For games, you'll want to have small file sizes for your textures so you can maximize texture memory. For normal maps, that means reducing gradients as much as you sensibly can. You can do that by setting each UV Shell to its own smoothing group / hardening the edges. You've already split the shell along that edge, so it's something you could do for free of cost performance-wise. Bake a soft edge cube and a hard edge cube down from a high poly beveled cube, and then compare their normal map sizes. You'll be surprised how much larger the soft edge one is. This process will also help your art look better in engine, as the normal maps are typically scaled down and compressed. If you have a lot of gradients in your normal maps that are being compressed, you're going to get obvious stair-stepping / compression noise in those gradients over a large surface, and they will have a hard time compensating for the soft normals of your render mesh in engine, which could cause some shading artifacting.
is it important to use sub-division to make high poly? i need to make a high poly model, my model already looks realistic without sub-division, so do i have to add sub-div or not?
This is only relevant if you're storing normal maps in certain formats - i.e., .png. In formats like .dds (An extremely common texture format for games), this detail is irrelevant as all textures of the same dimensions and compression in .dds have the same file size due to how they compress.
Don't forget to mention: 1. Channel packing for textures. 2. Trim sheets. 3. Using overlapped UVs depending on situation like toon shader or having the ability to put overlaps on a negative number position/axis/atlas to bake light map information. This is to prevent light map artifacting. 4. Using Texture Atlases to save on drawcalls in game engines. 5. Using alpha maps/ opacity in materials. (Not to use too much because of performance). 6. Having seperate UVs sometimes for light baking in game engines. 7. Having multi-materials for one piece (multi-sub materials). 8. Having a baking version of models. 9. Industry file systems and naming conventions. Like m_SceneObjectNameMat 10. Difference of using more geometry which affects gpu performance and texture maps that use cpu performance. 11. Making a collision model or material for objects. 12. Breaking apart models for baking (depending on baking software). 13 different ways to model: spline modeling, boolean modeling, displacment modeling, and model upwards from texture/ texture planes...ect 14 pipelines: highpoly/sculpt to retop-low poly or low poly to highpoly...ect. 15 exporting assets to game engines. For example the difference between tangent space and binormal.
As a beginner, what you've said is just too deep for me, but I'm especially excited to find your additions in the comments , it's a treasure of great value to me and I'll be making these the ones I want to master, thank you soooo much!
I gotta say this honestly is a great video for people starting out in the 3D arts. You are covering so much info they anyone can really use out there and it's so helpful. A job well done with this video. I'll be looking forward to the next one!
crucial info on how all this works! I've been messing around with Blender for years and have never come across a video that explains how the workflow works like this!
I like the approach of Star Citizen for hard surface models more, where they dont use a high poly model for baking, but rather a relatively detailed low poly model that uses texture tiling, weighted normals and decals for the details (like bolts, dirt and text). Looks great even at very close distance, doesn't use a dedicated (baked) texture for every model and can change materials on the fly. This approach is really scalable to have high resolution scenes with many different objects, without using much texture memory.
Great overview! Just a note - there are a lot of stretched polys in your model that are causing lighting artifacts (see top paneling and side panel lighting at 12:30). Could possibly be coming from the normal map bake itself, but you'd have to troubleshoot.
This video encapsulates the first 4 years of my payed education in modelling into 15 minutes, and is almost as in depth. Great video, guys! Wish I knew about flipped normals before I started, it would’ve saved me so much time!
I think in this particular instance it would make sense to bake maps without positioning out meshes, but rather explaining on how to split it to groups for Marmoset. Because in this case you will actually get a completely different AO.
is it important to use sub-division to make high poly? i need to make a high poly model, my model already looks realistic without sub-division, so do i have to add sub-div or not?
And here i am making game assets for someone I know with no knowledge 🤣 only a lot of ideas on how things should be done but dunno if its right way or not. Right I Finnished all modeling stuf and puting simple textures and materials to the model I'm trying to learn baking texures but it seems their is something that I missed that is having a high poly and low poly model that you said so that's the reason why their is always a high and low poly for game assets. thank you very much for this will research more and redo all the stuffs that i made so far.
