When it comes to choosing the right speaker system for your edit bay, home studio, or even pro studio, it’s a jungle out there. Over two-hundred possibilities from eager manufacturers big and small – and that’s just what’s represented on the floor of the NAMM show. Most of what’s available is – well – crap. And anything below $100/ speaker is – as of the writing of this article – false economy when it comes to serious listening. I’m not talking about listing to music for fun or having a party. I’m talking about being able to mix or edit the audio of a project listening on speakers which will have that work translate universally to any speaker system. That’s a tall order given the current demands of audio these days. What follows is another Cinema Sound research segment that takes a few nearfield monitor possibilities which we believe are most viable in the market and let them shoot it out for top dog at high noon.
Over the years we’ve worked with all kinds of nearfield monitors – and not all of them were good. We’ve seen all kinds of gimmicks to get hapless mixers to buy a speaker system: colors, special tweeter enclosures, weird casing shapes, overpowered amps and a host of other things that really just shine turds. In the end, the most successful speaker is one whose output is flat enough and snappy enough to represent what our audio files are going to sound like anywhere. It doesn’t matter the price, configuration or size. If it does that, then I call it a success. There haven’t been that many monitors which can claim that in my mind.
Now, of course, you can get used to nearly anything once you know what a speaker is doing and not doing to the sound. What’s more, some speakers are more well-designed for certain uses. For example: the venerable Genelec 1032B is a fantastic monitor to track or record music or sound effects in studio because of its strong high-mid frequency shelf boost. I mean, you hear EVERY little nuance and every little artifact with those speakers. But I can also tell you that mixing on them in 5.1 for more than 6 hours has me want to take a nap and has my ears ringing at 80 db. They’re just too “edgy,” and my mixes would turn out dull in the high frequencies.
Another example – which is in this Shootout: the venerable Mackie HR824 MK 1. I know you say you didn’t mix on them – but we all did. I did proudly in my small studios for years. And while to my ears the passive “subwoofer” in the back made for great reproduction of frequencies down to 37 HZ – as long as they were sustained. If you had to reproduce anything transient below 80 Hz, it was like listening to prize fighters punching sponges. Woof, Woof instead of Pow, Pow.
Each speaker, as long as it meets the base standards of frequency response and loudness – which to me is 20-20kHz response with ±5dB variance from flat from 50Hz to 15kHZ, and can do 105 dB at 48″ – every speaker has a use.
But we’re not concerned with that. We want to know what monitor brings us the most translatable sound overall, for the least money and the sexiest look so our clients feel like we’re pros when they walk in. We also want to bias this choice towards multimedia mixing purposes which means these speakers are going to take a lot of abuse and must be pleasing to listen to for dozens of hours a week. Let’s begin.
What Really Matters
After seeing all the gimmicks over the years, what we’ve categorized as very important features of a professional speaker are the following:
- Two-way speaker. There’s a high frequency speaker and a low frequency speaker. Any more than this and it’s too hard to find the sweet spot of the speaker in order to stay in phase with the drivers. Any fewer and we’re sacrificing too much.
- The Tweeter has some kind of horn or enclosure which passively helps expand the otherwise very directional high frequencies. The simpler, the better.
- Active amplification. While it’s true that amplifiers these days are a quantum leap over where they were even 10 years ago, a speaker which is designed around its power source appears to always be superior.
- The woofer is a circular speaker – not some weird shape.
- The tweeter is a circular shape – not a reflex system or whatever. This rule is not because other systems are inferior. Quite the contrary. We’ve heard some contrarian tweeters which are far superior. But right now no one has them, and what you hear from those speakers in the ever important high frequencies will not translate to anyone else’s system.
- Both drivers are phase aligned.
- The cabinet is some form of wood.
- They’re sexy to look at.
We never buy nearfield speakers that don’t have these qualifications at minimum. Please don’t confuse nearfield speakers with PA speakers or large format studio speakers as those requirements are far different than nearfield speakers. For example: we’d never consider a simple enclosure for a tweeter system for a PA speaker or limiting it to only two drivers. Similarly, we’d never consider an active system for a large format studio system. This article is for nearfields only, and so we jump into what we’ve discovered.
Taking the above rules for nearfield monitors, unlike our Battery Shootout article where we took every Amazon Prime shippable battery available, we chose speakers which we already had on hand, or we’d seen other Cinema Sound Members use. We didn’t take into account price or size or age or availability (as in the case of the Mackie HR824 or JBL LSR 305/308 MKIs). Why not? Well, to do an exhaustive test of dozens of speakers would be way beyond our budgetary constraints (but we’d be happy to do it if someone wanted to sponsor it), and we also believe it’s not necessary. We know that the majority of you aren’t going to spend $2,000/speaker on your next system – or even $500. Instead, based on our interactions with you, you’d rather not spend more than $200/speaker. That said, there are very few speakers worth dueling with when one of the least expensive speakers in a Shootout like this is the JBL LSR 305 MKI – which is one of the highest value, best sounding speakers around – especially in its updated MKII format – which for $260 a pair?!? Are you kidding me?
