

[neat spindle]

There's a CCTV system on the second floor

[broken scaffold]

do those monitors point to that room at any point?

[neat spindle]

Depends on how you define "point to". The CCTV camera position is in a room on the second floor.

[broken scaffold]

Well basically is that room visible from the CCTV

[neat spindle]

Then no

I don't know if this helps, but if I go back out of the hallway area the objects stop showing through the walls. That might just be object culling, though.

[left grove]

inital test for the ceiling texture

looked a bit better on instamat

tbf this one is for skyline appartments it should look nice, not scuffed

[amber atlas]

nice start! I can't help but feel like it might not be 'tiling' enough? like the pattern is very large, i would increase the tiling amount and make those details more numerous and smaller, also it kind of looks... inside out in vampire, from what your albedo map is showing there, or maybe it's just an odd angle?

[left grove]

I believe I haven't gotten the height right either

[left grove]

I am aiming for this type of texture, but maybe that's the wrong ceiling, since I don't know what american ceiling are like

maybe I need some research. but any feedback is good!

[left grove]

may be the angle yes

[amber atlas]

yeah on the albedo it looks like the they are divots going in, but in game it looks like they are popping out toward me

[left grove]

ah I see

maybe I did get it wrong

so

[amber atlas]

could be just flipping the green channel in the normal map

so maybe a directx vs opengl normal map value thing?

[left grove]

I did set it to directX even on instaMat

I am just curious as to what maps to what

1 sec

so after ingestion I got the Basecolor.a.rtex.dds, height.h.rtex.dds and Normal.n.rtex.dds

Basecolor.a.rtex.dds -> Base Material / Albedo
Normal.n.rtex.dds -> Base Material / Normal map
height.h.rtex.dds -> Displacement / Height Map

this is a question ⬆️

[amber atlas]

to be fair also, the way remix does heightmaps is through recession, so white is your highest point and anything after that goes INWARD

so its possible that could also be affecting it

some rendering engines do gray as the middle point and do inset and outset

but remix does inward only currently

[left grove]

oh I see

oh my inwards displacement is set to 0.01 because 1 was too much and that seemed best

should I minus it?

[amber atlas]

i think that's probably a fine value!

but im saying the map itself might be pushing things in you don't expect them too, it might not be

[left grove]

I see

worth mentioning I have no idea what I am doing with my material

[amber atlas]

would you mind putting your non ingested normal map in here real quick?

i want to see where the green is at 😄

[left grove]

ah sure

my bad, I went for a huge break and got side tracked

I have changed things up now tho, this is what I got

It looks like a hive pattern now

[amber atlas]

these look inset to me in an opengl way

in open gl green is the 'top' of a surface

in directx green is the bottom

so if you ingest as opengl it should look the same here, or you could flip the green channel and ingest as directx

[left grove]

oh I see

[amber atlas]

i just work in opengl, since it's the more common format and remix processes it

[left grove]

so probably something I need to figure out

[amber atlas]

when you ingest you could just set it to normal map in the top section and try it out and see if it looks more correct

if you ingest this as directx those little dots will be popping out, maybe thats what you are going for though?

[left grove]

I ahve this

oh I do want them to pop out

I think

[amber atlas]

ahh, i feel like you have your surfaces inverted

[left grove]

I am aiming for something like this

[amber atlas]

yeah, the things that pop out arent the wide flat areas

they are the thin peaks

[left grove]

yeah, I mean I have got it quite wrong atm

I am trying to figure out whether I should make it taller per say and then remove a bunch so it only has tiny bits popping out

or if I should just add bits to pop out, except I haven't figured out how to do the latter

bear with me

[amber atlas]

to me its almost the same as making little mountain ranges

[left grove]

yes... except I have no clue how to make little mountain ranges

[amber atlas]

i don't see it as bearing with you lol, i feel like its just, giving each other feedback etc!

[left grove]

I am guessing adding some pattern that looks like it as multiply or maybe adding

thanks

I mean I knew nothing yesterday and look at me making textures

[amber atlas]

yeah you are picking this up super quick its nuts

[neat spindle]

I didn't realize OpenGL was the norm, I wonder if that's what InstaMAT exports Normals as by default.

I might have to re-export some textures when I get home.

[left grove]

thanks! @neat spindle picked it up super quick I aim to be as good as he is

I did set my height map on InstaMat to DirectX

[amber atlas]

yeah you guys just suddenly are making textures out of nowhere its wild

im not sure as far as instamat goes i haven't used it so im not certain on output

freedoms flame knows better there

also

i think this is more of what you are looking for?

[neat spindle]

Free software tends to have that effect, haha.

[left grove]

woah that's beautiful

[amber atlas]

my old version of substance designer doesn't have voronoi so i mixed some stuff to make something similar

just a cells noise blurred at different amounts and blended back together and then a little black and white spots noise

i think this will get you a shape closer to that reference you posted

i think you're super super close with what you have

[neat spindle]

Using a Levels node to crush out the lower values should help achieve the effect of just having peaks too.

[amber atlas]

yeah! levels nodes are... the best

you can also blur results too, so you can really get in there and add or remove detail when needed etc

procedural workflows are funnn

[left grove]

thanks, just trying to process what I am reading here and trying to replicate what's above

I used chells and blended with clouds

[neat spindle]

I think the step you were missing was the Invert.

[left grove]

oh invert

going to try this, see if I can get these little pertruding things

but it does look funky

[neat spindle]

Like Safemilk said, levels nodes are your friend. You can adjust where your midpoints and peaks are if it's too harsh.

[amber atlas]

yes, basically this is just mixing more things in

same base

so im just pushing values up and down

on top of the base that i have

if i wanted to take it further, i might mask out the higher frequency noise from the peaks to smooth them out

i also use some of this + other grunge maps to do roughness variation

and i convert the height map to a normal map and adjust it's values as well

i think the new version of substance designer just comes with voronoi noise which would make my life easier, but the same effects can be built!

[left grove]

damn that looks amzing

[amber atlas]

just a few nodes! you got this!

[left grove]

can you screenshot the diagam closer please?

[amber atlas]

sure!

[left grove]

thanks!

but the fact you made that in like no time

amazing, I am always amazed by people that have mastered a skill

[amber atlas]

oh dude i'm DOGWATER at material making, the stuff they are doing on the half life 2 team is the REAL stuff, can't wait for everyone else to see what the material makes have done, they are dang heroes

[left grove]

I think you are decent

if not amazing

[amber atlas]

U ARE

also, to loop everything around to the discussion freedom and i were having the other day

[left grove]

wait decent sounds like an insult in this case

[amber atlas]

hahaha

im very pleased to be decent let me tell you what

[left grove]

you are great at this don't undersell yourself

[amber atlas]

this is the 2 worflows coming together!

blender model to texture bake

then pulled it into substance designer

and im combining it with the wallpaper texture you've already seen from earlier

[left grove]

woah, amazing

[amber atlas]

so id have to build the wood material and then mix it in using that color mask i made

this way, we don't have to spend hours trying to recereate complex shapes in the procedural tool

i dont know if you can do it in your guys tool you've been using, but any graph i make can be used as it's own node, so if i make a wood graph, i can bring it in wholesale as it's own node to add into this one

so you can essentially build a library of nodes that make variations

like how freedom has the stone texture, if he makes the lower wood texture, he could keep them separate and combine them as he sees fit

[neat spindle]

Yeah, you can add materials to a package and use them as nodes.

[left grove]

Different lightings, look at what I have got 😎 and criticise my diagram

going to import to game now

still pretty crappy compared to yours @amber atlas

but getting there

[amber atlas]

oh you're almost there!

