I could describe and thereafter scrutinize the craftiness and creativity of Elsweyr’s architecture, but it’s impressive enough that displaying a few images will do the trick. Level design in ESO has been a hit-or-miss dilemma, but this is surely one of Zenimax Online’s best works. Some of the scenic imagery here adds some serious visual value to the game as well.
Without some static and continually observable stimuli to document for empirical repeatability gauging entertainment value and gameplay intensity becomes far more challenging. Clearly those with the resources to record data on the matter with a high degree of reliability as far as their recording techniques go have the upper hand. However there are a few starting points and tools that can be used to get a better sense of what is more entertaining and optimally challenging.
An intuitive conception of what sort of notion is worthwhile as far as entertainment goes is inquiring whether the user has stopped playing the game prior to the typical length of time one remains attentive to a task that requires learning. In long-winded games, for instance, a user that stops playing twenty minutes into gameplay is likely not a satisfied one and most likely one who has not completed the game. The same user can play a different game for hours, and through data it can easily be shown that the characteristics and preferences of the player matter little as compared to the entertainment value of the game, measured in this case by the duration of play. Thus there is something to say about the quantification of entertainment as a value and there are some distinct indicators that can be adjusted to compliment gameplay as needed.
On the elementary level a risk-and-reward system must be created in order to take advantage of the primitive behaviors which partially undergird game interaction. Thus there must be a clear way in which the player interacts that offers a risk or punishment mechanism and a reward mechanism. These systems vary greatly but the evident value of a game dissipates without one or the other. Imagine, for instance, a shooter game where nobody dies or a football game where nobody keeps score. Chiefly among the characteristics which influence the understood value of an interaction is timing. Various parts of the brain contribute to risk-and-reward and participate in recording delays in some manner, such as the frontal and parietal cortex, the amygdala and the striatum . Generally, the greater the delay in reward the less of a response is given to stimuli acting as reward-predictors. However in the case of the amygdala, for instance, neuronal response varies based on the population of neurons and the probability of instantaneous reward. Despite the complications the primary obvious issue is the relevance of timing in a reward system and the variation of the response according to the duration of the delay before the reward is given. Any game whose rules force the delay between action and reward to last far beyond the duration of the activity should not be expected to have the same type of satisfaction as those that moderate the proportion of investment value to delay.
Alongside this ought to be taken into account the relevance of user-control and the degree of control, since this appears to have a relationship with entertainment value or compliance; data showed that, comparatively, a more user-controlled environment lead to greater compliance measures than an automated environment . Providing freedom and efficacy within the constraints of a game should generally lead to a more engaged player. This includes such seemingly inconsequential details such as the choice of reward and operation of play within the rules of a game.
There is also something to say about the juncture between the risk-and-reward system and the level of intensity or challenge of the game, as dopamine may be relevant in relation to gameplay performance . Perceived difficulty has been recorded to increase monotonically (consistent increase) with measured difficulty and there is a threshold beyond which the goal of the game in question is rejected by the majority of users . Thus the best sure method of intensity optimization is experimentation with a group of individuals within the population of the target audience. The design objective is to maintain a challenging encounter and goal acceptance and a high probability of success while conceding some leeway when it comes to probability of success.
In most cases it is not possible for designers operating on a small scale to gather the resources to experiment with control groups of a population so as to optimize gameplay performance and entertainment value but there are various human-universal characteristics which may contribute to the creation of a foundation on which smaller-scale testing can occur.
 Bermudez, M. A., & Schultz, W. (2014). Timing in reward and decision processes. Philosophical transactions of the Royal Society of London. Series B, Biological sciences, 369(1637), 20120468. doi:10.1098/rstb.2012.0468
 Nagle, A., Riener, R., & Wolf, P. (2015). High User Control in Game Design Elements Increases Compliance and In-game Performance in a Memory Training Game. Frontiers in psychology, 6, 1774. doi:10.3389/fpsyg.2015.01774
 Koepp MJ, Gunn RN, Lawrence AD, Cunningham VJ, Dagher A, Jones T, Brooks DJ, Bench CJ, Grasby.(1998). Evidence for striatal dopamine release during a video game. PM Nature. 393(6682):266-8.
 Erez, Miriam & Zidon, Isaac. (1984). Effect of goal acceptance on the relationship of goal difficulty to performance. Journal of Applied Psychology. 69. 69-78. 10.1037/0021-9010.69.1.69.