This study comprises the following speakers:
- Mackie HR 824 MKI (discontinued)
- Tannoy Reveal 502 $118
- KRK Rokit 5 G3 $145
- Presonus Eris 5 $149
All of these speakers are 5″ woofer variety except for the 708, 308 and HR824. I’ve thrown these in, because I wanted to show how an 8″ cabinet would compare to a 5″ cabinet. And two of these speakers are long gone from retail (but are still available on Ebay) – and as far as I’m concerned the 708ps are the top nearfield monitor in the world regardless of price. I also included the 305 MKIs, because many of you own them (on my recommendation), and I wanted you to see what the difference between the MKIs and MKIIs are. So really, this Shootout comes down to the 5″ers in the form of the Tannoy, KRK, Presonus and JBL 305MKII. And, hey, if you want to consider the 708s, or the 308 MKIIs (which aren’t plotted here but have similar improvement as the 305s) or you’d love to find an old pair of Mackies, wonderful.
We chose speakers which matched the criteria above, which we had or we saw Members using and which we believed could possibly beat the 305 MKIs for the goals listed below.
We took each speaker and placed it in our soundproofed studio (Cinema Sound Studio B), on a padded bench without a recoil stabilizer. We placed an Audix TM1Plus Measurement Microphone 36″ away from the front plate of the speaker and ran pink noise at 80 dB. We recorded the Audix on the Zoom F8 @ 192kHz/24bit. The resulting recording was placed in iZotope RX and scanned on the Spectrum Analyzer and then placed into Adobe Photoshop for analysis.
Please note: our “soundproof studio” is far from reflection proof. We deliberately did not do this test in an anechoic chamber. As a result, there are a host of reflection and phase issues as well as comb filtering and delay problems from being in an enclosed space. Even a well-treated room will have all kinds of reflection issues. Why did we do this? Quite frankly, because none of you will be mixing in the anechoic chambers that the speaker manufacturers tested their speakers in. And the only real way to test a speaker is in a real environment.
Each speaker was tested with the microphone on-axis, 45º off axis both sides, 90º off axis both sides, and 180º off axis in order to test dispersal patterns of high frequencies. If a speaker is too directional, it is useless, because its sweet spot is too narrow to hold one’s body within.
We then took the RX spectrum analysies and placed atop each other, colorized for clarity.
The second test was for bass frequency transient response. Here, we took each speaker into a corner of my living room and placed the Audix 36″ inches away. We recorded the same way on the same recorder, and brought the files into RX and have returned the results visually below.
What you’ll see might be otherwise disturbing at first glance. We usually want to believe that a speaker has a very flat frequency response – like the manufacturers tells us they are (the same for microphones). And while I have to believe this can be true in a perfect testing environment, in the real world with real reflections this is far from true. However, don’t be dismayed. It will all make sense by the end!
The goal of this Shootout is to find a speaker (that you can afford) which brings the flattest frequency response with the best transient response from those tested. As you’ll see, it’s not always an easy choice, since not all speakers are good in every area. Even the most expensive speaker in the group, the JBL Series 7, 708P has some things to consider before dropping $4,000 on a pair.
Still, we believe the results below should give you enough understanding and visual evidence to help you make a great choice in your next nearfield monitor experience – and maybe have you feel great about what you have or want to go out and upgrade.
The Frequency Plots
What you’ve been waiting for: the evidence. As I mentioned before, these graphs are going to freak most of you out. “How can a speaker be THAT bad!?!” I promise you that if in an anechoic chamber any speaker pulled a frequency plot like what’s below, they’d use it firewood. Again, in a real room with real reflections and architecture, speakers behave differently. And depending on the design of the speaker, those issues affect the sound less or more badly.
A standing wave is a reflection that goes to-and-fro from one architectural place to another creating a medium to strong amplification of a particular frequency. And since different frequencies have differing wavelengths, when a full bandwidth signal (like pink noise) is put into a room, “comb filtering” occurs. This is where a series of frequencies are both amplified and attenuated from those standing waves and other reflections in a way that looks like a “comb” on a frequency plot. You can also have wide-range dips and boosts in the lower frequencies. All of these are “reflected” in the results below and I’ve marked the most obvious offenders so you don’t expect that it’s a problem with the speaker.
