12v Nick

Base Audio, Sound Package Plus(SPP), Standard Audio, ASK Audio, or non-Bose — whatever you call it, the zero-cost sound system option in the 992 is objectionably terrible. I've heard a lot of factory audio systems over the years and this is by far the worst I've experienced in a European car.

I recently got my hands on a 992 Carrera T with the base audio package and decided to do some testing to find out why the sound was so bad. I compiled those results and decided that sharing the information here could prove useful for anyone looking to make improvements, or simply to understand a bit more.


LISTENING TEST




The first logical step was to do some listening to evaluate the performance with some familiar recordings. I set the the EQ flat(bass and treble at 0), balance to the center and volume to 3/4. Here are my track choices and listening notes.


Under The Tallest Tree
Anette Askvik — Liberty



The vocals in this track are often shrill and overbearing. Which I don’t find them to be quite that in this car, I don’t find them to be very warm or inspiring either. It sounds like a lack of power leaves her sounding lifeless. Not surprisingly, there’s also no center image. My suspicion is that they’re not using any technology to optimize the sound specifically for the driver.

When the bass comes in, I can barely hear it. After boosting the bass to +9 it’s more prevalent and somewhat balanced. That boost adds warmth, but there’s still no immediacy or impact to the bass. Similar to what I’ve heard from other non-amplified systems. Porsche claims the SPP is a 150-watt system, but the dynamics are indicative of much less than that.


Blue Bayou
Linda Ronstadt — Simple Dreams



One of Val Garay’s productions, the Simple Dreams album is full of richness and warmth that’s very enjoyable when listening to Blue Bayou on a great sound system. In this car, I had to again boost the bass to get anywhere close to the warmth I’m used to hearing in this track. I expected the vocals in the chorus to be overbearing, but they were rather tame and laid back. At the same time, they also lacked any life or richness. This recording typically sounds large and inviting. Here, it sounds cold and anemic.



Babylon Sisters
Steely Dan — Gaucho



Arguably one of the best albums of all time. I love this song in particular in spite of its complexity. There are a multitude of different sounds all overlapped, but somehow still with enough separation to stay balanced. Those instruments are far less discernible in this system than they are in others. Even the Bose system can pull out more musical information than this. Again, bass is hollow and anemic. The key strikes are quite shrill as well.



Theatric Nat Tricks
Smilk — Last Rainforest



This is a really fun track to give your system a workout, especially in the bass response. While the synth notes weren’t as boring as I expected, the bass was incredibly disappointing. This track has enough dynamics to give you a light tap in the chest to the tempo. Not a single bit of that came through, even with bass at +9 on the PCM. The click at the beginning of the track also has none of the snap the original production has. The most underwhelming track so far in this 992.



’Til The Dawn (feat. Sugar Free & Bosko)
E-40 — The Ballatitian



Not every Hi-Fi nut is into hip-hop, but those of us who are can usually be found stuck in the 90s-00’s era. This track from E40 is a surprisingly fascinating production that starts out with a thick synth bass followed by a powerful kick and some great tricks with the clap track. After the last song I didn’t have confidence that this was going to be enjoyable, and I wasn’t wrong. The bass had no authority or impactfulness and the difference between the synth and kick was nonexistent. The clap was separated but had zero bite to it.


Final Impressions

Underwhelming and disappointing. The sound leaves a lot to be desired in terms of dynamics and presence. It sounds underpowered and over-processed using cheap equipment, which is exactly what I would expect for a zero cost option in a $20k car — not in a $120k car.

Up until now, all of this testing was done with the windows up and the engine off!!! I took the car on a brief drive and decided the system is unlistenable while in motion. To overcome the road/ambient noise(which is not NEARLY as much as some have made it out to be), I had to turn the system volume up to 100%. The bass didn’t increase in volume at all and the treble on slightly from 75%. I had nowhere near enough volume to hear any of the subtle information in the music. I could tell what was playing, but only by picking out bits and pieces.

Okay, all that said, it’s time to take a more objectionable look into why this system sucks.


SYSTEM ARCHITECTURE

This is straight from Porsche’s literature





The system is comprised of eight speakers, and as outline above, that includes a 3-way component system in the front with a pair of Wideband drivers in the rear. No subwoofer, no center channel and no rear tweeters. It’s also powered directly by the PCM on only 6 discrete channels. The rear speakers and front woofers have their own independent channels, but the door midrange and dash tweeters share an output channel per side. This is a pretty common practice in base audio systems to save money.

