Lake People G103P and G109P Headphone Amplifiers

Introducing Lake People
Don't feel bad if the name doesn't ring a bell, Lake People is not a high profile company compared to many of the big players on the market. They started in 1986 making gear for recording and television studios, and have been making headphone amps since 1989. Despite this pedigree, their focus on studio gear combined with their German origins kept them fairly obscure as far as the North American market is concerned.

That changed some when they launched the Violectric line a few years back. Violectric is focused on the audiophile market, and is very Head-Fi friendly with their compact enclosures. The line consists of several headphone amps, a DAC, and a phono stage. I use Violectric gear in my reference setup and quite enjoy it.

The actual Lake People branded models have always been very utilitarian in form and function---basic sturdy enclosures or rack-mount configurations, mostly XLR inputs, that sort of thing. But with the newly redone G-series of headphone amps, Lake People are now reaching out to the home user with attractive but low key black aluminum enclosures, choices of RCA or XLR inputs, and reasonable pricing. In this review I sampled the lowest and the highest priced models in the range to see how they stack up.

Lake People Models
G-series models are available in Standard or Pro configuration. For the entry level G103 model, Standard means RCA inputs only, while Pro means XLR inputs. The unit is so small that I don't think they had room for both types. For the top range G109, the Pro adds XLR but keeps the RCA inputs. Pricing is a bit complicated - here's the list I was given, in Euros rather than US Dollars.

  • G103S - 245 EUR
  • G103P - 295 EUR
  • G109S - 445 EUR
  • G109P - 495 EUR

To an an extra wrinkle to the equation, those prices include 19% VAT, which would not apply to orders outside of that area. If we take away VAT and convert EUR to USD using the current exchange rate, the G103 and G109 come to approximately $250 and $450 respectively. Pro configuration adds roughly $50 to each price. These amps are so new that there isn't a North American distributor yet. Those duties will likely fall to the same distributor who currently handles the Violectric line. (UPDATE: the new North American Distributor for Violectric and Lake People gear is located HERE)

Design Theory
Before we get into specifics of each amp, I'd like to discuss the general topic of headphone amplifiers. There are two main camps in the headphone world: those who feel that all competent headphone amps sound identical, and those who think each amp has a unique flavor of its own. Interestingly, Lake People/Violectric has managed to become well respected among both types. NwAvGuy has repeatedly used Violectric as an example of well engineered gear, yet there are plenty of subjectivist users with expensive power cables and interconnects who have also given Violectric praise. It's not often that a company can bridge that gap and earn respect from both sides.

I asked Lake People CEO Fried Reim about his design philosophy, and about headphone amp design in general. About what makes amps sound different from one another? Most any well engineered amp will have a practically ruler flat frequency response assuming it isn't being over-driven beyond its capabilities. With frequency response being identical, how can two amps possibly exhibit any differences in sound? Fried agrees that amps can in fact sound different, and he gave me some thoughts on how or why that may be the case. English is not his first language but you should have no problem understanding his point:

"You are right, a properly made headphone amp from any brand will most times measure ruler flat concerning the frequency response in the audible range and far above. But here is where the differences begin. Every designer has his own approachabout how far the frequency range should spread. The unwanted regions are cut by the use of capacitors.

"As you know, the frequency edges are defined at -3 dB. And at this point we also are faced with a 90 degree phase shift. Cutting the low end at, lets say, 3, 5, or 10Hz is of minor interest in my opinion. You can't hear that low with a headphone anyway, so frequency or phase shift issues are really negligible.

"All this is more critical when cutting the high edge of the frequency range. A limitation to 50 kHz (-3dB) will affect the 20 kHz level by about 0.5 dB (which is not critical) but also by a noticeable phase shift at this point and below. So somebody might argue: why not kick off all these upper frequency limitations ... and in fact, some manufacturers do so. But in times of massive electromagnetic interference from everywhere it is not very intelligent to design an amp which is capable of amplifying long wave radio frequencies. Also, any CE or FCC certifications will be out of reach.

"Even when there is no specially designed frequency limitation, real life audio electronics are limited by component issues. These are described as GBW (Gain Band Width) and slew rate. The higher the feedback gain, the lower the GBW and the slew rate. And here we come to another reason for sonic differences: although the frequency response is flat and (measured) distortions are low, low internal gain (+8 dB, a factor of 2.5 with Lake People/Violectric models) not only keeps the noise floor low, but also allows for high GBW and high slew rate.

"But in my opinion most sonic differences are caused by impedances. The output impedance from the amp will always interact with the impedance from the headphone. This will have more or less influence on the frequency response and thus affecting the sound of the connected headphone. In some cases this will support your personal preferences, in some cases not. In former times when most premium headphones had high impedances this was not a big issue. But many of today's premium headphones have low impedance.