[left grove]

just going to see how it looks in game before I make any more changes

[amber atlas]

at that point i think you're just using levels nodes to dial it in

[left grove]

I just plopped them in, I am pretty sure I am barely using them

the process is just slow, takes a good 3 minutes or so to export the pngs, ingest them and then add them on the remix editor

"slow"

[amber atlas]

yeah is rough, since its not whole engine like unity or what have you, i wouldn't mind if they had some kind of automation, where you could export over your files and it would auto reingest, but it can't ingest if the folder is the same as the destination etc

[neat spindle]

So if you leave the Remix Toolkit open it'll keep your textures in the ingestion queue. Since they get exported with the same file name(s) from InstaMAT you can quickly iterate just by clicking "Add to Queue" again.

[left grove]

ah I see that is helpful tbf

welp, it's a little crappy atm

I am going to have food, I'll brb

[amber atlas]

nice, yeah i think it's just a matter of using your levels nodes and you should be good

[left grove]

is that so?

ah still looks awful, I am not sure what I am meant to do with the levels, tried to dial it in, but I honestly don't know what I am trying to go towards

[neat spindle]

I would shift your midpoints and peaks down.

So take the white and gray and slide them toward the black.

[left grove]

oh

that worked

a bit

but it's still getting overly softened

I wonder if I need more colour towards the top

either that or ray tracing is oversoftening it

[neat spindle]

Try the opposite then. Reset your gray and white and then slide your black slider toward white. That will crush out the lows into a flat surface while preserving the geometry of your peaks.

[left grove]

I believe it's the ray tracing

that's having a hard time with my texture

[neat spindle]

Oh, I think I know what you mean, actually. I ran into this with my brick texture too. Old-school textures have ambient occlusion maps that fake shadows in the recesses, but with RT lighting you don't have that. You're used to textures essentially being oversharpened.

[left grove]

oh

[neat spindle]

You'll notice that at more oblique angles (further from the light, basically) you'll see more of the shadows popping up.

[left grove]

yeah it definitely happens like that and it has no shadows basically

it's just rough

[neat spindle]

#1095817237631348736 message

You can see it here, compare the top left to the right side closer to the light.

[left grove]

hum I see

[neat spindle]

Are you just using a solid colour to make your Base Color?

[left grove]

blending white and a gray

bit tired now so might get back to it tomorrow

[neat spindle]

No worries. If you're using a Gradient Map, maybe darken your low points a bit more to add contrast.

[amber atlas]

@left grove @neat spindle I think I just got the view model guns working

a couple of setttings, I'll post them here so you can update them on the github!
or at least test them

testing it with all the guns

but i think it's working

only problem though...

[left grove]

Ayooo what

That's amazing

[amber atlas]

wait, might not be a problem... might work just fine haha

yeah i think it might be ok, please test as well

that just means the muzzle flash lights are the last real problem with the combat i think? other than maybe some particle texture tagging i'm guessing?

[neat spindle]

The feed animation effects are important too. The ultra cheese strategy is to crank Brawling and then drain everyone dry.

[amber atlas]

yeah that light on feeding!

[left grove]

yeah

freenzy is kind of weird too I believe

but feeding lighting is deffo off (overly bright)

[amber atlas]

aug scope might be an issue too

[left grove]

oh interesting

are you free to hop on a call? we could hangout and do bits for a bit

[amber atlas]

the graphic isn't there but the zoom works

no 😦 i was about to hop into a meeting for work work

[left grove]

I was almost ready to go tackle the texture again, but I may go and make those rtx.config changes and get those commited

no worries!

[amber atlas]

let me know if it works well for you!

i think it's consistently working

[left grove]

I'll have a test

[amber atlas]

the aug looks kind of funky when moving though haha

[left grove]

I may also document the process of adding textures and stuff

[neat spindle]

When did viewmodels get broken? The only issue with guns I remember was the muzzle flash thing.

[left grove]

can't detail much the material/texture creating side of things but I can definitely link the video

[neat spindle]

There's already full documentation on that in Nvidia's pages

[amber atlas]

oh maybe you had them fixed and i missed that?

[neat spindle]

Or they were never working and I just forgot about it, haha. It's been a long time since I've done anything other than mess with lighting and textures.

[amber atlas]

left is how they have been

with those settings they are now rendering in the correct FOV on the other rendering plane in front of everything

so no clipping through walls and no elongation

due to high fov of regular gameplay camera compared to low fov of weapon space camera from source

[neat spindle]

Ah, okay, I guess I just never noticed that.

[amber atlas]

sick

[bleak barn]

Oh I was wondering what those blobs on the normal map were for. It looks incredible 🙌🏻💙

[amber atlas]

Thanks! I'm not quite happy with them just yet, but I'm glad it was clear what they were, I wasn't sure if they looked like what I wanted them too haha! Lots left to do on this, the wood needs A LOT of work...

[bleak barn]

They look perfect to me. The screenshot you posted looked so realistic it's crazy

[neat spindle]

Deciphering the geometry on the interior pillars was a massive pain in the ass, but it's done!

[amber atlas]

That looks great!

Definitely feels like a believeable interpretation of those posts

[nimble arch]

sorry to disturb y'all , i was wondering if someone could help me set up the rtx remix , i followed the instructions on git hub but nothing work

[left grove]

I can try to help you

[left grove]

forgot to update: But Galaad was under the assumption that it was fully working and wanted to do a playthrough with RTX. Also I can update the repo to include some instructions on how to add the launch args when using GOG

[broken scaffold]

That reminds me @left grove , @neat spindle and @amber atlas I'm gonna send you the last build of the Path Traced Skybox stuff. It includes config options for each different variation. I think for VTMB you guys can get away with using the source engine algorithm but with a fixed scale of 16, since that's what most of your maps seemed to be, and that'll improve the performance a little. The algo isn't SO costly but it's also good to make sure we're running our best!

[amber atlas]

awesome sounds good

also, sorry i've been awol

im in a weird... like... i guess a race against the rest of my company?

im trying to do a bunch of content before they outsource it to students lol

im like no no dont worry we dont need to do that

see its already done!?!

so i've been getting slammed

[neat spindle]

That sounds very unsustainable

[amber atlas]

yeah, i get where they are coming from, they want to accelerate and intern some students who are about to graduate

not a bad idea one one hand

but the amount of training and catchup they'd need is... not negligible

i do have a vtmb asset in the works

but hl2 needs a turn i have to finish up an asset for that project

[neat spindle]

As someone trying to self-train and catch up for this project I feel that, haha.

[amber atlas]

ive got a lot of plates spinnin'

[neat spindle]

This wood texture for the paneling has been giving me hell for the last week

[amber atlas]

yeah, but you are learning so fast and are already kicking ass, honestly i'd rather outsource to you than college students haha

oh dude wood paneling is insane, its one of the hardest imo

getting the grain patterns right?!

ugh

[neat spindle]

Unintentionally picked like the hardest textures to make first, lol.

[amber atlas]

haha its all ez pz after these ones then

https://youtu.be/HVkExICVQes

i like this one a lot

[neat spindle]

Should probably take a break and do some floors or something so I don't feel like I'm just slamming against a brick wall

[amber atlas]

im sure it can be adapted to the program you have been using

haha or you could make a brick wall!

dont slam into it though

[neat spindle]

Oh, actually, while I have you here what is your process for baking displacement maps from Blender? I followed some tutorial but it's blurred everything and came out all melty, as well as having a bevel along the outside edge.