I already kinda knew most of this, but its always good to refresh a little or check to see if you missed some basics. So thanks for making the video, always great to see people make informative videos even for absolute beginners. Also fun to see how many dutchies (sorry if i assumed incorrectly) are in this work field.
The baking stage is a little confusing. Are you unwrapping both the high and low poly objects to have the same UV space before you do that? Or can you leave the high poly UVs untouched and the software does the work for you when it bakes the high poly mesh details onto the lower unwrap?
Only the low poly requires a UV, the high poly normals will then be projected outward onto the low poly and written onto the texture based on the low poly's UVs.
Nice explanation of the basics. There are many tutorials on each of these subjects, but I always seem to miss tutorials about the process that brings everything together. I'm trying to create my own racing game, I know some modelling, graphics, programming, ... . But I miss the right workflow. I'm creating a track, but how much and what do I need to make in a 3D modelling software and how and what in the game-engine itself. How much in a level is considered an asset and how much is considered as part of the set piece? Currently I have a small part of the track of a rocky beachfront, also with some animated falling rocks, all made in Blender. Did I make too many rocks in Blender? I actually made around which I spread around with a particle modifier, applied it permanantly and now have meshes of 'grouped' rocks. But do I need to convert them to assets or let them be part of the set-piece? Especially since I'm creating with VR in mind, I really need to be mindful of the resources. Btw, you sound either Dutch or Flemish. Am I right?
Hello!!!! Sir last time we buy a course from you related to this topic but i lost my account not remember it so, i have downloaded tutorial upto UV mapping so, i need access to tte texturing and baking part so, i will buy again te full course
do the lowpoly model needs good topology inorder to use its as a final game ready asset? i am a beginner and im always worried about topology, when modelling something. At which part of the game asset workflow that good topology is necessary? plz help me to clarify this one.
When it is a static asset (not being animated/deformed) it is more important for a model to be low poly than to have nice topology. The only reason you might want nice topology in a static game model is if you want to bake it from high to low poly. You'd need to have good topology before you start doing any UV unwrapping, baking, or texturing. I hope this helps!
@@kukkutdas4621 Though this is mostly true, topology also can have a big impact on the shading of the model(especially on curved and smooth surfaces. Just a good thing to know if you are interested. Good luck with your modeling!
Because game engines will triangulate your models. If you baked your normal maps before triangulation, then once you import it to game engine you will see that normal map is ruined.
Don’t listen to the comments, most of these people think they made a game because they modeled a crate for ArmA. They don’t know shit. It’s a good video, these workflows are legitimate, can confirm.
The sad reality is that when you ask on YT, you almost never get an answer. All the noobs (like you and me) only watch such videos without experience (otherwise you wouldn't be here?), the pro's spend their time elsewhere. It is better to approach some youtubers in person, pay them money (if necessary) and they will answer your questions.
Im learning 3d cause me and my friend is doing a passion project so we have something to show for when we apply for jobs. When modelling, do we have to use smooth preview? I assume that smooth preview is how it would look as high poly? or is it just named high poly because there's more polygons on the object?
Het was best grappig om te horen dat je halve wegen sneller begon te praten. Wou je de lengte van de video kort houden maar had je meer te verstellen dan gedacht?
hi, I have a question, could be that topology right for a game asset, or it`s just okay for renders. Because i use to see similar cuad forms and good looking topology.
Hey, i don't know if I got the question right, but an optimized topology does matter more for game assets than CG renders. Game assets have to be (realtime) rendered everytime, but a CG render is just an image. You can get away if it takes hours to render. However, an optimized highpoly object will be calculated faster, so a CG render will take less time ;) The quad, tri's and n-gon thingy. Correct me if I'm wrong, but for gaming it doesn't matter if you use tri's or quad polygon models. You will see mostly tri-models online for games. In a game engine i used a long time ago, it would convert my quad model intro tri's. Quad polygon's can easily be sliced into half (tri's), n-gon polygons should be avoided. Also because your model can be messed up when a smooth modifier or whatever will be put on these faces.