That said, even with the standing wave and comb filtering issues of the room (and BTW, this is a professionally tuned and deadened room – imagine what the graphs would look like in you living room) it becomes clear what the average of the frequency plots are showing us. If you’re unsure, just imagine drawing a line between the “bumpy” areas of up-and-down plotting and you’ll see clear trends.
In this first graph you can see what the “perfect” plot of pink noise would be. Here, we ran straight pink noise from Adobe Audition’s generator directly into RX (the brown line). This gives you the nature of this graph for “flat” response. The red line reflects the frequency response graph on the box of the Audix. We bent it around to fit this plot (since on the box it’s across a horizontal axis). This represents what the response of the microphone is as juxtaposed against the perfect pink noise representations. Thus, the closer a frequency plot gets to the wide red line, the more “flat” it is in the real world – as recorded by the microphone.
This plot shows the response of the JBL 708P. As with all the following graphs, green is on axis, blue are the two sides of 45º, purple are the two sides of 90º and small red is 180º (or completely facing away). The wide red Audix frequency response line is there demonstrating what “perfect” would look like.
On the whole “squiggly” lines demonstrate standing wave issues in the room or other reflectivity issues which are not the fault of the speaker. The weird bump at 10K is not a 5 dB boost coming from the speaker, I promise. The same with the large rise in the mid-hundreds.
Although it’s not clear from this first plot, the 708s have by far the most uniform frequency response. If you were to draw a line through all the squiggelies you’d find that it pretty closely resembles the Audix line. There does appear to be a bit of a high frequency loss of about 1-2 dB from 12K up, and a strong ultrasonic fall off which no other speaker in this list has. But by and large, you get what you pay for with the 708s and I can tell you as much from my own experience. The transient response test demonstrates that powerfully.
It’s also important to note the incredible high frequency dispersal from the 45º off axis plots. This is incredibly important, because the wider the sweet spot the more space there is for you to move and work – and to have a client listen next to you. A bad off axis response shows that a speaker’s high frequencies are too narrow – OR if the frequencies are up-and-down across the off axis spectrum it shows that the waveguide is inferior.
I’ve had 308 MKIs for some time. Generally, I find them a bit strong in the 2k zone, and I’m glad to see that my ears weren’t lying to me. Although the room doesn’t help with the bump at 1.8k, along with the strong ultrasonic bump, the 308s are great for tracking, but perhaps not so great for listening. The good news is that the update for these speakers have largely solved this issue. There’s also a bit of low frequency roll off. However, like the 708p, the off axis high frequency dispersal is excellent which makes this a great speaker to have in studio for tracking or Voice Over recording.
I’ve been dropping 308 MKIs into studios for nearly 5 years with great results. In fact if you look up at the Avenger’s STATION exhibit in Las Vegas, you’ll probably see our install of over 14 of them there. Like their big brother 708p, the LSR 305 has a wonderfully flat frequency response with the exception, in this case, of some loss around 8-12k. It has incredible off-axis dispersal, and has a crazy-good bass response down to 50 Hz. For a 5″ woofer, that’s pretty awesome. It is a bit under powered if you’re really trying to get loud in a mix, but for reasonable mixing levels it’s fine. There is a bit of an ultrasonic boost which is not represented in off axis dispersal. This may have mixes feel too harsh in the highest frequencies, but for the money (now only available on Ebay), this is a great speaker.
I only have a short relationship with the MKIIs, but everything I know about them so far is exceptional. You can see how much they’ve fixed between Mark I & II when comparing these two graphs. And while there’s still a dip at 12K, everything else seems to have been resolved. This speaker, interestingly, has an added twist that 180º off axis has a strong mid-low frequency fall off which wasn’t present in the MKI. The 45º off axis is nearly identical to the on axis – which is nearly impossible to do – and which makes this a premium speaker to have where multiple listeners need to be present. But JBL figured out how to do it. It’s rare that a MKII speaker is better than the first (as long as the first is good), but in this case, without question this is an improvement.
Perhaps more than any speaker I’ve used, my love affair with the original HR 824 has gone on the longest. Not until I bought my first pair of JBL 6328ps did I find a more easy-to-mix on speaker. And as you can see from the graph, with the exception of the high frequency shelf, it’s quite flat even though the low frequencies are similarly missing. Of course, thanks to its passive “sub woofer” if you push it against a wall, you’ll get more lows than you’ll know what to do with, but in open air, the 824 lacks bottom. Still, and maybe in part to the high frequencies being less, they’re wonderful to spend a 17 hour day mixing on, and I rarely experienced fatigue. However, the horrific off axis response reminded me that they were only good for solo listening where one is directly in front of the speakers. The sunken-in tweeter design did nothing to help disperse highly directional high frequencies. The further irony being the big passive sub woofer in the back didn’t help 180º off axis bass frequencies.
