Spent some time browsing the different speakers over at the celestion site e.g.
https://celestion.com/product/155/neo_creamback/And I noticed that all the frequency response charts are just about the same. Coupled with if they all had the same spec being 16ohm, 75hz, 60w, I'm wondering what else in them makes the audio difference and how is it measured so punters can determine which they'd like to use?
(I'm guessing the answer is cone shape, cone material, magnet material, magnet strength but surely of the frequency response charts are the same then it will sound the same right? )
Comments
If the charts were drastically different then some of them won't sound like guitar speakers any more.
The best way to compare speakers is to listen to comparison videos, rather than studying the frequency response graphs.
Because of the common physical properties between most guitar speakers - 12" metal baskets, cones of approximately the same substance etc - there are going to be broad similarities in the overall frequency response curve. But if you look at them closely, you'll see that there are differences. Pick, say, 3kHz, and compare the frequency response there. You might well find differences between models as high as 10dB (6dB is approx. half/double volume).
I just hopped over to the Celestion website and did some comparisons between a few speaker models. In the following, the first number is response at 2kHz, 2nd at 4kHz;
Alnico Blue; 105, 105
Alnico Gold; 103, 108
Vintage 30; 105, 105
G12m-65; 101, 98
A Type; 103, 102
So, the Alnico Blue and Vintage 30, purely in frequency response, have marked similarities, and that's interesting because the V30 was actually an attempt at a more modern alnico blue midrange, according to some. The Gold, meanwhile, is 5dB louder at 4k than at 2, and it is indeed a more fizzy, reedy speaker. The m65 is quieter overall, as you'd expect due to its lower sensitivity. At 4K it has a noticeable notch dip, meaning it's 3dB quieter there than at 2k, though either side of 4k it's louder.
The A-type is interesting, because unlike the other speakers mentioned it doesn't have a deep broad valley between 1 & 2k; the rest are up to 10dB quieter at 1.5 than at 2.5K. And that's probably a physical thing - the wavelength of 1.5k is probably one that produces subtractive cancellation as it bounces back from the edge of the cone towards the middle (more on that in a minute, while 2.5k resonates on the surface of the cone better.
So, even though the general shapes of the curves are similar, there are noticeable differences nonetheless. And that's just with the test tone & measuring a static frequency response.
In the real world, what you find is that the measurements are usually dynamic - they change with volume. And frequency response is only one aspect of the sound. There is also;
Efficiency. This is how much of the electrical signal being put out by the amplifier is turned into sound energy by the speaker, and how much of it is wasted as heat energy. More efficient speakers will be louder at the same amplifier volume setting.
Transient response. This is how quickly the speaker reacts to a sudden change in signal - say, a spanky clean strat sound on the one hand, and a fat palm-muted chug on the other. Some speakers respond faster of slower. Some speakers are more heavily dampened, so stop as soon as the signal ends. Others will overshoot, like worn car suspension, and the speaker cone therefore carries on resonating - making noise for a short while after the signal ends. The speed of response and the length of time the speaker carries on resonating after such a signal has ended will be different across the frequency spectrum, and will change depending on what electrical situation the speaker is in - speakers wired in series help each other resonate, speakers wired in parallel try to stop each other resonating, and exactly how they do it also depends on the amplifier circuit they're plugged into - some will more firmly resist the resonating than others.
Thermal Compression. As a guitar speaker heads into the top half of its power rating, it starts taking more and more wattage for each additional decibel of loudness. This is because the further you push it, the less efficient it becomes - you start to get more physical resistance as the cone reaches the extremes of its safe travel back and forth, and that resistance produces more heat. This thermal compression is one of the reasons guitar speakers can sound and feel so good to play through at high volume.
Cone Distortion. Hi Fi & PA style speakers usually try to make the cone move, as much as possible, as one coherent mass - so the centre and edges are doing the same thing at the same time. Often that means the surround, which is the bit between the moving cone and the still speaker frame, is made of cloth or rubber which is very pliable and lets the edges of the cone move quite freely. Guitar speakers aren't like that at all. The middle of the speaker moves, pushed back and forth by the voice coil. The surrounds of a guitar speaker are just corrugated paper, the same kind of paper the cone itself is made out of. So at some point, the movement in the middle of the cone is going to clash with the fact that the edges of the cone can't move at all. That means that sound vibrations radiate out from the middle through the paper cone like ripples on the surface of a pond. And just like in a pond, when the ripples reach the edge of the cone which cannot move, they bounce back towards the middle. This means that every frequency produces its own complex pattern of standing waves on the surface of the cone - so in some areas the cone moves a lot, in others it's totally still, like when you play natural harmonics on a guitar or like occurs on the surface of the metal plate in this video;
The overall effect of this is that the cone is doing different things in different places, and to varying extents the physical resistance of the cone to doing all this produces distortion because the cone can't faithfully reproduce exactly the signal being fed to it. This can be subtle (and, incidentally, these standing wave patterns etc occur at any volume) or it can result in audible cone cry or sub-harmonic rumbling.
