If you read the previous chapters on Recording Electric Guitar Part II, you know there are a few microphones that are considered the most common in the field. Those are the AKG C214 and many other large diaphragm variants, the Sennheiser e609 and similar variants like the SM57, the Royer R121 which is insanely expensive - I cheat with the Slate ML-2, and the Sennheiser MD421 - again a cheat with an ML-2 emulation.
But what the aspiring audio nerds want to know is why are these mics so coveted for recording Electric Guitar? Unfortunately there are a lot of answers to that question, and most of them come down to a subjective opinion like “I’ve always used them and they always sound great!”
But I don’t want to stop at face value. Being a bit of a nerd myself, I wondered if I could I provide a scientific explanation as to why these microphones end-up in the mic locker of so many Engineers as their go-to electric guitar mic?
ALL MICS ARE NOT CREATED EQUAL
Before we even get into this analysis, I think it is important to highlight one key factor: Every microphone is different. Not only are the brands and models different, but no matter how advanced the technology, no single microphone is exactly like the other.
Take two SM57s and test them side by side. There will be differences in their frequency response. Then compare an SM57 with an e609, bigger differences. So we must know starting this examination that my results will vary slightly or greatly from your own.
SETUP
My basic setup consists of an AKG C214, Sennheiser e609, Royer R121 emulation, and a Sennheiser MD421 emulation.
Starting left to right, the first mic is a Slate ML-2 running the MD421. It is pointed 55 degrees away from the center and aimed directly at the mid-cone.
The next mic is another ML-2 running the Royer R121 emulation. It is also 55 degrees off axis but points to the mid-cone on the other side of the speaker.
On the right I have an e609 and a C214 both positioned with the center of their capsules at the dust cap edge.
I setup my DAW to generate Pink Noise - which is essentially a full audio spectrum - 20hz -20khz - of sound. I re-amped the output directly into my amp which is broadcasted out of my Vintage 30 speakers and into each respective microphone listed above. All dials on the amp are set to noon so the only color added to the signal would be from the circuitry, tubes, and speakers.
Let me be clear here, I am not trying to reproduce the frequency analysis by the vendor. I am trying to identify the frequencies picked-up by each of these mics in their present position and without pedals, EQ, or anything else that produces color getting in the way. Thereby allowing me to anlyize the effectiveness of each microphone.
My amp will add color. So will my cabinet. So the first thing I will be able to do with my analysis is compare it to the frequency chart from the vendor. We should see a consistent application of boost or cut in specific frequency ranges as a result of mic position and the amp & cabinet, which will also give us an indication of how they shape the sound.
In addition, mic placement will follow the standard positions I highlighted in the previous chapter: Recording Electric Guitar II. So we have continuity.
Once we are able to see the frequencies picked-up by each of these mics, I am going to bet the qualities we discussed in the previous chapter like warmth and crisp distortion are going to be quite visible on our captures.
FREQUENCY CHARTS
Before we get into the results, we should do a quick review of what a frequency chart looks like so we can discuss differences against our results.
If you look at the frequency analysis from an SM57, there are three key areas:
First, notice the bass roll-off starting at 200Hz. This is primarily the reason SM57s are not able to capture punchy lows and sub-lows. It is simply not built that way.
Second, between 200-2kHz we remain pretty flat. This means the mic is not really adding or removing color from the source.
And finally there are distinct boosts at 6 & 8.5kHz followed by a sharp high end cut that crosses the median or 0 at 13k. This means the SM57 is very much focused on High Mids, but has little need for High information (>10k).
Based on this spectrum, we would expect the SM57 to be very natural sounding in the Low-Mids (200-1kHz) but more focused in the High-Mids (1k-10kHz). And given the cuts in the Lows and Highs, we should not expect to capture much in these areas.
If this frequency analysis were flat across the board, more like the one below, we would expect the microphone to add or cut very little from the source.
But because it is not, we should expect the capture to have noticeable punchy High-Mids, and otherwise be somewhat unimpressive (lack representation that we hear with our ears in a natural room) in the Lows and Highs.
Compare this to a Beta SM58 which has somewhat similar boosts in the upper mids at 4.5k and 8.5k, but an extremely rich low-end response depending on the proximity to the source.
At just 2” from the source, it has a 5dB boost around 200Hz, where the SM57 actually starts to roll-off. So we can assume the 58 is going to have an excellent representation in the mid-highs & lows with also some better capture in the Lows and Highs than the 57 due to the wider response range.
So using some of these basic perspectives, we can now hypothesize what we expect to see from our guitar mics and compare it to the actual captures.
BASELINE ANALYSIS
Placing an analyzer on my Guitar bus, which is really just receiving pink noise through all of the mics, here is where I landed.