@@maartenvanderveen5144 AFAIK - game engines usually turn quads into triangles automatically so there's no need to manually do it for the most part. Only in some critical places because automatic triangulation can cut quads in the wrong direction. Real-time topology is often very weird because they utilize every possible trick to decimate the mesh while still having enough resolution to maintain the big shape. As long as it looks good when shaded and deforms well - it's fine. Different studios have different standards tho. In blizzard, for example, every character in WOW has a perfectly horizontal loop coming from the corner of their mouth. It's literally just scaled 0 on z axis.
is it ok for game engines to have so many triangles instead of quads and hanging geometry? Or I should go from high to low poly with retopo or it will just increase poly count and ruin performance?
triangles are fine, i was once mistaken by someone that i should absolutely avoid anything but quads, unti,l i started learning hard surface modelling for real, and discovered that ngon workflow is way better than quad workflow for hard surface.
Can blockouts be easily 3D printed to actual size? The Blockout creation/rendering looks like it would be a great starting place for making custom jewellery with oval and natural shaped stones. I'm looking at some software to help me with 3D jewellery cad. Blender looks really good.
3D printing requires a bit of a different approach when it comes to modeling and prepping a print - we'd encourage you to have a look at these course if you're interested in diving deeper into the specifics :) flippednormals.com/downloads/prepping-characters-for-3d-printing/ flippednormals.com/downloads/splitting-keying-your-sculpts-for-3d-printing-using-zbrush/
with baking PbR,Sub D to Low poly modelling , sup.loops, with without cage, LOD, mip maps, UV technic, etc.etc.??? p.s...wO.Ow thx..thx...thx. ..thx.....
Despite it's really useful tutorial the asset contains so many mistakes making it awful... I appreciate what you do really much, and I would want not to write this comment, but unwrapping looks weird, normal map after baking looks glitchy with broken normals so people after the tutorial could get wrong understanding of the correct workflow
The asset is created by Senior 3D Artist, Emiel Sleegers, who has over 8 years of experience working on AAA titles such as The Division 2 and Forza Horizon 3. The techniques and workflow he's showcasing here are the exact ones being used by major games studios around the world. The course is meant to teach artists and hobbyists how 3D assets for major game titles are being created :)
@@fnmarketI’m always grateful to Artists, like this, who break the taboo of thinking you having to create “perfect geometry.” The results speak for themselves right?
Honestly comments like this really big me, especially when trying to learn. I've been put off from doing things by comments in the past until I realize most comments are completely incorrect. You should do a video of you think you know a better way. Baking assets is a legit method.
I don't understand these videos. You keep saying, later on I will do this and that, and never come to it? Is this a promo video for the ACTUAL TUTORIAL?
We always put a link to the full course in the description. This is a supercut of material from the full course we're giving away for free here on TH-cam :) With all of our FlippedNormals Exclusives, we release a trailer along with a free chapter from the course on this channel. Hope that clarifies things!
Oh man... N-Gons. And you expect this to be used in a game? Sir, the reason why quads are advocated for is because 2 tries can represent a quad and they are better interpreted by both us and the machine for animations and platform transitioning. Do try your best not to ignore this rule because although it looks good, the topology is not and thus the functionality will suffer.
For things that are planar and don't animate, n-gons are a perfectly valid option. Your final triangle count will be much larger without the use of n-gons in most cases, and for little or no benefit.
Man, I'm so happy flipped normals exist. Love these videos, relatively condensed and pretty informative.
Thank you so much! It's great to hear you enjoy what we do :)
Pro trick: you can watch series on kaldroStream. I've been using it for watching lots of of movies during the lockdown.
@Andres Aries Yea, I've been watching on Kaldrostream for months myself =)
@Andres Aries Yup, been using kaldrostream for years myself =)
For games, you'll want to have small file sizes for your textures so you can maximize texture memory. For normal maps, that means reducing gradients as much as you sensibly can. You can do that by setting each UV Shell to its own smoothing group / hardening the edges. You've already split the shell along that edge, so it's something you could do for free of cost performance-wise.
Bake a soft edge cube and a hard edge cube down from a high poly beveled cube, and then compare their normal map sizes. You'll be surprised how much larger the soft edge one is.