Tannoy has historically been an excellent manufacturer of speakers, and the Reveal 502s – especially for the sub $120 zone – are quite good. The frequency response is a bit ragged, but of all the speakers it appeared to handle the room reflections and comb filtering the best. Unfortunately, there is a pronounced high frequency shelf from 8K and up – which will doubtless have you turn down your high frequencies and end up with dull mixes. There’s also a pronounced low frequency loss from 55 Hz and lower. These are certainly NOT flat speakers, and from this plot alone should be disqualified. If I had a pair of these, I’d sell them and get something else.
Similar to the JBL 305 MKIs, if you look in the Avenger’s STATION exhibit you’ll find a few dozen E5s creating sound effects. These are speakers which have an incredibly loud output. I mean, these things are half way to Yankee Stadium PA volume. They do have an extremely brutal low frequency cut off at around 60 Hz. It’s not nice. There’s also a pretty good-sized high frequency dip. What’s interesting is that their 90º frequency response is more flat than their 45º response. We used these speakers specifically because of their volume characteristics, but we would not recommend them for mixing of any kind.
I’ve seen KRKs in studios for years. Never liked them. I understand that their initial marketing was for music studios where there was techno or hip hop, because they were supposed to have “big bottom.” Now, why anyone would want that while mixing escapes me, but as we can see from this graph there’s certainly none of that in these speakers. The off axis high frequency response is really wonky which means you must really stay in front of the speakers to get any kind of good response. There’s also really wonky high frequency response both dipping and boosting in the high to ultrasonic range. Also notice the mid frequency dip. These are not flat speakers. They’re probably not even good to listen to music for enjoyment.
Putting all the plots together, you can quickly see some powerful trends. Letter “A” reveals how brutally the reflections of the room can affect each speaker differently in the high frequencies. Between the highest boosted high frequency Tannoys to the highest cut highs in the Presonus, there’s over 6 dB of difference – over double the volume difference between them. Again, the nature of sound in a room radically affects the sound of a speaker. But from knowing this, we can draw trend lines and discover that there’s at lest 4 dB of difference between these two brands in the high frequencies. Speakers which can be seen between them should be considered far more nominal.
Letter “B” shows the obvious standing waves of the room affecting the graphs. Interestingly, on the whole the 8″ speakers suffered more from room issues than the 5″ speakers did.
Letter “C” shows the deep dip in upper-mid frequencies for the JBL 305 MKI, but also shows how much the MKIIs solve that issue.
Letter “D shows that most of the 5” speakers fall short of having flat low frequency response and the expected 708Ps came through with flying colors here. In fact, my guess is that the 705 variety would have scored equally well in low frequency response.
Letter “E” shows, conversely, the strong ultrasonic fall off of frequencies from the 708s. Coupled with its slight reduction in high frequencies may have it able to be listened to for longer periods of time without fatigue but may also require more highs to be added to mix than are heard.
For us, the clear winner here is the 708ps, and for the 5″ speaker the 305 MKIIs.
The Transient Responses
For the transient response graphs we’re looking at iZotope RX Spectrograms with Waveform overlay. This music material is dubstep with orchestra – so it’s really complex. It’s not just a simple transient. We’ll want to see:
- lots of separation between the 5 bass drum hits and the large snare hit
- separation between the big bass waves of the bass drum hits
- large dynamic range spanning the tallest transient to the highest low transient
- good level sustain from the bass frequencies/amplitude of the very loud/transient attack snare drum – so that its waveform resembles a block that decays instead of a bunch of jagged needles. This shows that the speaker can jump instantly to a loud position and transduce its sustain well without the mechanical physics of the cone trying to pull the speaker into other positions.
- No dark areas or bright areas in the spectrogram which are not present in other speakers
- pinpoint/pinprick low frequency transient peaks.
Above is the direct audio file run straight into RX. You can see the 5 deep bass drum hits followed by the snare and followed by orchestra sounds.
Again, we need to take into account the issues of the room adding its own reverb and flubbiness to the sound. But with the KRK, there’s a loss of high frequencies, and the transients themselves are all seemingly compressed. This can sometimes demonstrate driver resistance – or that the drivers themselves don’t really want to extend far enough to accurately reproduce the dynamic range of the source material. The snare drum sustain can also be seen to be sketchy in both the waveform and the spectrogram. However, the low frequencies appear to generally be reproducing a quick response – whereas their dynamic range is reduced.
