That’s very interesting. Shame that device is so expensive though. Given that class D amps draw less power than others I wonder if because the noise is a larger proportion of the total signal it poses more of a problem? And because the power draw is less I wonder if the power conditioners that are a problem for dynamics with big amps would work well for these speakers.
It would be great if D&D could make recommendations about power supply.
I’d also like to get an opinion from them about maximum gain for the parametric eq filters. They suggest trying to boost valleys in the frequency response might not be good. Will the signal clip if you use positive gain?
Very pleased with the Dutch & Dutch statement about firmware update issues. Maybe regular updates on progress would be a good way to keep customers happy.
Regarding gain, in digital (as you probably know already) there is an absolute max level of 0dBFS. Anything above this is clipping. With no DSP anywhere, playback of a digital file will be below 0dBFS. If you apply digital gain for EQ you must also lower the total digital volume accordingly to be sure clipping does not occur. In Roon the Headroom config is there for this reason. Also the clipping indicator can be enabled to help you see if clipping occurs. I don’t know if there is a clipping warning in the 8c.
Note that this is not related to how loud you play, just the digital domain level (where all signal processing takes place)
Excellent points. So I’m wondering how the 8c handles the positive gain and if it has strategies in place to prevent digital level clipping and what those are. If there aren’t any then I guess positive gains should be avoided completely.
I corrected my speakers for room with the latest firmware. I haven’t done this before as configuring a source was difficult for me. It’s easy now and the results are awesome! Well, blind A/B comparison, was not done. But I will take it, subjectively. I chose the wide area, everyone benefits procedure. One hint: if you need to redo the procedure, you need to remove all the previous filters from the speakers.
Glad you like it David! This is exactly what we hoped to achieve with our collaboration with REW.
One hint: if you need to redo the procedure, you need to remove all the previous filters from the speakers
Actually, if the 8c’s are selected as the default equalizer before you do the measurements, the effects of the active filters are taken into account. Go to the Equalizer environment and you’ll see all previously programmed REW-filters that were active during the measurements. You can turn them off, change them manually, or make them available to the optimizer again.
hi @Dutch_Dutch_Dutch (Martijn) – thank you for posting on here more frequently in recent weeks. As someone who hopes to become a customer soon, I wanted to express that it has made a difference to me. I also appreciate the important steps your latest firmware has made to improve sound quality (via REW integration) as well as laying the foundation of better firmware rollouts in the future.
Can you clarify if the REW integration ultimately sends convolution filters to the 8c or or parametric EQ filter settings? If the latter, any plans to add support for convolution filters? Or is that beyond the processing power of the built-in DSP in the 8c? I am curious as I currently use convolution filters in Roon and find that to sound better to me than using PEQ settings.
I’ll try to check in from time to time and answer some questions.
The 8c’s actually have an convolution filter that makes the phase-response flat within the 8c’s passband. It is used to correct the phase-distortion caused by the 4th order LR crossovers at 100 and 1250 Hz. We flatten the phase response because flat phase response is better in principle and we had the option of doing it without having to pay any real price for it. Is the difference audible? Depending on program material, room acoustics and listener training it might be.
By design the 8c has a very clean band-passed impuls response and thus flat amplitude response. It also has a very even dispersion pattern and it interacts with the room to a lesser extent than most other loudspeakers. Therefore by tinkering with the signal you risk degrading some of the things that make the 8c special. I’ve been a proponent of room correction for years, but please don’t overdo it, do no harm. Above the Schroeder frequency two ears and a brain to a large extent hear what the speaker is doing, not the room’s contribution. Basically, if you have an accurate loudspeaker, in all but the worst rooms you shouldn’t do any corrections above the Schroeder frequency.
In the bass the room behaves like a resonator. We pretty much hear the steady-state amplitude response, not the time domain behavior. A room resonance peak is essentially minimum phase. Fix the amplitude peak and you’ve also fixed ringing in the time domain. This is quite straightforward. But don’t fill narrow dips. They tend to be very localized and non-minimum phase. Fixing a dip will likely lead to a strong peak at that same frequency a short distance away. Also, keep in mind that 6 dB of gain means the speaker draws four times the amplifier power and it will have twice the woofer excursion. Therefore by filling dips you’re likely to overload the woofers and the amplifiers real quickly. For these reasons and because to the ear dips are not nearly as objectionable as peaks, don’t fill them. As a side note, blown out drivers are not covered by our warranty. Please be sensible.
Most room issues that are worth dealing with by means of DSP - such as room resonance peaks and boundary gain - are essentially minimum phase. Of any issues that are not minimum phase, it’s doubtful a non-minimum phase filter does a psychoacoustically more sound job. For instance, because FIR filters can have arbitrary phase response, by using clever FIR filtering you might fool a microphone and an analysis program that a reflection has magically disappeared, but you don’t fool two ears and a brain quite as easily.
FIR has some disadvantages, as well as some advantages. Some of the disadvantages are potentially audible pre-ringing, high latency and/or insufficient resolution at low frequencies. An important advantage to many is that with FIR designing a filter is straightforward, whereas IIR based EQs are usually a matter of trial and error. But with the REW optimizer designing a set of good IIR filters has become just as easy. Unlike an FIR filter it’s almost fool proof and it sounds great with the 8c’s. Based on the psychoacoustic science, I believe FIR is generally less suitable for room correction than IIR. Therefore we’re sticking with IIR and we’re not planning on adding support for convolution filters.
I’m curious if anyone is having an issue with REW finding their 8Cs? I’m unable to see them in the drop down on REW’s measurement interface (control m) but I can access RMS, test.lanspeaker.com, and lanspeaker.com from my browser and interact with those apps no problem. Is anyone experiencing this or has anyone figured out the issue? It’s happening on my MAC and my win 10 machines. I’m using the latest D&D and REW fw.
@Joel_Karp – you need to go in EQ and select the 8cs in the Equaliser menu. It does say this in the guide but it is easy to miss. I missed it at first also…
Thanks @Craig_Palmer for the tip but I followed the guide as well. I’m able to see and choose ‘8C’ as an option next to choosing the stimulus in the measurement interface - e.g ‘From REW’, ‘From File’, and ‘8C’. When I click 8C the choose ‘Speaker’ drop down is empty.
Thanks @Didier. I forwarded to Martijn as well but wanted to post here in case anyone else had the issue - or better yet - had a fix for it! Fingers crossed we can get this resolved soon. FWIW, I also let Martijn know that I’m experiencing no difference between ‘Linear Phase’ and ‘Low Latency’ mode. Per measured step response, I don’t think my speakers are engaging Linear Phase mode.
Dutch & Dutch confirms that the 8c studio has the REW integration and this is an addition to the already present parametric EQ (same as in the 8c in other words). The main difference is then down to the MDF cabinet for the 8c studio and the 8c’s ability to receive a stream via ethernet. Digital into the 8c studio is only possible via AES/EBU, which is actually quite good when you think about it. The only thing missing now would be a word clock input… what about a 10MHz input
