A Comparative Analysis of Game Asset Creation Using \nConventional and AI Methods
Bibliographic record
Abstract
The rapidly evolving Artificial Intelligence (AI) field is having a significant impact on many industries, as well as individuals. ChatGPT and image generating applications are undoubtedly among the most well-known and widely utilized AI tools. It is interesting to explore how these AI tools may be used in video games, specifically for asset creation, and how they compare to more traditional methods. \n \nThis thesis investigates the creation of video game assets using traditional techniques and AI tools such as ChatGPT and Midjourney. A group of 34 participants, each with varying levels of experience in different techniques and asset production, were given the task of creating basic game components using both traditional and AI tools. The goal was to rank both methods according to the overall rating, the satisfaction of the end result, and the ease of use to determine which one is preferred. The sample scene was a simple 2D platformer game, similar to Mario, with which most people are familiar and most likely played at some point. The participants were involved in both the creation process of game assets, as well as their evaluation, which brings a new perspective on the matter. It was discovered that, on average, AI tools are rated higher and are simpler to use than traditional approaches. Participants were highly satisfied with the results. However, the content created in such a way may not be as creative or as tailored for the specific needs as content made by people. Using these technologies makes it difficult to maintain a consistent style or create exactly what the person envisions, as compared to producing them manually, when the artist has complete control over every step and detail. Both methodologies have value, and depending on the project's goals and available resources, one may be preferred over the other. A hybrid method, which combines AI efficiency with artist creativity, may be the best option.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.004 | 0.005 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".