D_____

Credit to @Starelementpokeand @Energycore for helping out with this long-awaited refresh of the old PSU tier list by Aniallation (who previously went several months without logging in), leaving the old list in need of help. This is simply an update of the list that was flawed in creation by the PCMR subreddit and in dire need of a refresh.
 
This list is subject to change as new PSUs come out on a regular basis and standards may change or issues with units on the list may develop. A list of PSUs awaiting a tier on this list is at the bottom.
 
Tier 1 - Best
Aerocool - Project 7 Antec - High Current Pro, High Current Platinum be Quiet! - Dark Power Pro P10, Dark Power Pro 11, Straight Power 11 Bitfenix - Whisper Cooler Master - V-series (modular, not to be confused with the VSM), MasterWatt Maker Corsair - AX, AXi, HXi, RMi, RMx, SF, grey-label HX (2017), 2017 TXM greater than 650W Enermax - Platimax, Platimax Digifanless EVGA - G2, P2, PS, T2, 1000G1, 850W and above GS FSP - Aurum PT Gigabyte -  Aorus AP850GM LEPA - G1600 NZXT Hale90V2 Seasonic - X, Platinum, Snow Silent, all Fanless units, Prime (all series) Sentey - Platinum Power, Golden Steel Power Silverstone - Nightjar Super Flower - Leadex Gold/Platinum/Titanium, Leadex II Thermaltake - 1250D-T RGB, Toughpower Grand/RGB 1200W XFX - Pro Gold, Pro Black, XTS Tier 2
Antec - EDGE, TruePower Classic, EarthWatts Gold Pro Azza - Platinum be Quiet! - Pure Power 10 400W or greater Bitfenix - Formula Gold Cooler Master - VSM-series (semi-modular) Corsair - Most RM variants, HX (old), 550W/650W 2017 TXM, Vengeance Enermax - Digifanless, RG, GX EVGA - GQ, B2, 550W and 650W GS, GD, G3, G1+ (not to be confused with the G1) Fractal Design - Edison M, Tesla R2 650W/1000W, Newton R3 FSP - Hydro G, Dagger SFX Gigabyte - XP1200M Kolink - Continuum NZXT - Hale82 N higher-wattage versions Riotoro - Enigma Rosewill - Quark, Fortress, Capstone Seasonic - S12G, G-series/SSRM, M12II 750W/850W, Focus Plus Gold/Platinum, Focus Gold Sentey - Solid Power SS Silverstone - Strider Platinum Super Flower - Golden Green, Leadex Silver Thermaltake - Toughpower Grand/RGB, Toughpower Grand Platinum, Toughpower DPS Platinum XFX -  Pro XXX, XTR, TS Gold Tier 3
Antec - Neo ECO II be Quiet! - Straight Power E10 Bitfenix - Fury  Cooler Master - MasterWatt Corsair - 2017 "Grey unit" CXM***, CSM, some RM variants, 2017 CX - 450W CX have one PCIe connection Cougar - GX-S Deepcool - DQ-ST Enermax -  Enermax Revolution SFX 650W*** EVGA -  BQ 750W and up Fractal Design - Tesla R2, Integra M FSP - Hydro X, Aurum CM, Aurum Pro Lian Li - SFX-L LEPA - G600 NZXT - Hale82 N lower-wattage units OCZ - ZX PC Power & Cooling - Silencer Mk III, Turbo Cool Riotoro - Onyx Rosewill - Lightning, Silent Night, Tachyon, Photon Silverstone - Gold Evolution, Strider Gold, SX650/700/800 SFX, Strider Titanium Super Flower - Platinum King Thermaltake - DPS Gold/RGB, London, BlueEvo 2.0, Smart Pro RGB, Smart series >= 750W Vivo - 24K 650W XFX - Core > 650W, ProSeries Bronze >650W Zalman - EBT Tier 4
Antec - EarthWatts/EA Green, Earthwatts Platinum, High Current Gamer, Neo Eco  be Quiet! - Power Zone, Pure Power L9, Straight Power E9, Pure Power L8 Corsair - Old CXM "Green unit" variants, Rev. 3 "Green unit" variants, GS-series Cooler Master - GM, GX Storm, MWE Cougar - LX Enermax - Revolution X't EVGA - BQ under 750W, NEX Gold, B3 Fractal Design - Tesla R2 500W Inwin - Classic Series PowerSpec - PSX GFM Rosewill - Capstone G Seasonic - SS, SSP, S12II/M12II, ECO 430W Silverstone - Strider Plus, Gold SFX Thermaltake - Toughpower Gold, Paris XFX - TS Bronze Tier 5
Antec - Basiq VP-F, Basiq BP be Quiet! - Pure Power 10 under 400W. Corsair - Rev. 2 CX "Green unit" variants, VS "grey units" Enermax - NaXn 82+ EVGA - 450B/500B/600B/700B, NEX-B OCZ - ZT, ModXStream Rosewill - Glacier Silverstone - Bronze SFX XFX - XT Tier 6
Antec - Basiq VP Bitfenix - BPA Cooler Master - B2 Corsair - VS "orange units" EVGA - 430W/500W/600W 80 PLUS units FSP - Raider Silver Fractal Design - Integra R2 LEPA - MX-F1 NZXT - Hale82 V2 OCZ - Fatal1ty Rosewill - Hive, ARC (M) Silverstone - Strider Essential Thermaltake - LitePower, ToughPower, Smart series < 750W Zalman - GS/GT Tier 7 - Worst
Cooler Master - Elite series EVGA - 400W non-80 PLUS certified FSP - Hexa Thermaltake - TR2 These are all crap  
Currently awaiting more info
 
