As this has been a group project it has taken a lot of co-ordinated effort to pull it all together and I’m proud with the result that we’ve managed to achieve.
Fluid and flexible teamwork have been pretty essential to this project, and I’m pretty happy to say that it’s largely been a success. There have, of course, been very minor problems with miscommunication, but they have had no bearing on the overall project and are minor enough so as not to warrant mention.
Communication – we have a Facebook group chat which is useful for text updates and sending screenshots that we have used every day throughout the project, and we have also have a Google Hangouts where we have spent many of our evenings working together. The screen-share option in Hangouts is very useful for obvious reasons, anyone with a technical question can be helped very easily, and it also enabled me to view Joe and Lee’s live execution of my concepts, which I could comment upon.
A note on team dynamics – Lee has fallen into a role of being project co-ordinator, which is natural, as the project was his idea to begin with. His management style has adapted to our group dynamic – he is fairly hands off for two reasons – firstly he trusts me and Joe to do our jobs, and also he has his own workflow to manage (as well as working freelance himself).
My direct involvement with Lee has been general thoughts and discussions about world-building and art style, as well as working on the Elf character, which went past the concept stage as I then went on to assist with the texture and alpha creation for the armour.
Outside members – Lee is in a better position to comment on the roles of Jan Kaluza (Technical Artist), Matt Jenkins (Character Rigging), Adelaide Coldham (Animation) and Sanna Kempe (Assisting Game Artist), although I have worked directly with Sanna on game assets where I sculpted the high-poly versions and she got them down to being game ready.
Honestly it’s been delightful to work with everyone – all the team members have been highly professional, and most importantly, very enthusiastic about their work.
I cannot comment on professional dynamics within a game studio, as my own personal experience has been from a freelance perspective, but I’d imagine that, as with other types of businesses, the management styles, corporate structure and production pipeline all vary pretty widely dependant on studio size, the type of game being worked on, and general scope of the projects.
It seems that the following image by artist Jack Stroud has attracted a lot of attention recently on social media:
Culturally speaking, there seems to be a lot of ‘demonisation’ around using references in illustrative work. I think it comes down to people not really understanding how the process works, and seeing it as more akin to ‘tracing’ than anything else.
I personally feel that it is true that you get to a stage with drawing skills where studying a photo it looks so close to if you had actually traced it (I know for myself that is the case with certain subjects, less so with portrait faces).
I think it’s key to consider that as long as references are used correctly (and that is another subject of debate) i.e. they are used as an element to inform a larger piece.
I’ll refer to a blog post I made back in January where I discussed my process for creating a similar image. These big cat studies were done from photo references as structural sketches to understand the variety of feline forms:
After this informed study session I was able to utilise these references as I had built my mental ‘visual library’ and could then create the following hybrid design that doesn’t conform to any existing species – but looks like it could.
Again, these are older images by me now, but I feel they address the study/reference/memory debate quite well.
Stroud, J. (2017). I’m here to tell you the truth about referencing in art! IT ISN’T BAD. YOU CAN’T LEARN WITHOUT STUDYING.. [image] Available at: https://twitter.com/PetrichorCrown/status/895386350605799428 [Accessed 15 Aug. 2017].
So what exactly is the forefront of Concept Art?
There is a very high professional standard set for artwork in this industry (not just concept). Breaking in and getting a job you have to be better than, or equally as good as someone already in employment, and you are competing with people who do this day in and day out.
Throughout the course I’ve taken this as an opportunity off full time job in Graphic Design to hone skills and in doing so I have learnt a variety of new techniques and programmes. Ultimately my job is design and ideas, so I have been working on improving presentational skills to showcase my ideas as clearly as I can.
I would argue that ‘The Forefront’ of this is basically the front page of ArtStation:
And to compare that to my own portfolio on the website:
As a self reflective comment, it’s not hard to look between the two and honestly and frankly say that my work is lacking in quality in terms of the top industry professionals. This is no surprise to me, and is absolutely fine. It’s a long road, and as much as I want to be up there on the top, the reality is that there’s a lot of work separating me from there.
Without wanting to be too depressing, the consistent improvement that I’ve made in the past (that is pretty evident when scrolling back through my Instagram, where I keep a daily post and have done for the past 502 days, at the time of writing.)
