Digital signal processing (DSP) can be separated into two categories: fixed point and floating point.

There are a lot of discussions and information available on the two counterparts, yet the concepts are still confusing to many in understanding their differences and how those differences affect real-world products. The main purpose of this article is not to go too deep technically, but instead to explore the difference of the numeric representations in a simple yet intuitive way. Since it is only fair to compare an apple to another apple, we’re going to use the same 32-bit length for both notations in this article as shown in Fig. 1 (below).

For Fixed Point Notation:
Integer Value = -1Sign x Bits
Sign = 0 (Positive value) or 1 (Negative value)
Bits = 231 possible values
Positive Min: 1
Positive Max: +2147483647

With simple scaling, fractional numbers can be represented as well. The represented values are equally spaced across the whole range. The gaps between adjacent values are always the same.

For 32-bit Floating Point (IEEE 754):
Value = -1Sign x [(1+ Mantissa) x 2(Exponent-127)]
Sign = 0 (Positive value) or 1 (Negative value)
Exponent = 1 to 254 (0 and 255 are reserved for special cases). Subtracting 127 gives -126 to 127
Mantissa = 0 to 0.999999881 for all ‘0s’ to all ‘1s’. Adding 1 gives 1.000000000 to 1.999999881
Positive Min: 1 x 2−126 ≈ 1.2 × 10−38
Positive Max: 1.99999881 x 2127 ≈ 3.4 × 1038

The represented values are unequally spaced between these two extremes, such that the gap between adjacent numbers is much smaller for small values and much larger for large numbers. This notation “steals” eight bits to become the exponent, which gives the extensive dynamic range, but to gain this benefit, it loses the “stolen” eight bits’ resolution for all values.

Fig. 2 (above) illustrates how the two notations differ in their numeric representations. In this illustration, we just focus on positive numbers (Sign Bit=0) to avoid complications. Then the eight most significant bits in the standard fixed point format are used as the scaling index to match the number of bits of the exponent in the floating point format. This leaves the remaining 23 bits to be matched for both formats. Once the bit-matching is accomplished, we can then generate meaningful results.

Exponent e can be any arbitrary number as long as all blocks are within the min/max range (-126 to 127). The Index=1 is mapped to e once the desired data range is chosen for the fixed point computation. The graph shows that both notations have the same 23-bit resolution at i=1. For i=0, the resolution of floating point is better; in fact, tremendously better. There are e+126 floating point blocks, while there is only one fixed point block (i=0), both covering the same numerical range.

For 1<i<256, the resolution of fixed point is better. Although it seems floating point loses ground at higher values, keep in mind that the fixed point counterpart can only have a total of 256 (28) blocks with 23-bit resolution per block. Once the value goes beyond the upper limit, the developer is forced to scale up further and live with a lower resolution in order to be able to carry out the assigned mathematical task. This is why floating point is dominant when the computation values are small, and can handle large numbers when fixed point cannot. To make things more complicated for fixed point DSP developers, they have to handle overflow and truncation errors as well.

In Summary:

Advantages for Floating Point:

  • Extremely large dynamic range
  • Applications with intensive computations
  • The smaller the number, the higher the precision and lower quantization noise
  • Possible computation for large numbers, beyond fixed point’s capability
  • Simpler and faster development without worrying about overflow and truncation errors

Advantages For Fixed Point:

  • Better resolution within a narrow range of 1<i<256 (7 out of 256 floating point Exponent blocks)
  • Simpler DSP silicon

Donny Chow is the Founder and Chief Engineer of Xilica. Following its motto of Passion Through Performance, Xilica manufactures class-leading networked and standalone DSPs for live and installed sound applications.
Article published by Professional Sound Magazine.

The ability to achieve flexible control of your DSP as well as a panoply of third-party devices is among the primary benefits of the new Universal Control feature. The latest addition to the Xilica Designer configuration and control software for Xilica’s Neutrino and Uno Series DSPs, Universal Control has been developed with an emphasis on accessibility and ease of deployment.

To this end, it requires no specialist third-party control protocol knowledge or secondary software applications to add control for non-Xilica devices. Instead, it allows users to take control of their devices using the same, one-step drag & drop configuration familiar from the configuration of Xilica’s own wired and wireless remote controls.

