The source of this "myth" is the commonly held belief that oval speakers are incapable of producing the high fidelity sound that a round speaker is capable of. Note that I'm not in any way stating that an oval speaker is in fact superior, but simply that in many if not most cases an oval speaker can equal and in some cases better it's equivalent round speaker.
The common argument is that the larger and/or odd shape of the cone causes flex in the cone itself and creates more distortion due to a round shape being inherently more stiff. Also, the fact that oval speakers create an odd dispersion pattern as they approach beaming, which they do, but it's not necessarily a detriment.
The problem with this is that it is somehow thought that the common round speaker is inherently very stiff which really isn't the case. It is also believed that an oddly shaped speaker somehow needs huge amounts of reinforcement to approach any level of cone stiffness, which is also untrue. Most "cone flex" is due to the design of the cone itself. If you take a look at the typical speaker, the majority of them use polypropylene or paper cones, which most of them you can easily "bend" or flex simply by pressing on the edge of the cone near the surround. In fact, the only cone designs out there that really achieve a level of stiffness close to achieving the goal of true pistonic behavior would be extreme composites like Kevlar and some metal cones that have a LOT of material density such as the SEAS magnesium cones or the one case of their L18 7" driver which uses a specifically designed cone profile to achieve more stiffness and push the breakup node to a much higher frequency. The other issue presented is the material density required to get such a stiff cone usually results in a heavy cone in and of itself meaning a hit in sensitivity of the speaker.
What sounds worse: a broad, large increase in even order harmonic distortion in the lower midrange and bass or sharper but narrow band odd order harmonic distortion in the upper treble? This is the classic damped cone vs stiffer cone debate. A very stiff cone generally has lower distortion in the passband but much higher as it approaches the treble region (right before the point that you'd usually hand off to a tweeter) It also means that you are going to have a breakup node somewhere. In extremely stiff drivers like a W18 or W22EX SEAS magnesium cone, this is very apparent to the point that a notch filter has to be designed and the crossover has to be low enough so that harmonic distortion doesn't excite this breakup node. Though not to the same extent, breakup nodes still occur even in the case of softer poly and paper cones. If the design is soft (damped) enough, there's less and less chance of a big nasty breakup node, but more chance of having more distortion overall in the passband due to less than pistonic behavior. But round vs oval cones in this area? Referring to distortion/resonances generated by the cone itself, round cones tend to have a single breakup peak due to being symmetrical. Oval cones, due to their odd/dissimilar cone profile and their not being symmetrical all around tend to lessen single distortion peaks and "spread it around". A lot of the time, they can have less overall distortion in the region where "breakup" would occur due to this.
The other argument is dispersion pattern. A round cone, being symmetrical, disperses more "evenly" than an oddly shaped cone does. People claim that an oddly shaped cone, say a 6x9 for example, "beams like a 9" driver" which is not entirely true. Take a ribbon tweeter for example, which is much taller than it is wide. The narrower horizontal axis tend to have much better dispersion than the taller vertical axis. So the horizontal off axis performance is very good, where the vertical axis has "poor" dispersion comparatively. Is this necessarily a bad thing? Depends on the application. Of course, the frequencies affected are the ones that the wavelengths are smaller than the cone area, essentially frequencies "confined" within the dimensions of the speaker itself. The 6x9, for example, will beam like a 6" driver on the 6" axis, and like a 9" driver on the 9" axis. You could potentially use this to your advantage, especially in the car or a highly reflective environment where you want a higher direct to indirect sound ratio. A driver that beams like a 9" on a given axis means surfaces in line with said axis are not reflecting as much content as the speaker approaches the treble region, where the 6" axis can be aimed more toward the listener. This is not dissimilar to a line array, where multiple drivers are lined on a given axis, usually vertical to minimize floor and ceiling reflections. Of course, a line array causes the drivers to have cancellation on a certain axis across the full bandwidth of the drivers in question.
The last point: the intended application of an oddly shaped speaker is to offer more cone area where a round speaker otherwise would not fit. Not many people are considering 6x9s in place of an 8", which both of them are pretty close to one another in terms of surface area of the cone. Most people are entertaining putting something like a 6x9 in a place where the only other option is a 6 or 6.5" driver and perhaps a tweeter. A 6x9 has very close to double the cone area of a 6" driver. This can be effective in a lot of different ways. Such a driver can be designed to accomplish a much lower bass response (around an octave). Such a driver can be around 3db more sensitive on the same power level across the same bandwidth. Cone area wins over excursion for performance every time. There is a reason that pro audio opts for gigantic midbass and midrange drivers. You can accomplish much higher output levels and much lower distortion levels. To do the same with smaller drivers, you would have to have huge excursion (which despite how exotic the motor, you will always increase distortion as the required excursion level increases and the coil is forced to move further away from center position and the BL decreases). You would run into issues with power compression. You'd have to have monster amplifiers to drive them.
