Getting your sound, science or art?

[Phil Starr]

And as you know any headline or bit of clickbait with a question in the title isn't going to offer an answer

 

For me this is a surprising question, I'm a lot of a nerd, Science trained and I spend my time looking at graphs, numbers, measurements and formulae. I've just hit 4,000 posts and most of them are about speakers and specifically about the technical side of bass amplification

 

I like to share the little bits i know and hope it helps somebody somewhere. I'm just wondering though whether those who do it by feel, by listening and twiddling and doing all the wrong things may have it right. Is FRFR the holy grail of bass? Or is there a technical reason why we are constantly chasing new and better gear other than just GAS? 

 

 

So here is why I think we have a problem. We spent a few hours this Sunday at the SW bass bash measuring amplifier responses. All controls set to noon, as you'd expect that to be more or less flat and the amps would all look the same, but far from it. We were only really interested in answering two questions; did the amps all have built in HPF and was noon really flat. What we found was that the amps we tested varied wildly. The Ashdowns all had built in top end roll off, The Trace Elf had a smiley face response, The Warwick Gnome and the TC BAM had a rising response and most of the amps had irregularities, lumps and bumps in the frequency responses that would be clearly audible. Not one of the amps we measured could be described as 'clean' or hi-fi although some of them have been. Never mind class D or AB the pre-amps alone aren't flat, the tones aren't neutral.

 

Now add in that bass speakers aren't neutral either, we've been measuring these a lot recently and even some really expensive boutique cabs are nowhere near neutral, or natural come to that; have you ever heard an amplified sound that fooled you into thinking you were listening to something real and acoustic? 

 

Now I've got loads of thoughts on this but let's look at one example. My Gnome has a crazy response a bass hump centred on 150Hz, a huge midrange suckout at 400Hz and an overall  that just rises and rises. Oh and the bottom end is rolled off at 80hz. Fortunately it has a midrange control centred at 400Hz so balance is almost restored if you turn it to 3.30. I'll show you.

 

 

Now a lot of speakers have a hump at 100hz and also have a mid cut , but not necessarily at 400Hz, they might have a crossover to a horn at 2,500hz which creates a dip there and that horn might also have a rising response up to 10,000Hz and  then fall off after that. So now you have two peaks at the bottom end two dips in the middle and heaven knows what at the top. Good luck sorting that out with bass middle and treble controls. And of course you might want a middle drop out but at some other frequency than either the amp offers or the speaker gives. Now here's the thing, I quite like my Gnome, I've been using it a lot though with the mids boosted on my active P and a boutique FRFR bass cab. I've been a bit worried about the amount of mid boost but it's sounded OK and once the gig starts you've other things to worry about. 

 

So is it better to know all this stuff, seek the 'truth' and hunt the perfect rig or to live in blissful ignorance and just max out the tone controls until it's time to trade your gear for new in the eternal quest for sound nirvana. What do you think?

 

Edited October 20, 2022 by Phil Starr

[Sparky Mark]

I'm 4 pints down at the pub right now so didn't read all of the above, just the first and last paragraphs, so treat this reply with the contempt it deserves.

I think it's important for us to know which frequencies we like accentuated and attenuated in our personal "sound". If the amplifier allows this, and we can work out which knob does what then that's perfect. In my personal experience it has taken a long time to figure it out, but now I have I can get my sound from just about any decent amp system. 

The analysis and science bit is irrelevant for me, hic!

Edited October 13, 2022 by Sparky Mark

[Lozz196]

Very interesting about the Ashdowns and the top-end roll-off, as I find I do boost the highs a lot on mine.

[chris_b]

Playing a musical instrument is an art. Making one is science (with some art thrown in).

Edited October 13, 2022 by chris_b

[Downunderwonder]

2 hours ago, Phil Starr said:

My Gnome has a crazy response a bass hump centred on 150Hz, a huge midrange suckout at 400Hz and an overall  that just rises and rises. O

I don't see it like that at all. I see a response tailing off below 80hz, a small bump at 150hz and a bigger dip at 400hz, taking the 'zero' at "36".

 

Are those dB volts or watts being plotted?

[Downunderwonder]

I saw a pedal steel web forum where the guy had goosed the bass, left the mids more or less noon but still a little up, and boosted the highs to blazes all supposedy to make his Elf "flat". Presumably with the gain down to compensate for all the boosting it might not be a million miles away, because science?

