Preamp comparison in SOS October 2012: the key!

[Sam Inglis]

In October's SOS, we set out to see how much of a difference the choice of mic preamp made in one real-world recording situation: recording a close-miked grand piano. Thanks to Jonathan Dodd's Realpiano service, we were able to use a Yamaha Disklavier player piano to reproduce the same performance very precisely over and over again. This was recorded 24 times, through three different stereo mic arrays and eight different preamps.

Subscribers can read the full article here, and the files are available for everyone to download here. The catch is that all reference to the name of the preamp has been removed from the files, to make it a true blind test -- until now!

Many SOS readers and forum users have now downloaded the files and given us their opinions. The overwhelming verdict is that the differences are very slight, and many people felt that they could not reliably tell them apart in a blind test. However, many other people did have favourites. Hugh Robjohns collated readers' opinions and created a rating system based on these positive and negative comments. Each preamp/mic combination is given a score, though it should be pointed out that this is not at all scientific! The table below also lists the number of respondents who rated each preamp/mic combination as their favourite, and summarises some of the subjective comments that have been made about the sound. Thanks for listening!

The table below also lists the number of respondents who rated each preamp/mic combination as their favourite, and summarises the subjective comments that have been made about the sound. Thanks for listening!

Round 1: Preamp letter codes for the Brauner mic files:

A - ART Pro MPA IIRating: 7.5 Chosen as Favourite: 5 times
Comments: Smooth tubey top end, neutral and rich, especially roasty, more warmth and body

B - AMS Neve 1073LBRating: 3.4 Chosen as Favourite: 1 times
Comments: Classy vintage sound, too strident in midrange, veiled top end, rich mids and lows, brittle, sounded grainy

C - API 3124+Rating: 2.9 Chosen as Favourite: 0 times
Comments: Vague, dreamy, a touch boxy, slightly harsh midrange, disliked, clean but smooth, strident

D - Mackie VLZ ProRating: 6.8 Chosen as Favourite: 2 times
Comments: Not much character, big bottom end

E - SSL XLogic VHD PreRating: 6.5 Chosen as Favourite: 4 times
Comments: Clean, accurate, neutral, more transparent, warm lows, more depth and evenness, brittle, less solid, little dull, most information and naturalness

F - GP Electronics PML200ERating: 4.2 Chosen as Favourite: 1 times
Comments: Dark, slightly plasticky, close and intimate

G - Maselec MMX-4XRRating: 6.4 Chosen as Favourite: 2 times
Comments: Clean, accurate and neutral, muffled top end, slightly distant, smooth character, sharper transients

H - Prism Sound OrpheusRating: 6 Chosen as Favourite: 4 times
Comments: Sounds more stereo, zingy, closed in, veiled or murky, prominent low end, smooth
character, more clarity and separation, smooth without sounding dull, disliked and would
avoid

Round 2: preamp letter codes for Sennheiser MKH20s

A - GP Electronics PML200ERating: 5 Chosen as Favourite: 2 times
Comments: No particular sound, warm clean and detailed, on the dull side, probably txfm-less

B - ART Pro MPA IIRating: 7.4 Chosen as Favourite: 3 times
Comments: Intimate, can hear further in, more depth and ambience, sounds more musical and dynamic (same preamp as Brauner A and Royer D?), most information and naturalness

C - Maselec MMA-4XRRating: 3.5 Chosen as Favourite: 1 times
Comments: Slightly hard midrange, Mackie/ART?, slightly dull

D - AMS Neve 1073LBRating: 5 Chosen as Favourite: 1 times
Comments: Neutral, natural, rich quality

E - SSL XLogic VHDRating: 5.5 Chosen as Favourite: 1 times
Comments: Intimate, great clarity, nothing objectionable

F - Prism Sound OrpheusRating: 7 Chosen as Favourite: 1 times
Comments: Warm, close and detailed, best overall balance, slightly dull

G - Mackie VLZ ProRating: 5.6 Chosen as Favourite: 0 times
Comments: Warm and intimate, detailed, ART/Mackie?, sharp in the mids, probably txfr-less

H - API 3124+Rating: 5.5 Chosen as Favourite: 1 times
Comments: Rich mids, nice frequency balance, not as detailed, possibly Mackie?