The idea of architecture entrenched in physical descriptions is traditionally what is espoused in the practice of architecture in academia and the application of such ideas are attended to in practice, but the greatest benefit in relation to architecture that level designers have is their ability to escape from reality which affords them a far less limited description. There are still rules in constructing a level, but a level designer need not pay attention to rules which prevent them from designing constructive things in the real world. There is no need to pay attention to physics or restrict the mode of design to the set of methods possible in the real world. Therefore we need a more abstract conceptualization of what architecture is that is useful to level designers. A more useful definition is that architecture is the visuospatial characterization of geometric details over base geometry, which is collectively the most basic forms of geometry needed to visualize location.
Despite level design being an escape from reality the machines that interact with the level are still human, and so the rules of being human still apply, which is why adopting the exploitation of phenomena discovered in neuroscience is appropriate. Intuitively architecture in this more abstract definition performs two interrelated purposes: telling the same story and doing so in a different way. It’s highly unlikely that a level designer will be able to build a novel template on which architectural detail can be added since there are not many different ways to lay out the foundational geometry that have not already been done before. One might even claim they have discovered a new template only to discover that it can easily be characterized and traced back to something that has already been seen. A church is the perfect example. The template is the same, the details are different. So structure in architecture should optimize the visuospatial characteristics of the template for the sake of evoking a desired response in the viewer as structural detail contributes to the phenomenon. So the difficult question that arises is “What makes better structure?” Paying attention to what visual stimuli are more capable of invoking a particular response or mood on the part of the viewer allows us to determine that architecture which is more influential in provoking a response along with visuals that coincide with a human notion of aestheticism is structurally better than architecture that is less influential in doing so. This doesn’t mean that architecture can’t be “ugly” in the sense that a level designer might want to invoke a dark mood or a sense of disgust, but rather architecture that is less able to do so and is aesthetically nonsensical is purely inferior to architecture that is not. In every case looking at the concepts behind the details give an intuition of what the human brain feels in most cases. Humans group objects based on visual characteristics like shape, color, and spatial distance . Architecture can then be analyzed in a hierarchical way, in which structures contain substructures and the quality of those substructures combined with the quality of the larger structures provide a sense of quality of the whole visual construct. Those structures are the visual things humans group together. Two key principles to pay attention to are the orientation of symmetry and the harmony of ordered wholes. There are various kinds of symmetry but structure has to maintain symmetry somehow for the maintenance of proper structure, given that the human brain pays an elevated level of attention to symmetry . The latter is something which I am in the process of figuring out, but it doesn’t seem to be the same thing as symmetry. The best description I can give to it is the geometric respect for natural order. You can find plenty of ugly buildings in the world but I have never seen a building which was built in a way which violates this principle, perhaps because it is in the very nature of humans to avoid it innately. There is a visual orderliness in the universe in anything which by its nature has structure. This may need not apply, therefore, to fluid, for example. A great violation of this principle may be imagined. Take a look at the Forbidden City.
Now, in your mind, for every set of n fixed units in a cubic region, take any geometry that exists within that region and rearrange it in any arbitrary manner, through rotation, extrusion, decimation, scaling, stretching, splitting, et cetera in a manner totally different from every other region you have modified. This would be an ideal opposite of harmonious ordered wholes, which may perhaps simply be related to chaos in a geometric sense. The whole point of structure is the maintenance of visual order and in architecture it is done hierarchically. Since we have an idea of what not to do we have a sense of what to run away from and how to do it, which at least gives us an intuition of how to qualify structure.
Within a foundational structure creativity can be realized in how the designer applied innovation to what has already been done and how they manipulated the visual template in order to creatively maintain its structure yet give it a unique personality. Tōdai-ji is in a geometric sense built upon the same template as the Forbidden City, yet the two have architectural details which contrast them so much that they are uniquely distinct from one another.