"It is very difficult to measure what really is input to the human ear. So frequency/impedance interactions are most times calculated theoretically and displayed in a 2-dimensional curve, or better (as there is also an SPL interaction) in a 3-dimensional "waterfall" model. Perhaps you have seen MLSSA measurements from loudspeakers in different rooms. The headphone listeners "living room" is between the transducer and his ear, but also affected by the individual shape of the outer-ear and inner-ear. And here the whole thing goes beyond perfect measurability ...and thus a subject for personal taste, tales, magic ...

"To limit impedance issues Lake People/Violectric models offer output impedances well below 1 Ohm and so the chances are low for impedance related interactions between the amp and the can."

This isn't meant to be an exhaustive treatise about the ins and outs of amplifier design (though I think that would make for an interesting article all by itself). Rather, it is an interesting look into the opinion of one designer in the objectivist camp who believes there are legitimate reasons for why his designs sound distinct from one another. The bit about measurement difficulties falls in line with what Tyll has been saying, and also backs up the idea that some headphones can sound better or worse than their measurements would lead you to believe.

Speaking of measurements, Tyll has been working on a test suite for headphone amps but it isn't ready yet. So at this stage all I can do is talk about the design of these amps, and tell you how I subjectively feel about the sounds they make. Though I have a feeling these types of amps will end up doing well in Tyll's battery of tests, rather than the "tuned by ear" audiophile variety.

Let's have a look inside, and listen to the amps ...

Lake People electronic GmbH
Turnstr. 7a
D-78467 Konstanz
+49 (0) 7531-73678

itsastickup's picture

Even the cheapest violectric is violently expensive when compared to the O2 and yet doesn't really do much more [edit:for the ordinary consumer]. I suspect that this new line is a response to the competition that the O2 must be providing. The problem with the O2 is that the finished hand-built product is violently under priced. The sellers seem to have not much clue how to cost things properly or are not doing it for a living. I can't see that lasting: building 200 O2's for a pittance is going to wear thin eventually. But then again, the design is open-source and easy to build even for the inexperienced; perhaps someone else will take up the dropped batton.

In the meantime it will be interesting to see some measurements of these new kids; see how they compare.

mikeaj's picture

These are considerably more powerful than the O2 though (also probably better built), but they do lose battery operation. Whether or not users actually make use of the extra power is a different matter.

Lake People's specs are here:

For the THD+N figures, they detail both the output power level and load (but don't specify frequency...but that's probably 1 kHz), which is the honest way to do it.

On another note, -98 dBu or -101 dBu A-weighted noise may be definitely audible with the most sensitive IEMs? Is the figure conservative, maybe worst case scenario with the highest pre-gain set? While we're making comparisons, O2 supposedly gets -112.5 dBu A-weighted noise on AC power. It's not like that's relevant for most headphones though.

I kind of doubt that anybody's pricing is related to new products by others out in the market, but if any amp can drive down the price of other audio equipment, that'd be fine by me.

It's a shame that the pre-gain options aren't switchable with a control on the front or even back panel, since that would help when going between most IEMs and the less-sensitive full-size headphones. Don't the Violectric models have that? I guess that's product differentiation for you.

itsastickup's picture

...there's almost nothing that needs more power than an O2; not even the HE6.

O2s are pretty solid; except for a poor soldering job I'm not sure that comparisons on build come to much. It does have a weak point as a portable however: lead-free solder is fragile and there is nothing in the O2 to soften a blow from a fall. Case impact will shoot straight on to the circuit board.

Other than a need for balanced connections in a pro studio perhaps, the violectrics are looking pricey.

"I kind of doubt that anybody's pricing is related to new products by others out in the market"

I don't know about that. There was a collapse in the headfi 2nd hand market for headphone amps when the O2 became available; and that was despite the best efforts of the admins.

FLAudioGuy's picture

I agree about the "build quality" of the O2. I have dropped mine (from waist height) and not even the batteries became dislodged. I do have a piece of foam to press against the tops of the batteries, however, so not sure what would happen if the foam was missing.