Some things on this list may not be perfect. Let me know if you believe a PSU is undeserving of its spot on here and provide links to back up your claim and it may be moved. 
 
Some helpful links:
 
How Many Watts Do I Need? - How much power will you need for your system? This guide helps you get a PSU that has more than adequate wattage for you.
80 PLUS Efficiency and What It Really Means - general guide to 80 PLUS efficiency and what it entails. Currently pinned on the Cases & Power Supplies subforum.
 

 
Ask any questions below! 
 
How do you even grade these things? ***
But I don't like Tier Lists!!!!111!!!!
 

Either garbage motherboard, garbage USB dongle or garbage ears.

He's setting up some sweet 5 PC LAN, fitted with cameras in multiple angles, etc. 
 
Anyway, he mentioned the $1,000 HDMI cable, so I asked him if he watches LTT often.
 
Well, he watches it occasionally. I thought that was cool. I suppose it's expected; Deadmau5 is a pretty tech-savvy guy. 
 
 

Hello all! Long time LTT fan, new to the forum. I wanted to share my build, as it's really been a labor of love for the past month. I trash picked it from a local computer store in December, and it's come a long way since then.

It started life in 21 years ago 1995 with a 486dx2 running at a mind blistering 66mhz, with a massive 4mb ram, and 1mb video memory. The 3.5 floppy disk could hold a maximum data capacity of 1.44MB. As of a few days ago, it now has a an Intel i7 6700k running at 4.3ghz, 16GB's of 3000mhz ddr4 ram, and 8GB's video memory. The 3.5 floppy drive can read up to 512GB of data from a single diskette. This was not a plug & play build. Massive amounts of cutting, fabrication, welding, grinding, a filing were required. It was awesome.

Here's the full build with gifs for proper demonstrations of the floppy drive in action: http://imgur.com/a/fvh1M

 