The Eris 5 really shows its inefficiencies with completely missing bass, horrible snare sustain, flubby bass response where one cannot even find any bass waveform clarity and positive AND negative transients all handily compressed into a perfect box. The Eris 5 scores the lowest in transient response of the speakers tested.
The Tannoy shows its expected hyped high end and provides a reasonable dynamic range. However, like the Eris, transients are handily compressed, the snare sustain is whacky, and the bass response is flubby at best. You can always tell flubby bass response by how separate the low frequency wave is from preceding and forthcoming waves. Here, the woofer isn’t doing a lot of bass individuation movement at all and the mid frequencies it’s also having to transduce are interfering with the bass frequencies.
The Mackie did what we expected which is reasonable sustains, and a flubby-turned-okay transient bass response. Due to its passive sub woofer, the woofer has a reasonable response, but it takes several milliseconds for its passive partner to start working. Thus, low frequencies below 65 Hz are going to be difficult to snap-to. Like the former two, low transients are limited in this speaker. This is the lowest scoring 8″ system in this bunch. I’m sad to have to admit it.
We were shocked to see how nearly identical the transient response of the 305 MKI was to it’s younger brother the MKII. With the exception of a slightly smaller dynamic range, they are identical.
The LSR 305 MKII has good clarity with its transient bass response and a reasonable dynamic range. For a 5″ system, this is outstanding. As seen in the frequency plot above, the mid range has a little boost while the sustained Snare Drum bass is passable. Because of the smaller woofer, it’s able to respond faster to transients, and because of it’s low-frequency design bias, it can put out strong and snappy lows down to 50 Hz. This is the highest scoring 5″ system for transient response.
It shouldn’t be a surprise that the 8″ JBL 308 does well at being snappy with bass and bass sustain. It also has no trouble creating the elusive pin-prick transients which give the “leading edge” a sense of clarity in mixes – the lack of which causes mixers to push volume too far. The dynamic range isn’t as large as we’d expected, but it’s reasonable, and of course the expected mid range boost can easily be seen in the spectrogram.
Many people make the mistake of expecting that the added value of a more expensive speaker is in its “flat frequency response.” However, transient response is where the designers must put a majority of their development. And the JBL Series 7s take the cake with a katana. Very clear low snappiness, wide dynamic range, pin-prick transients, deep snare sustain and big lows. The only issue is the slightly diminished upper high frequencies. There’s a reason why you spend $2k each for these. Compare this image to the original above and you’ll see how much closer this is than any of the other speakers.
The Cinema Sound Opinion
What’s the Cinema Sound opinion? Well, ultimately, no matter what speaker you choose, you’ll need to get used to how it sounds and audition your mixes on new speakers on as many systems as possible so you can know what to expect. This is also true after you treat your room in any way. All in all, given the response on the frequency plotting charts and the transient response, I can’t recommend the JBL LSR 305 MKIIs enough. We’re vindicated as well, since we weren’t really sure how this was all going go – but hoped the MKIIs would win. We didn’t doctor anything and even though Harman/JBL is a sponsor of Cinema Sound we went at this with the purpose – as always – of bringing our discoveries to you unvarnished. And – sorry haters – it’s JBL by a mile.
|Frequency Plot||Off Axis||Transient Plot||<$149||Value|
|JBL 305 MKI||6||8||6||Yes||8|
|JBL 305 MKII||8||8||7||Yes||9|
|JBL 308 MKI||6||7||9||No||7|
Sure. There’s other speakers out there. And, no, we’re not going to test them all. Why not? Honestly, because our cursory research and experience show us it’d be a waste of time. Although we are interested in your thoughts on other speakers you like.
And again, it’s what you can get used to. But if you could get your hands on a speaker that scored an average of 8 and was $1,850 less than a speaker that scored 9.5…? Now look, if you can you should get yourself a pair of 708s. Trust me. They may not be the best value, but there’s nothing that sounds better – and you pay for those last 5% a lot more than you’ll pay for the first 90%. But for highest value and bang for your buck? JBL LSR 305 MKIIs.
The ability to mix killer sound in immersive surround environments will be non-negotiable for all media in no time. When will this happen? It already is. On Cinema Sound we show you how to use partner plugins to create everything from 5.1 out to 24.2 mixes for delivery on headphones to any channel on Youtube. But if you don’t have that experience or you’re not a member, is there a simple quick start guide to at least be able to get to an intermediate level in surround mixing?
Well, you can spend your time searching the internet for credible sources and just get more confused, or you can purchase “Quick Start to Mixing in 5.1 and Higher” and get it all handled.