Round 3: Preamp letter codes for Royer SF12

A - Mackie VLZ ProRating: 3.8 Chosen as Favourite: 0 times
Comments: Less detailed, slight edge, not as detailed, slightly dull, Neve/API?

B - Maselec MMA-4XRRating: 3.5 Chosen as Favourite: 0 times
Comments: Shrill top notes, slightly dull, bad!

C - Prism Sound OrpheusRating: 6.8 Chosen as Favourite: 2 times
Comments: Warmer bottom end, delicacy, nicest so far, very accurate and fast, deep and tight lows, neutral, Maselec?

D - SSL XLogic VHDRating: 6.6 Chosen as Favourite: 3 times
Comments: Delicate, Intimate, Richness, more detail, least natural (same preamp as Brauner A and MKH B?), delicate, precise, detailed and warm

E - API 3124+Rating: 3.8 Chosen as Favourite: 1 times
Comments: Less detailed, slightly dull, nice colour, valve preamp?

F - GP Electronics PML200ERating: 5.6 Chosen as Favourite: 1 times
Comments: Detailed yet neutral, midrange richness, slight edge, slightly dull

G - AMS Neve 1073LBRating: 6.4 Chosen as Favourite: 3 times
Comments: Less detailed, slight edge, clearer, bright and clean with more dynamics, Neve/API?

H - ART Pro MPA IIRating: 5.6 Chosen as Favourite: 2 times
Comments: Shrill and slightly harsh, bright clear and dynamic

[Phonix]

Well, it seems to me that one of the reasons for having a high end pre is that you can get a lot of different tones out of it depending on how you drive it. API's for instance are famous for having a warm mildly distorted sound, if you drive them hard using the pad, which is why they're great on rock vocals, and if you don't can be very clean if that's what you want. So, not really getting the point of this 'test'.

[Hugh Robjohns]

The 'point' as was made pretty clear in the article and the follow-up piece, was to show that even budget preamps perform extremely well, and are certainly not a hindrance to capturing excellent recordings.

The impetus for the feature was the continual claims seen, and 'advice' posed, that you must have an API or a Neve (or whatever) preamp to make decent recordings. And that claim just doesnt stand up to scrutinee. The biggest hindrance is more typically the recording venue acoustics, the recording techniques, or the performance itself. The preamp is never the weakest link these days, although sure, some nice preamps do nice things when deliberately pushed (abused? ), and that point was recognised.

H

[dadking]

Very interesting results
having just glanced through this months magazine I was interested in the article on miking up drum kits, I was especially interested in the higher end mics that were used on the kit, I cant afford 1 AKG414 let alone 2 used on the snare so what about doing a similar test using lower cost microphones on a drum kit.

[Matthew Ottewill]

I'm a bit (very!) late to this thread but just wanted to thank Sam and the SOS team for the mic pre comparison. A superb article!

I should also point out that I am sitting in my studio where I have 7 different mic pres and have long known that ergonomics aside, they are the least important link in the mic recording chain.

Here's my list of factors that contribute to the sound of a mic recording, most important first ..

1. Performer / instrument (new strings/reed, in tune, etc?)
2. Where you place the mic
3. Mic choice
4. The environment/room
5. Mic pre choice

What order would you choose?

[Hugh Robjohns]

Here's my list of factors that contribute to the sound of a mic recording, most important first ..

1. Performer / instrument (new strings/reed, in tune, etc?)
2. Where you place the mic
3. Mic choice
4. The environment/room
5. Mic pre choice

Close, but I'd argue not quite right (just like your balanced interconnect video -- the only one I've watched so far.. )

I'd go for:

1. Music & Arrangement
2. The acoustic environment/room
3. Performer / instrument
4. Mic Placement
5. Mic choice
6. Mic pre selection

H

[yada]

Cool test. I like stuff like this. Interesting. There's also another idea that circulates about needing x amount of gain to power x microphone. Think i've actually been guilty of passing that idea onto someone myself but i'm not convinced (maybe i'm ignorant). Like why do i need x gain when i have x voltage (for condensers)? ...and why do i need x gain anyway, is the cable used shoddy or something so it's 'leaking' (or something...).

...i think i just thought out loud. Maybe i shouldn't and should instead delete my rambling!