Since rear speakers are fairly inconsequential when it comes to what we hear in the driver’s seat, the following tests are all going to focus on the front 6 speakers and the 4 PCM outputs that power them. Speaking of power, let’s start there.


POWER MEASUREMENTS

This system sounds severely underpowered, but there could be a lot of processing making it sound that way. With an oscilloscope, we can measure the A/C voltage output at a certain frequency, combine that into a formal with the impedance of the speaker and get the approximate power output. Here are those results:





Wow… it’s starting to make sense. I had to triple check these measurements to be sure I wasn’t doing my math or test incorrectly. With only 7-10 watts per channel, these no way we’re going to get anywhere close to an enjoyable sound. This power is equivalent to what we used to see in base audio systems from the ‘00s. Again, we didn’t measure the rear, but we can assume we’re getting similar results.

Lack of power yields two major handicaps.

1.) POOR DYNAMICS — Dynamic range refers to the amplitude difference between the loudest and quietest notes in a recording. Good dynamic range makes things sound more natural and realistic. With low power we have to turn the volume up higher to get the same output. By the time we max the volume out we’ve reached the limit of the quiet notes but we don’t have enough power to make the louder notes any louder.




Once we factor in the road noise there’s barely enough output left to discern what you’re listening to, much less enjoy the music.



2.) DISTORTION — When we run out of power we run into distortion quickly. This is from feeding the system more signal than it can realistically amplify. At that point, the sine waves start to distort and eventually clip. Not only does this sound terrible but it destroys the equipment.




You can see from my initial measurements above that the sine wave in the SPP never distorted, even at full volume. That means they’re internally limiting the amount of volume to avoid distortion. This system would get substantially louder if they allowed for a little distortion. But, it wouldn’t sound very good and it’s already pretty bad.


ACOUSTIC MEASUREMENTS

What does all of this add up to in the driver’s seat? Let’s find out!




Playing pink noise(completely flat static noise) is a great way to run tests. We play this signal through the system via USB and, using an omnidirectional microphone, we take a measurement of the acoustic response.





In a perfect world, the response would match the red target curve. This is generally accepted by most audio tuners and car audio enthusiasts as a pleasing frequency response. We can see just how far off we are. The bass is heavily attenuated with lots of major dips and peaks. The upper midrange is highly boosted and the treble is mildly boosted, again with lots of peaks/dips in the response.





Adjusting the bass to +9 and the treble to -9 we get something that looks better on the RTA, but still sounds very uninspiring in the car.





One step further I adjusted the EQ within the app on my phone to attempt to compensate even further. This is as close as I could get to the target curve. Sounded terrible in the car. The hypothesis is that there’s just too much processing and too little power. If we want a pleasing response, we would need more than 4x the power and the ability to control those non-linearities in the response.


ELECTRICAL SIGNAL MEASUREMENTS

Now that we’ve determined the system is underpowered and lacks controllability, let’s see what they’re doing to the original signal in an attempt to make the sound suck less. To do that, we need to play pink noise again and measure the signal the PCM is sending to the speakers. If there were no processing, these signals would be completely flat with only crossovers to split the signal between woofer and midrange/tweeter.

This first measurement shows the signal from the front woofers and midrange/tweeter signal. It looks like the woofers are playing 40Hz-400Hz and the mid/tweet signal overlaps a bit at 250Hz-20,000hz(20kHz). We also have a decent amount of pre-equalization applies to both channels. This is likely their way of compensating for major inefficiencies in the SPP speakers or their locations.



To continue investigating the factory signal, I first need to flatten those two responses out. I use a powerful DSP to do just that and we end up with what you see above. The woofers output seems to be slightly lower than the mid/tweet, but this does not take impedance into account.



Now that they’re flat we can combine those signals into one “Summed” signal. This overlapping/merging will let us know if there’s any time delay or phase alignment done to either channel. If they do not sum together perfectly and remain flat, we have more processing to defeat.



Woah, so they certainly didn’t sum perfectly. Before I continue, I will match the output levels on the two channels to make it easier to identify the non-linearities.



Now we have a better picture of the summed signal. Looking at the dips at variable octaves, I applied time delay to the woofer channels until the response was what you see below



This indicates that the midrange/tweeter channels have .85ms of delay applied. This is likely an effort to prevent cancellation in their overlapping frequencies. The adjustments were identical for left and right, so they’re obviously not doing this for the driver or passenger individually. With the one large remaining hole, we can assume this is an all-pass filter applied to further help cancellation in the pass-band.