So who would I consider to be at the forefront?
Artists tend to rise to popularity via social media and online exposure to their work. The large bulk of concept art is never really shown to the public for good reason (studio secrecy, NDAs etc), and so we are exposed to the most prolific artists with an online presence because of the nature of the internet. So there are many top tier industry professionals and art directors that don’t need an online presence because they are already very well known within their own studios or respective fields. They already get consistent work on the back of years of experience and so do not need to worry about self-promotion.
Obviously each artist has their own speciality – so this ‘forefront’ would depend on the field of concept art (games, film etc. and even deeper into that – creatures, characters, environment, props etc.)
It’s slightly unreliable to say that it’s just the front page of ArtStation, by its very nature (user generated likes) is going to be slightly generic, as it showcases a ‘hive mind’ mentality of very high quality art that can have a tendency to fall into ‘tropes’ – i.e. really well rendered anime girls, toaster head mechs & zbrush alien busts. As such, the only real way of identifying concept artists who are at the forefront is by looking towards those willing to share their techniques. Websites such as patreon, gumroad, twitch and youtube have played an important part in education about concept art, as well as online courses such as Learn Squared. Artists such as Anthony Jones and Jama Jurabaev (to name just a few) are ones who share techniques and push the forefront.
Processes and Technologies at the Forefront
Nowadays, there seems to be much more of an emphasis of 3D in workflow, tools such as ZBrush, 3DCoat, Modo, Keyshot used to create images, and photoshop used to overpaint and present. I believe that this is, in part, due to a shift towards more powerful computational power available to average person, which has brought not just 3D packages, but also rendering ones, into a wider commercial setting.
Ultimately these are still just tools though, and what should always take precedence within my field is good, clear, well presented design – and that can be done with pencils, markers and copier paper.
Current Debates and Ideas Affecting the Forefront
One big issue that I feel the industry can have is with creating original designs. I can understand why this is an issue – studios may not be too keen for their designers to really ‘push the bar’ and create something that may be considered too ‘wacky’. Sure it might be original to have a Dragon that breathes bubblegum instead of fire, but it’s a difficult idea to ‘sell’ to your audience. At the end of the day, this is the creation of a product, and products need to sell and be understood by their target audience, thus there tends to be a lot of cases where designs are ‘played safe’.
Design trends are very much prevalent, and there are a number of generic ‘tropes’ within concept art – crashed spaceships, spiky mountains etc. One obvious parody example here from the anonymous artist known as ‘Mr Concept Art’:
(Mr Concept Art, 2017)
This is an obvious parody of the designs that many artists in the community were emulating using photo-bashing techniques to create sci-fi military near future soldiers.
How the forefront you have identified relates to commercial or usual practice (think about change and the adoption of new methodologies).
I have largely already covered the commercial practices of this field, but to recap, 3D skills much more prominent in concept art pipeline – many job listings have this as an “added bonus”, or even as an “essential skill”. I believe that this just highlights the importance of ‘sharpening the axe’ through continued progression and self-investment.
ArtStation. (2017). ArtStation. [online] Available at: https://www.artstation.com/ [Accessed 15 Aug. 2017].
Mr Concept Art (2017). Toaster Heads. [image] Available at: https://www.mrconceptart.com/ [Accessed 15 Aug. 2017].
So although I’ve been using KeyShot for WIP renders throughout the whole of this project, I was getting frustrated with some of the ‘cons’ that I’ve detailed in an earlier post. The time taking for it to generate high-resolution images was becoming more irksome, and I was spending more time playing with settings than doing my own artwork.
I had watched a tutorial video about nice presentations of WIP sculpts in ZBrush that had some good information in it, particularly the ‘render passes’ option that I was unaware about. I realised that I could use the information here, and combine it with my photoshop skills to create a final concept image, it just took a few extra steps from what the video provided as a decent base to work upon.
Render passes from Zbrush:
Depth – assigned to ‘Alpha1’ under the ‘Channels’ tab in Photoshop, can be used to inform the ‘lens blur’ effect, as well as some other features.
Mask – assigned to mask out the black area so that I can change background easily.
Render – basic render of ‘Basic Material 1’.
Shadow – set to blending mode multiply at 50%, can add colour to this in PS to add richness to the shadows.