Averaging-out the peaks and dips from each track in the aggregate, this looks like a raw guitar track. Smiley-face EQ curves are usually quite pleasant sounding on a midrange heavy instrument. But a guitar is not generating the sound, just pink noise.
In fact, if I do an analysis of only the pink noise, no amp, no speaker, no mics, it looks something like this:
So we are generally starting with a flat curve, although the bump from 500 to 40 may result in some low-end bias. We will see.
RESULTS AND ANALYSIS
MD421 - The Sennheiser 421 has a factory spec frequency chart that is somewhat flat between 90-1kHz. For a mic that is supposed to be very good for tom toms, this drop is somewhat concerning because it does not appear to be capable of providing a focused capture on the more spongy, sub-lows like a 58.
Looking at the other end of the spectrum, there is a gradual boost starting at 1k that somewhat returns to 0 around 8k but really doesn’t get there until 12kHz. Though it is subtle looking, it actually adds 8dB, so looks are deceiving. The impact on the mid-highs and highs should be noticeable.
This bump, however, is reminiscent of the bumps we saw in the 57 and 58, although instead of sharp bells this boost appears more like a shelf. Shelfs are going to represent a more realistic capture than bells, so maybe this is where the 421 got its reputation. Especially for vocals!Overall, this curve appears to be similar to the 57 in range, but we should expect it to be a more natural, flatter sound in our capture.
Before we get deep into the 421’s response, let’s recall a few things. First remember in Recording Electric Guitars: Part II, I pointed the mic approx. 55 degrees away from the speaker center and directly at the mid-cone. So we should see the high-end information somewhat dampened in the analyzer. The 421 was the right-side mic.
That said, remember the high-end bump on the 421? Well there it is in the flesh, but it isn’t as perfectly flat as the spec:
NOTE: We will only be referring to the Red Line on the PAZ Analyzer which represents peak. The Yellow line is very active and only represents a moment in time.
If we take the median line, approx -40dB as our zero, we see this mic picks-up a lot of high-mid to high-end information between 5-7k. Which is what we expected from the spec, although not as flat or wide as the spec, which tells me the position on the cone and the speakers are probably encouraging the additional bias.
Interestingly, however, we also have a lot of high-end info beyond 12k where the 421 is supposed to drop-off dramatically. This could be caused by the Slate emulation that extends the high frequency response more like a shelf than a cut or drop-off.
On the opposite side of the spectrum, we expected to see the 421 drop-off the lows around 80Hz, and looking at the analysis, we started a little deeper around 70Hz, and the drop actually crosses 0 around 100Hz. But the results are what we should have expected.
MD421 Conclusion - Given we are not using the exact same source as the manufacturer, we didn’t expect our analysis to be identical. And because we have both an amp and speakers coloring the sound, which may have caused the bumps around 125 and 500Hz, in addition to mic placement, we also expected the response to be a little different.
The only surprises were the two bumps around 10k and 15k, but again, this is likely the emulation at play here. Is the additional high-end bad? Not at all. You can easily cut fizz from a track.
R121 - If you look at the frequency chart for a Royer 121, you would expect our results to be very similar to what our ears perceive in a natural room.
There is very little fiddling with the source from this microphone because it is capable of capturing most frequencies from 40-10kHz without change. Below 40Hz, there is little roll-off. Above 10k there is also a gradual roll-off but it starts to recover around 16k. It doesn’t look like it can capture anything above 18k.
Based on this factory spec, the 121 should provide a very realistic capture of our source. Of note, since there are no specs on Proximity Effect, I am unsure how this mic will perform at near distances. Depending on the results, we may dive deeper.
Also recall this mic was angled away from the center of the speaker, toward the opposite side of the cone. So it should cancel-out a lot of the high-end information from the dust cap, and yet grab some good tone from the mid-cone where it is aimed. It is the mic on the left in the picture below.
So how did it perform in the pink noise test? Well let’s just say I am seeing a much flatter response than the 421. My speaker is definitely tuned to boost some frequencies.
Remember, we should expect to see the R121 cut-out around 18k. Instead we see a sharp decline starting at 6k. We got more high-end info from the 421.
Moving down between 1-7kHz, we expected a pretty flat response. While the ML-2 does seem to handle the high-mids from Vintage 30 cone more smoothly than the 421, the 421 was also tuned for a boost in that range, so the 121 should not be as big of a bump, and yet it is.
For the most part, the boost in the upper mids is an acceptable response since we are only looking at a 4-6dB swing. Again, very manageable.
For the rest of the spectrum, we were expecting a fairly flat response all the way down to 40Hz where it should start to dive out. Our results aren’t so flat likely due to the Vintage 30s enriching the 125 & 500Hz bands.
R121 Conclusion - We got some of what we expected and some that we didn’t. My guess is the angle and position, including proximity, of the mic and the low-mid bumps are not its fault. Same with some of the mid-highs.