So... frequency response, efficiency, transient response, thermal compression and distortion, are all characteristics of the speaker. And they're not static, they change depending on how loud you play and what combinations of frequencies you're sending to the speaker.
Finally, it's worth noting that especially at the low end, the response of the speaker is hugely shaped by the cab it's put in - open or closed back, internal volume, cab construction... this can easily be demonstrated by listening to a speaker either out of a cab altogether. They sound terrible.
Bandcamp
Spotify, Apple et al
Even in the esoteric world of very high end hi fi and studio monitors, little notice is taken of response graphs, loudspeakers are subjected to a battery of tests including impulse and waterfall plots to attempt matching of pairs for top stereo imaging. The very best makers will have very detailed records so that drive units can be replaced that keep a pair very close in sound. Even then some firms will only supply matched pairs of tweeters should one blow since they feel it is the only way to guarantee the result meets original specification.
Guitar speakers? No hope! Mk1 lug is the only way.
Dave.
If they were my speakers, I'd probably find a nice horizontal oversized 2x12 to try them in, it'd probably be a bigger sound than most 4x12s!
Bandcamp
Spotify, Apple et al
This is a perfect illustration...
Even ignoring the fact that taking just two points on a very complex curve is meaningless, that would imply that the Blue and the V30 sound similar, and anyone who has compared them will know that's nonsense.
It was an attempt at a G12H-30 - ie a "vintage 30", hence the name even though it's a 60W speaker (70W originally). That's why it has a ceramic H magnet. According to the original ad copy, it was developed using laser interferometry to measure the movements of a G12H-30 cone - I'm not sure where the idea that it was meant to sound like a Blue came from, but it's wrong on both counts.
"Take these three items, some WD-40, a vise grip, and a roll of duct tape. Any man worth his salt can fix almost any problem with this stuff alone." - Walt Kowalski
"Only two things are infinite - the universe, and human stupidity. And I'm not sure about the universe." - Albert Einstein
http://celestion.com/speakerworld/guitartech/7/66/Vintage_30:_A_Tonal_History/
"In developing the Vintage 30, Celestion made pioneering use of Laser Doppler Interferometry to analyse the cone behaviour of an original Celestion Blue."
Argue with them if you like, but both speakers do have big midrange, with tight lows and complex but not harsh treble. It's just that the blue almost immediately compresses, fills out in the lows and cone breakup piles on in the highs, and to my ears is less damped/controlled which means in practice, it's a totally different speaker.
Bandcamp
Spotify, Apple et al
But if it's based on a Blue, why is it called a "Vintage 30", when the Blue is a 15W speaker and the new one was a 70W speaker (at the time)? And why is it an H-magnet ceramic speaker?
They're not remotely similar - the V30 is all rough midrange, the Blue is clear, compressed and slightly scooped. The Blue also has far more deep lows than the V30, and is nowhere near as tight - and I have directly A/B'd them with the same (very different - an AC30 and a Mesa Trem-o-verb) amps, so I'm certain the difference in the speakers is consistent and not related to the amp. The extra low-end of the Blue is surprising if you think of it as a 'bright' speaker - which it is, far brighter than the V30.
Although the V30 doesn't sound like a G12H-30 either, it's far closer.
"Take these three items, some WD-40, a vise grip, and a roll of duct tape. Any man worth his salt can fix almost any problem with this stuff alone." - Walt Kowalski
"Only two things are infinite - the universe, and human stupidity. And I'm not sure about the universe." - Albert Einstein
And it's also worth remembering that the there's V30s and V30s. Mesa V30s are quite different than celestion V30s, while Marshall G12 Vintages are brighter and looser. They've used different cones too.
Bandcamp
Spotify, Apple et al