[amber atlas]

hmm i usually bake them in toolbag 4

i find blenders baking to be kind of limiting

another tool you might try is handplane

https://handplane3d.gumroad.com/l/znpF

you can set up multiple meshes and groups and bake multiple mesh types including material id maps

this was ALL i used until substance painter and toolbag4 came out

before that it was xNormal, but I wasn't as big of a fan of the interface

this is abandonware unfortunately, but it got to a pretty solid place

[neat spindle]

I wonder if there's maybe some sort of Mesh baking tools in InstaMAT itself. It's such a new program that there's hardly any documentation on it online.

I know it has some mesh-based texturing stuff.

[amber atlas]

that would be nice, i dont think designer has that at all, only painter

so instaMAT would have a leg up

[neat spindle]

Settings for the new 3D Skybox DLL

Figured out what was causing the weird rasterized surfaces! In interiors with no skybox the Sky Auto-Detect was taking the first textures that got loaded and treating them like sky. Switching to "By Camera Position and Depth Flags" fixes that.

[broken scaffold]

You've just reminded me I need to put the sky validity checks I had before for VTMB as well

No problem

[left grove]

I am also away on a work trip but haven't gotten much further with making textures myself. I still intend to push some updates to the repo, just have those changes locally for now

[left grove]

hello hello, I am starting my weekend in about 1 hour. I intend to work more on this this weekend. I shall add the latest finds both from @amber atlas that help with the guns and @neat spindle's find for the popping out of textures.
I wanted to give a nod to @neat spindle's find because it was something that was plaguing me for a while and I cannot stress how big of a find and fix this is

not sure if anyone wants to pair on trying to get some more textures working, I still have a lot to learn and I still haven't been able to get that test ceiling texture to work

[amber atlas]

Noice! I might have some time tomorrow to chill in a call and troubleshoot the ceiling texture working with you, but freedom knows that program wayyt better than me haha

[left grove]

that would be great! I'll also update our docs to include guides on how to make textures and some resources

[amber atlas]

I've been getting slammed by work lately so it has been hard to get to this and hl2 stuff 😦

I have a few things halfway done and they are just floating around in my brain haha

[digital badge]

This game runs on one of the oldest versions of source engine

[amber atlas]

A pre hl2 release version no less!

[digital badge]

Pretty crazy: https://developer.valvesoftware.com/wiki/Source_2004

[amber atlas]

Half of our discussions in here are how to figure out wtf to do about model compiling limitations 😦

Any animated props and characters

Bug limitation we aren't sure how to handle yet

[digital badge]

RTX Remix would probably work alot better if someone ported it to Source 2013 MP: https://developer.valvesoftware.com/wiki/Source_2013 or Strata Source (Pipe Dream) : https://developer.valvesoftware.com/wiki/Strata_Source

[left grove]

I believe the game is completly closed source and that would mean we couldn't port it

I'd like to be wrong here, but I think that's the truth sadly

[digital badge]

yeah that's a bummer

This could help: https://www.moddb.com/mods/vtmb-unofficial-patch/downloads/bloodlines-sdk

[left grove]

we do have this, you can optionally install it with the unofficial patch

[digital badge]

oh sorry I didn't look at the pins thank you

[left grove]

no all good

we have the repo, it should help you get it up and running, it's still far from perfect

I also have come across someone else that had it on GOG and I should update the repo to cater for those people. We all just happened to have it on steam

and feel free to contribute however you can

[bleak barn]

Oh that's nice to know that there are differences between versions since I do have the gog version myself

[neat spindle]

I think the only difference is that the GOG version comes with the Unofficial Patch already installed

[amber atlas]

I understood that to be the case as well

[amber atlas]

Not only that but Activision still holds the rights to bl1 and they go after people who do anything related to it other than modding, I could see them classifying an engine port and outside the bounds as it is technically modifying the binary itself etc. they love doing that.

[left grove]

I don't believe there are big differences, it's just how to get the launch args working

and well it comes with unofficial patch preinstalled

As much as I'd love to chat to Activision about Bloodlines, I'd rather it not be some legal battle

[amber atlas]

Yeah I think it's worth sticking to the modding path, it's the most open and safe vector

Plus the amount of time and resources needed to port forward would be years of dev time and people with know how we simply don't have access to

[bleak barn]

Project Vaulderie flashbacks

[amber atlas]

😦 yeah, heartbreaking, the work was amazing

[left grove]

I am getting this for some reason, trying to figure out why

unless this is intended and we just need to replace the textures here or set things as decals

it seems to only render the skybox for me

I am using the "latest" d3d9.dll that King David sent me a few days ago

[neat spindle]

Don't set things as decals unless they're actually decals, that's a bandaid fix that will cause more issues down the line.

The apartment is an interior that has a skybox, so that might be part of the problem.

[left grove]

yeah I don't want to just set random stuff as decals, was just a question rather than a "I am about to do it"

it seems so, I am just curious as to what is different between us if it fixed your issues on yours

it is possible I have other changes I am not aware of right now

[neat spindle]

I'll have to check that specific map, it just fixed the issue in interiors that didn't have skyboxes for me.

[left grove]

got any example of where, I might be able to go quickly check it out

will try the skyline appartments

[neat spindle]

Confession is still my test map, but I checked the Empire Arms too and didn't see any issues.

[left grove]

I'll give that a test

skyline was pretty much fine too

Confession seems fine too, from walking around

but any other map with skybox suffers a lot

[neat spindle]

Hmm, maybe @broken scaffold will have some insight on what Source quirks might be causing this.

[left grove]

perhaps

Thanks, I'll have to sleep for a bit soon, but I'll be here all weekend so ping me if needed

Downtown LA appears fine too, it's just Santa Monica that goes a bit funky with only rendering the skybox

[broken scaffold]

I have been activated

What's up gentlemen

[left grove]

hello

[left grove]

basically this

Changing Sky tunning > Sky Auto-Detect from By Camera Position to By Camera Position and Depth seems to cause that, when using just By Camera Position We get that funky bit where it changes to rasterization

[neat spindle]

We've been having issues with some textures being defaulted to rasterization in interiors with no skybox, and setting the auto-detect to By Camera and Depth Flags fixed it (presumably they were being flagged as sky textures because they loaded earliest), but now there are issues in interiors that actually have skyboxes.

[broken scaffold]

Have you tried messing around with the minimum z to see if it goes away

Really just an exploratory suggestion, not home yet

[left grove]

I can test that

now I am getting different behaviour, because I teleported in with the console it was like normal, only with the bit on the steps where it changes back to rasterization

going to load back the save from when I boot the game

[broken scaffold]

Interesting

So only when you pass the area portal in there does it actually break?

[left grove]

from my understanding it's the transition to showing the skybox

I got video evidence

the view model also shows up and disappears but the transition is what causes that weird texture pop out and in

it's almost as if as soon as it starts rendering the skybox it stops rendering everything else and then resumes rendering everything else, doesn't happen when we leave a sky box area however

that's the intriguing part

only when moving to a place in the map that has a skybox

[broken scaffold]

The camera is definitely locking your sky cam as the main cam

I don't know why it'd make that mistake only when it re-enables

The sky usually always renders first

It's possible the DRAW SKY bool that it would usually return needs to consistently return true to prevent it from crying

[left grove]

Interesting, I'll be back in 8 hours. I can try to do some more investigation

[slender tiger]

yo @left grove sry toping i see u typed opengoal

does RTX or will RTX support any Jak series?