[BigRedX]

As I said in another thread, unless you are strictly a solo player, what makes the bass guitar (or any instrument for that matter) sound good, is how well it works in the overall mix with the other instruments. That sound when solo'd is rarely the sort of sound you would want to spend listening to on its own, but put back all the other instruments and everything sounds fantastic.

 

Science, if you understand it fully can get you to roughly the right sonic space, but ultimately getting the perfect sound for a band/song will require someone with great ears at the controls.

 

Many of my favourite bands feature a musician with an unconventional sound or style and the other instruments often have to be completely re-balanced to make the overall band sound work.

 

And finally from a personal PoV, I play in two bands with different instrumental line-ups. The sounds I use on my Helix for each band are very different, and not at all interchangeable.

[stevie]

The deviations from flat on the Gnome are broad band (i.e. over several octaves), which means they'll be quite audible. It looks to me like the amp has been "voiced" to sound good with a cheap, small speaker cabinet.

 

The rolloff below 100Hz will help protect small speakers from overload, whilst the 150Hz peak is a typical ruse to make the sound subjectively more "bassy". The midrange dip will counter the usual "shouty" mids found in many smaller cabs and the HF lift will compensate for the lack of top end and off-axis rolloff found on most bass cabs.

 

So I guess there is a logic behind it.

[Beedster]

1 hour ago, BigRedX said:

Science, if you understand it fully can get you to roughly the right sonic space, but ultimately getting the perfect sound for a band/song will require someone with great ears at the controls.

 

I think the ability to use a combination of experience and perception to modify the environment to a specific state is the science. The art bit is the performance itself 👍

[JPJ]

Having been playing for over forty years the thing I found hardest to learn was that what I consider to be a ‘good’ bass sound hardly ever translates to a ‘good’ band sound.

When I started mixing bands about ten years ago I found that you have to ‘place’ every instrument in the overall mix and it’s very hard to place the bass guitar if all you are getting is a big fat 150hz signal and no mid-range at all. 
That said, manufacturers know that testing of their amps (and cabs) will most likely be conducted in isolation from other instruments at low volumes  in the local music shop (remember them?) so it’s no surprise they voice the EQ to sound ‘good’ in that situation. 

[Merton]

4 minutes ago, JPJ said:

Having been playing for over forty years the thing I found hardest to learn was that what I consider to be a ‘good’ bass sound hardly ever translates to a ‘good’ band sound.

When I started mixing bands about ten years ago I found that you have to ‘place’ every instrument in the overall mix and it’s very hard to place the bass guitar if all you are getting is a big fat 150hz signal and no mid-range at all. 
That said, manufacturers know that testing of their amps (and cabs) will most likely be conducted in isolation from other instruments at low volumes  in the local music shop (remember them?) so it’s no surprise they voice the EQ to sound ‘good’ in that situation. 

Mark Gooday at Ashdown admits that their preshape button is "the wrong way round" so the amps sound great at low volume in the shop!

[Beedster]

3 minutes ago, Merton said:

Mark Gooday at Ashdown admits that their preshape button is "the wrong way round" so the amps sound great at low volume in the shop!

 

....which for a lot of people buying gear constitutes the largest audience they're ever likely to play to   

[Merton]

19 minutes ago, Beedster said:

 

....which for a lot of people buying gear constitutes the largest audience they're ever likely to play to   

Me included

[Phil Starr]

31 minutes ago, Merton said:

Mark Gooday at Ashdown admits that their preshape button is "the wrong way round" so the amps sound great at low volume in the shop!

I do feel sorry for manufacturers, they have to do these things even if their engineering instincts tell them not to. Though to be fair a 'sounds good when practicing' button isn't a bad idea

 

Below is the response of the Ashdown RM500, you can see the rest of the measured responses we took HERE 

 

So when looking at this I ask three questions. Is there a low end roll off/HPF (where and how quickly)? Is there a high end roll off (same questions)? Are there any irregularities that make the amp non-flat. So roll off I take from the -3db points and the rate of roll off I calculate over 2 octaves from there. The highest sensitivity of this essentially flat amp is around 13db so the minus 3db points are where it crosses the 10db line and the roll off is more or less 6db/octave. 

