WinPCD Results Different Than Xsim/VituixCAD

[dcibel]

I said simulator, should have said emulator. You can run SoundEasy (SE) as a DSP crossover emulator, much the way the Ultimate Equalizer (UE) is. It's an active DSP unit for auditioning, what I find the most useful in SE.

Yes, I know, and exactly what my comment about VituixCAD and APO EQ conveys. APO EQ is the DSP engine, intercepting all audio on your PC for processing before output. It provides additional benefit over SE's implementation, because it applies the EQ process to all audio playback from applications on your computer, you don't have to feed it an external audio source to the soundcard input. Just follow the instruction I provided, open your favourite audio player and hit play.

Of course, I never debated that differently. What I did and do say is that the 3-measurement method provides the most accurate modeling and is required to get a the best relative positioning of each driver, so I find minimum-phase measurements to be ideal.

Well, I'd disagree a bit on that one. Your comment in post 13 was that minimum phase is essential and required, which it isn't, which started this whole debate. I've mentioned before and I'll do it again. Do your 3 measurement process, do it once with min phase and HBT, and once without, you may arrive with a different delay value simply due to modified phase, but with each delay value implemented in the crossover design, results will be just the same.

I've not seen a way to determine relative offsets accurately otherwise. Even measured phase with correlated phases between drivers requires that the model position the driver with all three (x,y,z) values if you want the off-axis to be correct. This would require knowing the true acoustic center, but that is unknown to high degree of accuracy for drivers. The relative offset method obviates the need to know that.

No, phase will be captured with constant mic distance to baffle surface, that's all you need for accurate results with 2-channel measurement (we're going in circles). True acoustic centre is irrelevant, relative phase between drivers is. I've already mentioned, off-axis is interpolated measured data with same measurement conditions for all responses (distance from mic to baffle surface and same window start reference time). We don't have to debate this, it's easily proven with real measured data. I encourage you to set up 2-channel measurement, follow measurement guide for VituixCAD with your measurement software of choice to generate the data and try it for yourself.

I take that to mean that you left the z-axis at zero. If that is so, then any off-axis result will have some degree of error, usually increasing the farther off-axis for most of the front hemisphere. That's been one of my main points all along. Any system can easily model the single design point with measured phase. I did that any number of times over the years when I wasn't very concerned with the off-axis. It was always a good, easy starting point. But I am more interested in the off-axis, especially as it relates to the power response. I'm not too concerned with 5 or 10 degrees horizontal off-axis change, but that's because I use felt liberally, so the on-axis are both much smoother. But there is little impact of felt on power response.

Don't model off-axis, measure it. Models can help with decision making for design, but final crossover design can and should use real measured driver data, this is the VituixCAD way. It doesn't matter if standard dome tweeter or ribbon or waveguide or horn or some experimental diffuser is used, driver with flat piston or weird shape, just provide the software with measured data to apply the crossover to, it will do it with the accuracy of the data provided, simply because it doesn't rely on any model for off-axis beyond interpolating the measured data and distance calculation for phase and SPL adjustment. If you want to keep discussing models with calculated off-axis data, I'm sorry but I have no interest.

Not in any software I've ever used.

It sounds like you haven't used VituixCAD ;)

For your modelled results using 2-channel measurement at single mic location that requires zero delay applied, you should try including whatever 3D positioning you are doing, just without modifying phase delay based on coordinates entered, because the delay is already included in the measured data it doesn't need to be added to the response file provided.

[dlr]

Yes, I know, and exactly what my comment about VituixCAD and APO EQ conveys. APO EQ is the DSP engine, intercepting all audio on your PC for processing before output. It provides additional benefit over SE's implementation, because it applies the EQ process to all audio playback from applications on your computer, you don't have to feed it an external audio source to the soundcard input. Just follow the instruction I provided, open your favourite audio player and hit play.

I may look into that at some point.