This all said, it is not indicative of the current offerings of the market as generally speaking, there are not many examples of "high end" oddly shaped drivers for car audio. But it is not due to an inherent disadvantage of the design itself, just due to a lack of good examples to base off of.
The common argument is that the larger and/or odd shape of the cone causes flex in the cone itself and creates more distortion due to a round shape being inherently more stiff. Also, the fact that oval speakers create an odd dispersion pattern as they approach beaming, which they do, but it's not necessarily a detriment.
The problem with this is that it is somehow thought that the common round speaker is inherently very stiff which really isn't the case. It is also believed that an oddly shaped speaker somehow needs huge amounts of reinforcement to approach any level of cone stiffness, which is also untrue. Most "cone flex" is due to the design of the cone itself. If you take a look at the typical speaker, the majority of them use polypropylene or paper cones, which most of them you can easily "bend" or flex simply by pressing on the edge of the cone near the surround. In fact, the only cone designs out there that really achieve a level of stiffness close to achieving the goal of true pistonic behavior would be extreme composites like Kevlar and some metal cones that have a LOT of material density such as the SEAS magnesium cones or the one case of their L18 7" driver which uses a specifically designed cone profile to achieve more stiffness and push the breakup node to a much higher frequency. The other issue presented is the material density required to get such a stiff cone usually results in a heavy cone in and of itself meaning a hit in sensitivity of the speaker.
What sounds worse: a broad, large increase in even order harmonic distortion in the lower midrange and bass or sharper but narrow band odd order harmonic distortion in the upper treble? This is the classic damped cone vs stiffer cone debate. A very stiff cone generally has lower distortion in the passband but much higher as it approaches the treble region (right before the point that you'd usually hand off to a tweeter) It also means that you are going to have a breakup node somewhere. In extremely stiff drivers like a W18 or W22EX SEAS magnesium cone, this is very apparent to the point that a notch filter has to be designed and the crossover has to be low enough so that harmonic distortion doesn't excite this breakup node. Though not to the same extent, breakup nodes still occur even in the case of softer poly and paper cones. If the design is soft (damped) enough, there's less and less chance of a big nasty breakup node, but more chance of having more distortion overall in the passband due to less than pistonic behavior. But round vs oval cones in this area? Referring to distortion/resonances generated by the cone itself, round cones tend to have a single breakup peak due to being symmetrical. Oval cones, due to their odd/dissimilar cone profile and their not being symmetrical all around tend to lessen single distortion peaks and "spread it around". A lot of the time, they can have less overall distortion in the region where "breakup" would occur due to this.
The other argument is dispersion pattern. A round cone, being symmetrical, disperses more "evenly" than an oddly shaped cone does. People claim that an oddly shaped cone, say a 6x9 for example, "beams like a 9" driver" which is not entirely true. Take a ribbon tweeter for example, which is much taller than it is wide. The narrower horizontal axis tend to have much better dispersion than the taller vertical axis. So the horizontal off axis performance is very good, where the vertical axis has "poor" dispersion comparatively. Is this necessarily a bad thing? Depends on the application. Of course, the frequencies affected are the ones that the wavelengths are smaller than the cone area, essentially frequencies "confined" within the dimensions of the speaker itself. The 6x9, for example, will beam like a 6" driver on the 6" axis, and like a 9" driver on the 9" axis. You could potentially use this to your advantage, especially in the car or a highly reflective environment where you want a higher direct to indirect sound ratio. A driver that beams like a 9" on a given axis means surfaces in line with said axis are not reflecting as much content as the speaker approaches the treble region, where the 6" axis can be aimed more toward the listener. This is not dissimilar to a line array, where multiple drivers are lined on a given axis, usually vertical to minimize floor and ceiling reflections. Of course, a line array causes the drivers to have cancellation on a certain axis across the full bandwidth of the drivers in question.
The last point: the intended application of an oddly shaped speaker is to offer more cone area where a round speaker otherwise would not fit. Not many people are considering 6x9s in place of an 8", which both of them are pretty close to one another in terms of surface area of the cone. Most people are entertaining putting something like a 6x9 in a place where the only other option is a 6 or 6.5" driver and perhaps a tweeter. A 6x9 has very close to double the cone area of a 6" driver. This can be effective in a lot of different ways. Such a driver can be designed to accomplish a much lower bass response (around an octave). Such a driver can be around 3db more sensitive on the same power level across the same bandwidth. Cone area wins over excursion for performance every time. There is a reason that pro audio opts for gigantic midbass and midrange drivers. You can accomplish much higher output levels and much lower distortion levels. To do the same with smaller drivers, you would have to have huge excursion (which despite how exotic the motor, you will always increase distortion as the required excursion level increases and the coil is forced to move further away from center position and the BL decreases). You would run into issues with power compression. You'd have to have monster amplifiers to drive them.
This all said, it is not indicative of the current offerings of the market as generally speaking, there are not many examples of "high end" oddly shaped drivers for car audio. But it is not due to an inherent disadvantage of the design itself, just due to a lack of good examples to base off of.
