Hi, on most speakers' tech sheet, the sensitivity can be found. This is measured with 1W at 1m. How is this values behavior when the input is increased ? And what signal is used? White noise? Pink noise? A certain frequency? Let's say we have a cab1 that offers 95db and cab2 that offers 105db. The second one will offer 10 more decibels, thus be twice as loud with 1W. Will this continue when the watts are increased until the speaker farts out?

Sensitivity spec useless. If they bother telling you they'll say it's measured at 1khz. 1khz is easy to do, to make it loud, Doing bass frequencies loud is where everyone is concerned. It's nothing more than a distraction. Ask for the frequency response charts. Then you can look at 1khz on the graph to see what the sensitivity is, you can also look at 100hz, to see if it drops off where you need it to still be solid for bass.

If there was a standardised frequency for a single number it would be more useful. Better to have an average over a range with limits at -3db and -10db. Even better is the full response chart with the power handling limit referenced to a sensible frequency. And there goes a flying pig.

How then are sensitivity ratings figured for cabinets with multiple drivers? For example, Ampeg rates the sensitivity of the SVT 810 at 100db, I'm not 100% positive but I think those 10's are rated somewhere around 92db each. On the other approach, what would be the total rating of a cab with 2 drivers rated at 98db?

Don't confuse individual drivers with cabinet specs. Doubling cabinets gives additional 3db sensitivity, somewhere along the line the advantage gets lost due to the size of the cabs. An 8x10 is theoretically 9dB more sensitive than a 1x10, also handles 8x the power for another 9db ultimate loudness.

If it's a 3-way, and the sensitivity is rated at 1khz, you're getting the midrange driver response. A cabinet may boast a high sensitivity but it may be at 3k or higher where the tweeter is at. Sensitivity is also only given on-axis. Check out Green Boy designs and you'll find full charts on and off axis. The science matches reality.

Different manufacturers use different nomenclature. This is what I do, not saying it's "right", but I do claim that it's better than useless: I use the word "efficiency" when referencing dB SPL for a 1 watt input (and yes, I know that "efficiency" should technically be expressed as a percentage, but that's quite cumbersome if one is trying to make comparisions to cabs rated in dB/1 watt). I use the word "sensitivity" when referencing dB SPL for a 2.83 volt input. 2.83 volts into an 8 ohm load is 1 watt, but into a 4 ohm load it's 2 watts. So take that into account when you're making comparisions that involve 4 ohm cabs, and make sure you know whether the manufacturer is referring back to 1 watt, or 2.83 volts, which would be 2 watts. So, efficiency = watts, and sensitivity = volts, and if you're making a comparision that involves a 4 ohm cab, convert back to efficiency (watts) in order to make a more apples-to-apples comparison. Finally, I usually come up with my dB figure based on the woofer's Thiele/Small parameters, rather than on the average SPL across the cab's passband (which is still reasonable), or based on the highest peak in the midrange (which imo is not reasonable). Sorry for all the background technobabble, I'll now try to answer your question (as I understand it) by using an example: Suppose we have a cab that is rated at 97 dB/1 watt. Each doubling of power will increase the SPL by 3 dB, and each ten-fold increase in power will increase the SPL by 10 dB (which is generally considered to come across as "twice as loud". So assuming the cab doesn't fartout, it it's rated at 500 watts and we drive it with a 500 watt amp, how loud will it go? Here's how I'd come up with an estimate: 500 watts is half of 1000 watts, so it would be 3 dB less than we'd get with 1000 watts. 1000 watts would be three ten-fold increases in SPL, or 30 dB. And 30 dB - 3 dB (because it's really only 500 watts) = 27 db. 27 + 97 = 124 dB, so that's our estimate (and yes I'm ignoring the effects of thermal compression in this example). Different manufacturers measure with different yardsticks, so the answer is, "it depends". If something sounds too good to be true, they might be using an optimistic yardstick. Yes! And assuming equivalent cone area, there's a very good chance that the 95 dB speaker will fartout at a higher SPL than the 105 dB woofer! The longer the voice coil the farther the woofer can move before fartout, but also the heavier the moving mass and therefore the lower the efficiency. So assuming both cabs' specs are measured with the same yardstick, we'd expect the 105 dB cab's woofers to have shorter voice coils, and therefore fartout sooner. For instance, the Eminence BP102 woofer has a modest efficiency of only about 92 dB, but it will go at least 3 dB louder than its fellow 10" Eminence woofers before fartout because it has a longer voice coil and therefore longer linear excursion. One possible issue that you can run into with uber-excursion woofers is, they don't give you much warning before they overheat. So if you know you're pushing it hard, pay attention for any signs of distress (like compression), and back off quickly.

That curve looks about 5-6 dB hot to me. What wattage was used, and what distance? I seriously doubt that woofer's broadband efficiency is in the 102-103 dB/1 watt/1 meter ballpark, as is implied if the wattage and distance aren't specified. Let's make sure we don't leave out critical data, so that what we're calling "science" does indeed match reality, otherwise it's just as misleading as an over-optimistic number from an overzealous marketing department.

Naa... this is a chart for a perpetual motion machine. All the fan boys know this box generates more noise than the driver can make, cuz it's Special. They also believe old fashion bass horns don't develop their loudness until 30 feet away. Kinda like a camp fire is hotter 30 feet away, than up close. Oh yeah, the Earth is still flat, too. And the fundamental is useless, not produced, or ever heard. Link to 3015LF Factory Response Chart Under real conditions, 3015LF does not exceed 100 dB until it reaches 1,000 Hz. The above chart is just another piece of simulation bullsh*t. Show me the RTA, not a simulation.