Skip to main content

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

Teppo Ylitalo

50
Posts
A member registered Jan 31, 2020

Recent community posts

(1 edit)

Hi there.

No, the level showcased wasn't unfortunately in your delivery. It did include a level called 'Asset zoo', which I'm basing my breakdown on that Technical Art regard.

It's cool. I read you had some tight deadlines to make the scenes to, right? I'm just stating things that I see, as I can't see 100% of the thought processes and what you know based on one scene; it seems like a better approach to mention such than to leave things and notice later that this person could have probably benefitted from something that could be fit into a single sentence. :)

Research & Development

Research shows nice high quality references. References have been divided to different sections. Did you lose the sight of main reference? You seemed to add quite a lot of very different elements to the scene.

Development of the scene is not super well documented in the pdf but video adds to it. Props are broken down well.

Creative Art

The scene showcases a nice snowy landscape with interesting buildings and supported time of year.

Assets could be pushed further with closer look at references, looking at how objects are built, adding irregularities and pushing the material work to read and feel more cohesive.

Technical Art

Objects have been scaled on the asset level, and probably also within scene, to look coherent and consistent. Polycounts seem optimized. Architectural modules could be pushed to the next level with irregularities and taking a closer look at how things are constructed. Ceiling modules are not waterproof, which feels a bit of a waste, as they read and look good, and could really use those elements not to break the immersion of real world and lose it to the realm of 3D. Lantern also moving all around with grass shader might not be the best call based on its structure.

Snow shader adds to objects. The fact the snow disappears up close is not a great thing. Pivot placement is a bit weird, and doesn't seem to follow a good logic to them.

Documentation

Different stages have been broken down well. It could be maybe done on a bit more general level, as now everything seems to be documented.

Final Presentation

Images are of high quality and show the subject matter. Video is a nice addition. The project file is missing the developed level.

Research & Development

There's nice writing for the topic; it feels very interesting. Different assets are shown well for the process of development. Research into real world and defined location could be maybe added, and choosing 1 image as the main reference.

Creative Art

The shrine reads, and there are interesting elements to the scene. Fruits are of high quality, and the scene introduces vivid colors with flags.

The scene could benefit maybe from an overpaint; you have organic elements in slightly blocky forms, and lots of small detail props. Try to make the next scene a bit more with big (>2m), medium (2m>x>20cm) and small shaped (>20cm) props for similar scene. :) Another thing could be to push the scene to be more concise and packed so you can push the quality rather than quantity; the altar itself could already present an interesting opportunity for the scene. If you want to create a bigger scene like this, I'd suggest researching level design for interesting spaces and how they lead player and eye to certain locations.

Technical Art

The assets show roundness very well, but smaller assets may be a bit too small to have that many polygons. Shrine asset has too few polygons to the roundness, while the crate has some polygons that don't add to the silhouettes of the objects. Add polygons to the rounded surfaces; Can you spot an object being 3D with sharp polygon lines? -> it calls for maybe more polygons to get the roundedness. For candle glass the roundedness might be a bit much; be more mindful of what player sees and try to be coherent there with that in mind.

Water works. Ground could maybe benefit from making use of Unreal's Landscape tools.

Post process volume with Min & Max Brightnesses set to 1 would help with delivering kickass lighting and material work.

Extra keywords to research Texel density (512, 1024), tiling textures (used already well, but maybe a bit more research would do no harm), trim sheets, face-weighted normals.

Documentation

Images are easy to look at. Text reads well. Layout could maybe be touched a bit to add to a more easy viewing experience.

Final Presentation

Renders showcase subject matter. Composition could maybe pushed a bit with implementing rules of third or golden ratios while keeping the main attraction or what you want to show well in the image.

Research & Development

Research showcases a lot of overall scene research in terms of lighting and feel. I'd like a bit more closer references as well, your objects might also benefit from getting some real world details to the as well for example material wise.

Pie chart color wheel works very well, I could adapt similar to my work.

Creative Art

The scene looks like an interesting futuristic office space on its own. Shapes and elements look interesting. Dang, do I like the shapes in there a lot, with mix of lamp's round shapes, rough rock wall, and flat surfaces. Canvases are also something I make a connection to modern spaces. Use of color and its ratios are mature and interesting.

The space a bit dim. Who has a place like this? Is it a company's office space? Is the owner some rich billionaire?

The shapes also, while being interesting, feel a bit off scale which might be due to them either not being done to what is commonly viewed as the object dimensions in real world, or just the  fact that there's not so much  ordinary/common stuff to compare it to.

The ceiling color is a bit maybe too eye catching with hot red wood to it; I'd suggest desaturating and darkening it to give the space a bit more of a relaxing feeling. Rock and concrete elements look intriguing, floor decision seems slightly odd (or the material on it does, maybe polished marble surface would do better), and the stairs could benefit from maybe a different material to it. Couch works well shape and color wise, but the material seems slightly beaten/something that would introduce dust into the air of the space and as such isn't exactly catering to the futuristic office feeling.

Modules could use edge decals to give breakage to the edges as well as some irregularities with polygonal usage to mimic the real world materials such as the plaster and concrete.

Technical Art

There are wasted polygons on chairs, railings on the bottom of chair are on high on polycount, cushion could use more polies to add more roundedness and reduce the 3d feeling. Materials use material instances (it would be nice if you showcase this also in the prefix of suffix, I use "MI_" for instances for example). Fuzzy shader is a nice addition to the carpet, but is a bit hard to read in editor. Lighting could use some optimization (I'm looking at you, dozen pointlights in the ceiling!)

Documentation

Documentation is well written and colorful. Words are well written; I'd like to get the fonts maybe a tad bigger for easier reading. Texts could maybe also do with shorter chapters.

Final Presentation

Images showcase the scene well. The video was a pleasure to look at.

Research & Development

There's nice research about subject and differences of western and eastern mechs. Research shows nice concept art about both. I'd like to see some real world reference for detail, technicalities and materials, there'd probably also be a plethora of real world reference for places used for storage, automated robotic lines etc.