“The whole focus with Universal Control is on allowing users to bring different brands and products under effective centralised control, without having to call on the services of specialists to create a workable configuration and therefore incur unwelcome delays and expenses,” says Xilica Regional Sales Manager, Barry Steinburg.

Xilica identifies two primary project types where Universal Control is destined to resonate. The first is in major projects that require control of the DSP equipment as well as expensive building automation control systems. In such instances, the hardware and software are typically very costly, as are the specialist programmers required to set-up and optimise these complex control systems.

The second target area is smaller projects where individual control systems would be an option, but an overarching control system such as Universal Control would provide a far more cost-effective, inclusive approach – and one that again does not necessitate the engagement of specialist programming support.

Steinburg outlines a typical scenario whereby a user “looks up a projector brand in the Component library menu and locates the specific model; drags and drops that model into the software’s view work area; clicks on the dropped module to open it; then finds all of the available controls for that unit. At that point the user simply needs to drag any or all of that device brand’s controls and drop them into any of our wired or wireless controls. It really is that simple.”

Once the control information related to third-party devices has been saved to the project file it is possible to control all of these devices from any of Xilica’s remote controls. The end-result is a control platform that affords users a tremendous degree of ease and flexibility, and one that can obviously be adjusted as the requirements of a given installation change.

Universal Control is part of the Xilica Designer DSP configuration and control software that replaced Xilica’s previous SolaroConsole software. It is available to download now from the Xilica website and provides many operational improvements in speed, features, looks and layout. As well as Universal Control, Xilica Designer also includes two other primary universal components: Universal Dante Configuration and Universal Platform.

“Universal Control is a very important addition to the Xilica range and is one that we believe has the potential to be a real game-changer,” says Steinburg. “Avoiding the need to invest in expensive third-party control systems, or opt for lower-value controllers that may afford basic overarching control but which still require specialist programming, Universal Control by its very nature offers an inclusive and very appealing approach to control. It bears repeating that absolutely no expert knowledge is require to implement Universal Control, and hence no costs will be incurred when deploying the system to take control of third-party devices. It really is as simple as drag & drop & save, and in the time-sensitive nature of today’s installation business that has to be a major asset – as well as one whose benefits are easily conveyed to clients and end-users.”

The extensive requirements of contemporary worship – which often includes performances by sizeable live bands and choirs, as well as the need for crystal-clear audio for sermons and background music – mean that many Houses of Worship (HoWs) now routinely invest in sophisticated DSP systems.

Xilica has certainly been a beneficiary of this approach, with its Neutrino and Uno Series products proving to be popular choices for HoW applications.

With such an abundance of DSP solutions on the market, it’s important for each HoW to select a system that is capable of handling their present needs as well as affording capacity for future expansion. Partnering with a specialist in HoW audio can make all the difference to a project, and it was with this in mind that Xilica spoke to Mike Sessler, who is a Project Lead at Olympia, WA-based CCI Solutions and runs the ChurchTechArts blog, in order to compile the following ‘top ten tips’ to be borne in mind when selecting an HoW DSP…

1) Make sure you have the appropriate I/O. It might seem obvious, but ensuring that a DSP system affords sufficient I/O to support multiple audio sources is absolutely pivotal to the success of a project. In Sessler’s experience, “in most HoW projects we’re not only providing DSP for the main PA, but also ancillary spaces such as lobby areas, green rooms etc.”

2) Input count is not usually a problem – but ensuring there are enough outputs is critical. In this regard, the use of Audinate’s Dante media technology – which is supported extensively in the current Xilica product range – can provide an attractive way forward. “In many Dante-based systems using Dante to analogue output boxes is a cost-effective way to leverage DSP horsepower and still add more outputs,” says Sessler.

3) If you are using a digital network, a DSP with that digital interface can be very useful. For example, says Sessler, with more and more HoW projects incorporating Dante for audio signal transportation, “being able to plug the DSP into the switch makes audio routing very easy. It also saves on unnecessary DA and AD conversions.”

4) Don’t over-focus on processing power. The fact is that processing power is often a secondary concern given that most DSPs on the market today will have sufficient capacity to handle the system tuning and distributed audio requirements. It is only for the larger-scale projects where processing power will need to undergo greater scrutiny.