With an Allpass filter applied @330hz to both woofer channels, we have a relatively flat and usable signal. This would be perfect for plumbing into an aftermarket amplifier if we were doing an upgrade. That said, there’s still one more data point here, and frankly this is the only factory processing I approve of.



Up until now, all of our testing has been done with the volume at 75%(23/30). Once we have a flat signal there, we can clearly see that there’s a dynamic loudness curve that varies with volume. This is actually a great way to get more dynamics at lower volumes and help help safeguard your speakers from damage at high volumes. While this does nothing useful for your factory system, it would be ideal for an aftermarket setup if you corrected the signal in all of the above ways.


THE SPEAKERS

So what does all this signal feed? Some pretty cheap-looking speakers.














Motor structures on the woofers and midrange are designed efficiently in an attempt to make use of the low power. Even these small coils and magnets need 2x the power or better to get moving. The tiny diaphragm on the tweeter is also a recipe for shrill and overly-bright response. Thier only saving grace is the fact that they aren’t given the opportunity to get loud enough to hurt.



TLDR— The SPP uses crappy speakers, only 60 watts total rms power and over-processes the heck out of the signal in a failed attempt to compensate for those shortcomings. What a terrible surprise for anyone that was forced to select this option thinking they’d be able to enjoy music.



Hopefully this helps answer some questions, but I’m happy to answer more about this system if anyone else is curious!



– Nick

Dr. Z

Are you going to be able to put together a “package” system that can be installed to replace this base system?
Thanks

VarTheVar

@12v Nick I found when I moved the fader all the way to the rear, those speakers were actually contributing a really bad sound, like it was coming through a telephone, and it sounds like there's a massive delay. I move the fader all the way forward for daily use which is obviously a pathetic fix.

Have you tried to isolate the rear speakers to hear what they sound like? Are the rear speakers even necessary for a 2 seater? If I unplugged them for now so I could move the fader back to the middle, would it damage the amp in the head unit in any way?

Edward911

What a great analysis! This is more like something I'd expect to find on AVSforum.

Are the Bose and Burmester systems performant at all, or would you expect aftermarket to be superior to them? Put another way, is it worth getting SPP for $0 and spending the equivalent of Bose / Burmester $ on aftermarket?

991.1 Guy

One thing I noticed immediately after the flash update of my PCM was the change in the quality of the sound in my base system. The low end was taken away probably in an attempt to keep from having to replace blown speakers and the rattle associated with them. The first thing I tried to do was adjust the base but nope. It won't go any higher than about half of what it was before.

Shogunade

This kinda passion and detail is what's lost these days in any field. Old man rant over. Kudos, Nick.

svkarl

Nick,
super impressive analysis, another reason that makes your team top notch!
best,
Karl

20C4S

Second that. Would love to know if BOSE is worth the optional cost.

Icegrill

For $1600 it’s more than adequate. For the true ‘ethos’ of the original “T” Carrera, the 992T comes with a 28w factory head unit as well as kraco 4” drivers in the doors

detansinn

Thank you for the detailed analysis. The base audio setup has never been commensurate with a six figure car.
I know that the "noise of the engine" crowd loves to chime in on it being superfluous, but if you're going anywhere other than cars & coffee, it matters.

Bose should be the base audio in a 911. For a Taycan, spend on the Burmester -- a spec sheet regret of mine.

MoeMistry

Nicely done Nick

Go Bruins

Get the Bose, and then get Nick's upgrade (he has three options, but splurge for Stage 3 if you can), and you'll have a system to rival anything on the road.

c1pher

This is great but it doesn’t take a rocket scientist to realize that you get the base system because you mainly listen to the car, want to save weight or only need speakers for phone calls or other audio alerts. Even the high end system isn’t perfect. I feel like getting a perfect sounding stereo in a car is a somewhat fleeting exercise. Can you spend a lot of money and get something decent? Sure but in the end, you’re in a metal box that’s not designed to be a sound studio. This is why I got the Bose system and just live with it.

detansinn

Bose is the Goldilocks choice for the majority of people.

6172crew

I have the base system chosen on my T because I wasn’t sure the Bose is worth $1600 and part of me hoped $1600 would be better spent on a set of speakers but now I’m thinking it might not be enough and I’m not willing to spend a lot on stereo stuff unless it was a nice as the one I had in my old stinger. That one sounded like a nice set of head phones.

I have until next month to lock in the build so I hope this thread gets filled with a ton more information.