Journals32
Newest
I haven't posted much these past years, but I did work in a few projects I'm really proud of during my last year of college.
I made a few videos in case you're curious:
Dynamic Windows: youtu.be/rtD4iKIYmQk
Combining aesthetics and technological development, the aim of this project is to experiment with space perception and interactions between humans and infrastructures, culminating in the presentation of a physical proof-of-concept prototype.
Dynamic Windows: youtu.be/rtD4iKIYmQk
Combining aesthetics and technological development, the aim of this project is to experiment with space perception and interactions between humans and infrastructures, culminating in the presentation of a physical proof-of-concept prototype.
Mooc Radio: youtu.be/KjXXQOODq5w
Combining mobile devices and MOOCs, the imagined service supports and extends education in ways not possible before in Africa through radio and a mobile app.
Virtual Actor: youtu.be/sKbcPvpnP6I
This project aims to mimic the emotive performance of an actor through expressive voice synthesis. The main tool, an intuitive editor and GUI, allows anyone to create their own virtual performance through prosody control.
Combining mobile devices and MOOCs, the imagined service supports and extends education in ways not possible before in Africa through radio and a mobile app.
Virtual Actor: youtu.be/sKbcPvpnP6I
This project aims to mimic the emotive performance of an actor through expressive voice synthesis. The main tool, an intuitive editor and GUI, allows anyone to create their own virtual performance through prosody control.
So... I really try to focus on the experience created by technology rather than the technology itself. Also, I like making cool videos instead of boring reports.
I'm currently doing an internship at a company that works on European research around the Internet of things, smart objects and building automation, so far it's cool.
I have to look for a Master's thesis subject to begin in February though, and I'd love to do something creative and artistic (as opposed to hardcore mathematical science or very serious software development.) I'd love to do something similar to those three projects in the videos, hopefully in a "design studio" kind of company... in Geneva ideally. Gotta start searching!
Alternately, if someone has an idea of how to convince a Professor and the school that making a video game is a good Master's thesis project, please do tell!
k thx bye
-c
Join the community to add your comment. Already a deviant? Log In
Nostalgic about what could have been.
2 min read
I'm a year of work away from a Master's in Computer Science. Everybody expects me to be a coder or an expert on some specific technical field.
I took as many design courses as I could and now that I have to do internships, it turns out as a bad thing (i.e., lacking typical strengths of a computer science student...)
I'm considering doing a Master's in Media Design instead of just working after this... but I'm not getting younger and I'm getting tired of living off my parents.
I never wanted to be *just* a scientist, but I didn't know I'd want so much to be an artist.
...
On a happier note I'm getting pretty decent on a ukulele and not so bad on guitar. I finally got into music this year!
And I drew this:
kthxbye
-c
Join the community to add your comment. Already a deviant? Log In
SugiStyle and the Choice of Unoriginality?
10 min read
Hi there, forgotten me yet ?
I was just "checking my messages" on dA and ran across this very interesting journal entry by fellow pixel artist and fakemon maker
: mojanbo.deviantart.com/journal… If there is something I've admired in her work over the years, it's her perseverance to keep artistic freedom and not give in to the global consensus known as "Sugimori style" in honor of the man who started it all, by drawing 'em all. But I have always been among those that kept pushing her in the opposite direction. Does that sound contradictory? My intention never was to repress her artistically, nor anybody out there I gave advice to on how to make fakemon. I will always value originality above everything else, but the exercise of fakemon spriting (or creating, in general) has always been to me more of a technical job. This you can see in how few fakemon I actually created over the years, as opposed to how many sprites I revamped for the (indefinitely on hold) CS Dex project. I'll explain below.
The question Mojambo and many others have been asking all these years is simple: why does it have to look like Sugimori's drawings to be considered a good fakemon? My answer to that question is short and simple: it's all about compromise.
I know, that doesn't answer the question. But that's because there's not a right answer. There's a consensus. I haven't been active in the "fakemon scene" in a long time, but it was mostly the same for all the time I was: people imagining fantastic creatures and dumbing them down (no offense) to, well, Pokémon. The way I approached fakemon creation is one that definitely helps a lot on that aspect (and one that most fellow SPPF members lived through at the time). It simply consisted in mixing sprites at first, and with time and imagination, we developed pixel art "skills" and started creating our own sprites from scratch. At that point, we were already familiar with the looks and mechanics of a Pokémon sprite, leaving little to room to originality in the "style" department. Many did manage to develop their own look and techniques, like my ol' friend
doing something completely non-Pokémon, while you can still feel the influence of his early works: fav.me/d1bkcxd One could give it a catchy name and call it a fakemon easily, but the style is personal, not Sugimori. Not many get there (or want/need to get there).