Development of assets is presented well. Thought process in decision making and furthering your research is a bit on the short side.

Creative Art

The scene reads well as a base of operations, with mech being the main focus. The setting is interesting, but could benefit from researching a bit more into game art techniques and real world locations and constructions.

Doing a mech in ~6 weeks is already a huge feat, I'd suggest cutting the project down to maybe doing a part of the mech on the floor, or on a construction site and concentrating a bit more into delivering high quality environment.

Technical Art

Rounded surfaces of things like table could use a bit more of polycount.

Mech could be done in a more optimized manner with tileables and separating elements to different materials to hit texel density of 512 or 1024.

Techniques such as face-weighted normals, tileable textures and trims could be further utilized to deliver a kickass environment first and foremost. Hard edges and smoothing errors (eg. on crate) break the immersion of the location actually existing.

Modular pieces mean breaking things down to smallest possible module, with which it's easy to still populate levels in level editors or modeling software (like you did with grates, not how you did with wall frames).

Scales could probably be revisited after doing some research.

As a workflow, start with boxy whitebox, where you start to block out shapes and feel the area. Then start adding interest to that by giving more of a detail layer to the models. After that, take the meshes and modules and make them into final assets, and polish them afterwards.

Documentation

Documentation shows different parts of development well. Texts are short and easy to read. Some thought process and reasoning behind calls is present.

Final Presentation

Lighting shows the main subject well. Images are of high quality. Things aren't being covered by effects etc.

Research & Development

The research showcases lots of cool concept art. I'd like to see an extra added bit of real world structure, detail and material work.

Creative Art

This scene has qualities I don't actually pay attention to that often. There are lots of shiny emissive details, the contrast is very strong with dark values taking most of the screen, and the bright values taking the second largest portion, keeping the mid values as the lowest ratio of the tree. The color scheme induce interestingly warm and futuristic feeling. The corridors have been color coded to help with orientation of the scene. The floating texts give a nice sense of using some sort of HoloLens device, or holographic projectors. Perlin noised energy pattern also adds a nice flare to the scene. Shapes of the corridor and door look intriguing.

The dark value could maybe pushed a bit brighter to show a bit more of the details of meshes, trims, decals and such. Material work could be pushed to the next level with reused texture spaces and hitting certain texel densities.

Technical Art

There's a nice use of trims and shader effect. Lighting read well, but might be done in a more optimized matter with spots and rectangle lights. Meshes seem optimized, but for example the door could use some nice baking or mesh splits to reduce the 30k polycount. Materials could be pushed to be used in an instanced matter out of a master material.

Documentation

Texts read well and there are lots of pretty pictures to make the viewing a pleasure.

Final Presentation

The presented screenshots show very well the generic feel and mood of the scene. Closeups and how everything it made on a closer level is easier to see on the project file.

(1 edit)

Research & Development

Research shows nice main concept. References for objects would be better, if they followed several images of one object (you could browse 2nd hand stores or using search words like 'used' to get some amazon/ebay or similar online stores as reference.)

Creative Art

The scene is a nightmare filled place with dream-esque lighting and atmosphere. Items, materials and scales all induce creepy shivers.

Dream-like lighting shows and reads well, even if it feels a bit more dream-like, and contrasting with the rest as such.

Objects could do with an additional look at the references, to capture the scales, widths, heights, negative space, detail and materials of objects. The overall scene might also be a bit too grungy, you could try introducing some flat surfaces in nice ratios, and the ceiling color is a bit funky.

Technical Art

UV packing seems optimized by most part (with the exception to lamp, what happened there?) I'd suggest reusing texture space, and maybe dividing materials to trims, tileables and uniques if necessary for bigger assets to hit the texel densities of 512 or 1024.

There's a nice use of volumetrics, dust particles, atmospheric fog. Cool texts on the decals ^^Light channel usage was also a nice thing I didn't know of.

Documentation

Documentation presents taken steps in a nice fashion with lots of pretty pictures. Asset breakdowns are done well.

Final Presentation

Images are of high quality and present the state of the scene. It's easy to tell what's going on.

Research & Development

Research shows a huge amount of references showcasing different aspects of the scene. Sketches are presented well to showcase the idea of what you're looking to achieve.

Creative Art

The final scene is a beautiful scene with fantasy and dream like feeling. There's a strong sense of foreground, midground and background elements in the composition, as well as light and dark. Color palette shows and works well, helping with the fantasy aesthetic. Objects have interest to their silhouettes.

Technical Art 

Polycount of the sword is very nice. Roots might adhere more to the fantastic feel if they were more round.

Volumetric lighting and fogs give the scene nice presentation.

There's nice adaptive tesselation on the ground, but the results might not be worth the high polycount of it for the scene.

Documentation

Lots of images and good text telling about different stages of development and reasoning behind made calls. Text is a bit chatty, but personal and shows deep understanding and desire to tell about the process.

Final Presentation

Screenshots show well atmospherics, readability and asset work in the scene.

(1 edit)

Research & Development

Research shows different world locations of similar places. I'd suggest picking a main reference and sticking a bit more to it for the time being. The structural correctness of modules shows, but might benefit from looking a bit closer at reference to get the things to read with better scales (how much things are sticking out)

Creative Art

The corridor reads as a abandoned asylum where spray artists have been exploring their artistic capabilities, with the main focal point being the bigger tag on the wall.

Technical Art

I'd suggest to look into texel density (512 for 3rd person, 1024 for 1st person), and check dimensions of real world objects such as tiles and doorways. Use this to your advantage when planning your texture, and try to reuse same texture space as much as you can without sacrificing quality. 2k maps are somewhat the biggest texture size you could be covering for individual assets, but you can also divide bigger assets into several materials (eg. tileable 1k material set, decals 256 material set, uniques 512 set, .. )

Lighting works well to present the main attraction.

In terms of polycount, round connections of gurney might have too high roundedness, while the module with rounded top might have it too low. In general, it would be a good idea in today's games to model separate elements of windows as separate elements, instead of extrusions and insets of a plane. Wheel chair has polygons that don't seem to add a lot to the model itself (eg.the laurels of wheel).