A lot of assumptions being made here.
We don't take money to do reviews. We never have.
We take money to do advertisements, which are never labeled as reviews.
NVIDIA provided the graphics cards (and Cooler Master sent cases and coolers, ASUS sent motherboards, Intel sent CPUs, Kingston sent SSDs and RAM, Swiftech sent water cooling gear) for our editing workstations for product placement in the whole room water cooling series of videos we produced featuring this hardware extensively.
In the grand scheme of things, $6000 of graphics cards (don't imagine that's their BOM cost on them either) is a great deal for 1-2M impressions on a series of videos about using them.. but those videos were NOT a review. They were about a clearly sponsored (we said it was more than once) outrageous project.
As for sending "keeper" vs "loaner" units of hardware, yeah for many products we just don't do loaners anymore, but more than anything else this is to do with me not feeling like it's worth my time to play email tag for shipping labels and spend an hour driving to FedEX and filling out stupid international bill of landing and commercial invoice nonsense over like a $100 item that costs more to send back than to just not bother with anyway.
That and quite frankly even a negative review (for some reason) often generates a lot of sales for the manufacturer.. so as far as I'm concerned they can eat the cost of the item and I'll do whatever I want with it when we're done reviewing it - whether that's continuing to use it personally (Pebble Steel and iPhone 6 come to mind) - or re-visiting it only when we need to re-do our comparative numbers (here's a lesson on why this is necessary and we can't just re-use old numbers)
And on the subject of the Pebble and iPhone 6 - these are both items that were NOT provided by the manufacturer. One was purchased for my own use and the other was provided by a separate sponsor who makes skins for phones (dBrand), but I didn't necessarily mention this at the time because who cares? It doesn't colour my impressions of the device.
Here's an item I wasn't allowed to keep and won one of the two editor's choice awards I've ever given out: https://www.youtube.com/watch?v=s1uj6wQTu58
Here's an item was allowed to keep that I ripped apart saying it was worse across the board than the two similar products from other manufacturers that I compared it against: https://www.youtube.com/watch?v=-O1aCNrjV64
Here's an item I was allowed to keep that I basically ripped apart on camera and sent it back anyway because it had so little value to me I didn't even want to give it to a friend: https://www.youtube.com/watch?v=c3AgcltUM9s
And I could do this all day..
You may notice that most of the items we review we are fairly positive about. There is a reason for this, and it has nothing to do with being paid to review it.
I don't generally bother to review stuff that I already know is stupid junk.
The bottom line is this. You can either trust me or not trust me. Ultimately that's your deal. My job is to do what I do, and the people who recognize it for what it is - my opinion being given to you - will follow, and the others can go find someone who maybe they think "hasn't sold out" but might just "happen to agree with them" to listen to. Enjoy.

Underage Elephants Riding Motorcycles.

Why the fuck is Gamerdude's post tagged as best answer?
Fuck this world.
 
 
yo i hurd u liek amps so i put a preamp in ur amp so you can preamp while u amp.

Here's a good question for you guys over at CM-Storm; any chance that you will  release a mouse with a similar design and grip to the Mionix Naos line-up or perhaps the Func MS-3?  Preferably optical, as well!

I do. Ask me anything.  Tesla T1 and LCD-2 here.  Long story short, headphones that expensive are only worth it to reply to silly posts like this.

The HD800s are bright. Sound is great, soundstage is great. I've had enough time with HD800 to do all sorts of crazy stuff with them, and I know the bass is very well controlled. But there's isn't that much of it. I've spent very little time with Alpha Dogs but from my time with it I was impressed. I felt they were pretty balanced. Nothing felt out of place to me. They are closed, so it's got good noise isolation, but it won't sound as open. You better be willing to try out the headphone in person or take a significant risk if you're going to get to that price range. And you're going off of the assumption that HD800 > HD650 or somethin'. I know multiple people that prefer HD650. Ignore the price, find one that sounds the best to you, and then look back at the price to see if the price is OK. Try not to let price influence your hearing (it will regardless, fact of human bias). You can't read your way into knowing exactly how a headphone sounds. And people often like to justify spending a mortgage payment for their house on their headphones, so they will over-exaggerate differences (if any) they actually hear.
 
HD800s should be more comfortable. They're going to be more lightweight. Alpha Dogs are closed and you have that leather which from other headphones, seem to make ears a bit heated after a long time. HD800s you can wear forever while gaming. Why don't you try out the HD600, 650, HE400, HE500, LCD2, Mad Dogs before you make a decision. You're auto-assuming you just want Alpha Dogs for HD800s. Audio is not for everybody. You can't just throw money at it at assume you'll be so impressed your balls will turn blue from the tease. You better try things out to make sure it's for you. And TRY to be honest with yourself. Do I like this? Or do I REALLY like this? Is price and bling factor affecting my judgement?
 
Mayflower suggested DT990s. Ehh, I guess so, but I think they should be kindda bright too. It's a U shaped FR, kindda. Boosted bass, boosted treble. I'm not a fan personally although I sure some people are. Sure, throw DT880 or 990s to the list.
 
Fuck sound cards. They need to go away. If you're going to upgrade your sound card, get an external dac and amp. Schiit stack is all you'll ever need and is a little cheaper than my favorite O2/Odac combo. Asus is betting that people like you have no idea what you're doing and you're going to listen to their marketing weasel-words.
 
Oh, and unless a headphone has just the PERFECT FR you are looking for, EQ is not a sin, it's a tool. Granted it's a tool most people have no idea how to use properly, but it's there and it makes a difference.
 