So, about making fakemon.
Character creation is a tricky field and something I'm not particularly familiar with, but the one common thing I've observed in all instances of it is that in the end it all comes to creating and fitting a (set of) character(s) within a given "universe" (which can be very loosely defined). That is, there are always rules and conventions to respect, or it will feel forced and be rejected by the targeted group (because yes, most of us make fakemon to share with others, we look for approval, it's human nature).
With Pokémon, the rules seen very much set on stone when creating one: it must fit the global consensus on what is a Pokémon, that is it must be defined mostly by notions representable in a Pokédex entry (types, moves, levels, natures, whatever they'll be adding next). The "research" part of the exercise consists in finding an idea and shaping it into a creature.
I will now consider the case of a "real" Pokémon, cubone. This is a cubone: 1.bp.blogspot.com/-WRa2pcqrw5I… One might argue the design is plain and simple, but that's its strength. It's a solid design. Plus it's official, so it is how a Pokémon should look like.
Now this is a "real life interpretation" of a cubone: ynm2nw.bay.livefilestore.com/y… Everybody will admit it definitely looks like one (although not necessarily like one might have imagined it, as we'll see later). On the other hand, this is definitely not a cubone: i3.squidoocdn.com/resize/squid… Why? Well, duh.
In both cases it's a mammal wearing "protective" headgear. The important difference is that the first one respects the design credited to the one man that can decide what looks like a Pokémon and what doesn't: Sugimori himself. The tricky part with fakemon comes (obviously) from the fact that they're not drawn by Mr Sugimori (interestingly, not all Pokémon are nowadays either if I remember correctly). That's where community approval becomes so important when making fakemon.
Let me compare this to AccurateRip, which I was reading about just lately. AccurateRip is a community based service which collects information (hashes) on ripped audio CDs and stores them in a public database for comparison. This database allows anybody using the service to compare their home made rips (be it MP3s, AACs for your iPod or backup FLACs) with those of others. This way, you can know if your files are "like" those of others. Statistically, if many people obtain similar rips, then these are likely to be of good accuracy (the probability of them being bad is reduced exponentially).
The analogy here is on how other people in the fakemon scene see your fakemon on how they stand against the community defined standards (the consensus). Put simple, the more opinions you get, the closer to an overall impression of your work you get. This tells you how to know if you're getting Sugimori style right.
Now let's get back to the issue at hand: originality.
Take a look at this: nightmarelover.deviantart.com/… That is one heck of a cool interpretation on the cubone concept. As you can see, its has sharp claws and no fur, it's definitely reptilian. It might not be to your taste, but I like it very much myself. But... it just doesn't look like a Pokémon.
To illustrate this, assume cubone doesn't exist. Now define it with all the usual information we know (from bulbapedia for instance), but with this one picture instead. Now, is that a good fakemon? Sure, but it just doesn't look like a Pokémon (this is my opinion, but I would be surprised if only few shared it with me).
This brings us back to my initial answer: it's all about compromise.
By setting the goal of making a "fake Pokémon", you're setting the "universe" in which your character design must fit: the Pokémon universe. Just like there is a set of types from which to chose from (no "light" or "nuclear" types available, sorry), there is also a look and feel associated with Pokémon. This takes away some of your freedom on how to represent your design. The whole exercise becomes a matter of representation, or rather interpretation. Sugistyle is imitation by definition. It's not possible to keep full artistic freedom while imitating, so you must compromise. You already do just by calling it a "fakemon".
The last few months I actually spent thinking about fakemon, which was one or two years ago, I came to realize how limited I felt in how I could design these creatures. This is something not only Mojango can relate to, but everyone who ever faced criticism surrounding Sugimori style. Such criticism might not sound fair from a purely artistic point of view, but makes perfect sense given the aim of the operation: making fake Pokémon.
[TL;DR from here] So, my two cents of advice for everybody out there making fakemon:
If you like the official Sugimori style, if you don't feel forced into it, that's great. If you're also getting it right and people love it, then that's awesome.