This process will also help your art look better in engine, as the normal maps are typically scaled down and compressed. If you have a lot of gradients in your normal maps that are being compressed, you're going to get obvious stair-stepping / compression noise in those gradients over a large surface, and they will have a hard time compensating for the soft normals of your render mesh in engine, which could cause some shading artifacting.
is it important to use sub-division to make high poly? i need to make a high poly model, my model already looks realistic without sub-division, so do i have to add sub-div or not?
no@@Kyoshi267
This is only relevant if you're storing normal maps in certain formats - i.e., .png. In formats like .dds (An extremely common texture format for games), this detail is irrelevant as all textures of the same dimensions and compression in .dds have the same file size due to how they compress.
Don't forget to mention:
1. Channel packing for textures.
2. Trim sheets.
3. Using overlapped UVs depending on situation like toon shader or having the ability to put overlaps on a negative number position/axis/atlas to bake light map information. This is to prevent light map artifacting.
4. Using Texture Atlases to save on drawcalls in game engines.
5. Using alpha maps/ opacity in materials. (Not to use too much because of performance).
6. Having seperate UVs sometimes for light baking in game engines.
7. Having multi-materials for one piece (multi-sub materials).
8. Having a baking version of models.
9. Industry file systems and naming conventions. Like m_SceneObjectNameMat
10. Difference of using more geometry which affects gpu performance and texture maps that use cpu performance.
11. Making a collision model or material for objects.
12. Breaking apart models for baking (depending on baking software).
13 different ways to model: spline modeling, boolean modeling, displacment modeling, and model upwards from texture/ texture planes...ect
14 pipelines: highpoly/sculpt to retop-low poly or low poly to highpoly...ect.
15 exporting assets to game engines. For example the difference between tangent space and binormal.
That'll all be covered in our full courses on FlippedNormals.com - this is simply a free sample we're offering here on YT :)
@@fnmarket bruh
@@fnmarket sir is that important to combine all part of same object when we are making game ready assets??
multimaterials... no...
As a beginner, what you've said is just too deep for me, but I'm especially excited to find your additions in the comments , it's a treasure of great value to me and I'll be making these the ones I want to master, thank you soooo much!
I gotta say this honestly is a great video for people starting out in the 3D arts. You are covering so much info they anyone can really use out there and it's so helpful. A job well done with this video. I'll be looking forward to the next one!
Thank you! Really glad to hear you enjoyed it :)
crucial info on how all this works! I've been messing around with Blender for years and have never come across a video that explains how the workflow works like this!
I like the approach of Star Citizen for hard surface models more, where they dont use a high poly model for baking, but rather a relatively detailed low poly model that uses texture tiling, weighted normals and decals for the details (like bolts, dirt and text). Looks great even at very close distance, doesn't use a dedicated (baked) texture for every model and can change materials on the fly.
This approach is really scalable to have high resolution scenes with many different objects, without using much texture memory.
Finally a guy that knows what he's talking about in terms of proper technique and industry. insanely rare on youtube
agreed!
Ikr.
Great overview! Just a note - there are a lot of stretched polys in your model that are causing lighting artifacts (see top paneling and side panel lighting at 12:30). Could possibly be coming from the normal map bake itself, but you'd have to troubleshoot.
A very good summary of the steps that need to be done for 3D assets. Thanks!
This video encapsulates the first 4 years of my payed education in modelling into 15 minutes, and is almost as in depth. Great video, guys! Wish I knew about flipped normals before I started, it would’ve saved me so much time!
Why multi resolution instead of subdivision modifier?
Great video! I'll be sharing this with people I know who want to get into 3d modeling, so they at least know what they are getting into.
Is it possible to sculpt a blockout, then retopo it for the low poly, instead of poly modeling the block out or low poly?
Almost every characters are made this way so yes. Just do it the most efficient way for the object you're trying to make.
Yes
I think in this particular instance it would make sense to bake maps without positioning out meshes, but rather explaining on how to split it to groups for Marmoset. Because in this case you will actually get a completely different AO.
My goodness I love the accent, I feel I've nearly drowned in American accents with all the 3D tutorials I've watched
Tip top job thank you
Great to hear you liked it! :)
What about the indian accent? I wish there was a filter for those.