[left grove]

LMAO

no I was referencing their way of tracking things done in an easy to diggest way

you really dug that one up, that was only for how they do the whole lines decompiled vs lines to decompile. We have textures we need to redo by hand here

I have no idea to that question, I'd guess not if it's not using an older version of DirectX. Probably one for their discord

[slender tiger]

nah i didnt rly dig anything up i just searched in discord for OpenGOAL

[left grove]

that is fair

[slender tiger]

shocked it went back far lmfao

but so no chance im reading?

oh u dont know my b

[left grove]

I don't want to give you hope

one to ask over on their discord probably

[slender tiger]

ye i heard any openGL is no go but i thought they updated something recently

but word

ty

[left grove]

No problem

[oak forge]

Well the Jak ports are later OpenGL so the odds go from slim to completely nonexistent.

[left grove]

I replied to a few of the github issues since they all seemed to have a follow up question to check with the latest remix runtime

[oak forge]

I saw the same thing. Thankfully two of the three issues that got said reply were actually fixed. The other, well, fixed in one case and not the other 💀

I only got them for very specific issues so I'm assuming they only asked if they actually attempted to resolve said issue.

[broken scaffold]

reminded me I gotta make the PR finally

[left grove]

I'll review it if you'd like even if I have no idea what I am reading

I do have a commit ready for the vtmb remix repo, but with testing that stuff I ended up jumbling some of my installs a bit

[broken scaffold]

If anyone would have a moment to double check the delta mode to see if it's still flickering I'd appreciate it

I think I reincorporated the fix I did specifically for VTMB in a much broader way

just send me a DM

[broken scaffold]

@oak forge and @left grove https://github.com/NVIDIAGameWorks/dxvk-remix/pull/73

i'll keep chugging away at stuff but at least got the PR up for feedback

[amber atlas]

Ooh nice hope that goes through, that will be big for all source games

[amber atlas]

Hello LA, you're up way past your bedtime.

been getting slammed at work with all kinds of weird stuff + hl2 remix work as well 😄

but trying to catch up a bit here!

[left grove]

Very nice! it's looking real good

On good news for me maybe, I got a new motherboard and it came with 1 month of creative cloud, so I may carry on with making textures

[cerulean wagon]

https://tenor.com/view/scooby-doo-anime-fred-jones-humping-gif-16758307

hot damn

[tame shard]

hey, do you guys have a list of models that need doing or access to any of your files available, or is all of that closed access atm?

[amber atlas]

So far I'm the only one doing the model replacement, and I have an internal list and folder structure I'm filling out. But all that is just me on my local hard drive (backed up on Gdrive)

[left grove]

I can try to help on having a publicly visible list that tracks what has been worked on and what can/should be worked on

[amber atlas]

Why what's up?

[neat spindle]

There should be spreadsheets for tracking models and textures linked in our documentation on GitHub.

[left grove]

that is true

almost forgot about that

[tame shard]

was just curious honestly, big fan of this game in particular and wanted to see how it was coming along 🙂

[left grove]

very slowly 😅

[amber atlas]

Right now I'm just working through the Santa Monica safe house map and doing all of the prop model replacements first

Then after that I'm going to do the materials and textures for that same map

And that'll be sort of a proof of concept of a small section of the game

I've got almost all the props replaced in that map currently and a few textures as well

Characters are still the big limiting factor. I have a Jeanette remodel that I've made but I can't inject it because of the way remix doesn't pull entire bone hierarchies

[left grove]

I honestly may need someone to guide me a bit through the entire process of changing a texture so it works well when replacing it in game, my ceiling still looks awful 😂

[left grove]

we could feature request

[neat spindle]

I've been doing something similar with Confession, albeit in reverse order. Lighting and textures first, then I'll figure out models. I don't have much experience so I'm having to learn as I go, so it's been extremely slow progress.

[left grove]

yeah, I am stuck at a single texture replacement, still haven't fully cracked that...

would be good if we did like a workshop or something to get us all some experience with the bits we are struggling with

but feel bad as it'd mostly be us pestering @amber atlas

[neat spindle]

I think I'm going to have to redo that paneling model... again. Apparently you can't have sharp edges in heightmaps because how it works is by checking for changes in edge slope, that's why it came out all melty.

Also been spending a lot of time helping King David test out his skybox fixes. He's done great work on that front.

[amber atlas]

I guess for me, I'm just like... goin'... haha, so we are all interested in the same stuff, and I'm always happy to drop tutorials and talk shop and game dev/art

[left grove]

I just need to streamline a workflow, once that is done I can regularly work on things, but in the mean time I am still slowly learning and I can't seem to even get a simple texture replacement in

[amber atlas]

I thought I saw you had some shots with the ceiling replaced?

Or am I mis remembering?

the stucco/popcorn ceiling

[left grove]

I did, but I wasn't happy with the result

[amber atlas]

ohhhh I thought you meant the ingestion process

you mean the overall process

gothcha

[left grove]

yeah... at least how to polish it and all

[amber atlas]

I did see a plaster tutorial for substance designer recently

its an older tutorial

[left grove]

oh 👀

may be useful

[amber atlas]

im sure most of it transfers to the program you are using

https://youtu.be/WcfpLUZRDMU

2 parter i think

[left grove]

I mean I do have 1 month of free creative cloud

[amber atlas]

he has a version without vertical leaks

[left grove]

thanks Asus

[amber atlas]

lol

[left grove]

last I remember I was tweaking my levels, but no clue how to do that properly

[neat spindle]

@amber atlas Any chance you'd be willing to have a call at some point to work through these heightmap issues I've been having? I keep trying to search for how to do it and I feel like I just don't know the right words to get the info I need, it's very frustrating.

[left grove]

I'd love to sit in and listen tbf

[neat spindle]

I've hit a brick wall here (pun intended)

[left grove]

I should be around all of this weekend, it's just timezones for us

[amber atlas]

yeah i should be good tomorrow in the morning/mid morning on my end, AZ time (11:02) am right now for ref

the height mapping stuff is pretty interesting, each engine/shader approach has some key differences

[left grove]

it's 7 pm for the UK

[amber atlas]

the major one i think is throwing you off is 'mid point'

that's where it calculates from, so if your mid point is gray, white comes UP and out, and black goes down and in

from what i can tell, rtx remix is a white mid point

meaning it only goes down/in

so white would be your highest point and gray down to black would be how stuff divots inward

i also think they are calculating it in like a very small measurement space so often times you'll want to put in values like .002 etc

[neat spindle]

I've got all that figured out, it's just that these points where I have sharp edges in my heightmap end up looking all melted when I import it.

It happens anywhere there's a sharp line between two heights, say a box of black next to white

[amber atlas]

hmm yeah im not getting those results on my end... at least as far as i remember

let me look real close at my hallway wall and see

could be down to a remix setting

these small ridges i expect

because i have heightmap in the wood grain

when i do the final pass, the wood grain will be directional to the planks and be less ridgelike

but it looks correct to me i think

mind posting your height map here?

[neat spindle]

This is the raw bake from Blender

[amber atlas]

the bake at least is clean AF

definitely no ridges on this

have you tried just slotting this in and seeing the result?

[neat spindle]

I can give it a shot, one sec. The snip above is from the preview window of InstaMAT

But it also shows up like that in-game

[amber atlas]

oh in instamat are you importing it and setting it as a non sRGB texture?

not sure if instamat cares

[neat spindle]

I drop it in and then have a conversion node to grayscale.

[amber atlas]

hmm yeah sounds like the usual procedure

at least in substance designer as well

[amber atlas]

@broken scaffold After messing with some settings on your sky features, I've also noticed that all the flickering during cutscenes seems to be just gone...

[left grove]

That's great

[amber atlas]

Still having crazy issues with my setup whenever I try to upgrade my remix toolkit, king_david was trying to help me but we dead-ended, so I'm asking in general.
If the answer is that I have to remake the project from scratch, I guess that's not so bad because I'm still off in my own little world, and I should get on the Catto framework.