If in the other hand you feel repressed as an artist, you may want to rethink the "Pokémon" part of it. Your amazing creature designs don't always have to be Pokémon. If it feels forced, it might not be worth it.
Ultimately, you must learn to compromise between your personal vision and a "product" for the targeted "consumer group". Pokémon is not just art, it's product design. In the end, no matter how much you push originality forward, it's statistically really really hard to reach a consensus where much freedom can be taken while trying to respect the strict (implied) "definitions" and "rules" of the Pokémon universe. And that is normal. You want Pokémon? Make Pokémon, make them like Pokémon, otherwise they wont feel like Pokémon. The visuals are obviously a very important part of the whole Pokémon "experience".
It might sound like fatalism, but it's not. It's acceptance. Once you accept that, you can move forward as an artist and be original. If you really want to make original Pokémon, reach a compromise you feel comfortable with. Do your best to improve towards your goal, but never, and I insist, NEVER force yourself to "do art" the "right way". Such a phrase is an absolute contradiction: there's no right way in art.
Now, the good news! You don't have to always "fit in". That is something my other ol' friend
understood long ago. Her amazing gallery, which is purged way too often in my opinion, has been the home of an impressive amount of original creations. Like me, she started simple, mixing sprites, but unlike me, she was not afraid of exploring, and she was inspired. This is a a good example of how to get Sugimori style right: matatonai.deviantart.com/art/T… This is an example of how to be original: matatonai.deviantart.com/art/f… it might not be among the greatest visual productions of hers, but it is 100% hers. At the time, she was working on alien creature concepts, life as it could have evolved in a different planet. There's detailed descriptions and illustrations on the biology of these creatures. But in the end I'm really just a sucker for her dragons: matatonai.deviantart.com/art/s… (wish she had a good scanner...)
All this to show you, not all your creatures need to be Pokémon, but if they are, they'll be expected to fit in the consensus. What you do with them is up to you, and it doesn't have to be Black or White (Pokémon Gray Version coming anytime?)
Hope I didn't bore you to death, this was meant to be a very short reply to Mojanbo's journal, but it turns out I really had a dissertation ready for the occasion buried somewhere deep inside my brain. So there you go.
Just remember to have fun. That's the most important thing. Draw for fun, enjoy it.
And don't hesitate to reply to this journal, even just to say hi
casti out
I was just "checking my messages" on dA and ran across this very interesting journal entry by fellow pixel artist and fakemon maker
: mojanbo.deviantart.com/journal… If there is something I've admired in her work over the years, it's her perseverance to keep artistic freedom and not give in to the global consensus known as "Sugimori style" in honor of the man who started it all, by drawing 'em all. But I have always been among those that kept pushing her in the opposite direction. Does that sound contradictory? My intention never was to repress her artistically, nor anybody out there I gave advice to on how to make fakemon. I will always value originality above everything else, but the exercise of fakemon spriting (or creating, in general) has always been to me more of a technical job. This you can see in how few fakemon I actually created over the years, as opposed to how many sprites I revamped for the (indefinitely on hold) CS Dex project. I'll explain below.The question Mojambo and many others have been asking all these years is simple: why does it have to look like Sugimori's drawings to be considered a good fakemon? My answer to that question is short and simple: it's all about compromise.
I know, that doesn't answer the question. But that's because there's not a right answer. There's a consensus. I haven't been active in the "fakemon scene" in a long time, but it was mostly the same for all the time I was: people imagining fantastic creatures and dumbing them down (no offense) to, well, Pokémon. The way I approached fakemon creation is one that definitely helps a lot on that aspect (and one that most fellow SPPF members lived through at the time). It simply consisted in mixing sprites at first, and with time and imagination, we developed pixel art "skills" and started creating our own sprites from scratch. At that point, we were already familiar with the looks and mechanics of a Pokémon sprite, leaving little to room to originality in the "style" department. Many did manage to develop their own look and techniques, like my ol' friend
doing something completely non-Pokémon, while you can still feel the influence of his early works: fav.me/d1bkcxd One could give it a catchy name and call it a fakemon easily, but the style is personal, not Sugimori. Not many get there (or want/need to get there).So, about making fakemon.
Character creation is a tricky field and something I'm not particularly familiar with, but the one common thing I've observed in all instances of it is that in the end it all comes to creating and fitting a (set of) character(s) within a given "universe" (which can be very loosely defined). That is, there are always rules and conventions to respect, or it will feel forced and be rejected by the targeted group (because yes, most of us make fakemon to share with others, we look for approval, it's human nature).