 

Ashdown clearly know what they are doing, -3db at around 50Hz and a gentle roll off is going to give you plenty of bass without overpowering and offers a little protection from idiots blowing speakers. Bass output is almost all in the 100-4,000 Hz range and that is flat with another gentle roll off above that. If you want a midrange dip there is a button and tone controls. The top end roll off will also tame one of the problems with single driver cabs (as opposed to ones with horns and crossovers) where most cone drivers have cone s breaking up in the low KHz region

 

 

 

[Merton]

Out of interest is that RM500 graph set with the EQ "in" or "out"? If it's "Out", what does it look like with the button "in" but all controls at 12 o'clock?

[Phil Starr]

6 minutes ago, Merton said:

Out of interest is that RM500 graph set with the EQ "in" or "out"? If it's "Out", what does it look like with the button "in" but all controls at 12 o'clock?

You'll have to ask John @Chienmortbb that. I was demonstrating speakers for most of the time whilst he was taking measurements. The aim was to find the 'flat' response though so I'm fairly sure it was flat.

Edited October 14, 2022 by Phil Starr

[Phil Starr]

10 hours ago, Downunderwonder said:

I don't see it like that at all. I see a response tailing off below 80hz, a small bump at 150hz and a bigger dip at 400hz, taking the 'zero' at "36".

 

Are those dB volts or watts being plotted?

You are perfectly correct, it could be described any number of ways, but I see it as a shaped response. I'm looking from an engineering perspective and trying to reverse engineer what is going on inside.

 

Any power amp will have a flat response up to whatever frequency that particular amp can manage and flat down to the point where capacitance and inductance filters out the bass. Generally this will be in multiples of 6db/octave. So 6/12/18/24db for ist,2nd.3rd,4th order filters are what I'm looking for. 

 

That means the response is shaped from flat, probably in the pre-amp stage and somebody has designed that shaping into the amp. In this case they helpfully quote 400hz as the centre frequency of the mid control so it is set deliberately down by around -4/-6db at 12o/c, turn  it up by that much (around 3.30 on the mids) and there will be roughly a straight line between 150 and 1.8khz. That leaves the rising response of 6db between the bass and treble. Clearly that is effectively a cut in the bass and a boost to the top end to give it a bright sound. 

[Chienmortbb]

1 hour ago, Phil Starr said:

You'll have to ask John @Chienmortbb that. I was demonstrating speakers for most of the time whilst he was taking measurements. The aim was to find the 'flat' response though so I'm fairly sure it was flat.

The RM500 was measured with the EQ off.

[Merton]

20 minutes ago, Chienmortbb said:

The RM500 was measured with the EQ off.

Would be ace to see it with the EQ on but all set at 12 o’clock 🤓

 

Would love to get my Glock tested in this way too!

[Dan Dare]

11 hours ago, Phil Starr said:

I do feel sorry for manufacturers, they have to do these things even if their engineering instincts tell them not to. Though to be fair a 'sounds good when practicing' button isn't a bad idea

 

Agreed. Much of their gear wouldn't make it out of the shop door if they didn't do it. That said, it's perfectly possible for them to follow their engineering instincts and make stuff that does the job in real-world situations when the "showroom appeal" button is not engaged.

[Phil Starr]

 

 

.

17 hours ago, Merton said:

Would love to get my Glock tested in this way too!

We had hoped to measure a lot more amplifiers. There were plenty there at the bass bash but we simply ran out of time. The result for the Gnome was so unexpected that we spent precious time checking everything over. I had to set up and run a speaker shootout in the main hall and couldn't act as gopher for John.

 

Anyway it was an interesting experiment I think we'll repeat and try to grab more data on other amps. There is free software to do this so if you have a simple two in and two out interface you should be able to do this at home so maybe a few of us could work together and build a database

[itu]

Just bought a used Bruel & Kjær unit. Have to refresh some old acoustics courses and measure my Glockenklang Soul, and the cabs I have. No, there's no anechoic chamber around, but certain measurements are still possible even in a room.

[Chienmortbb]

21 hours ago, Merton said:

Would be ace to see it with the EQ on but all set at 12 o’clock 🤓

 

Would love to get my Glock tested in this way too!

I will do that next time I visit @stevie. It won't be soon as I have Covid 😁

[Merton]

All the best for a speedy recovery and no nasty lingering effects!

[Chienmortbb]

2 hours ago, Merton said:

All the best for a speedy recovery and no nasty lingering effects!

Thank you Sir.