Quality and ergonomics page's image titled Feature Logo End to End High Quality 128w 128h

Quality and ergonomics

From the design phase, through production and delivery, we focus on quality and ergonomics, so you can focus on creativity.

Digital ergonomics

Split sound files

Every one-shot sample comes into a separate sound file.

Drag-and-drop import and exchange sounds with perfect sync, a productivity booster for any workflow.

Game audio

Easily Import and use procedurally

Quality and ergonomics page's image titled Still Fmod import 360p

Film sound

Align visuals with audio fast

Quality and ergonomics page's image titled Still DaVinci fast insert into markers 360p

Audio sequencers

Sync and automate with precision

Quality and ergonomics page's image titled Still Reaper replace sound in item 360p

Music production

Create your signature sound

Quality and ergonomics page's image titled Still Battery multisample import 360p

HDR end-to-end

From acquisition to production to delivery

A common problem with many audio assets out there, is that are made with inferior pipelines, using processors and algorithms that do the barely minimum.

In sound design, this is a problem, as we are called to layer many sounds and processors to creatively alter our audio signals.

Our pipeline is build with hardware and software that uses 64-bit floating point calculations end-to-end, featuring 32-bit IEEE floating point for our acquisition and intermediate files, in ITU RF64 raw PCM broadcast wave file containers.

This gives us power to manipulate our sounds to extremes without sacrificing quality for creativity. It is the equivalent of using raw camera sensor data when working with textures of many megapixels.

Any transformation on the time, amplitude, and frequency domains, produces a natural sound with no added artifacts. This is very important in sound design workflows, where many transformation algorithms are used in-series to shape new sounds.

In such a high-end pipeline, transforming sounds, adding creative effects, and compositing with countless layers, is free from any technical problems, so we can be free to focus on the creative.

This is a chart showing the difference of resolution steps available as we raise bit depth in digital audio signal processing. If it was a texture image it would be the equivalent of having more steps of each color channel, to translate the original color on the digital domain.

Quality and ergonomics page's image titled Icon for Success
That means that we achieved a production pipeline that provides us with more than 1500 dB of dynamic range, to be precise 1528 dB.

To put that on perspective:

  • The launch of a Saturn V rocket can reach 220 dB.
  • The infamous Tsar Bomba reached probably 224 dB, and accounts for the largest human-made noise. Until then, it was a Manowar concert.🤘
  • The eruption of the Krakatoa volcano probably reached 230 dB on 27 August 1883. This was so powerful that the sound wave circled our planet four times.
  • The Tunguska meteoric event of 1908 is calculated to have reached 300 dB. This massive explosion flattened 80 million trees over an area of 2000 square kilometers.

So, you can understand that 1528 dB of dynamic range, is more than enough to handle any kind of acoustical sound that will ever be captured in a computer file.

Up to 1536 KHz render precision

Saturation, harmonizing, distortion and other non-linear processors used as creative tools in sound design, generate higher harmonics on the signal. Those processes are known to introduce what is known as aliasing, a common artifact of the audio signal with negative results (we mean bad sound).

To counter that quality degrading effect, we perform our processing with oversampling oversampling, which minimizes this error to its lowest possible value. It has a big impact on the CPU and it takes to render forever, but we gladly sacrifice time on our render farm to provide the best.

Although good for harmonic adding processing, oversampling can be inferior by itself and lower audio quality at higher sampling rates, at SoundFellas we use our brains and our ears, and double-blind critical listening tests to determine the best settings for each case.

96KHz / 24-bit mastered deliverables

Our audio assets are finally mastered to 96 KHz / 24-bit.

Apart from the numbers, what we achieved with our scientific precision is to offer you sounds with high contrast between the foreground and background features, not over-compressed, with enhanced detail on spectral nuances, in a format that will play on all cases and can be further converted to your final product with ease.