So all the hashes are the same etc.

[amber atlas]

strauss note replaced

doing a test with the pad, would retexture from scratch

but for now i just did a screengrab of the pad ui and modeled it out from that

to see what it would feel like

i dont know if this sheet model shows up anywhere else in the game off the top of my head

if it does... i could just make it a blank pad the way the sheet was

[left grove]

Ah I see, I could do captures

And have them catalogued. if you still have all your source textures you should be able to add them to the repo

And reproduce the changes

[amber atlas]

Yeah I'm mostly just wanting to align hashes in general

If you provide that definitive framework anyone in the future could decide to remaster a mod for vtmb and they could fit it together with the vase remaster hashes

Which would be sick

[left grove]

Ok I'll work on that

[amber atlas]

I was looking at the repo but I don't recall if you had captures up yet!

Would be cool to be able to push captures back and just build up a good catalogue of universal captures etc

[left grove]

no I don't, I was thinking of a good way to do them

I have a naming scheme and everything, it's a shame they all have to sit on the same folder, or has that changed?

just finishing up some tidying up, I can get started on something like that

[left grove]

the only "issue" we have here is we'll have captures for just Vanilla Unnoficial patch, for the plus patch we'd need to recapture, but that's an issue for later

[amber atlas]

Well that was my whole thing from earlier

The test I want to do is capture from plus patch, put out a test mod with asset replacements. Load it into a non plus patch and see if it works.

Then add plus patch back in and see if that inherits any plus patch changes as well

Plus patch changes will probably be compatible I think, since the plus patch mostly changes brush work and then only model positions and texture changes. New models and maps would be their own hashes

That's my theory

If that holds true it would be double the work to remaster separately.

But there's a chance that it could actually cut down on work and we'd get plus patch setup to remaster for the future

[left grove]

Interesting

That's a good theory, I could help test that but for now I may just start all the captures

at least get most captures done for Santa Monica

the rest I can do a bit later

[left grove]

is it a lot of work to port your haven work onto the new captures?

[amber atlas]

I'll probably have to do it either way

Since all that is done on my crappy old first test project

So I def want to do it maximum after haven proof of concept

[left grove]

sure

I have looked at it and we do have a few captures on there already, mostly for the tutorial. I can try to do all of the ones that are relevant for now, and maybe even a bit around Santa Monica

[amber atlas]

I think the patch plus test is probably step one. Before doing a bunch of captures etc

[left grove]

ok

[amber atlas]

Trying to strategize how to approach that

[left grove]

yeah, I can deffinitely capture both, so do a vanilla capture and a plus patch capture

[amber atlas]

Could be a throwaway project that changes 2 things and just confirms what happens on vanilla vs patch plus

Well my concern is that the hashes will be different, but I guess it's totally possible they'd stay the same

I need to take another look at patch plus folder and see what's actually replaced

Vs what's new

[left grove]

I believe what's the same will remain the same

but if they change then they hashes will be different

[neat spindle]

Aren't captures pretty large? I'm not sure of the practicality of uploading all the captures needed to cover the entire game. I'm pretty sure hashes don't change with different captures, that's why when you take different captures that overlap objects still have the same ID.

[left grove]

at least that's the logical way I can see it work. With the captures then it'd just link the correct texture files

Captures themselves aren't very large couple hundres maybe thousands of KB

it's the assets that get large

these guys here

[neat spindle]

Yeah, I was including the assets since captures are unusable without assets.

[amber atlas]

Yeah if hashes are project set regardless of captures then no need to upload any to the repo

[left grove]

they are all only a few KB each, the only thing is there thousands of them

[amber atlas]

Hmm

[left grove]

I have a few captures and it's 380MB

as it needs a lot of data to go along with those captures

ah a chunk of that is the thumbnails for each capture

and the remaining stuff is the Textures folder

[neat spindle]

Yeah, it's essentially decompiling everything in the scene out of the efficient compressed packs the game normally uses. Bloats up fast.

[left grove]

yeah, I do hope it's efficient in the sense that the more captures we have the more will share the same textures, and won't resave the same ones

[amber atlas]

When does it calculate the hashes for everything?

[neat spindle]

Since everything is hashed I'm pretty sure it doesn't duplicate.

[amber atlas]

Just on first launch? Or on mod file creation?

[neat spindle]

If I had to guess it's probably generated based on file structure in the game? The asset gets pulled the same way every time.

I don't know how else you would get a consistent hash.

[amber atlas]

Yeah true

I'm getting deja vu lol

[left grove]

well, I wouldn't mind to capture everything in the game myself

for the sake of consistency

since I still need to learn a lot about material creation

(definitely not me procrastinating it even more)

[neat spindle]

Just from my experience with Confession, I think it's best to create captures as you're actively working on a scene, so that way you know what angles you need. Something that came up a lot for me was that the game has really aggressive distance culling for detail objects so I had to take more captures than I expected.

[left grove]

yeah that's what I am wondering, I took what one of the tutorial videos said and just ran with it, which is to take a few from a few angles and use that

and one from out of bounds (in case it's needed)

it would be cool if we could merge these captures, like have 3 or 4 and that way you could see the entire area in one go

but I believe if I do a first pass a lot of the hashes will be assigned, that should help with consistency at least, if you need a different angle that should retain the same hashes

or so I'd hope

[neat spindle]

Yeah, I followed the tutorial initially and then realized all my captures were unusable because of distance culling, haha

[left grove]

yhup... that's what screwed me over in mutiple occasions

[amber atlas]

@left grove

Supress Light Keeping

[left grove]

is that an option?

[amber atlas]

finally read about what it does

turn that on and then shoot the guns

[left grove]

wait where are you?

I have never been to that place

[amber atlas]

oh its a map i made to start putting all assets in it

[left grove]

amazing

[amber atlas]

the option seems to help with the muzzleflash actual light, but there is still some ghosting on the volumetric effects

its not perfect, but its much more bearable

not sure what other knock on effects that might have though, so it will be good to test

[left grove]

I still need to do a lot of stuff, I have been busy with life stuff. As well as a family member not doing so well, so apologies for being less active

[amber atlas]

oh no worries at all, i just leave this stuff here for you!

[left grove]

thanks

[amber atlas]

anytime i find some other weird thing or option out 😄

[left grove]

other than getting captures of Santa Monica for now, is there anything else you'd need from me?

you have been doing great lately, you keep finding all of these obscure options that actually do a lot

[amber atlas]

no, im just trucking along, more stuff on the way, but its slow going because of work etc

haha im doing that thing i should have been doing where i read the options tooltips D:

[left grove]

hey that's fine, you are doing plenty, and this is slow burner project, sadly everything else takes priority

[amber atlas]

ok so i got a little side tracked

Some shots from marmoset for comparison

[somber sundial]

Gorgeous stuff! I feel like you could step up the weathering on it a fair bit to stay in tune with the original

And maybe consider tinting the metal just slightly blue? Just to retain the original feel

[amber atlas]

maybe that's what this guy meant?

haha

definitely does feel different than the OG!

i did toy with making it more blue, but it felt strange since i added the missing components

still worth experimenting with im sure

[neat spindle]

I think with the blued steel of the original texture it makes the edge weathering show up better, adds more contrast.

That said, I don't think Micro Uzis typically have a blued finish, although yours is maybe a bit too flat.

