Passive vs active DSP speakers KEF and Genelec

[echopraxia]

Do actives really increase bass extension? I'm sure some do. It seems like many just EQ more extension in, which(imo) doesn't actually increase extension, it just limits the max output of the other frequencies. You could do the same with passive speakers, it's just that you have to do it yourself.

Bass extension is simply defined as having frequency response +-X dB extend down to Y hz. It doesn’t really matter how it’s accomplished. Of course the laws of physics always apply, though there are many ways of squeezing more efficiency out of a system.

Then there is the separate question of what SPL is achievable at different frequencies. Perhaps this is what you were thinking of?

Bass extension and SPL capabilities depend on many factors that go into the design. There is no single reliable way of predicting them in either active or passive speakers aside from just actually measuring them.

For example, my Ascend Sierra 2EX bookshelf speakers (passive) actually have way deeper bass extension than my Revel F206 towers! But I’m certain the towers are capable of far more bass SPL before the drivers start bottoming out. DSP didn’t cause this situation.

There’s really no such distinction as “real bass extension“ vs “EQed fake bass extension”. They are the same thing. What you are probably looking for is SPL vs frequency measurements, which I admit are sadly rare to find.

 

[richard12511]

That’s what it means to increase extension, though — to have the frequency response +-X dB extend down to Y hz. It doesn’t really matter how it’s accomplished.

Then there is the separate question of what SPL is achievable at different frequencies. Even in fully passive designs, you’re not always going to be able to reach the same SPL at your deepest bass extension limit as you can at something like 1khz.

Ultimately, the bass extension and SPL capabilities depends on many factors that went into the design. There is no single reliable way of predicting them in either active or passive speakers aside from just actually measuring them.

For example, my Ascend Sierra 2EX bookshelf speakers (passive) actually have way deeper bass extension than my Revel F206 towers! But I’m certain the towers are capable of far more bass SPL before the drivers start bottoming out.

I guess I see those questions as very much connected, given that you can easily pump up the bass with EQ. What matters to me is max output *after EQ.

For example, you say your Revels can play a lower note louder than your Ascends? I have no idea if this is true or not, but assuming it is, then to me that means the Revels have more extension, as you could easily apply a REW filter that would make it so.

I guess if you don't have the ability to EQ, then manufacturer EQ is more important. That may be the case for some folks.

 

[echopraxia]

I guess I see those questions as very much connected, given that you can easily pump up the bass with EQ.

For example, you say your Revels can play a lower note louder than your Ascends? I have no idea if this is true or not, but assuming it is, then to me that means the Revels have more extension, as you could easily apply a REW filter that would make it so.

I guess if you don't have the ability to EQ, then manufacturer EQ is more important. That may be the case for some folks.

You are using different terminology than is standard. I understand what you’re talking about, and it IS important. But it is confusing to call it “bass extension”, because you are redefining that phrase differently than how it is defined and used by everyone else.

What you are talking about is max SPL vs frequency for a given distortion threshold. This is definitely not called bass extension, unless I’m completely mistaken.

I definitely agree we could use more SPL vs frequency plots, especially for the more compact speakers. It’s super important, and sadly quite scarce data.

 

[maverickronin]

I do wish AVRs were capable of digital out, or TVs has digital out with volume control — if only for convenience. It seems there is no technical reason the TV industry shouldn’t be capable of supporting this.

AFIK they can have digital out but the content cartels artificially limit it 16/48.

Given that some form is actually permitted I have no idea why it isn't more common either. I think there were supposed to be some JBL modles coming out with digital out.

 

[richard12511]

You are using different terminology than is standard. I understand what you’re talking about, and it IS important. But it is incorrect to call it “bass extension”.

What you are talking about is max SPL vs frequency for a given distortion threshold. This is definitely not called bass extension, unless I’m completely mistaken.

I definitely agree we could use more SPL vs frequency plots, especially for the more compact speakers.

You're not mistaken. I'm definitely using a different definition for "extension" then what is standard. I guess I just disagree with the standard, assuming you have the ability to EQ.

I guess my problem is that there's no standard spl for which we define extension. The standard for each loudspeaker is relative to itself. For example, the JTR 215RM has two 15 inch bass drivers and has "standard" extension to 32Hz. The Genelec 8351b has 2 8 inch? bass drivers and also has "standard" extension to 32Hz. By the Genelec's spl standard, I could use EQ apply a +15db boost at 16Hz to claim 16Hz extension. You'd have to lower the max output spec, like you said. I guess I just don't like the ambiguity of the "extension" spec. Genlec does a great job of listing both extension and max output, so I don't have a problem with them, but other manufacturers often don't. I'd much rather see a max output frequency response showing the max output at every frequency. To me, that would be a better display of true extension.