Chrome/Metal – basic renders of some different materials, in PS cranked up the levels to only show white and used them as highlight/specular layers.
Clown – really useful one for masking out each individual element quickly with wand selection tool in PS, made it easier to assign gradient masks and photo textures accurately.
Rim – assigned a new material with a rim light effect to overlay/blend in with the PS composite.
Theban – a flat colour render that just shows polypainted info. I made an alpha for the symbols and applied it to the cloth subtool in zbrush, then with the flat colour render it only shows up the black (and it conforms to the folds and form of the fabric). Then in PS I can use that info as another mask and paint on a gold material.
The final result (with some overpainting and atmospheric fog)
I feel that this was a really effective process that was fast and easy to accomplish. As I had already invested a fair amount of time into creating a nice detailed sculpt that I was happy with I wanted more control over the final presentation aspect. I’ll definitely be using this method in the future for more polished pieces.
This method of ‘render stacking’ is fairly common industry practice in concept art particularly, where ultimately it’s about the design, and so the production of a 2D end image (utilising 3D techniques) is justifiable. I’ve found some examples (one from as far back as 2010) that show although this is not a new technique, for concept art it is still very much a stable and reliable one that many, many artists at the forefront utilise.
Athayde, D. (2013). Passes for BPR Render. [image] Available at: http://www.zbrushcentral.com/attachment.php?attachmentid=348100 [Accessed 15 Aug. 2017].
Collings, N. (2010). Making Of ‘Orc Maori’. [image] Available at: https://www.3dtotal.com/tutorial/zbrush/character_creation_orc_maoari/orc_2.php [Accessed 15 Aug. 2017].
Having just watched a promotional video of the new features in KeyShot 7 there was one in particular that caught my eye
Configurator/Presentation Mode (2:23 in video)
With my own workflow incorporating a lot more 3D elements, and one of the core aspects of my job is to provide variation, the ability to create these variations quickly using PBR render qualities is going to be a huge time saver vs taking screenshots of each element and manually compositing them together to scrutinise them.
So, that’s one of the new features that I’m excited about, and can easily see me using it in my own work – for example, if I continued work on the Plague Wraith and sculpted x10 different masks and x10 different texture patterns to go on the cloak, then this tool could be utilised to quickly showcase a presentation sheet of those variations in 3D format.
This brings me on to an evaluation of Keyshot, as I’ve recently (with the last renders for the Plague Wraith) opted for the in-programme BPR render that comes with ZBrush, for reasons that I’ll detail in another post. I think I can summarise fairly bluntly in a list of pros/cons:
- Unarguably fantastic looking results (thanks to some advanced ray-tracing), picks up information from HDRi spheres so you can preview model in a set environment (e.g. if you imported an HDRi sphere from game, then it would be a more accurate result of what something would look like in game)
- Comes with a library of physically accurate pre-made materials (I’m not a technical artist, so this is a huge aid)
- Fairly intuitive and easy to learn
- Works with a wide range of 3D formats
- Time! This is a pretty big one – it takes a while to create good quality images, especially as the resolution gets higher. May affect deadlines.
- CPU intensive – rendering takes a lot of horsepower, so it’s not really suitable for laptops/working on the go. Even with an i7 5960x (8 core, 16 thread) processor I find it to be slightly too slow for my liking.
- Laziness – I noticed that my tendency to rely on the huge library of pre-made materials was limiting my options, and although I made some efforts to learn and tweak settings (see earlier post on the wraith skin translucency), I don’t quite feel the same ‘tactile’ sense of creation as when painting.
- Not game engine – despite the ‘cheat’ workaround of importing an HDRi from a game, it still doesn’t represent the same materials/qualities that you’d get from an in-engine screenshot.
There are a few other minor pros/cons that aren’t really worth mentioning, but ultimately it comes down to a suitability score, it’s down to the artist to take on board this information and learn about new tools, at which point they can make an educated decision as to what works best in their workflow.
Luxion (2017). KeyShot 7 Top 5 New Features. Available at: https://www.youtube.com/watch?v=1hnTRR7OkWw [Accessed 15 Aug. 2017].
One of the things that I really wanted to achieve with this concept is the skeleton shining through effect that I had in my original 2D concept
After much tinkering I was able to replicate the effect in KeyShot, although I still will need to adjust the hue of the skin to be more yellow.