@@edenassos why ?
is it important to use sub-division to make high poly? i need to make a high poly model, my model already looks realistic without sub-division, so do i have to add sub-div or not?
I would like to know the answer to this question as well!
And here i am making game assets for someone I know with no knowledge 🤣 only a lot of ideas on how things should be done but dunno if its right way or not. Right I Finnished all modeling stuf and puting simple textures and materials to the model I'm trying to learn baking texures but it seems their is something that I missed that is having a high poly and low poly model that you said so that's the reason why their is always a high and low poly for game assets. thank you very much for this will research more and redo all the stuffs that i made so far.
I already kinda knew most of this, but its always good to refresh a little or check to see if you missed some basics.
So thanks for making the video, always great to see people make informative videos even for absolute beginners.
Also fun to see how many dutchies (sorry if i assumed incorrectly) are in this work field.
The baking stage is a little confusing. Are you unwrapping both the high and low poly objects to have the same UV space before you do that? Or can you leave the high poly UVs untouched and the software does the work for you when it bakes the high poly mesh details onto the lower unwrap?
Only the low poly requires a UV, the high poly normals will then be projected outward onto the low poly and written onto the texture based on the low poly's UVs.
One of the greates videos.
hey where i can find full tutorial of this course i couldn't find it online
Just discover your channel and i'm very happy to leanr those is the way you present it ! Thaks for your great work
Welcome to the channel! Glad to hear you enjoy our content :)
Nice explanation of the basics. There are many tutorials on each of these subjects, but I always seem to miss tutorials about the process that brings everything together. I'm trying to create my own racing game, I know some modelling, graphics, programming, ... . But I miss the right workflow. I'm creating a track, but how much and what do I need to make in a 3D modelling software and how and what in the game-engine itself. How much in a level is considered an asset and how much is considered as part of the set piece?
Currently I have a small part of the track of a rocky beachfront, also with some animated falling rocks, all made in Blender. Did I make too many rocks in Blender? I actually made around which I spread around with a particle modifier, applied it permanantly and now have meshes of 'grouped' rocks. But do I need to convert them to assets or let them be part of the set-piece?
Especially since I'm creating with VR in mind, I really need to be mindful of the resources.
Btw, you sound either Dutch or Flemish. Am I right?
Hey! Emiel Sleegers! This tutorial is definitely going to be good.
love your accent, thanks for the valuable vid
good explanation. thanks
Excellent video, very good pace.
When you talk about "I will talk about this further down the line", what's the course that follows?
You can project the screen on in substance painter. However I wold probably design it in photoshop or illustrator.
Hello!!!! Sir last time we buy a course from you related to this topic but i lost my account not remember it so, i have downloaded tutorial upto UV mapping so, i need access to tte texturing and baking part so, i will buy again te full course
do the lowpoly model needs good topology inorder to use its as a final game ready asset? i am a beginner and im always worried about topology, when modelling something. At which part of the game asset workflow that good topology is necessary? plz help me to clarify this one.
When it is a static asset (not being animated/deformed) it is more important for a model to be low poly than to have nice topology. The only reason you might want nice topology in a static game model is if you want to bake it from high to low poly. You'd need to have good topology before you start doing any UV unwrapping, baking, or texturing. I hope this helps!
@@fnmarket thank you so much.....your videos help me a lot. keep making awesome content like this, guys...all the best👍👍
@@kukkutdas4621 Though this is mostly true, topology also can have a big impact on the shading of the model(especially on curved and smooth surfaces. Just a good thing to know if you are interested. Good luck with your modeling!
@@squishybluecat thankyou so much for sharing the knowledge...
I tend to use the decimate tool on my high poly models. I know it’s a bit lazy, but it seems to work well for what I do.
Isn't that the way to do it though? Or dropping the count on the multiresolution lol
im scared
@@squishybluecat why ?
For LODs its fine, but for closeup objects (like a gun) you would have to manually work to create the low poly version.
Vry good info.. thanks u so much 👍
please tell me why you are modeling with triangles in topology?
Because game engines will triangulate your models. If you baked your normal maps before triangulation, then once you import it to game engine you will see that normal map is ruined.