Having covered how to roughly qualify architecture in level design we can focus on the specific ways in which the human mind can be influenced by the spatial and visual structure of architecture. Symmetry, for instance, can be used to highlight differences and similarities. A focal point in a symmetrical space can be used to orient the viewer in order to aid them in absorbing a scene wholly. Repetition of visual motifs allows simultaneous processing of the visual figures being repeated. There is also some work that may suggest that facades themselves influence mood, and perhaps if a relevant portion of the cortex that is stimulated is the same portion which is involved in the processing of human faces, there may even be a more grounded sense of what distinguishes aesthetic facades from unaesthetic ones. Some work has been done attempting to use machine learning for face recognition (SVM) in order to extract from facades those inputs that would otherwise be recognized in faces and assign to them particular moods , albeit this may not truly relate the data to the phenomenon through which processing faces occurs in the human brain the phenomenon of pareidolia has been studied for some time.
So it’s evident that architecture is a profoundly relevant factor in level design and can be used to further the very purpose of level design as an interactive art.
 “Gestalt Principles”. Web. http://graphicdesign.spokanefalls.edu/tutorials/process/gestaltprinciples/gestaltprinc.htm
 Symmetry activates extrastriate visual cortex in human and nonhuman primates. Yuka Sasaki, Wim Vanduffel, Tamara Knutsen, Christopher Tyler, Roger Tootell. Proceedings of the National Academy of Sciences Feb 2005, 102 (8) 3159-3163; DOI: 10.1073/pnas.0500319102
 Simulating Paredolia of Faces for Architectural Image Analysis. Chalup, Stephen K., Hong, Kenny. ostwald, Michael J. International Journal of Computer Information Systems and Industrial Management Applications (IJCISIM) Vol.2 (2010), pp.262-278
The combined scale and grandness of a big idea brought to life with aesthetic and intricate architectural designs grazed with rich detail are what makes this DLC stand out among the rest. The Clockwork City blows every other zone in the Elder Scrolls Online out of the water. This makes what I’d argue is the runner up, Orsinium, look like a little village.
As I’ve summarily explained in a previous article, having no framework leads to unfair review practices in a very technical sense of the word. On top of this we should consider that unlike systems that are powerful in that they express their elements truthfully, those that approximate a temporary molding together of personal influence and bias alongside sentiment and guessing are easier to use because humans do it all the time. They act as heuristics by which people operate in the world in a correct manner. But if the whole point of a process is to measure quality then that sort of ambiguous heuristically-driven approach doesn’t fare well at all. This is not difficult to realize. Find any reviewer on the web and find a review they made on a game level. They may have described their experience in playing the level. More analytical approaches provide elaborate essays detailing particular events and things that stood out. But in the case wherein no defined framework is provided you cannot trace their reasoning back in a logical way to conclude how they went from a description of sensory experience to a number, which so happens to be the very element of ambiguity. This is true both practically and logically; English is an ambiguous language and I don’t know if there has ever existed a human-developed language that is not ambiguous, in reference to the definition of ambiguity used in formal languages, a linguistic and computer-scientific study.
We can at least begin with a framework that has an outer structure. In the cases where each aspect of which the framework is composed of is well defined, readers and observers can more easily decrypt the reviewer’s reasoning and isolate particular components in order to more adequately figure if the reviewer’s rating makes sense, which is to determine the degree to which the rating coincides with the reasoning supporting it, which is collectively an approximation of the quality of that particular element, and being the very point in question, it suffices to say it is the whole challenge motivating the framework in the first place.
In order to build a framework however, we need to figure out what we want to get out of a work of art, which roughly translates into the need to define quality. This is dangerous territory because so many fields of study come to mind with partial answers. I’ll provide two different answers. The first is a general rule that may give a thoughtful and intuitive sense of what the goal for this framework is. I hold the philosophical presupposition that art ought to express and evoke the human spirit. The author ought to express it, and should endeavor to construct their art in order to evoke the spirit of others. This should be taken with a grain of salt, because what I mostly mean by this is more grounded in scientific concepts. This is where the second answer comes along, and it’s a long winded one. Beauty –the beauty we speak of, or perhaps artistic quality– is a property constructed by human interpretation. Hence we have fields of knowledge in every direction which may contribute to the problem. Beauty is undoubtedly influenced by phenomena studied in psychology, evolutionary biology and neuroscience. It has also been well-established, for instance, that certain stimuli in controlled environments yield a deterministic response from the human brain. Change blindness is a popular example. Scientific experiments have demonstrated that changing a particular element on a still-image in a sufficiently slow manner does not catch the attention of the human viewer, provided there are no disruptions . There have also been quite a few many studies done documenting human male preferences of females based on deciding their level of beauty. Physical appearance is a culturally-independent property relevant to humans in seeking a mate, and certain features are found to be attractive across cultures . It has also been demonstrated that combining visual and auditory stimuli resulted in activity in a particular cortical region , although it is more difficult to scientifically extrapolate more meaningful conclusions out of such experiments. Given that there are particular stimuli that evoke deterministic or chaotically deterministic responses in the human brain and that beauty is heavily influenced by human evolution we can at at least be certain that there is an human-contextual objective description of beauty, and so a framework which seeks to measure beauty as a quality of artwork ought to aim to identify aspects which reflect the degree to which beauty is properly expressed.