When I look at the O2 compared with the LP/Violectric/DAC1 I am not so sure that I can attest to a higher "quality" in the latter with regards to actual parts quality. They are all made from quality manufactured parts as far as I can tell. What can actually separate them? Is the LP/Vio PCB higher quality because it is shiny or is it because they chose a different solder mask? Does a 4-layer or even 8-layer board mean higher quality? I think not. It may mean more routing options and internal ground plane, to be sure, but is there higher intrinsic quality? The O2 may be more compact and parts density is visibly higher but put that board in a heavy Hammond case, 10mm thick front panel, engraved/filled/lasered nomenclature, nice feet and voila! you have a typish "high-end" amp. All that accouterment doesn't add to the sound though. Does it make the amp higher quality? If both amps have a service life of over 10years (more than many TV's and yet-to-be-seen) what differentiates their quality. Please fill me in :) I see more expensive power socket, XLR connectors, a lot of machining and labeling, shiny PCB and a nice layout (dubious enhancement)... to my eyes, I see $$$$.

In my estimation, many people misunderstand "quality". The amps probably measure very similar with the O2 even beating it and the DAC1 in one or two spots. I cannot hear ANY noise in the O2 in either gain setting (2.5/6.5x) with my Super.Fi 3 Studio's or Triple.Fi 10 Pro's. Is that not intrinsic quality?

As far as more power, it has been stated many times on NwAvGuy's blog that there may be a few headphones whose needs may be beyond the O2's venerable capability. It can produce 7VRMS while the DAC1 produces 10VRMS and the LP is higher as well. I don't see it as a realistic shortcoming.

I do like the XLR input option as I have need to interface with studio gear such as Avid's MBox Pro and C|24, neither of which have decent HPA's. The lack of balanced inputs on the O2 has not been much of a loss actually. I have built O2's for friends that use IEMs on stage via body packs and the battery operation does come in handy. Even Aviom's sound better with an O2.

Good reading review of the Lake People product(s), I know they will be proud additions to anyones home audio system or be the basis for a sweet mini high-end headphone system. Cheers!

John Grandberg's picture
I ran the Lake People models on the lower gain settings, and could not find a trace of noise with any of my IEMs. Perhaps bumping the gain to the highest setting would reveal some - it would also give way less usable volume range, so I would never run it that way. I personally didn't notice the price of amps on HeadFi being driven down. Maybe I missed it.
John Grandberg's picture

These are basically refreshed versions of existing Lake People models. I just think Lake People realized that they were potentially missing out on the large "home user" market by making the older G-series models look so studio oriented.

I agree about the pricing situation though. It's not really fair to compare value for a DIY project and a commercial product. But the companies who are selling them pre-made have to be making very little margin on each one. That doesn't seem like it would be sustainable.

I do recall some comments from Lake People CEO Fried Reim regarding pricing: he specifically stated that he wanted to be able to pay his workers a good wage, so that they can earn a comfortable living rather than having to work a second job in their spare time. Happy workers are always good, period, but this also leads to better build quality and QA, which leads to less issues once the gear is in the hands of the customer. This philosophy requires pricing to be more substantial than the typical "make it for as cheap as possible" approach taken by many others in the industry. I suppose both methods have their place, but obviously some folks are willing to pay more.

On functionality - yes, the Violectric models have the pre-gain adjustment right on the back panel. It's more convenient. They have other benefits too, some more tangible than others.

Regarding the O2 having plenty of power, even for the HE-6: I'm not so sure about that. See NwAvGuy's post "more power?" and look at the chart. He lists 85 dB/mW as the lowest sensitivity (the HE-6 is a bit less than that) and shows it requiring about 1000mW to hit 115dB peaks - the Lake People and Violectric models can hit that number, the O2 can't. Not trying to start an argument about the O2 (that's already been done to death), just pointing out an example of where the added power would come in handy.

mikeaj's picture

I haven't checked anytime recently, but as far as I can tell, if you're making lots of O2s (by hand), BOM cost should be definitely under $60 per unit including enclosure, front panel, batteries, and AC/AC adapter. It's probably more like $50 if you're making hundreds. It shouldn't take more than an hour or so to solder and assemble if you're reasonably fast, which should be the case if you've been making a lot.

These are selling for a bit above $150 with batteries and adapter, so I'm not convinced that the margins are really that thin.

John Grandberg's picture
For whatever reason, I could have swore it was almost $100. In which case the company like Epiphany is making $50 for their hour of work, minus packing supplies and shipping costs. Maybe that is an outdated figure by now. Even so, it's hard to compare a DIY project to a commercial model. Distribution adds significantly to the cost of most commercial products, from socks to washing machines to headphones.
mikeaj's picture

Agreed, costs are difficult to compare in some sense. BOM costs are interesting, but there is a whole lot more to it than that. I certainly wouldn't begrudge Lake People for their prices.

However, from the consumer's point of view, operating expenses, materials costs, and everything else don't really matter. The bottom line is the purchasing price and the quality of the product.