  Before I explode, I can tell you're new here (to the audio sub) and to audio in general. Headfi requires that you already have a lot of knowledge in headphones otherwise you're going to believe the truckloads of shit they are going to try to sell you from magic dacs to magic cables. It's forum full of people suffering from placebo and expectation bias. You are making the problem worse by recommending a newbie that website. I see you mean well, the forum IS big after all, and big usually suggests authority. If I were in your position I would probably say the same thing. But the answer is NO. Just no.
 
​
I'm having a streak of bad temper this week and you're getting close to setting off one of my triggers. You don't know anything about us. We have our own private little Audiofeelz Thread where we have private sexy parties. $1000 headphone is just the norm. Stop acting like you know anything, enough to even know what to suggest to seek help because clearly you don't.

You act like hailing from Headfi makes you MORE credible, which is just fucking rich. And seriously? "resident audio elitist"? Who do you think you are, LOL. I am open to criticism from reasonable people who have actual points instead of 'hurrrrrrrrr, go to Headfi meets' (which I've BEEN TO, by the way). Turns out it was a terminology usage problem, sue me. I've asked paragraphs and paragraphs of questions, but they were to reasonable people, not wannabe cool kidz like you.
 
And again: No shit sherlock, of COURSE you can alter the sound with a tube amp. There are a gazillion ways to alter the sound, from taking a hammer and banging it against your amp and thus destroying it, decreasing fidelity via tube amps, getting any other non-transparent SS amp, blah blah blah. NO SHIT. Are you THAT eager to prove everybody wrong to go picking out technicalities with a guy with a gaming laptop trying to get his headsets to work well? What, now you're going to recommend a tube amp to the guy? Are you out of your mind?
 
 
Suuurrrre. Your ears defy science. Powering a headphone is not arcane knowledge, in fact it is very well understood. Tube amp is old tech, there's a reason why engineers left tubes en mass, until of course, some hipsters decided to go back to old tech. Good for them, but don't be a moron when using it. I don't care how many irrelevant engineering degrees you claim to have on the internet, it means nothing. Oh, you're trying to prove how much of a pro elitist you are? Baby, I have the single largest Haswell OC thread on the internet, and I've written a 300 paged book on religion. Is this what you want? A dick measuring contest? You happy now? Last I checked you don't win even if you claim to have a monster 9 feet long cock.
 
Headfi is a cesspool incest orgy of confirmation bias and placebo and sponsorship corruption. So sick of these placeophile audiopeelz that want to justify spending their second mortgage on a few glowing tubes, so they make up new problems that defy science. Once a headphone is driven with correct levels of power and gain to achieve good loudness (factoring in preferred listening levels and dynamic range of the track), with correct FR, low distortion + noise (including no clipping), and low output impedance, THAT'S IT. Mission accomplished. As far as fidelity is concerned (aka, hi-fi, aka hi fidelity), mission 100% accomplished. And don't you even bring up the fucking subjective euphonic distortion argument, this was never about that. Or, you conveyed your frustration so poorly nobody could pick up on that.

>implying we're all already not doing drugs
 

Anyways, still my point stands: I think it's unlikely that the Game Zero would exhibit lack of bass with no other symptoms. People keep thinking power = bass, and it's coincidental that it also happens to be what Jay claims is missing when it's not powered well enough. He should have done a basic blind or even a sighted tested with an a/b switch. He comments here saying that DB drags destroyed his low frequency sensitivity (assuming he's partially deaf), but in another video he mentions that he's hyper-sensitive to bass and gets a headache. Which is it? I don't think he cares.
 
So, I don't think Jay's video counts as evidence. The original response was that it doesn't take a headphone expert to understand what something sounds shitty. But this is a type of mistake I see inexperienced audio reviewers making (apart from placebofiles): Saying that lack of power = less bass but nothing on clipping. He let his preconceived notions of what powering a headphone means to color what he heard, and his inexperience allowed it to continue.
 
There's a chance I'm wrong, but I think more likely than not I am right.