[amber atlas]

yeah ive been looking for blue finish stuff, and i do see lots of cod camos

i also figured the 'isn't blue' aspect would cause some issues

its interesting too because i see many of their guns are just direct recreations

like... uzi is just straight up an uzi super accruate, same with steyr aug, all the pistols except the deagle

i think the mac10 and deagle were made by a different artist, and i dont think they were necessarily concious choices, at least when i look deeply at the files themselves

like the changes i plan for the revolvers are next to MINIMAl, same with the uzi and aug, they were so dead on, but the mac10 looks to me like someone said, quick, we need a mac10, just get something together before the next build

this and the deagle are probably the ones that will change the most, because im trying to make them actually look like their real life counter parts like the rest of the guns in the game

though i do understand i do that with risk 😄

also that flamethrower is something i think they kit bashed so i'll probably have to stick pretty close to what they had since it doesn't really exist in real life

[neat spindle]

I think IMI is probably busy right now, lol, doubt they have the bandwidth to go after random modders (/spicy chat)

[amber atlas]

there are no markings so i think i'll be ok

im very interested to see how some of the guns and the characters play out... i think that will be the are that has the biggest friction with observers

i already have seen someone say this is not going to be a good project because the lighting is more contrasted in the early shots

so im definitely going to be watching how reactions are to changes in general, people are going to very much have some reactions as i get into the props that were built weird or phoned in so to speak

[neat spindle]

Fuck 'em, lol. For my sanity, I made the decision to mostly ignore feedback from people unless they've directly contributed to the project.

Opinions, assholes, etc

[amber atlas]

Haha yeah I definitely love all the perspectives though, you guys have already steered some of my pieces for the better and given me great feedback in here.

Hopefully the world can heal from the mac 10 maybe being blue... maybe being gray in the release

[neat spindle]

Something I'd pay attention to as you're updating models is to keep track of the timeline. A lot of the weapons hitting the streets are going to be coming from police armouries, so when looking at references you'll mostly want to look at specific police variants of the guns.

[amber atlas]

yeah! i was looking at the 38 special and trying to match it with what model it probably was

and it seems like it's a detective/police variant of it, but sometimes it also looks like a long barreled version of a navy variant

also that aug model is definitely the older version so i have to make sure to stick to that

less rails, no MLOK systems etc, gotta go for the classic models

[neat spindle]

Another common thing is to have civilian variants that are converted to full auto. That's why the TEC-9 became so iconic, it was easy to convert.

So legal purchases get converted to illegal firearms with auto sears and high capacity magazines. If you want to add extra detail you can add some subtle material differences between parts that are aftermarket and parts that are stock.

[amber atlas]

interesting, i think the mac10 in this case is mainly used by the mafia guys... i don't even remember where you get the aug or the uzi for instance...

[neat spindle]

The Aug is used by the Society of Leopold, and I think the Uzi is late game too.

[amber atlas]

flying around SOL, they have augs, mac10s and hunting rifles

gotta stop myself from modeling an aug right now

[bleak barn]

https://tenor.com/view/what-the-hell-gif-10739159

[neat spindle]

I just realized I was calling the MAC-10 an Uzi, lol. That's what I get for trying to talk about things while at work. That said, there is an Uzi in the game, the Lassiter Killmatic, and you could probably tell a fun story with ita design by explicitly making it a converted model.

[neat spindle]

Oh, minor detail, but I think you're missing the button on the bottom rear of the receiver that extends the stock. Also the fire selector switch should have some bumps to catch the switch.

[neat badge]

https://tenor.com/view/mods-tie-him-up-ltg-low-tier-god-kill-yourself-now-kys-gif-97316207949095273

[neat spindle]

Did a little more reading on the MAC-10 and I was right, the one in the game is a converted civilian model, which is why it didn't have the stock and grip. It's also missing some other important surface details, but the lack of stock and grip are actually correct for what kind of weapon would be on the streets of LA at the time.

[amber atlas]

Funny story that's in the high poly. I just forgot to export the low poly.

Also today I took some more time and did some masks for imprints like the words and such on the side

[stray whale]

said what i was thinking out loud

[neat spindle]

Case in point

[amber atlas]

I guess I'd encourage you to think of how to transform that into something that's more akin to feedback, something that's more actionable? I think you'll find there are a million ways to interpret something being "soulless", so maybe you could provide better context to how your feeling about the remastered piece?

[stray whale]

give the old model a trigger

then leave it be

the #1 problem i have with vtm weapons isn't how they look

• the model on the left was clearly made in house with some sort of intended creative decision

[amber atlas]

curious about the phrase 'in house' here, does it feel like my version is not made by me when you view it?

[stray whale]

100%

[amber atlas]

where... do you think it came from?

im a bit confused sorry

i def modeled it by hand using reference

[stray whale]

csgo or some 3d modeller stock asset site was my assumption

[amber atlas]

ouch

well to be clear, i refuse to use stock assets, and ai stuff

i like to make things by hand

it's why i went to school for game dev and its why i got into games

[neat spindle]

They're both replicas of real life objects. What you see as "in-house decisions" are just replicating a real life gun.

The one in-game is based on a civilian semi-auto variant, the one Safemilk made is an older full-auto one.

[stray whale]

kk but if i had the choice between the two i'd go with the old one still

the shitbox aesthetic i like more

[neat spindle]

I actually also agree for the reasons here

[stray whale]

left looks like it was a shitty first smg which it is + brutal metal design making it looks awkard and violent to shoot

[amber atlas]

i guess im seeing a disconnect here, because i could just turn off the stock and fore attachment section and turn it blue, but i think you'd still dislike it for other various reasons

guessing this still doesn't fit what you are wanting

copying and flipping the front of the trigger guardto make a trigger on the old model

i guess this comes back to the 2 stages of remastering i'm talking about

1. is, bring int remix and lighting, relight and fix up some textures for it and tag things
   this gets it running, looks pretty much identical it just has good lighting

2. remaster individual assets, rebuild them (from scratch in my case) and replace them overall in the game
   this is more of like... a remastrering of the actual full game and it's assets

version one we basically have the framework for it right now

like that's pretty much blocked in, and it would just be a matter of going through every scene and tagging textures

then fixing all albedos with emissive to not have the emissive in the alpha channel

like thats pretty much it

game is bare minimum remastered at that point

which is cool!

and catto basically has that up and running so if literally anyone wanted to contribute and push changes back and pull request them, that whole portion of the work could actually be open source crowd built

[neat spindle]

https://static01.nyt.com/images/2023/01/26/us/monterey-park-shooting-guns-promo/monterey-park-shooting-guns-promo-videoSixteenByNine3000.jpg?year=2023&h=1690&w=3000&s=916f0e9effad8da772ea1f3ecb344956cb1fd80bb7387ee743f526a01e775c4c&k=ZQJBKqZ0VN&tw=1

[amber atlas]

i think if its set up in those two stages then it kind of keeps it in like a sort of...
inoffensive
and more offensive

2 stages of work

it's also why i was advocating for having the framework be a separate moddb page that is like "Catto's Remaster Framework" and then other modders base it off of that

[neat spindle]

Maybe adding a little more surface wear to make it look less new? The lack of fire selector switch on this is interesting, too.

[amber atlas]

oh sorry this is the W model

the view model has a fire selector switch

[neat spindle]

I was talking about the converted one in the picture I linked.

[amber atlas]

right but if your goal was to match with whats in the game, then techincally it does have a fire selector switch

just not on the w model

[neat spindle]

For sure. I think deviations from the exact model in game are fine, I just personally think it should still be a converted civilian model (no stock/grip, maybe a lack of fire selector) to fit that black market aesthetic and make it look less military FPS.

Not that the MAC-10 is a serious military weapon in any capacity, haha

[amber atlas]

well thats a mac11 as well

you can tell because the receiver is vertically shorter

just for posterity, here's the V model

[neat spindle]

Ah, true. Was looking more at the surface wear than the shape.

