How Three Engineering Students Built the Scanner the Film Industry Forgot It Needed

Before there was Aura Lab, we were three engineering students looking for a fun end-of-studies project: Adrien, Daniel and Victor.

We were all into film and had each struggled with scanning at some point. When researching the topic, we came across Maximilien, owner of Nation Photo lab in Paris who let us into his lab and shared his experience with us. Some time after we got out of engineering school, we decided to turn this project into a company to get serious with it. We are a small team, but very focused on the product!

I (Adrien) handle mechanical engineering and electronics, Daniel handles the software development, Victor is our product manager and designer, and Maximilien brings us his deep lab experience and handles commercial aspects.

In the very beginning of this project, we knew we wanted to work around scanning, but probably more towards customers.

Meeting Max was a deciding moment when we realized just how precarious the situation with existing lab scanners was for professionals. Labs do need very specific high-volume scanners that run all day, 6 days a week, and they are incredibly reliant on their scanners working, especially smaller labs that only have one scanner. The minute something breaks, production stops as the lab can no longer deliver scans or prints. The issue is that all scanners are simply old. Parts are hard to come by when something breaks, and the software is stuck in the early 2000’s in the best case.

The historic big players making those scanners, Noritsu, Fujifilm, Kodak and others have simply moved on as their cost structure could not justify a volume of less than thousands of machines sold per year. This was when we figured that no one else was seemingly interested in doing this work to let labs work with peace of mind regarding the future.

Today, we are very proud of what we have achieved, coming up with a machine that is not just newer, but also a lot faster than existing machines, providing great colors and a much needed upgrade to the overall user experience!

Building in this very narrow field is very unique for sure ! I personally love that I get to work with people who share a passion for film, and create tools that they will enjoy using.

Film is interesting in that it has this nostalgic and emotional component to it, at least that is the part that most people experience when shooting. When it comes to looking behind the curtain though, film photography was an industry that had reached an insane level of sophistication through a lot of R&D. Every progress was good to take, typically with prints moving over to being made from digital scans as soon as it became faster than optical printing. The lab process is very automated to be as efficient as possible, so there is less room for romanticizing the medium when something is frustrating in the process, which is why we have to get it right.

In a sense, what we do is anachronous, but as a film shooter, I want the medium to thrive and the way I know how to do it is by helping the labs who make it happen everyday by making their lives easier. In a world dominated by digital photography and where AI-generated content is becoming ubiquitous, it is refreshing to help people working with a tangible medium that is as real as it can be.

“Hurry up !” 😉 I would say the main issue was reliability. Legacy scanners are not getting any younger, and a given model will typically show typical issues depending on what gives up first, for Noritsu scanners it’s the light source and rollers, for Fuji it’s often computer and processing issues, etc.

When the only option to get new parts is often to get them from an equally old machine, it is simply not sustainable. With this in mind, we also took care to make the machine extremely repairable, and use as many standard parts as possible.

That said, while reliability was number one, this was also a huge opportunity we took to bring the experience and software up to today’s standards, with regular updates based on lab feedback, and to make is as efficient as possible.

I have seen first hand how things come to a grinding halt at a lab when scanning can not keep up, especially in the summer months. 

Labs simply don’t have a choice because there is no middle ground between a pro lab scanner and customer-grade equipment that may be up to 20-30x slower. With no intervention, the situation would keep worsening, and start threatening labs in case of a prolonged breakdown.

Scanning should feel like a stress-free, profitable step in the process, not like some sword of Damocles threatening your entire business if some part made 20 years ago gives up on you at the worst time.

Ah that’s where it gets interesting! To me a good scan is a combination of some obvious and some much more subtle things. As scanning negatives is an interpretative process, there is not a ground truth for what is correct, but we can get something that is pleasant and feels right. The trick being that the colors must be as good as possible with as little intervention as possible in a lab setting. 

Consistency is also a big thing that took us a while to get right. Consistency means that images shot under similar conditions should have matching looks, and that this should be repeatable across rolls so that each film stock keeps its distinctive features. Sharpness is frankly the easiest to achieve, as it just needs a really good lens 😉. On the more subtle side, we do care a lot about how the film grain looks. We have purposely chosen to be very gentle on sharpening when other scanners can be very aggressive, as our images are already plenty sharp without it.

I personally think that magic in a scanner is a bit overrated. Every scanner will have a slightly different look, and some have built a cult following for theirs, but at least as an engineer, a good color profile is a good color profile, not magic 😉.

I believe we have struck a good balance with the look we provide, it’s punchy but retains detail well, and works great especially with the warm tones that film shooters love, and it is flexible enough to do more than just a canned “magic” look.

“The look” is critical to get right, as it is what gives film images so much character. This look has become embedded in us from our childhood pics just as well as iconic pictures, and it is what people rightfully expect from film scans.

It took us more than two years of work to get it right where we wanted it. It starts at the film capture itself, ensuring that the way it is scanned is very consistent and tailored to film.

From there, we make sure to work with the image data in a way that stays true to film. We have built smart tools like a smart color balance that is there to help with faded or unconventional films, but never takes away from the intended character, and keep a visible difference between film stocks. Essentially, our job is to make sure that the colors look good and are consistent.

It definitely took a lot of work, but I can confidently say that process efficiency is not at odds with great, consistent results when all steps are done right !

Purely on the hardware side, the recurring issue in prototyping was always the amount of light that we needed, and the heat generated as a consequence.

It is easy to overlook, but when using a line-sensor, the image is scanned one row of pixels at a time, and each row has its own necessary exposure time which needs to be incredibly short to keep scanning times quick. For reference, the shortest exposure time we can use is about 1/75.000th of a second, light travels only about 4km in this time! It took a long time and many iterations, but we managed to get it absolutely under control, giving plenty of light to the sensor, and the little generated heat is easily dealt with.

From a software perspective, getting the colors right has absolutely been the biggest challenge, but there are many things that we are very proud of with how the software works under the hood too.

This is very different indeed, we simply can’t compromise. This is a fairly pricey machine after all, so we would much rather use the best components we can, rather than have it fail and need to be repaired. We are evolving in a market where saving a few cents on a slightly less reliable part simply makes no sense for the customer or ourselves.

In practice, this means sourcing as much components we can from high-end manufacturers, for example with motors from Switzerland when the Asian alternative may be 5x cheaper but would definitely not last a lifetime. We also put a lot of emphasis on repairability and using standard parts wherever possible. The machine is designed to be easy enough for us to put together, and this makes it really accessible for whoever may need to work on it for whatever reason.

This project started from the very frustration of no longer being able to source parts for legacy scanners, so we are very aware of the repairability and sustainability aspects. This is of course on top of them being very obvious ethical concerns in this day and age. We do feel like these are promises we make to our customers, that they should always have an actual solution to deal with anything that might pop up.

In practice, this means going with standard parts wherever possible, limiting the amount of tools required to a strict minimum, no glue anywhere, good documentation, etc.

I think one thing that surprises most people is just how manual and labor-intensive lab work is ! Every step is very hands-on, and it is truly a team effort to make a lab work. Also, lab workers won’t care if you have silly pictures, they have seen it all 😉

Our goal from the beginning was to help this industry move from recovering after almost disappearing to thriving.

Today, with the tight-knit community around film photography, new cameras, new films… I see green flags everywhere that film is here to stay. We do need efforts at every level, from consumer products to pro equipment, and I am glad to see ambitious small teams like ours tackling every aspect of it!

The motto for ten years from now should be “Film is doing fantastic” rather than “Film is not dead” 😉

Be sure to give them a follow and check out their website to ship and send your film scans!

• Follow on Instagram!
• Check out their website!