Playing the build, the character is either half the size of common 2m, or the objects might be double in scale. Maybe there was a conversion issue?

There's probably something slightly funky with the packed file, too, as it seems to be almost 2Gb of size while the project unpacked out of it is something around 500 megs.

Documentation

The documentation presents images well. Text is by most parts easy to read. Progression and thoughts behind decision making could maybe be more present.

Final Presentation

Presentation, although being a bit dark, shows the scene and doesn't try to hide anything. Delivering build instead of uproject was a nice touch, even if it meant judging models I needed to import models to Marmoset Toolbag. All in all, nice delivery.

Hey, please note that you do not have to take all of the advice to heart if it doesn't sit right. Also, if you feel like I could maybe help a bit in the future, don't hesitate to reach out to me in through other media, such as twitter or linkedin. Cheers, and all the best!

(1 edit)

Research & Development

Documentation shows interest towards subject and making research a bit into real-life and a lot into other artist's interpretations of the subject. I'd like to see a bit more research into how constructions are made in real world for objects such as stairs and floor.

Creative Art

The presentations show an interesting contrasted scene which you want to take a closer look at. The material scheme mostly consisting of metal surfaces and color scheme comprising of whites and and blues give the scene a cold and cryogenic feel.

To me, the scene feels to be a bit too fed with detail, which results in it geting a bit heavy to look at for a longer while. Introducing more organic elements might soften the scene a bit from the untouched look. As an idea, you could maybe add some frost as decal or material blend. Small emissive lights of another color might also add a bit more spice into the mix, and giving structure, such as stairs, some real world structure to them as a base which could afterwards be more tweaked to cater the scifi feel. The props have interesting feels to them, but might use a bit of polish either through face-weighted normal method or bakes, and the wall modules would look better without broken normals.

Technical Art

It's a good practice to put it into a scene, set the volume to cover the space or check the unbound extent of it, and set the Lens - Exposure - Min and Max Brightness values both closer to 1 to get the default 'eye adaptation' feature of Unreal Engine to not interfere with lighting and material work. As you seem artistically capable, I'd suggest also experimenting with Color Look-Up Table feature, and maybe screen space AO settings as well to give the scene a bit more polish. Using Master material as well, and using material instances of it for rest of the materials is also a good practice. I personally use one for landscape master, one as prop master material and one as architecture master.

In terms of revisiting the subject matter, I'd like to also advice taking scales of different objects into account, as the walls for example feel a bit big with the default First-Person Blueprint. Some pivots are placed in a bit interesting fashion as well, which I'd like to see a mention of in the documents if done so for a purpose.

Where the props might need a bit more polycount or baking for extra interest, the wall pieces have quite a big of a polycount to go with them.

I like the pipe glow you did, and the light profile is something I should pick up myself, too.

Documentation

Documentation shows interest towards subject, shows different aspects of development and has nice closeup shots and reasoning to go.

Final Presentation

Final presentation consists of high resolution screenshots that have a nice level of intrigue to them.

(1 edit)

Research and Development

There's nice research, and a trip to the place. Judging by the image, there are lots of screenshots that show nice detail of scene, mood, objects and materials.

Creative Art

The scene is a beautiful, stylized market which is buzzing with life, supported with lots of vibrant colors. Feel of location is supported with nice detail such as volumetrics and shadow colors, adding to the gorgeous stylized feel. The scene is simply stunning.

There are mostly smaller things you might do better after this point, like materials might have a bit too dark values thanks to object's ambient occlusion and don't fit that well with the rest thanks to it, or doorways or the wooden balcony bits could have further structural correctness that would play together with UV mapping, roof planks could be pushed with geometry to add an extra layer of interest, some objects are clipping with each other.. Minor polishwork, really, as the scene looks very nice and has well-made and all round supported structure to it. Layout wise everything feels slightly too close to you to maybe give a bit of a claustrophobic feeling, but it might as well be because there are so many things to look at, with no clear spaces for your eyes to rest.

Technical Art

Post-process, lighting, atmospheric fog, bubble and water give a lot to the scene. The moving clothing is also nice with simple grass wind effect. Material instances for objects also add nice variation to play around with.

The landscape and marketplace floor has a huge tesselation factor going on without adding that much to the end result. Flat surfaces, such as on marketstand_covered mesh, also have lots of unnecessary polygons. I'd like to see pushing those materials into a single prop master material, of which all could be instances of.

Documentation

The work is well presented and goes through research to high detail. Images and texts are easy to read and well polished to make for a great learning experience.

Final Presentation

Images are of high quality, and showcase area as a whole. I'd like to see images that showcase a certain aspect of the scene maybe next time, like prop cluster or similar, as these images feel a bit funky, as the center of attention, or what you'd like to probably showcase, wanders a bit off the images.

(1 edit)

Research & Development

There's good amount of reference and inspirational images. To help with taking the scene to the next level, I'd suggest having some more high quality references that showcase more intricate detail of objects and material work, as well as how structures are built in similar places.

Creative Art

The scene reads as a nice dungeon-crawling endspace with dark fantasy-esque feeling. There's a nice usage of decorative runes and ornaments. The lighting showcases interesting areas, but feels a bit dark at times.

Technical Art

There's some nice vertex blending going on at the surfaces to introduce the moss to the models. Emissive mushrooms also add an intriguing flare to the scene.

The polygon usage is a bit on the lower end for rounded surfaces. I'd suggest diggin a bit deeper into texel density (512 is common for 3rd person games, 1024 for 1st person), and taking it from there to UV mapping and using tiling textures and shared texture space. Having a look at some wireframes from this generation console titles might prove useful as well; https://www.artstation.com/artwork/r0466  https://www.artstation.com/edgar_a_martinez

Documentation

The scene development is broken down well in text and image format.

Final Presentation

Images are of high quality and showcase well the subject matter.