Don’t listen to the comments, most of these people think they made a game because they modeled a crate for ArmA. They don’t know shit. It’s a good video, these workflows are legitimate, can confirm.
Is it necessary to bevel low poly edges?
1:05 is it a cyberpunk level? Looks like a location from the final mission with panam and the basilisk tank
looks more like the dam from the black ops 3 campaign.
hehehe my guy!! je engels Is funny maar me waardering is hoog, liefde.
Nice video, I've been doing modeling in blender for a bit more than a year and I still didn't have a clear knowledge of these things.
Glad we could help!
So what was the polycount on the low poly?
The sad reality is that when you ask on YT, you almost never get an answer. All the noobs (like you and me) only watch such videos without experience (otherwise you wouldn't be here?), the pro's spend their time elsewhere. It is better to approach some youtubers in person, pay them money (if necessary) and they will answer your questions.
Way more than necessary.
@@virxest low poly used about 15k tris
This is very informative thank you!
Glad it was helpful!
Just press Ctrl+1 or Ctrl+2 and it adds subdiv surface modifier in Blender. Multi-Resolution is a sculpting modifier.
Very infromative, thank you!
Glad it was helpful!
Really informative
6:49 I wasn't paying attention since my phone rang, and for a second I heard ''madafaka'' LOL
Im learning 3d cause me and my friend is doing a passion project so we have something to show for when we apply for jobs.
When modelling, do we have to use smooth preview? I assume that smooth preview is how it would look as high poly? or is it just named high poly because there's more polygons on the object?
Thanks
i have a question should we have to give spotting loops in low poly or not
A well deserved like and sub
I needed this :D
7:50
What app called
So he is talking about Marmoset Toolbag for baking, but not mentiones Marmoset Toolbag 4 for actual texturing?
Het was best grappig om te horen dat je halve wegen sneller begon te praten.
Wou je de lengte van de video kort houden maar had je meer te verstellen dan gedacht?
hi, I have a question, could be that topology right for a game asset, or it`s just okay for renders. Because i use to see similar cuad forms and good looking topology.
Hey, i don't know if I got the question right, but an optimized topology does matter more for game assets than CG renders. Game assets have to be (realtime) rendered everytime, but a CG render is just an image. You can get away if it takes hours to render. However, an optimized highpoly object will be calculated faster, so a CG render will take less time ;)
The quad, tri's and n-gon thingy. Correct me if I'm wrong, but for gaming it doesn't matter if you use tri's or quad polygon models. You will see mostly tri-models online for games. In a game engine i used a long time ago, it would convert my quad model intro tri's. Quad polygon's can easily be sliced into half (tri's), n-gon polygons should be avoided. Also because your model can be messed up when a smooth modifier or whatever will be put on these faces.
@@maartenvanderveen5144 AFAIK - game engines usually turn quads into triangles automatically so there's no need to manually do it for the most part. Only in some critical places because automatic triangulation can cut quads in the wrong direction.
Real-time topology is often very weird because they utilize every possible trick to decimate the mesh while still having enough resolution to maintain the big shape.
As long as it looks good when shaded and deforms well - it's fine.
Different studios have different standards tho. In blizzard, for example, every character in WOW has a perfectly horizontal loop coming from the corner of their mouth. It's literally just scaled 0 on z axis.
Can you make a video about character modelling workflow without sculpting? I'm struggling alot with it
my problem is always to decide which bevels i do in mesh or in the normal map
is it ok for game engines to have so many triangles instead of quads and hanging geometry? Or I should go from high to low poly with retopo or it will just increase poly count and ruin performance?
Game engines convert all quads into tris when it's read by the engine, so triangles are no problem at all :)
triangles are fine, i was once mistaken by someone that i should absolutely avoid anything but quads, unti,l i started learning hard surface modelling for real, and discovered that ngon workflow is way better than quad workflow for hard surface.
Uit welk land zou deze man komen?
Quick question: do I need to bevel hard surface low poly models?
If you're talking about weighted normals then yes, unless the object/edges are very very small :)
My dream is to find one tutorial about molding a 3D object like this one, for games. But I can't find
I wish that I had this video when I was first learning lol.