[amber atlas]

i grabbed the W model since it wasn't attached to a hand bone and floating at an odd angle the way the v models do

[neat spindle]

I do like the deviation from the blued steel, though, I think it was an odd choice in the original.

[amber atlas]

I very acutely remembering as a kid in highschool playing this and wondering why the mac10 and deagle looked so off, and how all the other guns look much closer, but at the time i was using all these same techniques and was still learning using a... 'student copy' of 3dsmax 7

like the uzi... i mean they just did actual photo collage with the actual uzi

all the other guns for the most part exluding the mac10 and the deagle do this, those two are built very differently, and not using photo collage techniques

it has been a mystery to me for many years

guessing a couple different artists did various guns

[neat spindle]

I think if you wanted to lean more into the shitbox aesthetic you could maybe add some spots of surface rust and a slight patina to it? Could also have some fun with modeling a homebrew foregrip that someone added to manage the recoil.

https://globalnews.ca/wp-content/uploads/2022/05/ALERT-May-6-gun-seizure-2.jpg?quality=85&strip=all

https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTQdvmneaSoANP21doR_kWSGb6kz3dsXxJQtgdeQ9VEkA&s

[amber atlas]

i guess thats the friction point, i'm personally liking what i built, but i can see it's not going over well in general so im trying to figure out how to reconcile that

the other thing to is, it's used across the game for multiple groups, so it's doing double duty for well armed mobsters, society of leopold, and gang members

so getting something that is flexible enough to kind of reach to all those use cases was something i was trying to accomplish as well

[neat spindle]

Ah, that's true. Maybe the cleaner look is the call, then.

[amber atlas]

if mod layering works well or it gets a bit more functionality
i'd love to lean into doing like... replacements or variations

like imagine if you could say:
I want the original mac10, the clean one, or the dirty one, or the one with or without a stock

and you can kind of customize it a bit

so you load up the mod with the stuff you like and skip the stuff you dont

man im going to get executed in the street when we get to the characters though omg

im going to have to do some like.... pre therapy to prep for that

[neat spindle]

Apologies if I'm contributing to the problem here, haha. I'm just having fun nerding out about what guns would be in the setting.

[amber atlas]

haha no no it's all a discussion!

but i definitely am curious about how things will go when i get to the assets people care more and more about

like i have an in progress jeanette and i feel like im going to get lit up for it

so how do i navigate all that

an interesting question i dont have an answer to

[neat spindle]

I stand by this, tbh. Make what you want to make, and if people can provide actually constructive feedback that's great, otherwise ignore the hate.

Tried working more on that wood texture last night and I think I'm making some progress. Converting SD tutorials to InstaMAT is a process, since every single node works slightly differently.

[amber atlas]

oooh you have any progress shots?

[neat spindle]

Also figured out that issue with my bake, turns out it's just an inaccuracy in the preview renderer and it looks fine in-game. My bake did have an edge bevel around the outside for some reason so I had to fix that, but other than that it's looking a lot better.

[neat spindle]

I can drop some tonight when I'm back at my computer. Feels bad struggling with literally one texture for weeks and you're knocking out entire fully-textured models, lol.

[amber atlas]

dude dont even sweat that, i've been doing this since i was like 13, ive got a process

you have been kicking ass

like how fast you've been picking this up is insane

im pretty envious of how quickly you are learning 😄

[neat spindle]

Oh, question: I feel like the displacement in the paneling comes out looking too shallow in the final product, would passing the heightmap through a node that increases the contrast make it look deeper? I'm happy with my original model.

Since it functions by recessing, I have the highest points on the paneling as white, with the recesses somewhere in the gray.

[amber atlas]

yeah if you push those grays down darker it should give you a wider range of depth

but you could also just increase the inset value of the material in remix too first and test it that way

[neat spindle]

I can't do that because the bricks have a value that makes them look good. I'm starting with my existing brick material and then overriding part of it with the paneling.

[amber atlas]

gotcha, then yes, push the dark values darker to deepen the inset

[neat spindle]

Would a Levels node be the best tool for that?

[amber atlas]

yeah you could do a levels node

histogram scan might also work

but thats more for sampling values that already exist

so if you dont have darker darks you cant really sample them i think

so histogram scan might not actually be that good for it

[neat spindle]

I don't love it, but it's a decent first pass at wood laminate

[amber atlas]

yeah!

if i were to think about where i'd take it, i'd say, just very faintly add in the wood grain to the roughness map and get some nice variation in there, and add the board separations into the height and the normal (if they are, punch that up way more)

then i'd layer on some slight VERY slight staining variations and grunge on the roughness as well

so shallow angles and lighting makes some nice shapes on it and nice variations

overall the boards look great and the color varation in the albedo looks real nice too

[neat spindle]

I left out the board gaps and grain intentionally on this one, I was referencing some images of wainscoting where multiple boards were combined together to form larger panels and it looked like they gap filled with the laminate and smoothed it. I'll see what I can do about adding random variation with grunge, though, that sounds good to me.

Here's where I'm at on the final with edge wear added

Left out board lines entirely since it looked weird on the paneling

[neat spindle]

@amber atlas How's this look?

[amber atlas]

oh yeah that makes sense on leaving the board gaps out, in that case, im not sure how much the boards themselves should be visible overal, like the wood panels should probably be single sheets?
not sure on that though

i'd push the contrast on the roughness pretty hard actually

overall that's looking nice

[neat spindle]

Whoops, actually forgot to plug it in to the roughness

There

https://i.pinimg.com/originals/a4/54/3f/a4543f0cb2bbc6c1c47171b77c19ec46.jpg

Here's one of the images I was referencing, you can see there are boards in the larger panel but it's been made into one large smooth surface

[amber atlas]

ok yeah that makes sense then, i was picturing them as one piece, good ref!

[neat spindle]

Alright, final step was to use an Ambient Occlusion map to add some dust to the recesses, I think it's pretty much done! This texture covers like 50% of the map, so it feels good to have it done.

[amber atlas]

Very nice I think you sold the surfaces of that panels really well

Would love to see it in the map!

[neat spindle]

@amber atlas

I wasn't kidding when I said it's used on like half of the scene, haha. The paneling from the wall gets used on the bar and booths. Pretty clever texture optimization, honestly.

[amber atlas]

nice, really catches the light in that second shot

[visual arch]

wainscoting

[neat spindle]

Yup

[neat spindle]

Floor tiles done

[modern bloom]

I've been wanting to contribute some models and when time allows I'll hopefully be able to work with your idea of individually remastering assets.

But to add my own two cents as someone who never really played the game initially.
I personally really liked the initial approach you had to the weapon design.
Not sure how the first one has more "soul" as that twitter post put it

[left grove]

Woah some really good progress, I was away for a few days, I am back now

[neat spindle]

Not me getting annoyed over a 20 year old misplaced texture

[left grove]

it's looking really good so far tho

so keep at it

[neat spindle]

This was just a basic test to see if my shapes matched the geometry. Turns out the devs screwed up and parts of it are put in the wrong place.

[amber atlas]

they are EVERYWHERE

[neat spindle]

With the flat textures it's not very noticable, but holy hell it really shows once you add in displacement

[amber atlas]

yeah :/

[neat spindle]

I wonder how much of this we could fix in the level editor? That would be a whole additional pass after we've finished, though.

[amber atlas]

should be super doable in theory, BUT

how will that interact with all the changes in the UP and UP+

if there are map fixes in there we'd overwrite them wholesale

[neat spindle]

Yeah, I think how that would work is we would have to rebuild every time UP updates.

[amber atlas]

which is... not great

[neat spindle]

I don't think it would be from scratch, you would just have to release a version that incorporates the changes. Relatively normal compatibility patch stuff.