[Hugh Robjohns]

Different mics have different sensitivities which can range from about 1mV/Pa for a dynamic vocal mic, up to 30mV/Pa for a high output capacitor mic -- although there are plenty of mics above and below those rough extremes.

The amount of preamp gain required depends on the SPL (acoustic volume) of the source, the mic sensitivity, and the required contribution level into the mix. So you might need 60dB with a dynamic mic for a gentle spoken voice, but only 40dB for a screaming vocal on the same mic, and maybe only 20dB if you switch to a high output capacitor mic! No two situations -- different sources, different mics, different mix requirements -- will require the same preamp gain settings

H

[yada]

Hell that was a good response i feel. Clear and concise. I guess i was ignorant... good to know. Cheers.

[joncoopermusic]

I have been Playing with a stack of new equipment including...

Korg M50
XDJ-R1 record decks
Kenwood amp (used as a pre amp)
Laptop - Running soundforge and Ableton (win8)

For 2 weeks i have been playing with pre amping the sound ran through the decks to add color effects
with pre amping i found that it was incredibly difficult to get the levels correct. To record keyboard to a laptop at least it would seem that direct to laptop is the way to go, but this means you lose the effects from the mixer....

You can here some of the recording i have done over on my site and also on my soundcloud
The clipping Extreamly noticable in some tracks recorded through the preamp.

Listen Here http://joncoopermusic.com the last 30 secs of the short mix in the header illustrates the pre amping problems from my keyboard at least.

the Long Mix again was pre-amped and you seem to lose level control to a certain degree as you will notice, and software would not normalize the levels as i would of liked.

Vocal recording has not been a problem through the desk as my AKG mic offers outstanding results in a portable sound booth $200 ish.

In my 20 years as a writer / producer it is sometimes wiser to record too quiet and use software to boost the volume after as well as make subtle sound changes.
Depending on the amp and the subtlety in which you use it, it can make all the difference.

Glad i found this site....i found alot of useful tips myself! TY

[Dennis J Wilkins]

Hello Sam (and Hugh),

Great test and article. Although I read it a year ago, I just last night downloaded the files and loaded them into my studio DAW (REAPER). They were listed only as A, B, C, etc and I had not looked at this "answer sheet" so I has no idea which pre was which. I found two issues with the files straight away -- they were not time aligned and the levels, as measured by REAPER's RMS metering were not within a dB in some cases. When I first listened and switched instantly between a couple files, I heard small but very discernible‎ differences. I quickly saw the time variation between the files and adjusted that, then looked at levels and found the small differences there, and not the same mismatch in left channel as in right channel in some cases.

After adjusting to within =/-0.1 dB (using REAPER's faders and pan pots so I didn't modify the original files) I found switching between any two files made pretty much an imperceptible difference. I find it best to loop a short section of the files so I can hear the same passage repeated, and listen separately to attacks, and to note tails for any nuances. As you said at the end of the article, the differences, at least at the levels you pushed any given preamp, are negligible. The microphone, and of course most importantly the performance (including mic placement and room acoustics), are going to have the greatest effect on the sound.

Still, I'd love to have a few Neve's (I've got some Focusrites!), API, or SSL just for the pleasure of owning such fine gear. And of course a few classic Neumann's would be great! But I can't complain about the affordable gear these days, even a RODE, AKG, or Audio-Technica mic driving a Mackie, ART or Focusrite preamp can do an excellent job when used properly.

I look forward to many more informative articles in Sound on Sound!

Dennis 'Sound-Guy' Wilkins

[Sam Inglis]

Hi Dennis,

Glad you liked the article, and thanks for the comments!

As mentioned in the article, we recorded a short piece of 1kHz tone from a mono loudspeaker at the start of each 'take'. I used this tone to match the levels between takes. On a peak meter reading they were all within 0.1dB of each other. So if you find that the actual piano recordings appear to differ in level slightly, this could be down to one of two things. One possibility is that some of the preamps have a frequency response that is not entirely flat, so matching at 1kHz doesn't necessarily imply level matching across all frequencies. The other is that there is a small amount of variation in peak or average level between different Disklavier performances. Given the difficulty everyone had in telling apart the takes, I think the latter is much more likely!

I didn't see a need to time-align the performances with perfect accuracy, so that was just done by eye in Pro Tools.

All the best
Sam

[Dennis J Wilkins]

Hi Sam.