Well, I'd disagree a bit on that one. Your comment in post 13 was that minimum phase is essential and required, which it isn't, which started this whole debate. I've mentioned before and I'll do it again. Do your 3 measurement process, do it once with min phase and HBT, and once without, you may arrive with a different delay value simply due to modified phase, but with each delay value implemented in the crossover design, results will be just the same.

I still say it's required for the best modeling that includes off-axis. I guess I wasn't clear in the beginning. Most software doesn't import a large number of off-axis points, so it's still the best to model in that case. Software that models is still what many prefer to use because off-axis measurements are time consuming. I also am curious as to how the power response is then calculated because it is not just a simple summed average of the off-axis points. Power response is probalby the second highest priority and not only my opinion.

No, phase will be captured with constant mic distance to baffle surface, that's all you need for accurate results with 2-channel measurement (we're going in circles). True acoustic centre is irrelevant, relative phase between drivers is. I've already mentioned, off-axis is interpolated measured data with same measurement conditions for all responses (distance from mic to baffle surface and same window start reference time). We don't have to debate this, it's easily proven with real measured data. I encourage you to set up 2-channel measurement, follow measurement guide for VituixCAD with your measurement software of choice to generate the data and try it for yourself.

You're focused on measured off-axis. I'm focused on modeled off-axis, much easier and reasonable since most listening is done at or near the on-axis. The off-axis is most important for power and that can be modeled rather well.

If you're using direct measurement phase, you have phase to the mic point. But the origin reference of that is not the baffle front, it is the actual driver acoustic center. If the off-axis is calculated based on the front baffle (x,y,z) = (0,yValue,0), then the off-axis points cannot be calculated properly due to simple geometry. If you're only interested in using all measurements, then that's moot. But for modeling software, what I am concerned with, it's crucial to have the correct offsets. For full accuracy, that is. Maybe it's sufficient with whatever errors must occur, that's a different debate. I guess that the particular caveats need to be included.

Don't model off-axis, measure it. Models can help with decision making for design, but final crossover design can and should use real measured driver data, this is the VituixCAD way. It doesn't matter if standard dome tweeter or ribbon or waveguide or horn or some experimental diffuser is used, driver with flat piston or weird shape, just provide the software with measured data to apply the crossover to, it will do it with the accuracy of the data provided, simply because it doesn't rely on any model for off-axis beyond interpolating the measured data and distance calculation for phase and SPL adjustment. If you want to keep discussing models with calculated off-axis data, I'm sorry but I have no interest.

I understand that, but modeling was central to my comments. I thought I had made that clear at some point.

It sounds like you haven't used VituixCAD ;)

Only briefly, but had no need to spend time learning it.

dlr

p.s. In any case, good debate.

[dcibel]

You're focused on measured off-axis. I'm focused on modeled off-axis, much easier and reasonable since most listening is done at or near the on-axis. The off-axis is most important for power and that can be modeled rather well.

Yes, and probably why we have gone in circles somewhat. We can agree to disagree, I'm fine with that. Off-axis model that is nothing more than a piston calculation is only accurate for flat piston drivers with crossover well outside of breakup range. So it only applies with any hope of correlation in certain conditions, less than half of my DIY speakers currently in my house, maybe 50% of overall DIY efforts over the years. I will assume you'll disagree with this, but that's where I'm at on that topic regardless.

I'm focused on obtaining the most accurate simulated result that matches as close to reality with as little effort as possible. 2 channel measurement and VituixCAD is the best I've used (yet), with CTA-2034 power & DI chart and directivity maps you can focus effort to on or off-axis and see the balance of the entire system with high reliability and confidence. There is greater accuracy in the "model" of the crossover using real measured data than any computer simulated off-axis model currently available.

FWIW you can apply the piston model using VituixCAD diffraction tool with "directivity" check box, but due to limitations of the piston model (breakup behavior and complex shape are not included), accuracy is low above 1kHz so it is meant for application of nearfield low frequency data primarily.