Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Tara from Chasing Sunsets was introduced to the world on April 3, 2020 and launched public content on Sept 24 2020.
College Kings launched in October 2 2020, or about a week later.
Our Tara predates any public record of her counterpart in that game by five months, and in released content by a week, for what it's worth.
To be clear, there is no allegation of ill-intent here: The number of female models available in DAZ are limited, and coincidences will happen, but in this case, we absolutely were first.
I thought the answer would be something like this,
Some Characters in games look similar to others in other games,
This will always happen. Especially when Dev's are using the same rendering program.
and PS: I read the FAQ after I posted my comment above about redheads....and it was only an observation and not intended as criticism...lol
Our product has been known in AVN circles for well over a year, and a huge argument broke out on another forum about Jaye's "orange" skin tone.
Out came image and art experts with their Photoshop filters and color calibration tools, and one commentor finally showed the math suggesting Jaye's actual skin tone is called "smoke tree" with very little orange in it according his color correction software.
It was such an epic thread that I paid homage to it in Chapter 3 when Jaye referred to her own skin tone as "Smoke Tree" as a bit of an inside joke.
That said, the darkness of her tan does create lighting challenges for us, as it's very easy for her to appear too dark for how she was envisioned initially if we're not careful. The scenes of her tanning on the yacht most accurately reflect our design objective as the overhead sunlight is white and about 4500K, which therefore does little to affect her natural skin color.
As stated in the FAQ: We know redheads typically have lighter skin. We like Jaye the way she is. Apologies if it makes suspension of disbelief impossible for some.
Man Jaye is perfect the way she is, it is totally acceptable for someone to be "tanned". Also she could dye her hair lol, anyways i LOVE the way she is. please believe in your decisions! haha
As for the 'identical model' issue, I've used DAZ at various times over the past decade, and I distinctly recall spending countless hours altering values and sliders to create totally unique facial/body structures from the 'base' models. It's not that it cannot be done (and you only have to do it once per model, not once per render), it's just that it takes a lot of time to do it properly (human faces in particular have a tendancy to look very odd when modelled digitally), and most of the assets used in the majority of VN's are 'stock models' anyway. Unless you're crazy enough to write your own software raytracing engine from scratch (like me), you're going to simply use the engines, tools and models available on the open market and not sweat the small stuff. I imagine most VN creators are more concerned about getting the story out there than nit-picking over competing character models.
Typically we take a "Is the juice worth the squeeze" approach. Characters we'll see a LOT of tend to be more customized than characters we'll see less of. It's all about prioritizing our development time and the overall impact for the player.
I've written my own physically-based raytracing renderer that is capable of real-time rendering, but at best (even with multithreading and a metric ♥♥♥♥-tonne of model/math optimisation) it is only capable of rendering in 800x600 at about 30 FPS on a low-end 6-core CPU, which drops to at least half that (and usually much worse) as soon as reflection/refraction becomes part of the scene. I could use it to render scenes like those found in VN's fairly quickly (if we calculated only based on the 'frames rendered' metric, the average VN could easily rasterise at 1920x1080 in a few days), but my engine currently only supports 'keyframe' models, since I haven't yet figured out the math behind 'rigged' models (which most non-static models in VN scenes are these days).
Even disregarding the use of potentially multi-million-triangle human models, rasterising things like hair, eyes and other multi-layer refractive and/or reflective surface groups slows rendering to a crawl for a simple static scene. Newer scene rasterisers have started adding sub-surface scattering to the already layer-laden scene tree. Now, add the time/design costs of setting up each individual scene, multiply all of that by the number of frames per second per animation sequence, add in the coding/setup (depending on how you do it) of the sequences themselves, and you're looking at... well, a lot of time.
Unfortunately, that is simply the cost of using the mathematics behind light-transport physics (much of that being costly dot/cross products; for the layman, more additions and multiplications per screen pixel than you can poke a burnt stick at). Even on modern processors, that takes time; a good kD-tree implementation helps, but there are only so many triangles you can cull before hard limits set in. Plus, it helps if you just happen to have more than one dedicated rig to do all this on, since any decent engine will usually peg your CPU utilisation close to 100% throughout scene rasterisation.
Bear in mind, that hasn't taken anything but the visuals into account.
Amazing how little most people understand just how much time, effort and (in the case of those of us who do it the really hard way) mathematics actually goes into creating virtually any form of digital art... and complain that it takes a while to get it done right.
The average render takes 45 minutes to process through iRay, not to mention posing, scene composition, lighting, and postprocessing. We have two development rigs running dual RTX 3090s, and one is used for posing while the other one renders in parallel.
One thing I do take pride in is that we're regarded well in the community for delivering regular updates.
We typically do a unique render for every line of dialogue, which is tedious, but we hope adds to immersion.
This is our first game on Steam, and we felt it was more honorable to go Early Access rather than follow the trend and release the game in separately-purchased instalments ("Book 1" and "Book 2,") and thus forcing the customer to pay twice to get the full story.
The unfortunate downside is that there is a delay between updates.