I built my single O2 for something like $80 during the first big PCB group buy, including shipping costs from 4 different vendors. (I don't mean to nitpick an honest mistake, but just to confirm the point, I really doubt prices for Epiphany, JDSLabs, Audio Poutine, headnhifi, etc. ever approached $100.) The parts populating the PCB are under $25 in single-unit quantities. Of course, it would cost more to get something with that kind of performance at higher power levels, but it's not as if something like The Wire at diyaudio (with even better performance, at least by the charts) costs a fortune either.

tomasz's picture

Thank you for the review John. I make headphone amps myself and I'm always interested in technology used by others. You write that in G109 the power stage is based upon 4 transistors per channel but from the photo I see only 4 transistor for the whole stereo amp which would suggest a standard push-pull architecture. I'm very interested what kind of transistor are used. Most likely it's a BD139/140 pair but maybe it's something else. And one more thought: in my experience I tend to dislike headphone amps with very low output impedance which is most likely done through severe negative feedback. Negative feedback is doing bad things with time coherence and resolution, but measuring very good in laboratory.

John Grandberg's picture

No, there are definitely 8 total. Look at the picture again - 4 are mounted flat, and 4 are standing up attached to heatsinks. They are not BD139/140 but rather high-speed video transistors which should be stable under all circumstances.

tomasz's picture

Ok John, but is it visibile what kind of transistors? I also use high bandwith video transistors like 2SA1538, but I also seek some new ideas.

Thanks for any help.

Nosgis's picture

Thank you for a good review of theese new amps. Not many of them out there.

I'm now one step to closer to ordering the G109 to use with HiFiMAN HE-500. Sounds like a good bang for the bucks, especially comapared to V200, which is way over my budget.

itsastickup's picture

"And one more thought: in my experience I tend to dislike headphone amps with very low output impedance which is most likely done through severe negative feedback. Negative feedback is doing bad things with time coherence and resolution, but measuring very good in laboratory."

I'm lead to believe that negative feedback as sonically bad is a very old myth. Perhaps the good measurements are telling us something.

High output impedance makes for flabby bass on low impedance dynamic headphones and weak and boomy bass with rolled off treble on low impedance armatures. Everything is a matter of taste; but that must be a rarer one.

FLAudioGuy's picture

I truly believe that "NFB=BAD" is so old-school audiophile myth. Time coherence and resolution are both very well measured in a lab, so saying it "measures well but sounds bad" is highly questionable. Subjective bias and LEARNED bias at play I think. There was a time, long ago (in audio terms) when perhaps there were some poor designs using NFB but as the audio engineering discipline matured and parts matured and refined, HOW NFB is best used became widely known. I don't think that the bygone years way of doing things is still at hand.

Since adding my first O2 (JDS Labs version), even with .5Ω output impedance I have noticed a sharpening of the bass control, especially in my BA IEM's. Gone are the little resonances here and there that occasionally popped up from the loss of control that the Avid MBox and C|24 (33Ω HOZ) exhibited with their built-in HPA's. The bass quality has improved in linearity, extension, impact and ringing. Overall, I sense more "responsiveness" throughout the audible spectrum, not just the bass region, regardless of which 'phones are plugged in. I wholeheartedly attribute this primarily to low-Z output.

John Grandberg's picture
I also agree. I've got a couple different higher end tube amps which allow switching of the output impedance via front panel selector. So it's easy to go back and forth between low and high. I have never once preferred the higher setting with low impedance headphones.
aravind's picture

how is the soundstage with the g109?..i have read elsewhere that violectric amps have a narrow is it especially with the audeze lcd2 which have a narrower soundstage

John Grandberg's picture

It's similar to the Violectric V200, meaning it is reasonably large, very accurate, but not stretched huge like some amps are.

I really do think the "small soundstage" meme has gotten a little out of hand though - even if I'm the one who may have inadvertently started it! The Lake People/Violectric presentation is just as spacious, if not more so, than maybe 80% of the other amps out there. It just isn't the main focus as it is with some others.

aravind's picture

Thanks for your reply...would you consider the g109 as one of the better amp choices at the sub 500$ price range? any other recommendations especially for the lcd2.2s?..
one more question..g109 'p' version specsheet shows 'Outputs with relais controlled switch-on delay' which is not there in the 's' version specs..please explain its meaning and significance..

John Grandberg's picture

It's right up there at the top of my list. Especially for LCD-2, which is known to pair very well with the Violectric/Lake people amps.

The relays (or relais) are there to help avoid any audible "thumps" when powering the unit on or off. I can't say if it's a big deal or not because I don't have the S version to compare it to.

aravind's picture

Does this output relay switch on delay provide protection to headphones against accidental volume rise to high levels? if its just to prevent audible thumps at turn on it may not be a big deal..
thanks for your replies..

