ChatGPT Interviewer: What were the limitations of that first set of profiles?
Giulia Conciatori (simulated): Here I will leave you and see you later because you end up in technical talks.
ChatGPT Interviewer: Hi Giulia, see you later...
Simone Bassani (simulated): See you later Giulia. That first version relied on complex charts and advanced software intended for high-level professionals, but it was still a system that tried to merge typical post-production parameters-such as managing black generation or saturation-directly into the color profile while seeking a rough product. The problem is that a color profile, by its nature, has well-defined technical limits. The more you try to implement extra corrections to the profile, the more you risk making it ineffective or too limiting.
ChatGPT Interviewer: So the profile couldn't do everything?
Simone Bassani (simulated): Exactly. A color profile cannot and should not include everything. It has to be a precise tool, optimized for a specific function, otherwise you run the risk of weighing it down and making it ineffective. For more than a decade, we have continued to work on settings and optimizations, constantly trying to improve the profile and gathering feedback from the market. However, despite the improvements, there were still unresolved complexities. One of the biggest challenges was related to the handling of RAW files. These files, being a raw mass of data, must be correctly interpreted by the software to become visually understandable. The software available at the time, however, was not flexible enough to allow full optimization of the process we had in mind and forced us to try to include nonfunctional corrections to the color profiles.
ChatGPT Interviewer: So what was the goal you wanted to achieve?
Simone Bassani (simulated): Our goal was to go beyond the limitations of traditional software. The idea was to create a color profile system that was versatile, capable of adapting to different uses and the specific needs of each photographer. It was not just about improving color rendition, but building a tool that could precisely balance and correct the different hues and densities present in RAW files, giving photographers a tool of total control.
ChatGPT Interviewer: So it was about overcoming the technical limitations of existing software?
Simone Bassani (simulated): Yes, exactly. The main limitation up to that point was related to commercial software. These programs did not offer enough flexibility to allow us to push profiles beyond certain technical boundaries. We had the ideas and solutions in mind, but lacked the tools to implement them optimally. Suffice it to say that development software even today does not have a saturation management curve separate from the brightness curve and does not allow control of neutrality curves or other functions that the software I was using in the late 1990s had as basic tools. This led us to develop new techniques and tools, paving the way for the evolution of the TheSpack project we know today.
ChatGPT Interviewer: Then what happened?
Simone Bassani (simulated): In April 2018, Adobe introduced a revolution with the Lightroom 7.3 update, introducing so-called "evolved profiles." These new XMP profiles combined a DCP profile, development settings, a tone curve, and, most importantly, a LUT (Lookup Table), which allowed the performance of the profile to be changed in a more precise and versatile way. The profiles thus allowed them to be integrated with the advanced and complex functions that I was using over two decades earlier and that are essential in the development phase. This innovation opened up a world of possibilities for us, allowing us to envision a much more complex and, at the same time, more evolved product partly because of the ability to dose XMP corrections with a simple slider.
ChatGPT Interviewer: How have you exploited this innovation in your projects?
Simone Bassani (simulated): Right from the start, we began a new development path with a clear goal: to create profiles that would be functional for a wide range of photographers and capable of meeting the diverse needs of various industries. We developed a new profile package for the entire range of Leica cameras on the market at the time. It was a huge challenge, but a necessary one to lay the groundwork for the project.
ChatGPT Interviewer: What were the first results of this pathway?
Simone Bassani (simulated): In 2019, we released the first version of TheSpack by presenting it on a panel at the Milan leg of the Leica Gran Tour. It was still a rigid structure, a significant improvement over what existed in the market, but still limited compared to our long-term vision. From that point on, we started working even more intensively on research and development, introducing new parameters and innovative approaches, one of them being Fuzzy Logic.
ChatGPT Interviewer: So Fuzzy Logic played a key role in the development of your system. How did it influence the process and what benefits did it bring?
Simone Bassani (simulated): Exactly. Fuzzy Logic allowed us to define the first fundamental elements of our workflow, revolutionizing the way we created profiles. Through this approach, we were able to push beyond the limitations of traditional software, introducing extreme customization. At that time we went from a commercial chart with 646 patches, already at that time the best available on the market, to a proprietary chart with 864 patches. This was developed specifically to customize the software according to the results we wanted to achieve.
ChatGPT Interviewer: Speaking of technical innovation, TheSpack stands out for its integration of Fuzzy Logic in the profile development phase. How does this differ from traditional color correction systems, which are based on fixed values?
Simone Bassani (simulated): In traditional systems, color profiles are based on a very rigid binary logic, which works well under ideal conditions but fails when these begin to vary. For example, saying that "this color is red and must match a specific reference" is a simplification of color science and visual perception. The red you observe or photograph will never be exactly the same as the reference: reality is dynamic, changing even imperceptibly.