Research & Development

Research seems to be based a bit too much around concept art instead of real life references. From what I know, concept art is mostly based on real world and built on top of that. The documentation is also a bit short on different prop and material research. I'd suggest doing a bit more research and searching for high quality reference to help with pushing the feeling  a bit more towards real-life and realism.

Creative Art

The exterior of the house looks and reads beautifuly in the blue an orange lighting. Interior is lit well to showcase the warm inn and a cosy fireplace.

To help push the scene a bit further, I would have a couple of things to note. The house looks great and interesting, but is a bit grounded in the realm of 3D with the silhouette reading a bit too much like straight shapes. On the interior, it would be nice to see a bit more of planks being modeled as 3 dimensional objects maybe, and having some material blending. The orange light on the exterior, as nice as it looks and reads, seems to be coming a bit out of nowhere, especially with the color that the scene is showcasing.

Having made the call of doing both interior and exterior in there might have been also maybe a bit of a not-so-good call; I believe the quality could have been pushed a bit further had you cut down either interior or exterior of the scene and concentrated on making higher quality content for just one or the other.

Technical Art

Use of blueprint for the candles seems clever. There's nice usage of tileable and unique materials.

Some rounded objects are a bit on the low end to get a closer look at them, and some of the normal maps might have a flipped channel (for example on the wooden floor) Also the scales of tables, chairs, doors and windows leave a bit of an unsettling and a toy-world like feeling. Also, please take away the extra tesselation out of the exterior modules, it's mostly used for landscape/terrain; the subdivisions you're showcasing is already good for vertex painting, and then you can take the modules further by giving it some love through modeling some indents for breakage and such, as was used also for Uncharted 4; https://www.artstation.com/artwork/r0466

Documentation

The document shows development steps throughtout the development with great, short descriptive texts and illustrative images. It was very enjoyable to read through, and gave a great sense of knowledge.

Final Presentation

Images are of high quality and showcase the results in a clear manner while not trying to hide anything. Kudos! :)

Research & Development

As the documentation is missing, it's a bit hard to feedback this area.

Creative Art

The thumbnails look great and interesting, the scene shows interesting lighting and use of neon color adds a punch and unique feel to the scene. Architectural modules look interesting. 

Having pallets, barrels and such in an interior scene that looks otherwise very polished and to be very clean feels a bit weird. Also the fire in the barrel is a bit off, as the place seems to be also running on electricity. A bit more polish into the material work and props might help with pushing the scene also to the next level, and adding irregularities to silhouettes, as well as material blends. The pillar ornaments feel a bit intriguing as well with shifted, diagonally cut pieces.

Here's how it reads to me; there was this fancy place which was taken over by Rage 2's bandits in the post-apocalyptic setting (the pink gives me the connection to that game). There's a lot of room to support this to make it feel so as well; chandelier could be fallen, ceiling could be broken and given dust layer to the ground, the floor might have had a carpet that's been torn from it, or just thrown to the side to add a bit more of color etc. In my opinion, there's a huge potential, and my creative juices start flowing from the images already of how to make it more like so.

Technical Art

Judging by your artstation link below, you use quite a lot of unique texture space. It's a good practice to use tileable materials and trims, and reuse same texture space as much as possible while keeping the interest in mind. When making unique texturing, it's good to make sure the texture is really unique from what could be replaced with tileable materials that could be used on many surfaces, or if bakes cut down on polygon usage by a big deal (eg. things with nails, screws, etc.).

Documentation

The documentation and scene files are a bit M.I.A., but I can see some interesting things here.

Final Presentation

Screenshots are of high quality and showcase the delivered  scene and level of quality so well that it makes giving feedback rather easy.

Research & Development

Research seems to have included western bars in sense of Red Dead Redemption, but the documentation seems to be a bit missing, so I can only make assumptions.

The research should comprise of real world reference, images that would fit the mood you want to achieve, lighting and atmospheric references. In terms of props, there could be a quality bar image, real world reference of the object and closeup shots of the materials that showcase intricate detail such as detail, wood grains, directionality of wood, possible wear and tear etc. Story elements and how for example a place looks after a fight could also be added to help with the creative part of the area.

Creative Art

Scene shows promise. I get a feeling of dusty, off the record gambling place with shady personnel playing there. Chair and other elements showcase nice modeling. The constructed ceiling support elements shows an interest towards real world building.

The scene could be pushed to the next level with modeling planks of the floor, adding irregularities to meshes to further get rid of the 3D feel, and adding good looking material blends. Also, the gun is actually very impressive, and could be lit better as the center piece, or maybe placed on the table as the price for the poker game, or someone felt a bit aggressive after losing a game and...and... The story bit could be maybe taken to the next level a bit, along with some extra material separation, as the scene reads as mostly wood right now.

Technical Art

The project file includes meshes which look nice. Face-weighted normals look nice on meshes, and most of the round surfaces look great with maybe the time running out on the SM_Door mesh' doors that look a bit whiteboxy. Modeling is definitely your strong suit. UVs, if done with unique texturing, could be further pushed to minimize the so-called empty, or 'wasted' texture space.

Documentation

The documentation seems to be M.I.A. :)

Final Presentation

Shots are of high quality and showcase the feeling and mood of the scene well. Darker areas are hiding a bit what's going on in those parts.

Research & Development

There's historical knowledge infused with something original. Both sides are well shown and spoken out about. Steps taken and thought processes are well captured in the texts.

Creative Art

The scene shows interest as an Egyptian tomb. The creative part of torturing devices and screaming man fall a bit short without a lot of other elements to support it. Lighting is beautifully crafted, the main attraction reads and shows well and the Color LUT settings add a lot to the final scenic feeling. To generate a bit more interest, you could push the silhouettes a bit with breakage and irregularities with geometry. The ominous aspect could also benefit from some wear and tear, as well as material blending. The paintings are maybe a bit stretched; given more time, they could be utilized to fit with the ominous feeling and story as well.