John Grandberg's picture
An email to Lake People will clear things up if you are really interested. Most people report quick and helpful answers, from the CEO of the company no less.
tomasz's picture


"I'm lead to believe that negative feedback as sonically bad is a very old myth. Perhaps the good measurements are telling us something"

I can't tell you what you like and what you dislike. As you said: it's all the matter o taste. If you prefer a push-pull with NFB amp I will never even try to convince you otherwise. I have such amps and I like them too. I prefer single ended zero feedback but still I don't think it's the only right way.

But if you justify the negative feedback with measurements I must note that all known measurements are telling us nothing at all about potentially negative impact of NFB. Shure, we can view the step responce, frequency responce (esspecially into a simulated headpone), squere wave and, offcourse, the distortion, but at the moment we cannot measure a micro-phase shifts that NFB most likely adds.

Thanks for your view.

aravind's picture

I will do that..

Can Crazy's picture

Thanks for the first review and detailed reference to Lake People and Violectric. I have been asking for a serious review of Violectric's Amps for ages, but sadly with no results. Until now, that is. Hopefully this will lead you to do a review on the Violectric you mentioned to have in your rig; thorough measurements would certainly be greatly appreciated.

I got so tired of asking everywhere for a decent review of Violectric Amps that I just oredered a V100 with 24/96 USB DAC. It was certainly going to be quicker than learning German to read the reviews that already are out there. A blind test of the V90, V100 and V200 would be fantastic, especially since NwAvGuy claims they should all sound the same -or at least impossible to tell apart in a blind test-, as they all meet the requirements for audio transparency and can drive virtually any headphone.

Again, I hope you will share more of your impressions and tests on Violectric and Lake People Amps in the future. Thanks for a really nice review and for a great first. 

devhen's picture

There are many reviews on

Tyll Hertsens's picture

And I'll add that if you take the time to read the communities take on something, generally speaking, it's quite accurate.  You gotta do a bit of sorting through random remarks, but I find it worthwhile.

John Grandberg's picture

Thanks for the thanks! As far as I know, this is an exclusive first review of these new Lake People models. I was also one of the first to cover the Violectric V181 and V200 amps over at HeadFi - links HERE and HERE. As you can see, many other very knowledgable Violectric users have since added their thoughts. 

Once Tyll gets the amp measurement routines all worked out, you can bet I'll send over my V200 and the Lake People amps if I still have them by then (those are here on loan). I expect the measured performance will be similar to the subjective sound - excellent. 

Can Crazy's picture

I just finished reading your reviews on Violectric's V181, V200 and V800 DAC; very thorough and a really nice read. Too bad I missed them on my previous lookouts for Violectric reviews. Nevertheless, very much appreciated.

In the meantime my V100 with 24/96 USB DAC arrived, and wow has my listening experience changed! This is my first desktop Amp since I am new to the hobby, and I have to say that although I had heard a couple of top tier headphones and even a couple of tube Amps out of curiosity, I never thought they would seriously rival my stereo. I have to admit I couldn't have been more wrong.

I must have listened to those headphones and Amps in very poor conditions and with poor recordings, since I definitely don't remember having heard what I hear now. Right now I've even had some crazy thoughts, like selling my whole rig and getting a pair of Stax SR-009 and a Blue Hawaii and forget about everything else, lol.

For now I have a pair of Ultrasone HFI-2400 and a pair of AKG Q701s, but I already ordered a pair of HD800. I'm so happy with the sound I might just as well get a V800 and sit back and enjoy that combo for a couple of years, until next big step.

Regardless of the fact that I already got my Amp, I'll be keeping my hopes up for any upcoming measurements and, and hopefully a review and measurements of Violectric's optional 24/96 DAC board. I own a Dacport LX, and I can'r really hear any difference between them so far, but I have a little click sound issue with Violectric's DAC when ever a tune starts playing, right after the silence between tunes. I already wrote to Fried Reim, but I guess his answer will be there on Monday (hopefully nothing serious or as result of the weeklong shipping).

In the meantime I'll be checking out some of your other interesting reviews, it's really refreshing to read something serious on Head-Fi among so much nonsense. 

Greetz from Peru


John Grandberg's picture

Glad I could be of some assistance. I got my start doing this for fun on HeadFi and I continue to review stuff there when I have time. Yes, there's a lot of "junk" over there, but if you know where to look there are some very knowledgeable folks too. I've learned quite a bit from that community and will continue to support it as best I can. 

Glad you dig your V100, and I'm sure Fried will help figure out what's going on with the DAC module. He's always really helpful.