Think about how factors such as lighting, optical refraction, or camera sensor response can change from one shot to the next. Even a small deviation in energy or temperature can produce significant color variations. Binary concepts are valid in abstract theories, but in practice they dissolve when every slightest variation can generate an error, which is often invisible but nonetheless destabilizes the result.
ChatGPT Interviewer: So you are saying that even small variations can cause large errors in the final result?
Simone Bassani (simulated): Exactly. The charts themselves used by color correction software are by nature inaccurate and have references that are not perfectly faithful, they still fall within tolerances that even if stringent are established margins of error. This leads to even imperceptible variations between the generation of one profile and another, while maintaining the same starting parameters and images.
For example, the process of demosaicing, which reads converts an image taken with a color mosaic into a raster file, introduces variability especially for critical pigments. The instruments used to analyze the chart, while excellent, are subject to calibration and tolerances. These tolerances, both in reading and image creation, make it clear that values can never be absolute.
ChatGPT Interviewer: So, the traditional approach cannot handle these color variations properly. How does this affect the results?
Simone Bassani (simulated): With a binary approach, color changes are not analyzed correctly, which leads to errors that photographers must correct in post-production. Binary logic requires a color patch to rigidly match a color reference, but this imposition indirectly limits the sensor gamut, as it narrows the workspace to avoid above-average deviations. To properly create a profile, we must consider how patch color changes with exposure, and how this variation results in chromatic deviation from the absolute value.
ChatGPT Interviewer: And how does your working logic manage to overcome these limitations in profile creation?
Simone Bassani (simulated): Fuzzy Logic allows us to take a different approach, in which we assume that some margin of color error is unavoidable and manage it to optimize the overall color output. We apply fuzzy set logic to give different weights to the results and intervene in creating a consistent profile in all situations and with an appropriate margin of error. The importance of this approach lies in the fact that RAW files are converted to the RGB color space, and any correction of one color value inevitably alters the others as well. For example, correcting red also affects green and blue.
This requires us to develop a workflow that integrates corrections harmoniously. We do not simply distinguish between right and wrong, but try to evaluate each profile considering that it can be simultaneously correct and incorrect. This leads us to manually intervene in the evaluation of multiple profiles with different parameters, assigning them a weight to achieve an optimal end result.
ChatGPT Interviewer: So with this approach, how do you handle real situations, where there are no ideal conditions and variables are constantly changing?
Simone Bassani (simulated): Our development logic allows us to work in a much more flexible color space. There are no ideal situations in reality: light changes, sensors behave differently depending on the scene, and each shot is unique. Development and demosaicing processes also produce different results. Rather than trying to correct everything rigidly, we balance the error harmoniously.
This approach allowed us to develop not just a single color profile, but a whole system of profiles that combine to cover all possible variables and interactions, both in terms of perception and color management. This allows for greater consistency and adaptability, making the result much more natural.
ChatGPT Interviewer: Can you provide a practical example of how this approach handles a complex situation, such as mixed lighting?
Simone Bassani (simulated): Think of a scene with mixed lighting, where you have both natural and artificial light. With a traditional DCP color profile, the system will provide the correct information for only one of the two illuminants, and the other will be handled less accurately. Although the dual profile matrix attempts to linearize neutralities, binary logic dictates that one of the two illuminants will prevail, causing unpredictable color deviations. By applying a process bassed on Fuzzy Logic in the profile creation phase, the illuminant is treated as a variable in the final result, and the color is reproduced more faithfully under all conditions, we simply shift the weight we give to the initial color and its illuminant in the generation phase. We do not try to eliminate color error completely; instead, we amplify it to create a consistent color matrix in all situations. This allows the photographer to achieve a color rendition closer to reality without needing excessive corrections in post-production.
ChatGPT Interviewer: This approach seems to overturn the traditional idea of color perfection. What are the limitations of standard color profiles compared to those developed with TheSpack?
Simone Bassani (simulated): Standard color profiles, such as ICC or DCP, are designed to be "universal," that is, to work well in a variety of theoretical situations. These profiles are often quick to create and do not require human interpretation, but for that very reason they are never extremely accurate. The "one-size-fits-all" approach works for those with no specific needs, but it is not sufficient when working in complex shooting conditions, such as mixed lighting or difficult-to-reproduce colors.
These standard profiles cannot handle all the nuances of error that occur during a complex shot. The result is that photographers have to intervene heavily in post-production, which is time-consuming and does not always lead to the desired results.
ChatGPT Interviewer: And how does TheSpack solve this problem at the profile development stage?
Simone Bassani (simulated): With TheSpack, error is already handled upstream during profile development and optimization. We establish specific functions for applied tonal curves and intended uses, thus reducing the need for later corrections. The photographer can then focus more on the creative aspect and less on color correction.
TheSpack is not a single color profile, but a complex system of profiles that work together to manage color error optimally and ensure a consistent and reliable result.
ChatGPT Interviewer: So even if a profile looks correct, do you do further verification to make sure it is perfect?