Technical Art

Color LUT and post process work great, lighting's beautifully crafted but could maybe be a bit more optimized with the implementation of spotlights instead of point lights. Use of decals for paintings works great. Polycounts seem pretty much spot on, and LODs work well for the pillar. There's a good use of tileable texture in the ground. The step could also use maybe tileable materials with vertex painted material blending. Material instancing would also be nice to see in terms of optimization.

Documentation

Steps taken and thought processes are well captured in the texts. Text is maybe a bit chatty, and a bit elongated as such, but gives a sense of character. Images show process well throughout the process.

Final Presentation

Screenshots are of High Quality and easy to look at. The compositions work very well.

Research & Development

There are nice high quality images of areas and props, as well as lighting and atmospheric feeling of the scene.

Creative Art

There's contrast and color in lighting of these, and the vending machine shines from the distance to snatch the focus.

Technical Art

I would suggest doing a bit more modeling first, get the shapes of objects to read as source materials and improve on ways to look at references. After that, start UVing your objects; test out a bit how the texture looks and have fun. Next step is to dig a bit deeper with the concept of texel density to plan out the scales of your UVs, and making high quality texturing work. Tileable textures, trim sheet and material blends come next. :) You already seem to have the basic post processing volume setting placed well, nice one!

Documentation

Texts are well written, especially in the moodboard section, and the thought process in decision making is captured well.

Final Presentation

The image is nice and crisp and showcases the work well.

Research & Development

References include screenshots and a concept from one game with a couple of real world references. Captions show well the interest toward level design and using environments to guide the player and give the place a certain feeling. There's a bit of lack on the references and concepts of props.

Creative Art

The scene feels warm and soft with dust, particles and bright light shining through building cracks along with wooden pillars, bookshelves and books taking the main stage. Use of tileables and trims make the scene look very cohesive. Books and props add nice smaller additions of color into well working color palette.

The end result rather close to the main concept so that it makes you compare those directly to each other, which makes me feel like the scene falls a bit short of the main concept, but shows promise and has many things that have been done well. Nature having taken the place over is one thing, but also the showing aspect of the organization's soldiers having been to the place are not super well present. Wood is a great element, but it contrasting so well with metal in the concept, I'd like to see a bit more of metallic items to the scene to strike the interesting balance a bit more.

All in all, it's a great scene that could use a bit more of polish and attention to detail for props and materials.

Technical Art

Masking approach for books works well in the scene. Made materials also showcase a great interest towards the subject and easy ways to develop scenes in rapid working environments.

Scales of things work pretty well in the scene, and polycounts are within game-ready budgets. Some things feel a bit out inaccurate with scale, such as the MLTs, and it might a have a bit high polycount. For bigger props, I'd like to see a bit more variation in the silhouettes with some well thought out geometry cuts. Pillars and shelves could maybe also use a few variations with different levels of breakage.

Documentation

The techniques and props are documented  very well along with material graphs. There are nice captions within images and the text is nice and easy to read.

Final Presentation

Renders are nice and sharp and showcase well the subject matter. Lighting works and thumbnails look great.

Research & Development

There are nice references of environments with proppings. Lighting, color and elements vary a bit. Pushing the research of real-world locations a bit further could  be beneficial in the future, as well as adding more detailed shots of things such as props and materials you want to make for your scene.

Creative Art

The scene feels ominous and dangerous with dominant red lights. Props work well with the location. The scene could be pushed to the next level with researching face-weighted normals, and using maybe more roundedness to the tunnel walls. Material work could also maybe be worth to put more time to, as well as looking into tiling textures and vertex blending to achieve better texel density. Props could also benefit from thoughtful irregularities achieved by adding more geometry to objects.

Technical Art

Materials are done in instanced fashion which is good. Lighting is done with spotlights and the scene showcases use of decals. There's also an unbound post process with also the breakdown about experimenting with it with lighting shows good knowledge about what polish work needs.

Some things, such as cables and ground are abit on the high end on polycounts, but mostly the scene is a bit low on polycount to today's standard.

Documentation

Text is easy-to-read and showcases good progress throughout the development.

Final Presentation

Post process and lighting show the scene well. Screenshots are of high quality.

Research & Development

Reference shows lots of cool scifi scenes from movies and games as a reference. The research into subject matter is a bit on the loose side, and the title making it a repair ship shows mostly in not-too-aggressive shape language and color palette usually associated with construction vehicles. From what I know, most of the design work also for movies and other games comes originally from real-world in terms of functionality and then gets a shape language of something else implemented into it.

Creative Art

The scene is packed with lots of detail, there are many things to look at, lighting shows many aspects of the scene. Assets look interesting and the design of space looks intriguing. Functionality and reason for the detail is a bit on the loose side. Ratio between smart materials works, and the color palette gives a cohesive impression

The scene works well, but is definitely asking for a bit more polish and diving deeper into materials and bakes and how they work in the engine. When it comes to polish, you might want to also revisit scales of things and make sure to have the nice ratios, that you have going on with palettes and materials, going on with also those and in population of that detail.

Technical Art

Polycounts are mostly on par with game art. Some rounded surfaces might have a bit too much roundedness. The delivered content shows a lot of Materials without Material Instance workflow, which affects shader compilation times. Most of the materials also seem to be using unique textures; it's okay to use only a couple of tiling textures and trims along with some unique textures for props.

Documentation

The documentation showcases creation of assets, steps to produce the final work and well written thought process and reasoning behind calls in the scene. It's worth maybe noting that it would be a bit easier to read through all of it if it was packed into a single document. The delivered project file also seems to be maybe missing the level file, from what I can tell, for closer look, and objects might have a hundredfold scale to them but they are cohesive and consistent of size.

Final Presentation

Renders are of high quality, the compositions work and everything reads well. Nothing is being tried to be hidden from the viewer, and it's easy to read what's going on.

Research & Development

There are scenic references with elements you want to use in the scene. Separate objects have been broken down to their own research in an easy-to-read manner. The final scene steers a bit away from how the reference board on the first page looks; and a lighting reference board might be of good use.

Creative Art

The silhouettes of trees look interesting and you can make out a path as a player to follow the lights. The scene is a bit dark, which can be fixed quite fast with for example a use of skylight, and a few tricks posted in the technical art section.