With Pokémon, the rules seen very much set on stone when creating one: it must fit the global consensus on what is a Pokémon, that is it must be defined mostly by notions representable in a Pokédex entry (types, moves, levels, natures, whatever they'll be adding next). The "research" part of the exercise consists in finding an idea and shaping it into a creature.
I will now consider the case of a "real" Pokémon, cubone. This is a cubone: 1.bp.blogspot.com/-WRa2pcqrw5I… One might argue the design is plain and simple, but that's its strength. It's a solid design. Plus it's official, so it is how a Pokémon should look like.
Now this is a "real life interpretation" of a cubone: ynm2nw.bay.livefilestore.com/y… Everybody will admit it definitely looks like one (although not necessarily like one might have imagined it, as we'll see later). On the other hand, this is definitely not a cubone: i3.squidoocdn.com/resize/squid… Why? Well, duh.
In both cases it's a mammal wearing "protective" headgear. The important difference is that the first one respects the design credited to the one man that can decide what looks like a Pokémon and what doesn't: Sugimori himself. The tricky part with fakemon comes (obviously) from the fact that they're not drawn by Mr Sugimori (interestingly, not all Pokémon are nowadays either if I remember correctly). That's where community approval becomes so important when making fakemon.
Let me compare this to AccurateRip, which I was reading about just lately. AccurateRip is a community based service which collects information (hashes) on ripped audio CDs and stores them in a public database for comparison. This database allows anybody using the service to compare their home made rips (be it MP3s, AACs for your iPod or backup FLACs) with those of others. This way, you can know if your files are "like" those of others. Statistically, if many people obtain similar rips, then these are likely to be of good accuracy (the probability of them being bad is reduced exponentially).
The analogy here is on how other people in the fakemon scene see your fakemon on how they stand against the community defined standards (the consensus). Put simple, the more opinions you get, the closer to an overall impression of your work you get. This tells you how to know if you're getting Sugimori style right.
Now let's get back to the issue at hand: originality.
Take a look at this: nightmarelover.deviantart.com/… That is one heck of a cool interpretation on the cubone concept. As you can see, its has sharp claws and no fur, it's definitely reptilian. It might not be to your taste, but I like it very much myself. But... it just doesn't look like a Pokémon.
To illustrate this, assume cubone doesn't exist. Now define it with all the usual information we know (from bulbapedia for instance), but with this one picture instead. Now, is that a good fakemon? Sure, but it just doesn't look like a Pokémon (this is my opinion, but I would be surprised if only few shared it with me).
This brings us back to my initial answer: it's all about compromise.
By setting the goal of making a "fake Pokémon", you're setting the "universe" in which your character design must fit: the Pokémon universe. Just like there is a set of types from which to chose from (no "light" or "nuclear" types available, sorry), there is also a look and feel associated with Pokémon. This takes away some of your freedom on how to represent your design. The whole exercise becomes a matter of representation, or rather interpretation. Sugistyle is imitation by definition. It's not possible to keep full artistic freedom while imitating, so you must compromise. You already do just by calling it a "fakemon".
The last few months I actually spent thinking about fakemon, which was one or two years ago, I came to realize how limited I felt in how I could design these creatures. This is something not only Mojango can relate to, but everyone who ever faced criticism surrounding Sugimori style. Such criticism might not sound fair from a purely artistic point of view, but makes perfect sense given the aim of the operation: making fake Pokémon.
[TL;DR from here] So, my two cents of advice for everybody out there making fakemon:
If you like the official Sugimori style, if you don't feel forced into it, that's great. If you're also getting it right and people love it, then that's awesome.
If in the other hand you feel repressed as an artist, you may want to rethink the "Pokémon" part of it. Your amazing creature designs don't always have to be Pokémon. If it feels forced, it might not be worth it.
Ultimately, you must learn to compromise between your personal vision and a "product" for the targeted "consumer group". Pokémon is not just art, it's product design. In the end, no matter how much you push originality forward, it's statistically really really hard to reach a consensus where much freedom can be taken while trying to respect the strict (implied) "definitions" and "rules" of the Pokémon universe. And that is normal. You want Pokémon? Make Pokémon, make them like Pokémon, otherwise they wont feel like Pokémon. The visuals are obviously a very important part of the whole Pokémon "experience".