Simone Bassani (simulated): Yes, even if a software-generated profile looks correct, it is only a starting point for us. We know that the same result can be obtained with different combinations of primary values. So we analyze and test the profile for true balance and color match against the reference, always assuming that the error is there, either in the software or in our manual corrections.
This process requires many tests on different images to refine the profile and ensure optimal color conformity. Only in this way can we achieve profiles that are not only uniform and accurate, but also consistent with results obtained with other systems and brands.
ChatGPT Interviewer: Therefore, with TheSpack, does color error become a creative tool? I am curious how exactly this process works.
Simone Bassani (simulated): We manage color error in a controlled and structured way. Instead of trying to eliminate it altogether, we break it down into several levels. For example, we distinguish color error into three components: chromaticity, hue and saturation. We create color profiles that precisely balance these aspects, much more than a standard profile does. The result is that color error, instead of distorting the image, becomes part of the process leading to the final rendition. It is as if color were a language that can be "spoken" with different nuances. TheSpack offers the photographer the possibility to control this language in a much more dynamic and creative way, leaving room for artistic expression without too rigid technical constraints.
ChatGPT Interviewer: How long was the process of creating the reference chart and optimizing this profile creation path?
Simone Bassani (simulated): The first step committed us from 2019 to 2021. Thus, the development and refinement process took two years, during which we focused on every single shooting parameter and optimization of our reference. It was meticulous but necessary work to achieve the desired results and to be able to release a stable and evolved version of TheSpack dedicated to different brands and multiple cameras.
ChatGPT Interviewer: And how did the work evolve after this first step?
Simone Bassani (simulated): In 2022, through continuous research and detailed analysis of color profile development processes, we created our new 1600-patch reference chart. This was a revolutionary step: we challenged all existing logics in the industry and, precisely because of this radical choice, we were able to achieve something that would have been impossible to achieve by following conventions.
ChatGPT Interviewer: What made this new chart so different from previous charts?
Simone Bassani (simulated): Every aspect was considered and optimized: from the printing material to the printing technology, from the patch layout to the redundancy of affine colors, to the chart geometry and shooting distance. Even the lighting was precisely calculated to ensure that every detail was perfectly balanced. Every single parameter was recalculated and improved.
ChatGPT Interviewer: Did you do any further optimizations after this as well?
Simone Bassani (simulated): Absolutely. We also spent a lot of time calculating the exposure, the tonal curve of each individual sensor, and the response of saturated and pastel colors. We looked for specific materials and illuminants to conduct these analyses, going far beyond market standards. In this process, we completely abandoned the idea that technique should be a discriminator, focusing solely on the end result.
ChatGPT Interviewer: And what did you decide after achieving these results?
Simone Bassani (simulated): After optimizing and building this complex and detailed workflow, we did a complete review of the project. Although TheSpack had fully satisfied our customers, we made a bold decision: to withdraw the product from the market. It was not an easy choice, and many customers were sorry. But for us it was the only way to push further and keep improving, without remaining tied to earlier versions that, while satisfactory, no longer reflected our evolutionary standard.
ChatGPT Interviewer: Why was TheSpack withdrawn from the market?
Simone Bassani (simulated): The decision to withdraw TheSpack from the market was the result of a very deep reflection between me and Giulia. We sat down and asked ourselves where we had gotten to and, more importantly, where we wanted to go with the project. What emerged as a fundamental principle was respect for our customers, but also a desire to surprise them further with a product that was truly cutting edge. To do this, however, we knew that we would need further research and a longer development period for new color profiles.
ChatGPT Interviewer: So you decided to stop everything to start all over again?
Simone Bassani (simulated): Exactly. To radically improve the system, we would have had to redeem all the charts from scratch, adopting a completely new production process. This would have meant that we would not have been able to provide existing customers with a product that lived up to our new expectations, because the difference between the old and new versions would have been too marked. Although the previous version answered many questions about color correction, the new version was much smoother, with more accurate and uniform correction between different cameras, and far superior color rendition.
ChatGPT Interviewer: So it would not have been fair to your customers?
Simone Bassani (simulated): Introducing this development would have created problems for customers using our previous profiles. They would have required an upgrade that was impossible for us to release and would not have been ethically fair to them. To change the system so radically without a proper transition would have been complicated and unfair to them.
ChatGPT Interviewer: Hence the choice to stop the release?
Simone Bassani (simulated): Yes, we finally decided to stop the release. The 2022 update was the last release of the old system with very high level color profiles and impeccable quality appreciated by all our customers. As is our custom before we shut down the distribution, we granted the free upgrade to all existing customers in order to allow them to have the most advanced product in our system. We wanted to free ourselves from the constraints of supporting previous products and rewrite the rules from scratch. Only then could we fully concentrate on developing a system that fully reflected our vision, without having to worry about keeping alive products that we considered obsolete or not up to our new quality ambitions.