Technical Art

I'd suggest you start making decisions on asset making based on texel density. Reusing the texture space is a common policy to hit texel densities such as 512 or 1024. That should help you define how and where to use unique textures and what surfaces could use tileable and trim textures.

The first thing you might want to do in your Unreal scene is to add a post process volume, with infinite extent (Unbound) checked.

Next, to not get the 'eye adjustments' in the scene, you generally want to tone the exposure values closer together, so you won't be faced with pitch black or overexposed images, as is pretty common with the default engine values. It can be found in the Post Process Volume - Lens - Exposure Min Brightness and Max Brightness.

From there, you can start working on your general lighting, values and other things.

Documentation

Text is well written, and breaks down separate asset research into different settings. Showcasing separate UVs of objects is a cool thing to do.  It would be nice if you had included also texture and prop renders next to the UVs to get a better idea of the whole. Looking at good, detailed references can help you push the material definitions and silhouettes of objects for some spectacular results. Use of 3rd party props should be probably mentioned in the documentation.

Final Presentation

Renders are of high quality. The scene is a bit dark, to maybe such extent that it's a bit hard to read how the objects, scales and such work together.

(1 edit)

Research & Development

Research is pretty closely

Creative Art

The scene is filled with magic - the scene reads, glyphs feel very interesting, effects are spot on. Metals contrast well with shiny tiles and wall materials.  To push the feeling a bit closer to God Of War's temple, you could introduce some organic elements (something to match the tree in GoW) and ornate trims to generate more interest to the floors, pillars etc. However, the results look absolutely stunning for short period of time frame scenes. To push the feeling a bit more, you could also add some mindful decay, wear and tear into some of the silhouettes, and maybe some extra decals and material blends. This could also give a bit more reason to show more props and how the place has been more shaped and used by its inhabitants. The sand piles might work also if there was maybe an opacity mask running along object's UVs, or maybe with use of Render Target to add interestin some sort of way (as food for thought)


Texture wise, the breakage of tiles is a bit extensive to me, and the baseColor textures could maybe use a bit of work on top of slightly notable gradient mapping effect. They work very well in the scene though.

Technical Art

There's good usage of tiling textures, trims, decals, lighting, blueprints, particles as well as custom shader effects. World-positioned sand material is also a nice show of technical aspects. The ceiling could maybe be a bit more optimized along with the portal surrounding rocks. Everything looks and feels well stylized and dream like with wisps and trails.

Documentation

Documentation is easy-to-read and images convey the steps taken to produce the end results. References and models used are shown very well.

Final Presentation

The center of attention is clear, images are of high quality and showcase effects well.

Research & Development

I liked a lot the images of boats and how the research developed further down the line to fit with the end results. Researching also lots of different series and movies for reference was also cool to see, and could be something I could pick up as well.

Creative Art

The shape of scene and the sense of space work very well and the lighting looks nice. I also like the bits of story telling with the fallen props and decay. The scene could use a bit more clusters of props and higher texel density.

Technical Art

Greybox stage looks interesting and nice.

It was nice to see you dividing bigger boat asset into separate materials. The polycounts can be a bit on the high end on rounded surfaces but they look rather good.

Breaking the scene into architecure modules, based on metrics, and assets/props, is a rather common policy in the industry. Reuse of texture space is a good policy as well to hit 512 or 1024 texel density. Tileable textures with material blending and decals are also a nice way to add interest into larger wall modules. After that comes adding irregularities in a sensible scale based on good reference images.

Documentation

Process is well written and research and its results are easy to grasp on. I also like links to databases and references to similar TV shows and movies make me want to go watch those. Also the top and bottom bars add an extra layer of movie enthusiast/writer feel to the documentation. :)

Final Presentation

Small thumbnails look nice and inviting. Renders are of high resolution, and you can clearly make out what you see in there.

Research & Development

The Research shows cool ideas and references about stained glass and props. Mood board is a bit interesting, as the lighting, colors and stained glass vary slightly from image to image.

Creative Art

The focus reads very well, the contrasting cold and warm lights add to the divine feeling. The milky glass is an interesting idea, but a bit tricky to pull off. As cool as the shapes in the glass are, it casting shadows impacts the readability a bit with the floor patterns, I think.

Technical Art

The polycounts are good. Candle holder might do with a bit fewer polygons as well as the pillar the book is sitting on.  The book being the main object has nice roundedness for the pages caught by the wind and it looks nice even if the pages are rendered in one-sided manner.

The use of tiling textures and other shared texture space usage might give you a better texel density, and as such improve the overall quality of the scene.

Documentation

I like the sketched props and taking the design from there. The progress is showcased well with the images and the text is easy to read.

Final Presentation

The scene reads well and the screenshots are of high quality.

Research & Development

There are cool images of old worn wooden structures and smaller props to fill out the scene. Bigger props with which you could easily populate the place quite fast, are a bit missing here.

There could also maybe be a bit too more research into how things are built. Concrete slabs as roof don't seem very accurate of a material to be put on top of wooden walls. The support structure also strikes out as a bit odd with pentagon in the middle, which could of course be argued to be a funny design choice.

Creative Art

Texture work looks great on tileable surfaces and clothing on most parts. The lighting is done in way to give nice contrast with lighter and darker spots.

To push the scene a bit further, I might suggest adding some sort of window and support  the structure a bit more. Structural interest is a bit missing with, for example, the roof frame sitting on top of clipping wood beams, and the meshes could use with some extra carefully placed irregularities to add mesh interest to papers, structures and reused props. Getting detail like wood grains to match real world directions would also add quite  a lot to the feeling real world objects.

Technical Art

Some of your materials include macro texture extras, which shows interest towards the subject.

Walls and floors look optimized. Objects are a bit on the higher end on polycount, and maybe tells a bit that the center of attention should be on the weaponry and clothing. 

Documentation

The documentation shows references and has peaks into the development of the scene. It's a good practice to block out more of the scene in one go and then dig deeper into working on the detail and props. Some basic materials might help with defining the looks.