You are insulting people when you call it an 'audiophile religion'. When you make such a broad statement, people are also liable to getting confused and think you're talking about them when you are not. Religion is about dogma. I was willing to discuss why my understanding of headphones and their need for power is incorrect. I was saying that only lack of bass is unlikely when the headphones are not driven properly (and should be damned well impossible given Creatip's planar magnetic example and his measurements there), and that people keep thinking driven well means good bass for some reason, maybe it is a linguistic thing. If your hearing is damaged it is important to say so in your reviews. A large amount of people watch reviews from mainstream Youtubers: LTT, you, Barnacules, etc and they get the wrong idea about audio. You guys have reached a large audience, congrats on that, but that comes with the burden to double-check your info more often.
 
If anything, I am the one not in the religion in this entire thread. Ironically LTT is the single most 'anti-audiophile-religion' audio subsection of any large forum I know of, including OCN, Headfi (lol), and the like. Maybe Logan's forum would do as well, I haven't been there in a while.
ANYWAYS, the point here is, you recommended an amp and a dac, I suggested sticking it to onboard and maybe trying EQ. I was suggesting the most ghetto solution possible. Not agreeing with you doesn't make this an audiophile religion. Or maybe you should explain what you mean by that in the first place.

I don't understand why a top of the line headphone has to have a distinct sound signature. In fact, to be an accurate headphone, it needs to be as flat as possible... not have its own personality.
 
The HD800s are hyped at the treble, that alone will make it sound less muffled, more in your face, more detailed, less forgiving, more fatigueing.
 
*Insert more BS about HD800s being picky here*
 
OP still hasn't replied or stated why he is picking the HD700s. Did he just look at his budget, look at the most expensive Sennheiser headphone he can afford, and decide based on that? Feels like OP is less involved than the people answering...

He-6 with O2 is still kindda sketchy. ODA to the rescue!
 
I was thinking of things to buy for the holidays. I was thinking of a Ducky Legend TKL, but a rep told me it won't come in TKL. So I was thinking maybe an ODA. If... it actually comes out for sale.  :rolleyes: @Mayflower hint hint wink wink nudge nudge JUST RELEASE IT ALREADY (JK, I'd rather wait for a complete product than buy an early, buggy version.)
 
Rosson only skimmed the thread at your request. There were juicy-looking pictures and lots of numbers. That's enough to get people to say "oh, looks legit", and then not bother reading it. Nwavguy has given as so much. Not just the Objective and its spinoffs, but that blog is a cornucopia of knowledge. And despite all that, many people don't know him and have never read his blog and the information contained therein. So, maybe I am pessimistic about the reach of my thread,
 
You know that I did more than just add the conclusion in my last edit, right? (You probably do, just making sure.) Part of what I say in the past have been jokes, some are actual opinions. There's a difference between having a different preference and denying reality. I tolerate people who deny reality all the time, and not just in audio. I've managed to post on Headfi without getting banned so far and I anticipate that to continue. But of course, our beliefs have consequences, and I see people repeat old audiophile myths about picky headphones and an amp "not being enough" for a headphone (not enough as in what? Because there sure is enough power.) People are being misled and basically their money are being stolen.
 
For the record, Nwavguy's own blog contains more words attacking tube amps and other things of that sort than my thread. Maybe I'll edit the conclusion. I dunno.
I spent so many hours on this thread and I learned so much. I was very eager to share what I had learned and how it all finally made sense. I didn't really see the same enthusiasm back though (positive or negative).
  Buying Audeze headphones on Amazon tends to suck (for some odd reason). That is why I bought it off Headfi instead of Amazon.

Looking for a super TL;DR guide to how to calculate power requirements? It's in here:
 
Calculating Power Requirement of LCD-2
A few of you were doubtful about the O2's ability to drive LCD 2. Here is my answer.
 
Sensitivity is dB/Vrms
Efficiency is dB/mw
 
To convert between the two, we need to know the impedance of a headphone. There is a formula for it, but just use this calculator (I've verified that it's super legit).
http://www.jensign.com/bdp95/headphones/
 
LCD2, being a planar magnetic headphone, the impedance is pretty much the same no matter the frequency.
As I mentioned in the past on another thread, Audeze's website lists the efficiency of LCD2 but calls it sensitivity.
 