At this point you can imagine that interactive quality has a similar background, although perhaps there is a more distinct difference in male and female brains when it comes to interaction and play as opposed to visual and auditory stimulation. Nevertheless this collectively provides us a foundation onto which we can explore more proper methods by which we can quantify quality, and with this kind of unexplored and mysterious challenge we’re sure to modify some things every so often, but the whole point is to get closer to a more universal reading of design quality.
The initial template we use is as follows and is given out of its sum value of 10 points:
- Architecture (2)
- Structure (1)
- Innovation (1)
- Atmosphere (2)
- Visual Immersion (0.75)
- Auditory Immersion (0.75)
- Detail (0.5)
- Gameplay (2)
- Entertainment (0.5)
- Intensity (0.5)
- Novelty (0.5)
- Flux (0.5)
- Visual Impact (2)
- Concept Impression / Grandness (1)
- Visual Awe (0.5)
- Visual Creativity (0.5)
- Character Development (0.5)
- Plot Development (0.5)
- Depth (1)
I will break down each component individually and elaborate on it in future articles. You may notice that Storyline doesn’t have the kind of immediate experimental support that other aspects hold, so it also begs investigation on a psychological level, although that is an entire topic that can be written about in the future.
 “Change blindness in the absence of a visual disruption”. Simons, Daniel J. Franconeri, Steven L., Reimer, Rebecca L. Perception. 14, July 2000. Vol. 29 pp. 1143-1154 DOI:10.1068/p3104
 “Maxims or myths of beauty? A meta-analytic and theoretical review”. Langlois JH, Kalakanis L, Rubenstein AJ, Larson A, Hallam M, Smoot M Psychol Bull. 2000 May; 126(3):390-423.
 “Toward A Brain-Based Theory of Beauty”. Ishizu, T., & Zeki, S. (2011). PLoS ONE, 6(7), e21852. http://doi.org/10.1371/journal.pone.0021852
Have you ever read or listened to a game review, or a review for anything for that matter, and wondered how on Earth the reviewer came up with their diagnosis? Maybe they gave it their highest rating, but you find yourself baffled at how they might have been able to reason it through in the first place. Perhaps you saw a mod with a “Best Of” award, although the mod didn’t seem particularly outstanding. In the overwhelming majority of cases I’ve seen, mod reviewers in particular employ a practice of judgment and reasoning which is only as sound as the extent to which they can properly judge and reason, which is not saying very much given that it’s difficult to say that a review based off of reason does much more than discreetly display the reviewer’s biases and preferences. Not to mention that they might have never exposed their methodology which they use in order to come to a decisive conclusion in their review. The collection of these problems makes things confusing.
Firstly, since there is no sufficiently rigid framework for what defines quality, a reviewer can “feel” as though one mod is better than the next. The reasoning by which they say one is better than the other can hardly be said to be identical if they can’t describe the framework through which they felt one was better than the other. The result in effect is two different ratings based off two different (un)reasonable conclusions. It would be the equivalent of doing the following: looking at two different works and judging the first solely on its visual quality, and the second solely on its auditory quality. The first gets a far better rating than the other. That’s not a fair rating because, technically, what the reviewer actually did is give one a rating based off System A, and the other based off System B, whose components are totally independent, so that there is no comparison to say which one is better, yet one got a better rating. What many reviewers do is similar. They have no rubric and as a result, they can step into all sorts of mistakes that skew their review process. It’s not particularly credible. In the end if you are comfortable with that kind of phenomenon, it’s fine, but realize it’s simply the “Person’s Opinion Awards” and not an endeavor for true quality control, measurement, and comparison.