Final Presentation

Presentation shows interesting angles of the space. Screenshots have sharp edges and could maybe showcase a bit more of the details put into textures and objects.

Research & Development

The visual interest shows well interesting buildings and elements of swamps. Style guide shows several different styles.

Creative Art

The color and stylized approach work well in the image. Material definitions could maybe be pushed a bit further. The turntable looks great with fireflies in engine.

Technical Art

The scene is mostly on gameart level polycount wise. There are a few things I'd like to mention, though; Seeing roof tiles in different scales on different roofs also feels a bit odd, along with the house feeling a bit like it's been built with rock, the round tree meshes are a bit on the high end on polycount in contrast to the wooden pier the house is founded on, which is a bigger mesh chunk. Lilypads could be maybe almost flat planes with applied alpha texture.

As you mentioned about the UVs, you can get pretty great results with automated UV packing in many of the modeling programs. It's mostly about knowing what texel density you'd like to hit, and working from there it becomes more of a problem solving type of thing!

Using tileable texture means having a texture, where if you place the same texture right next to it, it would be a seamless transition between images. Using such texture with a bigger surface calls for scaling the UVs of the mesh in your modeling package.

Use of engine's linear gradient is something I probably wouldn't have thought about, which is super nice, and actually produces a cool effect. For your next scene, you could try looking for similar gradient feature in Painter to save a bit of shader instructions by baking it into the texture. :)

Documentation

Documentation showcases well different parts of production. The text is easy to read, and filled with pretty pictures.

Final Presentation

The presentation showcases different aspects of the scene well. The images look nice and crisp. Water, lillies and fireflies are a nice addition in the scene.

Research & Development

There are cool moody concepts in the subject matter, and the photos from a trip are a great addition.

Creative Art

The main focal point sticks out very well as a dark silhouette rising in contrast to the lighter gray sky.

Technical Art

It's nice to see there's not a lot of wasted texture space.

The light could get a bit of intensity, make it look for example like a moonlight lighting. Putting also some subtle spotlights to draw attention to key assets would also help to draw attention to details.

I'd recommend looking a bit into tileable textures and trims in the future. Getting certain texel density was the key for me to start thinking more about using tileable, trim and unique textures. For 3rd person games, it's most common to target for 512, while 1024 is more common for 1st person perspective games.

Documentation

The documentation covers steps taken to produce the results and showcases progress very nicely.

Final Presentation

Squinting at the images, you can clearly make out the main attraction. Screenshots are of high quality which makes it easy to read what's going on.

(1 edit)

Research & Development

Reference showcases nice and organized way of reference gathering and planning.

Creative Art

The end result is a stunning, simplified scene with lots of character. The use of color, contrast, shapes and putting detail to where it counts is remarkable work. The result reminds me of Dishonored/Prey. Details of materials coming mostly from roughness variation, and not pushing a whole of detail into the normals other than basic shape and form makes it easy to read. The video with sounds on adds a lot into the horror feeling, which is not all the way present to such extent with the easy reading scene.

The end result looks great, but I'd love to see maybe how the scene would look with a bit of carefully placed grunge into the normals in form of decal or vertex paint to see if it would add another layer of horror into the scene, as it now reads more of an easy-to-look-at scene with art fundamentals being used to a very high extent. Adding a bit of interest into the silhouettes by adding irregularities might also work, but we might be steering the scene towards a different style if these sort of changes would be implemented.

From narrative perspective, you could maybe add some elements to support the feeling. For example, you could have a torn children's book with a title such as "My favorite chair" to make a connection between the teddy's owner and the chair to give a slight added scare factor, with planks creaking behind, in the corridor.

Technical Art

The scene is very optimized with texture usage. The use of polygons seems a bit high with smaller details such as door knobs and wallplugs. Use of decals might also be a bit high if the scene is not supposed to have a key moment or meaning.

I like the use of special effect shaders; they pack a serious punch to the scene.

Documentation

I like the showing of organized references, Notion and using cinematic scene as an inspiration for lighting. The breakdown of materials, sims, props, lighting and post process are also shown and written well.

Final Presentation

The screenshots look nice, and the connection of doll and tv, after seeing the clip, becomes more apparent. The screenshots are nice and sharp.

Not at all, please do if you feel like I can be of help!

Research & Development

The research showcases different assets to be used on the scene. Exploration of narrative possibilities, lighting and material closeups might call for an extra layer of research.

Creative Art

The final image looks easy to read, with large surfaces and mostly subtle variations. The bakes have also very nice soft, round edges. The focal point is a bit on the storefront, but also the vending machine with lights on, that the shadow's line are leading towards.

The surfaces could use a bit larger polycount to make use of vertex blending and shadows could be made a bit lighter to sell the feeling of open air. I might also suggest looking at thicknesses of objects, as everything has that extra cartoony and rounded feeling with exception to architecture. Most of the props also feel more exaggerated on widths rather than height, which adds to the stylized feeling. Adding polygons to add interest to silhouettes would also be considered a plus, as most things feel a bit thin of irregularities, which gives a bit of a 3d feel to the image.

Technical Art

The modular pieces are well optimized. Some of the objects could be pushed a bit with on polycounts; planks on benches could be made with geometry and the plastic containers could have a bit more shape to them.

The lighting settings and post process showcase you put time and effort into adding an extra layer of polish into the scene. From what I've been able to tell, those two are the things that make scenes feel the most connected; as in, objects belong into the environment along with architecture. Were you missing skylight from your scene? It's an easy way to add brightness to the shadowed areas.

The rounded edges look nice with unique bakes. I'd suggest baking with face-weighted normals to get rid of gradients as they can mess up LODs of your objects. You might be more prone to tangent basis errors as well. You did the low polies with Max smoothing groups, right?

Documentation

Images showcase the scene and props very well. I would have liked more progression shots with lighting, whiteboxing and greyboxing stages.

Final Presentation

Screenshots are of high quality and showcase different aspects of the scene.