To get voltage requirement: 
10^((dB peak-sensitivity)/20) 
 
To get power requirement: 
10^((dB peak-efficiency)/10)
 
LCD2's impedance and efficiency has been different in the past. The best good data I can find which shows an old LCD2 revision is from Inner Fidelity's measurements here, from 2012: http://www.innerfidelity.com/images/AudezeLCD2sn53211704circa2012.pdf
Unfortunately, Inner Fidelity's measurements are listed differently than the format I use. It needs to be converted. 2012 LCD2 requires 0.209 Vrms to reach 90dBSPL according to their chart.
90 + 20 * LOG ( 1/0.209 ) = 103.6 dB/Vrms, & @ 59 Ω = 91.3 dB/mw. The efficiency of 2012 LCD2 is 91.3 dB/mw. The impedance and sensitivity are from Tyll's measurements (which BTW, seems very accurate, props to the man. His measurements were dead on for HD800).
 
Now let's figure out the efficiency of LCD2 with Fazor from Tyll's latest measurements as listed here: http://www.innerfidelity.com/images/AudezeLCD2sn5423021Fazer.pdf
LCD2 w/ Fazor requires 0.114 Vrms to reach 90dB SPL.
90 + 20 * LOG ( 1 / 0.114 ) = 108.86 dB/Vrms, & @ 72 Ω = 97.4 dB/mw.
 
Let us compute the required amount of power for a headphone now we know the impedance and the efficiency. Don't forget the formulas to figure out power and voltage requirements which I already listed.
 
2012 LCD2 based on IF (91.3 dB/mw, 103.6 dB/Vrms):
105dB:
23.4mw
1.17v
 
110dB:
74.1mw
2.1v
 
115dB:
234.4mw
3.72v
 
120dB:
741.3mw
6.6v
 
Latest LCD2 based on Audeze's spec sheet (93 dB/mw, 104 dB/Vrms):
105dB:
15.8mw
1.1v
 
110dB:
50.1mw
1.89v
 
115dB:
159.5mw
3.35v
 
120dB:
501.2mw
5.6v
 
Latest LCD2 based on IF (97.4 dB/mw, 108.8 dB/Vrms):
105dB:
5.75mw
0.65v
 
110dB:
18.2mw
1.15v
 
115dB:
57.5mw
2.04v
 
120dB:
182mw
3.63v
 
Looking at O2's Capabilities
O2's max output (from NWavguy's blog):
Max Output (33 Ω)   613 mW 
Max Output (150 Ω) 355 mW 
Max Output (600 Ω) 88 mW
 
P (Power in w) = V (Voltage RMS) * I (Current in amps)
I = V / R   (Ohm's Law)
Therefore: 
P = V * (V / R)
Therefore:
P = (V^2) / R
 
0.613 = (V^2)/33Ω
V = 4.5  ; Therefore, we know that the O2 can power 4.5V 
Since P = V * I...  0.613 w = 4.5 x I  
I = 0.14 amps.
​If you do this calculation for all impedances:
15Ω,   337 mW - V = 2.2, I = 0.15
33Ω ,  613 mW - V = 4.5, I = 0.14
150Ω, 355 mW - V = 7.3, I = 0.048
600Ω, 88  mW -  V = 7.3, I = 0.012
 
Voltage swing is simply how much voltage a headphone can put out from one situation to the next. The O2 can dish out 7.3v for most situations.
As you can tell by matching the specs of the O2 with my calculations, the O2 can drive all three possible LCD-2s to 120 dB SPL. Well, except maybe the 2012LCD2. That really depends on how many volts the O2 can dish out at 59Ω, which isn't specified by Nwavguy. Either way though, it can power it to above 115dB. If we assume the O2 can deliver 7.13v still, we get something like this:
P = (7.13^2)/59 = ~903mw. (The highest peak wattage O2 can output out of all possible impedance is at 49Ω.)
 
...Which means, the O2 would easily be able to power it. It's important to note that for the purposes of this particular headphone and calculation, we're arguing between 1-2 dB here... something like 118dB vs 120dB. Every extra dB requires that much more effort to achieve.
 
 
How Loud is Too Loud?
The threshold of pain is at 130dB. Hearing loss begins at 85dB which sets in after 8 hours of continuous music. Refer to this chart below:

120dB is such an insane number, IT'S NOT EVEN LISTED IN THE CHART! At 120dB you will suffer hearing loss in approximately 10 seconds. What use is an audiophile with busted ears? Imagine rocking out to 120 dB for an hour a week. Imagine what that does to your hearing.
 