5) Not all software is equal. Some is much more intuitive than others. And at the purchasing stage, the ability to demo software can be extremely useful if the choice comes down to two or three units.

6) The ability to customise parameters is very useful. Many HoW customers will require standard presets that allow them to change basic parameters without the potential to cause disruption to the basic day-to-day operation of the DSP. “Speaking as an integrator I would opt for a software package that would allow me to build custom parameter set pages that let the customer see what they need to see without letting them break things!” says Sessler.

7) Effective control is a must. “Any DSP that cannot be controlled and completely configured via a computer software interface is not going to be considered. The ability to save, recall and archive configuration files is a must, so if a DSP can’t do that, it’s off the list,” says Sessler.

8) Avoid unnecessary conversion. Streamlining the workflow ensures that audio quality remains high and over-complexity is avoided. “A lot of people think that AES inputs and outputs are necessarily better because they’re digital,” offers Sessler. “But if you have to convert from something else to get to or from AES, it not only defeats the purpose but introduces conversions that don’t need to happen. Match the I/O of the console and amps to the DSP.”

9) Remote control capability is crucial. The ability to add remote control stations can be a significant bonus, particularly in larger HoW installations where audio is being distributed to various worship spaces, back-offices, and so on.

10) Engage specialist support from the beginning of the project. Successfully implemented, modern DSPs can bring tremendous flexibility to HoWs. Selecting a specialist installer or consultant can make all the difference, and fortunately the number of those with a focus on the specific requirements of HoW audio continues to grow.

The shift from point-to-point connectivity towards fully networked solutions has been one of the great narratives of professional AV since the turn of the millennium. The ability that networking affords to move audio and video flexibly around a building, or within a broadcast environment, has meant that more and more customers are now willing to enter into this brave new world.

Although networking technologies that require the installation of new switches have gained trac-tion in some areas, it is IP-based technologies that are able to use existing infrastructures that have enjoyed the most rapid adoption. None more so than Dante, which was developed in 2006 by Sydney, Australia-based Audinate and delivers uncompressed, multi-channel, low-latency digital audio over a standard Ethernet network using Layer 3 IP packets.

Improving upon previous audio-over-Ethernet technologies such as CobraNet and EtherSound, Dante’s remarkable trajectory in recent years owes much to its enthusiastic support from third-party developers and licensees, allowing an extensive community of products to be established. As of 30 June 2017, the number of Dante-enabled products on the market had increased by 36% to 1,182 when compared to the total of 872 products available at the same point in the previous year. Simultaneously, the number of OEMs adopting Dante has also continued to grow, rising from 310 to 369 in the year ending 30 June 2017.

Although adoption of Dante continues to grow across a number of verticals, commercial install remains the largest market for Audinate. It is in this space where the benefits of audio networking can be seen to be particularly acute, including: the cost and time savings from leveraging existing Ethernet cabling; the ability to route audio over long distances without signal degradation; and the flexibility of routing audio with the click of a mouse in software instead of changing out physical cabling.

We spoke with Joshua Rush, Vice-President of Marketing, at Audinate, who believes that there are several primary factors informing the ongoing Dante success story. “Firstly, and most im-portantly, is that Dante was thoughtfully engineered from the ground up to make it easy to use. Features like automatic discovery of devices, native device naming, and the Dante Controller software all contribute to a literal ‘plug and play’ experience. Secondly, we’ve had a wave of products that have come available on the market. We know from our research with integrators and end customers that the number one criteria when selecting a networking protocol is the number of products available on the market. Over the past three years, Dante has become the clear leader in available products, eclipsing the next closest networking protocol by nearly 5X the number of supported products.”

The other major piece of the puzzle is market adoption, which has been encouraged by a significant focus on training and education. Rush continues: “We recognise that in this industry our customers are used to being the technology experts, and while we pride ourselves in how easy Dante is to use, we also want to make sure integrators and end customers have the technical knowledge to give them peace of mind when selling and supporting networked audio solutions. To that end, we’ve rolled out an extensive, three-level Dante Certification course that is available both in-person and online, and have had nearly 20,000 people participate in the programme to date.”

Xilica has been an Audinate partner since 2011 and offers extensive Dante support in product families such as Neutrino and Uno. With the latest Xilica Designer software, Xilica has integrated control for a growing number of Dante OEM partners — meaning that the end-user experience is improved right from the initial installation of the Dante device to the physical routing of audio.