I wish you the best, and hope you could make out some parts what I was trying to convey through my text. If there's something that you'd like to ask about, you can also reach out for me through other medias such as ArtStation or Twitter.

Cheers!

-Teppo

It's great if I can be of help. If you want to adjust the changes or would like to know more, you can also DM me through other medias such as ArtStation or Twitter.

Cheers!

-Teppo

I'm glad if I can be of help. If you want to adjust the changes or would like to know more, don't hesitate to reach out for me through other medias such as ArtStation or Twitter.

Cheers!

-Teppo

Research & Development

There are lots of beautiful images diving into the subject matter, with lots of high quality reference images.

Creative Art

The scene captures the Christmas Market feeling well. Contast, color and composition work well together. In today's workflows, you can basically model all the visible structures and how real world objects are put together.

Technical Art

To get an extraordinary leap into next level, I'd suggest couple of terms you could research a bit more into; texel density and tileable textures and their usage through scaling the UVs of the mesh. To break too obvious tiling and to create interest, we also can add material blends, that's maybe for the future. In the industry we use a lot of the same texture space with our UVs to try to optimize the memory usage.

Branches of Bell and Christmas Decoration might benefit from using the same material with opacity mask as within the tree.

The snowflakes are a very nice addition.

Documentation

Text is well written, and you can make out the progression from the document.  Used software is also listed nicely with assets, and the wireframes and textures are showcased in a very clear manner.

Final Presentation

The screenshots are nice and sharp, the compositions work and the results are showcased well. The detail and passion for the theme shines bright like a diamond.

Research & Development

The research into the subject matter could be a bit more thorough. I see a couple of reference images without a separate mood/reference board or research into how the things were builtand formed.

Creative Art

The use of blue and purplish red along with cyan works wonders on the scene. The scene feels energetic and colorful.

Pool, lanterns and the fire pits add a lot into the space. The weapons also look remarkably well done.

To push the scene further, you could divide the floor into difference spaces by using different decorative trims and materials. The sides could be composed of different materials, for example. The meshes could also be pushed a bit further with added geometry and vertex blends. It might look a bit better if the metallic trims would also be a bit less shiny and new as well. The shape of trims seems a bit contrasty compared to the pillar.

Technical Art

Considering you started 4 months ago, it looks very good.

Reuse of texture space and use of tiling trims and tileable materials is something we use a lot in the industry. Firepit mesh showcases a few polygons that don't seem to contribute to the final look and feel of the asset. You can use vertices for material blends, to control shading or give shape or form to the object, but on a flat surface without material blends, it's a bit of a waste to render those polygons individually by our graphics cards. Some of the rounded surfaces you have might have a bit too many polygon however, but they do look nice and round. :) The sense of scale might also come a bit more across objects we see in our everyday life would be of correct sizes such as door.

One key to start thinking about possible use of tiling materials and trims is to be able to hit the correct texel density for a scene. 512 is common in today's games for 3rd person games, 1024 for 1st person titles. Hitting those calls for some thought process. It's common for games to also make use of vertex blending and decals.

Documentation

Texts tell about the progress in a very nice manner, and the images are placed in an informative manner next to the subject matter.

Final Presentation

The final presentation looks good. The images show the result well and accurately and are of high quality. The composition works and the focal point is obvious. 

I would have liked to see more closeup shots of the weapons you put into the scene, I only more closely noticed them when opening the scene myself.

Research & Development

The research is a bit subtle there for how Roller Coaster Stations look like. I'd like to see a bit more research into why there are structures involved, such as the metallic support, that you are using also as a truss system for lighting. In some of your reference, it's used as a support to hold up the concrete slabs or corrugated metal plates of the ceiling, but seeing as how small your station is, it might be a bit of an exaggeration like this.

I'm also missing a bit of the narrative comearound here. A themed Roller Coaster Ride is a cool idea, and you could introduce some very intriguing ideas into the space with cutouts, flyers and using decoration and architectural materials of similar kind.

Creative Art

The space reads as a Roller Coaster Station. The lighting works well, and thumbnails look good.

The thickness of rails is one thing that I might suggest on making thinner, they feel a bit thick right now. The scene could also do with places for the eyes to rest; right now there's either detail from materials to interfere with rest or interesting silhouettes and color contrast of train carts. making the end result feel a bit like everything is fighting for attention.

On a more general level, to take the scene to the next level, I'd suggest you add carefully more interest into the silhouettes of objects with geometry, like you can see in titles such as Witcher and Red Dead Redemption, for example.

Technical Art

Texel density seems rather coherent and consistent. The scale of things feels a bit off, the door is a bit too small when playing the level and the relative size of fencing and cart seem a bit too big. The lamps look good  with geometry, but they are a bit small to have that many polygons, same applies to carts in contrast to the walls being simple planes without subdivisions to make use of vertex blend material. You might profit also from looking into Material instancing, as now all the materials are really separate materials. Control panel also has unique textures for very many separate buttons, and could benefit from reusing the same texture space.

Documentation

The written document is easy to read and follow. It showcases progress of props and beginning the layout with a sketch. Shader breakdown for the Parallax mapping is a nice thing. I would have liked to see a bit more of the prop and module progress.

Final Presentation

Contrast and lighting is there. Metals contrast with painted carts as well. The renders look sharp and showcase the results well.

Research & Development

There are lots of images to showcase for the visual research. I'm missing a bit of the structure research.

Creative Art

I like the use of colors a lot, they bring very magical and fantastic feeling. The lighting only adds to that, with warm shadows cast from warm light. The quality looks very nice.

Technical Art

Polycounts are a bit on the high side for such scene; I believe the scene could look almost the same with halved polycounts. You could also consider using trims, tileable materials and reusing the same texture space for different pillars and bridge planks to get the most out of your textures space, even if the end results do look spectacular even as they are now.

The lighting looks very nice.

Documentation

There are lots of nice reference images. The process is well written and shown with texts and images. I like the show of lineart approach.

Final Presentation

The diorama shots read very well and showcase the use of color very well. The renders are high resolution and show the subject matter in a nice fashion.