I got a chance to listen more carefully (and not past midnight!) and found one single file that sounds very different from any of the others -- the reverb characteristic seems significantly different than any other preamp or mic for that matter. It's the Brauner mic with the Prism Sound Orpheus preamp (Brauner H). Looking at the wave forms shows an envelope shape different than any other mic/pre combo. Rather odd, and at this time nothing anyone needs do about it, but just found it a strange anomaly.

Anyway, very informative test and article which underlines the importance of the performance, room and mic over the more purely electronic signal path. I look forward to many more useful articles, tests and reviews.

Dennis 'Sound-Guy' Wilkins

[Sam Inglis]

Ah... There *was* a problem with the Brauner H file, but I thought a fixed version had been uploaded. Let me check it out when I'm back in the office on Tuesday.

[Jeraldo]

Today, in mid 2014, while cleaning out a desk (!) I discovered my listening notes I made when this whole experiment took place. To my surprise:

For the MKH20 files, I picked, in order of preference:
1. GP Electronics
2. Prism
(3-very distant API)

For the Royer files, I picked:
1.GP Electronics
2.Prism
(3-distant Masselec)

I was helped by having a lot of use with MKH20's. I was quite certain I had the same two preamps chosen for Royer and Sennheiser.

But for Brauner - Here I digressed - and completely.
I didn't like anything, but strongly preferred ART, which seems to go with the majority.

My listening chain was very simple and very low tech.

I don't think I'll be buying the preferred preamps, though........

[Jim C]

In October's SOS, we set out to see how much of a difference the choice of mic preamp made in one real-world recording situation: recording a close-miked grand piano. Thanks to Jonathan Dodd's Realpiano service, we were able to use a Yamaha Disklavier player piano to reproduce the same performance very precisely over and over again. This was recorded 24 times, through three different stereo mic arrays and eight different preamps.

Sam - Just caught up on this article and the demo files. It's BRILLIANT to see and hear this kind of testing. More!! - Jim!

[Guest]

This is Great , loved it , Thanks

[AudioWonderland]

The table below also lists the number of respondents who rated each preamp/mic combination as their favourite, and summarises the subjective comments that have been made about the sound. Thanks for listening!

Round 1: Preamp letter codes for the Brauner mic files:

A - ART Pro MPA IIRating: 7.5 Chosen as Favourite: 5 times
Comments: Smooth tubey top end, neutral and rich, especially roasty, more warmth and body

B - AMS Neve 1073LBRating: 3.4 Chosen as Favourite: 1 times
Comments: Classy vintage sound, too strident in midrange, veiled top end, rich mids and lows, brittle, sounded grainy

C - API 3124+Rating: 2.9 Chosen as Favourite: 0 times
Comments: Vague, dreamy, a touch boxy, slightly harsh midrange, disliked, clean but smooth, strident

D - Mackie VLZ ProRating: 6.8 Chosen as Favourite: 2 times
Comments: Not much character, big bottom end

E - SSL XLogic VHD PreRating: 6.5 Chosen as Favourite: 4 times
Comments: Clean, accurate, neutral, more transparent, warm lows, more depth and evenness, brittle, less solid, little dull, most information and naturalness

F - GP Electronics PML200ERating: 4.2 Chosen as Favourite: 1 times
Comments: Dark, slightly plasticky, close and intimate

G - Maselec MMX-4XRRating: 6.4 Chosen as Favourite: 2 times
Comments: Clean, accurate and neutral, muffled top end, slightly distant, smooth character, sharper transients

H - Prism Sound OrpheusRating: 6 Chosen as Favourite: 4 times
Comments: Sounds more stereo, zingy, closed in, veiled or murky, prominent low end, smooth
character, more clarity and separation, smooth without sounding dull, disliked and would
avoid

Round 2: preamp letter codes for Sennheiser MKH20s

A - GP Electronics PML200ERating: 5 Chosen as Favourite: 2 times
Comments: No particular sound, warm clean and detailed, on the dull side, probably txfm-less

B - ART Pro MPA IIRating: 7.4 Chosen as Favourite: 3 times
Comments: Intimate, can hear further in, more depth and ambience, sounds more musical and dynamic (same preamp as Brauner A and Royer D?), most information and naturalness