Professional Sound Magazine features an exclusive profile on Xilica’s Chief Engineer, Donny Chow.

Special thanks to Professional Sound Magazine‘s Editor-in-Chief, Andrew King, for including us in Professional Sound Magazine’s December issue.

Growing up, Donny Chow had a curious fascination with electronics – specifically, electronic musical instruments and sound systems. It was that evergrowing interest that eventually led him to leave his longtime home of Hong Kong, where he was born and raised, and travel to Canada to continue his secondary and then post-secondary studies.

After high school, Chow enrolled in the electronic engineering program at the University of Western Ontario in London, ON. During and following the completion of his studies, he witnessed first-hand the analog-to-digital domain revolution of the late ‘80s and ‘90s and kept pace with the non-stop technological innovations it spawned.
Having earned his degree at Western, Chow stayed in Canada and landed his first full-time job with Adamson Systems Engineering in Port Perry, ON. “That was my first taste of the pro audio industry,” recalls Chow. “I was a member of Adamson’s engineering team, pioneering the development of the company’s first DSP loudspeaker controller system.”

In 1993, Chow founded Arctic Design (the name being something of a tribute to his very seasonal new country of residence) and went into business for himself. Over the next decade, he and the company completed several performance-critical custom design projects for major award-winning Broadway productions like Show Boat and Sunset Boulevard in addition to a host of other theatrical and live event applications.

Then, drawing on his years of experience designing and developing new technologies, Chow founded Xilica Audio Design in 2002. He and his team have since been producing high-end networked audio DSP systems and control solutions for the pro audio and integration markets. Earlier this year, the company celebrated its 15th anniversary of operations and also partnered with Contact Distribution to boost the brand’s presence and profile in the Canadian market.

Currently, the team at Xilica is building on the success of its existing XP, XD, Uno and Neutrino series of DSP products with the next-generation Solaro series. “We’ve bumped up the channel counts, improved the Dante capabilities, and redesigned the modular cards methodology,” explains the company’s director and chief engineer.
Another product currently in the works is the Universal Media Controller, which enables users to easily configure and control AV systems – audio, lighting, projection, and more – via an IP/serial/IR interface. He excitedly adds that both new innovations feature completely redesigned touchscreens for user control.
Like many in the professional audio industry, Chow says his most significant ongoing challenge is keeping up with the ever-changing market and the technological innovations it spawns; however, coming up with new solutions to help Xilica’s expanding customer base makes it all worthwhile. “It’s the customers’ enthusiasm for our products and technologies that keeps us active and wanting to improve and continually achieve a higher standard,” he says.

When he’s not tinkering away with some of the most sophisticated technologies the audio industry has seen to date, the longtime musician enjoys playing and composing on the guitar and keyboard from his family home in the beautiful Whitchurch-Stoufville area of Ontario, just north of Toronto. He also enjoys being able to simply relax with his family, though he now has the opportunity to interact with his daughter, Andrea, during the business day since she now works at Xilica in its customer service department.
On that note, Chow says he loves the small and intimate nature of the professional audio industry – the fact that, often, coworkers, customers, and suppliers are also close friends. He also enjoys sharing his still-avid fascination with audio technologies with those friends, as he did when he presented a keynote on the mathematics behind fixed-point and floating-point architecture in DSP systems for the Toronto chapter of the AES in October.

Chow is looking forward to working closely with some of those aforementioned partners and friends in the New Year as Xilica launches its Solaro DSP platform and Universal Media Controller internationally.
As they have in previous years, the team will be exhibiting at the early-2018 ISE Show in Amsterdam, offering many in the industry their first look at these upcoming innovations.
“I hope that 2018 is indeed our blooming year,” Chow says in closing,
and as the company’s reputation continues to grow on a global scale with new products in the pipeline, there’s no reason it can’t be.

Written by Andrew King, Editor-in-Chief
Professional Sound Magazine
December 2017 Issue

Professional Sound Magazine is a Canadian magazine for audio industry professionals and serves the Canadian recording, live sound, broadcast, post-production, and audiovisual integration industries with up-to-date news, in-depth application stories, product reviews and showcases, and much more.

Download the full article here: Xilica_PSMProfile_Dec2017Issue