[amber atlas]

if evey UP map is released with it's actual hammer editor file it wouldn't be too bad, but otherwise you'd have to go in and find source files for each map, then compare with UP, and that's only including visual fixes that might or might not be there, if there are any gameplay fixes we'd have no idea what they were

but then we dont have parity with UP after that point

there is no amount of file diffing at all with source maps so its either one file or the other

[neat spindle]

I'm not super clear on the specifics, but I think the editor basically decompiles the game however you have it installed? So since the editor comes with UP, the map file you get out has all the UP changes integrated.

I think.

[amber atlas]

you mean the hammer editor?

[neat spindle]

Specifically the modified hammer editor that was built for VTMB, yeah

[amber atlas]

hammer editor doesn't have any decompiling functionality, but i do think there might be a decompiling tool in the sdk alongside it

[neat spindle]

Yeah, that's what I mean. You don't have anything to edit until you decompile the game.

[amber atlas]

yeah there is a bsp decompiler in there as well as bsp editor

[neat spindle]

No clue how versioning would work, though, I know essentially nothing about Hammer other than that I hate using it every time I've had the urge to try and learn.

[amber atlas]

there is no versioning, and i dont think there is any reasonable way to merge changes using git or anything like that

its just... vmf file is a vmf file

[neat spindle]

So if you built it for UP 9.X.X then there's basically no way to update it to UP 10.X.X without building from scratch on re decompiled game?

[amber atlas]

im saying from version to version of an individual map, im not sure about if maps are compatible moving forward to a new compilation of the actual mod binary

but

i dont know what UP changes in terms of the original binary in general

or how large the changes are in between updates

or if wesp has to recompile every map or if it has forward compatibility

mainly im just saying that if you fixed that texture issue and then the next UP patch comes out and fixes a quest or something in there, you'd have no reasonable way to merge those changes

[neat spindle]

Yeah. wesp has integrated fixes from other people into UP before, but I have no idea what that process looks like.

So if we went through and corrected texture alignment we could maybe get that merged in.

[amber atlas]

yeah, i think it has to happen up stream there if we want to maintain partiy/compatibility

[neat spindle]

For now, though, I guess we can just deal with alignment issues

Since it requires level editing to fix

[amber atlas]

yeah, in the end if we do determine a path, doing those texture fixes is super easy

you just use the little texture alignment controls

set position/rotation etc

i certainly don't miss it 😄

but i can see those controls in my sleep even to this day

[left grove]

the latest build of dxvk-remix allows us to run in fullscreen I tested it from a build, it should make it onto the next release of the rtx remix runtime

[amber atlas]

nice

[left grove]

small caviat is that you need to add -fullscreen as a launch argument if you ever ran it with -window previously, no clue why

[worthy bobcat]

hi guys

[amber atlas]

Sup duuudee

[left grove]

hello

[worthy bobcat]

thanks for this project ❤️ i love u guys ❤️

past on youtube, guys masterpice of vampire game coming thanks for great job (i from brazil sorry for limited inglish

[left grove]

I am European Portuguese, if you want to talk in Portuguese feel free to Direct message me, I understand it 👍

and progress has been slow, I can do some more soon, but I have been busy lately

[amber atlas]

I have a TON of stuff I've been putting in the game... I just haven't been posting

but I'm almost done with all the models in the actual apartments

Monitor, PC Tower, Mouse, Keyboard, Both Letters on the desk, folding table, orange chair, mailboxes etc.

others

finishing the hallway plant now

Also snuck in a few texture tests as well

like the threshold metal barrier

[neat spindle]

Hell yeah, nice. I finished the catwalks and have a good sense of how to make metal materials now (the way they interact with RT looks very different from the rasterized previews in my program). Currently working on some glass and that's a whole mess since Remix expects transparent materials to be discrete from opaque ones and that obviously isn't how any old games did anything.

[amber atlas]

yeah the glass setup is... tough

i have a plan for windows for the most part i think?

but the specific glass on the confession floor presents a pretty big issue

i have SOME early thoughts on how to tackle, but i'm not sure how that would work until i did some tests myself

as far as santamonica goes, as well as the apartments, most of the glass is inset into windows, so i may actually be able to include it as a discrete model as part of the frame, then put the 'interior' behind it on the texture

but im not sure yet

trying not to let myself get distracted by that 😄

[neat spindle]

When I get home I can send some screenshots of the progress I've made. I think I can get away with treating it like an opaque texture with alpha masking, but I need to do more testing on what opacity looks good.

[amber atlas]

there are some oddities with some settings i found in the actual remix toolpanel in game that really changes the way that particular technique displays, in one case it looked okay but broke other cases, but it could be cases that are fixed by other means

[snow osprey]

Hello, sorry for the tactless question, do you manually create 3D models yourself?

[amber atlas]

its a weird ecosystem

yeah

everything i do i do from scratch

using the original models as guides

so i set the old low poly models in blender and then get ref of objects in the world that represent what the low res textures were sort of trying to communicate and use that to sort of reason about what i could do to remodel them

so then i do a hi poly, and a low poly, then bake and texture in toolbag 4

the plant is a PAIN because foliage is hard as hell but i think im getting an ok result

[neat spindle]

I think the issue with using alpha settings is that it won't have proper reflections/diffraction/etc. the way glass should, but I'm going to see if I can get something approximate to start. It gets used on the booth at the entrance too so I'd like to avoid creating a model and stick to material replacement.

[amber atlas]

i thought the hl2 group had a custom shader for transparency, so it might be possible to write our own

however difficult

[snow osprey]

How cool, This is incredible, After working with a blender, I feel killed and you are doing an even more difficult job than me, my respect to you 🔥

[amber atlas]

Just years of practice! Blender is great so if you stick with it, I'm positive you'll kick ass too!

[neat spindle]

I do appreciate the amount of detail Nvidia's RT simulation has for transparent surfaces. Honestly never thought I'd actually get to apply that optics course I took in uni, lol

[amber atlas]

haha excellent

i was hoping that we could do some fun stuff with newtons rings but that's not in there D:

[neat spindle]

Yeah, I think for simplicity (and computation requirements) all of their settings assume parallel planes.

[amber atlas]

toolbag 4 has a fast aprox fake version of it that lets you change film thikness and line frequency and its so snazzy, but i don't even know how to begin with that in terms of hand writing a shader

[neat spindle]

Ooh, that sounds way cool.

Bit of a tangent, but I've been thinking about all the windows in the hubs. I think we can get away with faking those using opaque textures and emissives like the original game did, since there's no modelled interiors that will need transparency settings. Thankfully there's very little actually transparent glass in the game.

One issue I have run into is that I'm pretty sure the Source engine broken glass uses a shader.

[amber atlas]

yeah that's part of what i was saying, BUT I have a test I want to do where I actually add a pane of glass and whatever bars etc to the frame model, then use the texture underneath to simulate inside the window

also yeah i think you're right on the broken glass part

[neat spindle]

Hmm, yeah, that could be cool, but is also significantly more work than just basic material replacement. It's adding an additional step of creating a model to every single building texture in the game.

[amber atlas]

only for the ones that already have frames that are models, and its just adding one or two extra details to those, then removing those details on the corresponding surface textures for underneath

since the underneath is a brush

[neat spindle]

Ah, okay, that makes more sense.

Are there many buildings that have frame models? I feel like a lot of them are just textures.

[amber atlas]

in SM_Hub_1 a lot of them are

the trick of course being that... in the second shot

the top and bottom windows share a texture

so removing the window frame cross bar and stuff from the texture would remove it from the top

BUT those dont have the frame model