C - Maselec MMA-4XRRating: 3.5 Chosen as Favourite: 1 times
Comments: Slightly hard midrange, Mackie/ART?, slightly dull

D - AMS Neve 1073LBRating: 5 Chosen as Favourite: 1 times
Comments: Neutral, natural, rich quality

E - SSL XLogic VHDRating: 5.5 Chosen as Favourite: 1 times
Comments: Intimate, great clarity, nothing objectionable

F - Prism Sound OrpheusRating: 7 Chosen as Favourite: 1 times
Comments: Warm, close and detailed, best overall balance, slightly dull

G - Mackie VLZ ProRating: 5.6 Chosen as Favourite: 0 times
Comments: Warm and intimate, detailed, ART/Mackie?, sharp in the mids, probably txfr-less

H - API 3124+Rating: 5.5 Chosen as Favourite: 1 times
Comments: Rich mids, nice frequency balance, not as detailed, possibly Mackie?

Round 3: Preamp letter codes for Royer SF12

A - Mackie VLZ ProRating: 3.8 Chosen as Favourite: 0 times
Comments: Less detailed, slight edge, not as detailed, slightly dull, Neve/API?

B - Maselec MMA-4XRRating: 3.5 Chosen as Favourite: 0 times
Comments: Shrill top notes, slightly dull, bad!

C - Prism Sound OrpheusRating: 6.8 Chosen as Favourite: 2 times
Comments: Warmer bottom end, delicacy, nicest so far, very accurate and fast, deep and tight lows, neutral, Maselec?

D - SSL XLogic VHDRating: 6.6 Chosen as Favourite: 3 times
Comments: Delicate, Intimate, Richness, more detail, least natural (same preamp as Brauner A and MKH B?), delicate, precise, detailed and warm

E - API 3124+Rating: 3.8 Chosen as Favourite: 1 times
Comments: Less detailed, slightly dull, nice colour, valve preamp?

F - GP Electronics PML200ERating: 5.6 Chosen as Favourite: 1 times
Comments: Detailed yet neutral, midrange richness, slight edge, slightly dull

G - AMS Neve 1073LBRating: 6.4 Chosen as Favourite: 3 times
Comments: Less detailed, slight edge, clearer, bright and clean with more dynamics, Neve/API?

H - ART Pro MPA IIRating: 5.6 Chosen as Favourite: 2 times
Comments: Shrill and slightly harsh, bright clear and dynamic

I am pretty amazed at how similar they all sound and how well the ART and Mackie units did in terms of rating etc. I have been debating where to upgrade next and have been reviewing a lot of different preamp options which is what brought me here. Considering how little difference there really was and how well my existing preamp's faired I think I may need to rethink my upgrade plans. I need more inputs for drums but it looks like it could be done and done well for a lot less money than I originally thought

[CS70]

The consensus seems to be that whatever differences there may be, they would be audible in a blind test only if the preamps were pushed. It'd be interesting to *that* comparison too!

[Martin Walker]

The consensus seems to be that whatever differences there may be, they would be audible in a blind test only if the preamps were pushed. It'd be interesting to *that* comparison too!

Agreed, that would be absolutely fascinating.

However, you'd have to push each preamp by at least three or four different amounts to see how gracefully (or otherwise) they gradually overloaded and moved into distorted territory, and probably with three or four different types of instrument, so it would be a HUGE comparison to undertake

Martin

[Sam Spoons]

It still might be worth doing though, comparing a low end preamp with a mid range and boutique one, keeping to to just three might make to manageable?

[Sam Inglis]

It'd definitely be interesting. I think the problem would be finding a suitably repeatable source. A grand piano is something you'd typically want to record as clean as possible. People are more likely to want to overdrive their preamps on drums, or rock vocals.

[Sam Spoons]

Yes, good point. Somebody in another thread suggested playing back a recording through a good monitor and then recording that would be the closest to repeatability but it's never going to sound the same as a live vocal, of course if you could get four examples of each preamp and a 1<12 way mic splitter

Maybe it doesn't have to be completely scientific across all four tests, as long as each preamp is represented at each gain level (only requiring a three way mic split) the the fact that the four performances were not identical would not matter too much as the performance at each GL would be identical.

[ManFromGlass]

shows me that I am ok in the pre department but I could use a disklavier! That would be cool!