 

[HammerSandwich]

guess I see those questions as very much connected, given that you can easily pump up the bass with EQ. What matters to me is max output *after EQ.

Sensible enough, because we all know that 20Hz isn't useful at 70dB.

The sophisticated approach is to add lots of boost for maximum extension, then roll off the low end as the signal rises beyond the speaker's limits. You won't miss the 20Hz quite so much when 30Hz is shaking the room at 100dB! I don't know how many speakers do it this way, but I promise it's zero passives.

 

[Senior NEET Engineer]

Do actives really increase bass extension? I'm sure some do. It seems like many just EQ more extension in, which(imo) doesn't actually increase extension, it just limits the max output of the other frequencies. You could do the same with passive speakers, it's just that you have to do it yourself.

Yes. Higher effective power handling, increased dampening, reduced and more consistent load, detection of system limits and intervention, implementation of feedback, driver and amp designed as one system. You will get higher SPL at lower frequencies with active design.

 

[steve59]

First, I’m intimidated posting on a science forum. The full range dsp speakers have proven to be easy to set up and get balanced sound from. Unlike my experience with the multitude of passive speakers Previous.

 

[echopraxia]

First, I’m intimidated posting on a science forum. The full range dsp speakers have proven to be easy to set up and get balanced sound from. Unlike my experience with the multitude of passive speakers Previous.

Yes, and I believe many people underestimate several factors relating to passive speakers that make them much less competitive in practice than they might be in theory:

• Some EQ is virtually essential for passive speaker solutions, for subwoofer crossover integration and because even the most exceptional passive speakers (e.g. KEF R3) are not nearly as flat as active ones (but could be with some EQ). Yet, like it or not, not everyone is going to go through the effort to research and purchase and install and configure room EQ solutions like Dirac for every speaker they have.

• And even those who do have Dirac setup, it's going to be very difficult to achieve performance anywhere near as optimal as an active speaker that is calibrated and EQ'ed to have near-perfectly flat anechoic frequency response. Maybe with a lot of research and study and careful use of e.g. moving mic spatial averaged FR measurements (or downloading Amir's Klippel measurements if you're lucky enough to have them for your speaker), educated use of hand-crafted filter parameters to cautiously correct high frequencies, etc. -- perhaps then it can be done. But it's completely absurd to expect any great number of people outside the "hobby speaker designer" crowd will do this successfully. I see a lot of those "hobby speaker designer" types here promoting ideas like this, perhaps not realizing how limited the set of people are that are both willing and able to do this.

• In fact, in most people's situation you simply cannot "just use Dirac". This idea that everyone is running every pair of speakers they have from a full PC is ridiculous and not at all realistic except to perhaps the most hardcore "tinkerer" type audiophiles. There are many different kinds of audio sources, and it's completely unrealistic to assume every one of them must support every kind of room EQ / DSP software built-in.

• Few people are going to have 400 watts per channel of amplification for full range speakers to get the most out of their bass capabilities, and those who do have to wade through lots of confusing reviews about overpriced amps etc. And those who visit this site will find that although HypeX amp modules seem the way to go, there is no clear easy way to buy a HypeX based amp “off the shelf” in most countries.

• For digital active speakers with digital inputs, even the question of DAC choice is removed. The product choice is maximally simple: you evaluate the speakers as a complete audio solution. Other than connectors to get the lossless digital signal in, there is no worry about whether signal chain component A, B, or C are preventing your speakers from living up to their full potential.

The last two points are I think why many traditional audiophiles and snake-oil salesman may be afraid of active speakers. They solve the entire problem of building a sound system in one product (or two, if you add a DSP subwoofer).

Some people don't like such a simple and effective solution. Some people WANT to waste time and money on tweaking and tuning signal chains, though I think that’s foolish: If you want to tinker, tinker with EQ digitally in a way that gives you infinite ability to tune the sound to your liking at $0 cost per adjustment.

And of course, those who sell snakeoil products and massively overpriced DACs and amps will certainly have a vested interest in pushing the false idea that a passive speaker that sounds worse than an active speaker just needs $100,000 in DACs and amps (which they coincidentally advertise) to magically "scale up" the sound quality of the passive speaker to exceed that of the active speaker, they claim.

And on top of this there are the inherent advantages you get from active crossovers as mentioned above. I think active speakers are the way to go (whether the electronics are internal or external), even from a cost effectiveness perspective.

 

[echopraxia]

Can you please do that. My main concern with that speaker would be its max output at say 3-4m or so. Crossed to subs would be an even better test(for my selfish curiosity ), as I'll have subs to take care of the bass, so it should give a truer sense of max output. Good thing with those is you don't have to worry about damaging them.