One thing that is being annoying is that when I put the cloak on the character, this translucency is lost because it’s covered up. It may be the case that I just rely on a 2D paintover when I come to the final presentation.
I’m beginning to not like KeyShot for this, as I feel a tendency to be lazy and rely on the huge library of pre-made materials. In a way this is limiting me, as I find I’m making selections from a series of pre-made effects. Tweaking some of these settings and making my own skin material has been good, but I’m not a technical artist and I don’t quite feel the same ‘tactile’ sense of creation as when painting.
A quick overpaint (and anatomy exploded view) gets a little closer to what I want, including the darker ends of the limbs to really accentuate the sharpness of those edges, as silhouette is so important in a game when the player has to make a snap decision about something.
I think that ultimately this may convince me that KeyShot is perhaps not the best tool to render this model, although it has been very useful in WIP production, the times for renders and my over-reliance on the stock materials have not been helpful.
So with this project I’ve been assisting with the creation of in game assets – as my familiarity with ZBrush through my own personal development has vastly improved, I’m now more able and confident to flesh out my concepts in 3D. I feel that this is a pretty useful skill that I’ve developed over the course.
I don’t hugely enjoy the process beyond the initial sculpt – at least until it gets to the stage when the model can be imported into a piece of software such as Substance Painter. I have gone through the entire asset creation process a few times (in earlier modules, with the Dunkleosteus and the Carnivorous Fungal Colony), and I feel that, as a concept artist, it’s very important to have at least a familiar knowledge of the pipeline. It means that you can design with the potential technical restrictions in mind (e.g. level of detail).
As such, I’ve assisted this asset creation process by doing the highpoly sculpts in ZBrush, and then handing the file over to be technically handled by (in both cases in this project this has been by our friend and fellow artist Sanna Kempe, who has been assisting on the project). I’ve then had a nice detailed highpoly model to do some renders and paintovers of that I can add to my portfolio of work as ‘concept sculpts’, and it’s also good to then show a literal translation of my work.
As a concept artist one of the key aspects of the role is creating clear designs for a game, and I have done that with orthographic views and sketches from front/rear/side etc. (like I had done with the Elf Armour) but these sculpts remove any doubt/room for interpretation when I want something to be described really clearly.
As Baj Singh (Lead Character Artist at The Creative Assembly) has said: “In most companies, the sculpt is only 50% of the job” – so it I, as a concept artist, can do the first 50% of a character artist’s job, then I become a more valuable asset to a company as it means I’m a more multi-disciplined artist. During crunch time I’m able to take an active role in their pipeline (as concept art is mostly done in pre-production). I can imagine this being a particularly useful skill in smaller companies where staff are more likely to take on multiple roles outside of their original job-description to get a game shipped.
I’m emulating the workflow of many professional artists at the forefront, most of which I have mentioned in this blog before, such as Peter Konig – who has done similar work on the game eVolve – see below image, where he worked on the highpoly sculpt for this character, as well as the original 2D concept art.
So, that said, I have been working on production highpoly sculpts for a few different assets in the game, one of which is a stone Dragon Bust for Joe to use around the city, and across the bridge area. Here is my initial highpoly sculpt:
And here it is having gone through re-topology, textures and made game ready by Sanna Kempe:
Of course, once that’s done, Joe is able to import it into his environment and begin placing it around, as well as doing extra work to it such as using his materials to paint snow on top of it where necessary.
We also had a similar process for my Gargoyle design where my highpoly sculpt (bottom left) was taken and reworked to become game ready by Sanna.
And you can see that in game, in an early WIP development image from Joe:
So basically my ability to work this way with others would hopefully put me in a better professional position than others who may not be comfortable with working in 3D. Although it tends to be the case that larger studios have very dedicated teams working 3D and may not need the extra help, certainly in smaller studios or indie teams, this is a useful skillset to have.
Hobbs, J. (2017). Orloth Screenshots. [image].
Kempe, S. (2017). Dragon Bust. [image] Unpublished [Accessed 15 Aug. 2017].
Konig, P. (2016). Evolve Behemoth. [image] Available at: https://www.artstation.com/artwork/Rzzwy [Accessed 15 Aug. 2017].