I've never heard anyone speak such good English yet, it completely sounds Dutch. Like I'm back at Ms Verlinden's class. Crazy.
Can blockouts be easily 3D printed to actual size?
The Blockout creation/rendering looks like it would be a great starting place for making custom jewellery with oval and natural shaped stones.
I'm looking at some software to help me with 3D jewellery cad. Blender looks really good.
3D printing requires a bit of a different approach when it comes to modeling and prepping a print - we'd encourage you to have a look at these course if you're interested in diving deeper into the specifics :)
flippednormals.com/downloads/prepping-characters-for-3d-printing/
flippednormals.com/downloads/splitting-keying-your-sculpts-for-3d-printing-using-zbrush/
very good
Can you provide the reference image please?
For the Pokedex? The concept is made by Justus Sluijter and can be found here www.artstation.com/artwork/zOJm4Q
@@fnmarket Appreciate it, thanks!
with baking PbR,Sub D to Low poly modelling , sup.loops, with without cage, LOD, mip maps, UV technic, etc.etc.???
p.s...wO.Ow thx..thx...thx. ..thx.....
ugh... why is it that english subtitles are never available in videos with difficult accents. it's frustrating.
its called a dutch accent
@@KaasTVNL Lol ok and ?
Nice
wow thankie workflow
after hearing the voice of the video im pretty sure the creator is dutch. am I wrong?
You are correct!
wow
NINE!!!!
Despite it's really useful tutorial the asset contains so many mistakes making it awful... I appreciate what you do really much, and I would want not to write this comment, but unwrapping looks weird, normal map after baking looks glitchy with broken normals so people after the tutorial could get wrong understanding of the correct workflow
The asset is created by Senior 3D Artist, Emiel Sleegers, who has over 8 years of experience working on AAA titles such as The Division 2 and Forza Horizon 3. The techniques and workflow he's showcasing here are the exact ones being used by major games studios around the world. The course is meant to teach artists and hobbyists how 3D assets for major game titles are being created :)
@@fnmarket that doesn't cancel the fact UV packing could be more effective either as unfolding, and shading is broken, that's obvious on 15:02
@@fnmarket the experience is not a vaccine from mistakes.
@@fnmarketI’m always grateful to Artists, like this, who break the taboo of thinking you having to create “perfect geometry.” The results speak for themselves right?
Honestly comments like this really big me, especially when trying to learn. I've been put off from doing things by comments in the past until I realize most comments are completely incorrect. You should do a video of you think you know a better way. Baking assets is a legit method.
man....You forgot about hard edges. This bake is very f**** up
Topology grid is not correct and not gameready!
Wish more tutorials had time stamps.
I’m Collecting video data for my AI software to read timestamps in generalizing workflow. The ultimate TUT AI
I don't understand these videos. You keep saying, later on I will do this and that, and never come to it? Is this a promo video for the ACTUAL TUTORIAL?
We always put a link to the full course in the description. This is a supercut of material from the full course we're giving away for free here on TH-cam :) With all of our FlippedNormals Exclusives, we release a trailer along with a free chapter from the course on this channel. Hope that clarifies things!
who did the captions 🤣
TH-cam bot
not any more ue5 changed everithing
AO map is a show!@#$@ i repeated this 1000 times and didnt get it 11:50
Soft shadows
Sure love tutorials that don't focus...
Je bent sowiezo Nederlands
it does not move. we dont need a tutorial for models that dont move. and by the way, who makes a game by himself with such high-poly models??
Oh man... N-Gons. And you expect this to be used in a game? Sir, the reason why quads are advocated for is because 2 tries can represent a quad and they are better interpreted by both us and the machine for animations and platform transitioning. Do try your best not to ignore this rule because although it looks good, the topology is not and thus the functionality will suffer.
For things that are planar and don't animate, n-gons are a perfectly valid option. Your final triangle count will be much larger without the use of n-gons in most cases, and for little or no benefit.
"Explained": We do stuff, and then more stuff... there ya go.
Very helpful. Thank you
we don't care if they are popular we need free ones
Getting sued by nintendo in under 16 mins!~