The R121 emulation on the whole is a flatter response than the MD421.
e609 - Interestingly, at distance, the e609 looks reminiscent of the MD421 though with a steeper high-end roll-off after 4k. It also has the same steep cut after 15k.
At closer proximity, and on the low-end of the chart, the e609 looks very much like the SM58 with an appreciable boost around 150Hz. So we would expect this mic at close proximity, say directly on the grill between 1-6” away from the cone, to produce some pretty luscious low mids.
Recall we positioned the e609 at the cone edge so we should see some good high information and based on the proximity, some good lows as well.
Did we have a similar experience as the other mics when we hit our e609 with pink noise?
Well on the high-end I would say yes. We did see a peak at 4k as expected, but also another around 6k. And then a lot more high-end than the spec as well; likely the e609 at Cap Edge. More high frequency information is going to be thrown at the mic from this position so I would say bumps in this area are expected. We will see if the C214 has the same experience.
Moving down the spectrum, we also saw some bumps around 125 and 500Hz which we also expected based on the last two mics. And with proximity, we should have seen more bass than the other two mics so far.
e609 Conclusion - Just as I had hoped, the 609 proved to follow pretty closely to the factory spec. In fact, this is the flattest response yet!
This is a gem of a mic! And comparing the factory spec frequency chart to the SM57, it has a much more dynamic low end, for the same price!
C214 - This Large Diaphragm Condenser by AKG a consistent capture across the 20-20kHz band due to the large capsule and design.
The C214 has a bass roll-off switch that we will not be engaging for our purposes, but again we see a substantive roll-off around 60Hz which is similar to the R121. This is also similar to most other high-end, large condensers, so we aren’t looking at an outlier.
The slight bumps between 1k-7k are also interesting, but amount to just 1-2dB, so they aren’t that significant. But the cut around 1.5k is noteworthy since 1.5k is where guitars and vocals - and anything else with a strong signal in this range - really cut through the mix. Again, the cut is not significant, but I can tell you from experience, this is one of the most dangerous bands for vocals and guitar. A few dBs too much and it’s like a gutter nail in your ear.
Where large diaphragm condensers are supposed to shine is in this high frequency bands, above 10k. And the C14 does not disappoint. Not only is there a notable boost at 13k, the bell or Q (Quality) is quite broad and continues thru 17k until it passes the median at 0dB.
Compared to the other mics, the C214 appears to grab the most low and high-end information and should be very realistic like the R121 spec.
Lastly, recall we placed the C214 at cap edge. High frequency information will be plentiful. But it’s those pesky bumps at 125 and 500 that I wanna target first! Are they the result of the mics or the V30s?
Drum roll please…
Yes, we see bumps at 125 and 500, so you can say with some degree of certainty the amp/V30s are likely causing the boosts here. But the C215 was hardly moved. In fact the dip at 250Hz is a pleasant surprise; usually a frequency range filled with mud.
The bump between 2-7k is also somewhat expected since we are at cone edge. But it’s nearly a 20dB bump! We didn’t expect to see anything at all like that given the factory spec.
The drop-off starting at 7k does not appear to be caused by the speaker. We got way higher detail from the other three mics. But I have to say I appreciate the control exhibited by the 214 in this area.
C214 Conclusion - You definitely can’t judge a mic by its frequency curve! While everything below 2k and above 7k looks good that bump between 2-7k is really stark.
But when I listen to this mic on a recorded guitar track, however, I have to say the high-mids are the best of all the mics. Punchy but not at all feeling like they could be distorted.
I also think this mic is likely an odd bird. Remember, two identical mics side by side may provide two different experiences. So this 214 may have a natural bias toward high-mids.
CONCLUSION
When I first began working with microphones I would obsess over the factory curves in an attempt to ensure the mic I was applying for the instrument.
But as we identified here, even mics that have an extremely flat response curve will pickup additional frequencies similar to other mics. So expecting a mic to “control” frequencies is putting a bit too much responsibility on the mic.
That and each microphone has its own character. They resonate and move differently based on proximity and different sound sources. You simply cannot judge them by their frequency curve.
If there is one rule about mics it is that you can’t really know for sure how much you will like them without hearing them with your own ears and looking at the results.
And like anything else, if you get good results from a mic, chances are you are going to use it frequently. I have found trying to determine if you are going to like a microphone based on marketing or frequency charts is a dead end. You really need to use it before you can make a confident decision.
Overall the ML-2 mics did produce some good results. But generally speaking, I prefer the capture I get from the e609 and C214.
I think mic emulators are a very versatile tool, but my preference would be to own the actual mic. Using the emulator, however, makes that decision a little less risky since I get to try all of them out first without the added expense of buying and selling a bunch of mics.
Commenti