It might sound like fatalism, but it's not. It's acceptance. Once you accept that, you can move forward as an artist and be original. If you really want to make original Pokémon, reach a compromise you feel comfortable with. Do your best to improve towards your goal, but never, and I insist, NEVER force yourself to "do art" the "right way". Such a phrase is an absolute contradiction: there's no right way in art.
Now, the good news! You don't have to always "fit in". That is something my other ol' friend
understood long ago. Her amazing gallery, which is purged way too often in my opinion, has been the home of an impressive amount of original creations. Like me, she started simple, mixing sprites, but unlike me, she was not afraid of exploring, and she was inspired. This is a a good example of how to get Sugimori style right: matatonai.deviantart.com/art/T… This is an example of how to be original: matatonai.deviantart.com/art/f… it might not be among the greatest visual productions of hers, but it is 100% hers. At the time, she was working on alien creature concepts, life as it could have evolved in a different planet. There's detailed descriptions and illustrations on the biology of these creatures. But in the end I'm really just a sucker for her dragons: matatonai.deviantart.com/art/s… (wish she had a good scanner...)All this to show you, not all your creatures need to be Pokémon, but if they are, they'll be expected to fit in the consensus. What you do with them is up to you, and it doesn't have to be Black or White (Pokémon Gray Version coming anytime?)
Hope I didn't bore you to death, this was meant to be a very short reply to Mojanbo's journal, but it turns out I really had a dissertation ready for the occasion buried somewhere deep inside my brain. So there you go.
Just remember to have fun. That's the most important thing. Draw for fun, enjoy it.
And don't hesitate to reply to this journal, even just to say hi
casti out
Join the community to add your comment. Already a deviant? Log In
JPEG (oh hi there)
2 min read
so, hum, anyone's got an idea on how to improve a badly compressed jpeg image?
for a digital photography assignment we're supposed to use all the knowledge we've acquired and come up with a way to improve the quality said image.
we've been shown how to "fix" capture issues like bad contrast and noise, but the thing is jpeg compression produces visual artifacts for which basic denoising procedures just won't help.
jpeg artifacts include (and i site some random google search result):
- "Mosquito noise"‚ commonly found around edges, such as eyes;
- Square-shaped noise distortion known as "blockiness" or "checkerboarding";
- Fuzziness and loss of edge clarity.
basically, it's all localized issues and the usual [blur, sharpen, and do some sort of addition of those images to compute a cleaner result] just doesn't seem to work too well...
and i doubt white balancing algorithms would help :/
i've got till friday midday (CET) to get it working. any ideas?
PS: all matlab code, i tried filters (low pass, high pass, gaussian, laplacian, and what not), it works for usual noise but not here.
for a digital photography assignment we're supposed to use all the knowledge we've acquired and come up with a way to improve the quality said image.
we've been shown how to "fix" capture issues like bad contrast and noise, but the thing is jpeg compression produces visual artifacts for which basic denoising procedures just won't help.
jpeg artifacts include (and i site some random google search result):
- "Mosquito noise"‚ commonly found around edges, such as eyes;
- Square-shaped noise distortion known as "blockiness" or "checkerboarding";
- Fuzziness and loss of edge clarity.
basically, it's all localized issues and the usual [blur, sharpen, and do some sort of addition of those images to compute a cleaner result] just doesn't seem to work too well...
and i doubt white balancing algorithms would help :/
i've got till friday midday (CET) to get it working. any ideas?
PS: all matlab code, i tried filters (low pass, high pass, gaussian, laplacian, and what not), it works for usual noise but not here.
Join the community to add your comment. Already a deviant? Log In
455
1 min read
deviations for me to check.
so i took the time to flush all those journals and favs (thank you) and reply to some comments...
you know about
's contest, well, funny thing is i had the concepts for the fakemon entries before the deadline, but only sketched them on sept. 2nd, and haven't developped the hardest one of the two (it's a stat-shape-shifter ninja-esque humanoid i've got yet to refine). so i completely skipped it!
remember to check
NOW!
so i took the time to flush all those journals and favs (thank you) and reply to some comments...
you know about
's contest, well, funny thing is i had the concepts for the fakemon entries before the deadline, but only sketched them on sept. 2nd, and haven't developped the hardest one of the two (it's a stat-shape-shifter ninja-esque humanoid i've got yet to refine). so i completely skipped it!remember to check
NOW!
Join the community to add your comment. Already a deviant? Log In