KEF’s new LS50W Active Music System contains an on-board Digital Signal Processor (DSP) that has been specially designed and implemented to give you the magical experience of articulate, spacious and pristine musical reproduction regardless of listening environment. We have combined the renowned performance of the passive LS50 with a DSP and amplifier set that perfectly matches the requirements of our drivers and enclosure design – all in an enclosure set that is barely larger than the passive original.
The LS50W outperforms multiple component systems costing hundreds (if not thousands) of dollars more and that performance is owed in large part to the on-board DSP.
But what exactly is a DSP?
A DSP is a computer (more correctly a processor) that does specific calculations and can do them extremely quickly at regular intervals. In audio terms, the DSP contains the crossover and any adjustments or changes made to the original signal.
To understand what a DSP is, we need to look at the two different types of computers (processors): data manipulation and mathematical calculation. Back in the early 1960s conventional wisdom had it that by the end of the 20th Century robots would be doing our cleaning, cooking, shopping, banking, and pretty much everything else we still get stuck doing on our days off. Rosie was a mathematical calculator who did housekeeping. So much for promises. The reason for this has to do with the enormous amount of processing power artificial intelligence processors require to do abstract non-repetitive tasks. Then as now, power equals cost.
Data manipulation processors can do things like store this piece I am writing on my hard drive, spell check it, send it to a printer or allow me to copy and paste the formatted words over to the blog program for final publishing. Basically, this type of computer takes blocks of data and moves it from one place to another.
Mathematical calculation computers do exactly what the name implies and are used in engineering, scientific and data signal processing. Becasue of cost, original DSP computers were limited to military purposes like radar and sonar, space and oil exploration and medical imagery – pretty much just the big, important stuff.
A processor that could perform both tasks – manipulation and calculation – were extremely expensive until the late 1980s and 1990s when computing costs began to plummet. By the early 2000s processors proficient at data manipulation and calculation became common-place. That’s why your local TV station has an incredibly precise Doppler Radar system and you can play League of Legends in your pajamas on the same computer you store your selfies on.
With a long history of latching on to new technologies and making them their own, the audio world took this processing power and applied it to the recording and listening experience.
A simple example of the use of a DSP in the audio world is autotune, which had its mainstream debut in Cher’s 1998 single Believe. Back then it was a quirky little effect run over an established singer’s voice. Little did we know what havoc that little trick would wreak on our poor ears over the next twenty years.
Autotune was originally developed by an Exxon engineer who was working on a tool to interpret seismic data. Make of that what you will.
However, We Use the Power of the DSP For Good
We use Digital Signal Processing to help the listener get as much out of our LS50Ws as is possible.
Our engineers spend thousands of hours testing and adjusting the output of our speakers to get them to the point that you are listening to what the artist originally intended and not the interpretation of the speaker itself or the enclosure it’s in. We strive for transparency because we believe listening to music as pristine as possible is the only way to go. Of course, the room you place your speakers in plays a huge role in how your speakers sound but through the power of our bespoke DSP we help you adjust for certain room variances that would otherwise spoil the experience (or make it sound like any other run-of-the-mill active music system).
We’ve broken our settings menu into two sections: Basic and Expert. Both settings adjust the same things, they just describe them differently.
A horizontal surface in front of a speaker will change the performance of the speaker at certain frequencies. If you're going to set up your LS50W speakers on a desk or table top with a horizontal surface in front of them, then you’ll want to enable Desk Mode. Basic Mode will ask you the distance from the driver (front of speaker) to the edge of the tabletop while Expert Mode allows you to make the same adjustment using dB instead of distance. You can toggle between Basic and Expert to track the changes you’ve made in either distance or dB. Experiment with the setting to find the sound that suits your ears and room the best.
When on a stand, the distance bewtween the back of a ported speaker such as the LS50W and the wall behind it will affect the performance of the speaker at certain frequencies, in particular the high-bass or low-mid frequencies. This can result in muddy or boomy sounding bass. In Basic Mode the app will ask you what the distance is from the rear of the speaker to the wall, and in Expert Mode you can adjust for this distance in dB. Expert Mode also allows you to operate in Wall and Desk Modes concurrently. With a little experimentation you’ll be able to dial in a setting that suits your needs the best.
The size, shape and contents of a room can also have a strong impact on the performance of a speaker, particularly in the mid-range.
Some quick definitions for Basic Mode adjustment:
Damped is the setting for a room that has a low ceiling, lots of fabric either on the walls or on the furniture, carpeting on the floors, etc. This type of room will literally suck the life out of the upper mid frequencies causing the room and your music to sound dull and lifeless. This is because the reflections that your ears interpret as “lively” are nullified by the room.
Moderate is the setting for a room that is somewhat acoustically neutral: there is some liveliness and bounce, but reverb is minimal and most reflections are absorbed fairly quickly.
Lively is the setting for a room with high ceilings, lots of glass (either mirrors or windows with minimal treatments), hardwood floors and furniture without a lot of fabric. In a lively room the reflections are not dampened and tend to bounce around the room long after the initial signal. This causes confusion and an overall sense of inarticulation in your music.
Decide which of these settings is best for your room and then experiment with the slider to find the precise setting that works best for you.
In Expert Mode this setting is called Treble Trim and can be adjusted by dB. Again, you can toggle between Basic and Expert to see your settings interpreted in different terms.
Our world-renown Uni-Q driver has been developed to provide a near-perfect time-alignment between the high frequencies (tweeter) and the mids and lows. This is how music is made – the sound comes from one source and moves to your ears. Standard speakers have inherent time alignment issues because the drivers that produce the different frequency ranges are physically located in different spots on the speaker baffle. This means there are slight variances in phase and timing when those differently produced signals reach your ears. At best, this results in a narrow “sweet-spot” where the time is aligned correctly and at worst an ill-defined soundstage.
We’ve taken our ability to compensate for this problem with the Uni-Q one step further by eliminating the inherent time delay caused within a cross-over. Simply put, when two frequencies (e.g. highs and lows) are processed through an analog or digital (passive or active) crossover the frequencies change phase in relation to each other, this results in similar problems experienced with a standard speaker that does not feature the Uni-Q.
Take a song that you’re really familiar with and turn the Phase Correction on the LS50W app on and off, cycling through the same song a few seconds at a time. You will eventually hear a widening of the soundstage in all three dimensions around the speaker with the Phase Correction on as compared to when it is off. Your music will now appear to your ears in 3D. It’s subtle, but it’s amazing when you latch onto it.
For a bigger room, or a room with a lot of dense fabric that likes to chew up bass frequencies, you’ll find the More setting to be preferable. In very small rooms where you tend to listen at higher volumes, you may find that the Less setting is more desirable.
Basically the Standard setting is the flat, default bass response of the LS50W. The Less setting will get you closer to the voicing of the LS50 passive speaker and the More setting will actually approximate the bass response of our Reference 1.
Take some time with your new LS50Ws and experiment with each setting. You’ll eventually dial in to a setting that is clearly more to your liking than any other setting. Don’t worry if you set things all out of whack, you can always just select the Default setting and start over.
Because of the way the LS50W uses the app for its processing power, you’ll need to set your DSP settings while in Wireless Streaming mode, but once set your DSP settings will be operational for all input settings.