BTW I still haven't done this test, as I may want to wait (if I can) for my stands to arrive. My hope is that in addition to SPL tests, I can do some blind listening comparisons between some of my speakers (e.g. Revel F206 vs Ascend Sierra Towers) and repeat this on the winner (vs Neumann KH310 and Genelec 8351B) until I establish a ordered ranking of these speakers.

 

[Steveysteve]

Noobie here. I have a question about using DSP active speakers with D amps and no digital inputs. That makes the chain from my digital source (with DAC) DAC-ADC before the signal is finally converted to analog again.

Under those conditions, is the work of the original DAC negated? Does putting an extra ADC in the chain degrade the signal?

 

[HooStat]

The original work of the DAC is not "negated" if you are sending analog to the active DSP speaker. You need a DAC to get the analog signal into the speaker, so a DAC is required. In theory, if you use a bad enough DAC on the input side, you could degrade the signal that way, but I assume that is not the question you are asking.

The ADC doesn't degrade the analog signal in any measurable way. However, you will hear anecdotal comments otherwise. But DACs are solved problems, especially in a closed system where the speaker manufacturer controls the whole AD, DSP, DA process.

Even speakers with digital inputs will adjust the digital signal to match its internal DAC.

You can worry about all of this, but I don't know of any evidence to support the worry.

 

[echopraxia]

Noobie here. I have a question about using DSP active speakers with D amps and no digital inputs. That makes the chain from my digital source (with DAC) DAC-ADC before the signal is finally converted to analog again.

Under those conditions, is the work of the original DAC negated?

It depends what you mean by “negated”. Many DACs are overkill in the sense that even the absolute best speakers in the world produce many orders of magnitude more distortion than even some of the worst DACs. So if your DAC was overkill with a passive analog speaker, it will still be overkill with an active DSP speaker. And if you were hoping that a extremely-overkill DAC was yielding audible improvements vs a moderately-overkill DAC, then you'd be equally wrong whether you're feeding a passive analog speaker or an active DSP speaker

Does putting an extra ADC in the chain degrade the signal?

Theoretically yes (*), practically no. The signal is technically degraded a bit, but when the physical speaker drivers themselves introduce many orders of magnitude more distortion more than these extra signal chain components, it really doesn't matter and definitely won't be audible.

(*) This is like asking whether attacking a housefly with a Tsar Bomb instead of a MOAB bomb increases our probability of success at killing the housefly. As a matter of theoretical physics, the answer is technically yes — a nuke dropped onto a housefly is more likely to guarantee the assassination’s success than would a less powerful bomb. But I hope it’s obvious that both are so much overkill, that the probability of either failing is not even something you ever need to worry about

Some numbers on how little you need to worry about DACs here (as long as they're "overkill enough", as any DAC recommended by ASR is).

 

[Steveysteve]

It depends what you mean by “negated”. Many DACs are overkill in the sense that even the absolute best speakers in the world produce many orders of magnitude more distortion than even some of the worst DACs. So if your DAC was overkill with a passive analog speaker, it will still be overkill with an active DSP speaker. And if you were hoping that a extremely-overkill DAC was yielding audible improvements vs a moderately-overkill DAC, then you'd be equally wrong whether you're feeding a passive analog speaker or an active DSP speaker


Theoretically yes (*), practically no. The signal is technically degraded a bit, but when the physical speaker drivers themselves introduce many orders of magnitude more distortion more than these extra signal chain components, it really doesn't matter and definitely won't be audible.

(*) This is like asking whether attacking a housefly with a Tsar Bomb instead of a MOAB bomb increases our probability of success at killing the housefly. As a matter of theoretical physics, the answer is technically yes — a nuke dropped onto a housefly is more likely to guarantee the assassination’s success than would a less powerful bomb. But I hope it’s obvious that both are so much overkill, that the probability of either failing is not even something you ever need to worry about

Some numbers on how little you need to worry about DACs here (as long as they're "overkill enough", as any DAC recommended by ASR is).

Interesting. So the DAC comparisons in reviews and discussions here are largely academic?

 

[richard12511]

Interesting. So the DAC comparisons in reviews and discussions here are largely academic?

I'd say the fact that most DAC "reviews" spend the majority of the article/video discussing "sound quality" is a testament to just how much our imagination adds(due to expectation bias/placebo) to the final sound that we "hear".

I don't think most reviewers understand that those subjective impressions would all fade under blind conditions. Its possible for a DAC to be so terrible that it does create an audible effect, but that's rare imo.