The hardest track to power is a special type of non-compressed music which has consistently quiet sounds with sudden, brief loud sounds (Sforzando notes). This is very different from a normal orchestrated piece, with sustained high levels, and even more different from modern, heavily compressed pop music. Nwavguy gives this guideline to tell us how much dB peaks we can expect from a type of track:
Highly Compressed Pop: –6 dB to –9 dB Well Recorded Pop: –9 dB to –12 dB Well Recorded Acoustic/Jazz: –12 dB to –18 dB Wide Dynamic Range Classical: –18 dB to –30 dB Let's assume the worst case scenario given by Nwavguy with the ridiculous 30dB peak. 85 dB + 30dB peak takes us to 115dB with an extra 5dB of "buffer" (aka overkill). Do you want to pick how much hearing loss you want? Now is your chance!
 
A few of you voiced concerns that going close to or even halfway of the maximum output power of an amp drives up distortion. Nwavguy himself has disproved this (well, for the O2 anyways):

Here we can see that distortion increases very little until you hit a certain point, where the distortion explodes. This is basically clipping point. This shows that we do not need an extra "buffer" on top of the 5db buffer for the 115db buffer assuming you listen to the most audiophile and hard to drive tracks and another buffer given that 115db itself will cause hearing damage in a few minutes. Worried that the chart is misleading because it lists voltage? P = (V^2) / R. Let's say the first clipping line shown in the picture is at uhm... 2.2v? That's P = (2.2^2)/15 or 322 mw. Do you remember NWavguy's listed maximum output power at 15 Ω? It's 337 mW. I was very close to guessing the maximum output of O2 at 15Ω simply by eyeballing the graph. It also proves that the large distortion jump occurs during clipping, not at some random amount before clipping. How much distortion are we trying to keep under in total? Under 0.005%. But because music itself masks distortion, the generally accept number is 0.01%. As you can tell from the graph by NWavguy, you won't even hit 0.005% (as long as you don't clip). From a human perception standpoint, we're not going to be able to distinguish small amounts of distortion anyways when the music is blasting 120dB SPL to our ears, even if the distortion was much higher. (Me, I'd be grasping my ears for relief.)
 
Miscellaneous Information
Let's do a less insane situation on how much power you will need. Say you listen at 85 dB with 15 dB peaks (still in audiophile jazz category, far beyond compressed Lady Gaga). That is 100 dB. Let say you're using the most inefficient version of LCD 2, with an efficiency of 91.3 dB/mw. That's 10^((100-91.3)/10) or 7.4mw of power required. Do you understand just how much buffering and insanity we've gone through to try to see if the O2 can hold up to push requirements up x100 the 100dB listening volume implies with the least efficient headphone out of the three? The Audeze's website recommendation for LCD 2 ranges from 1-4w.
 
To calculate maximum SPL from a given amount of power we use this formula:
Efficiency in dB/mW + 10 * LOG ( Pmax in mW)
91.3 + 10*Log(4000) = 127.32 dB, which should cause hearing damage in approximately 1.875 seconds.
 
Some audiophiles want tube amps for some reason, and that makes their life harder because tube amps have inferior amounts of distortion as power output goes up. They also have higher output impedance, and when that is too high in relation to the impedance of the headphone (this can easily happen with low impedance headphones like LCD2 or LCDX), that causes some fuckery with the amp's ability to output power. Such high output impedance also distorts the sound (but people get tube amps to get those distortions in sound because they find it euphonic). So: Tube amps will make your life harder, especially if you want to calculate the amount of power the amp actually needs to give out. If you're looking for a simple, cheap, powerful, and dead-on accurate amp, a solid state amp like O2 can drive the LCD 2 without any problems. If you're using a tube with very high output impedance, I can see why Audeze's 1-4W power rating actually makes sense. But it doesn't make sense to people using the O2.
 
Impedance Swings
Planar magnetic (aka "Orthodynamic") headphones like the LCD-2 have the same impedance across all frequencies, so the argument that impedance changes with frequency played doesn't even apply here even if it was a valid argument.
 

 
In a dynamic headphone this isn't the case. The higher the impedance a headphone is at a frequency, the easier it is to power at that frequency. To figure out the worst case scenario we need to look at a dynamic headphone at its lowest impedance. The frequency in which the impedance of a headphone is at its maximum is called the 'resonant frequency', which is where the headphone is easiest to drive. Thankfully, like every dynamic headphone and their mom are easier to power than the 2012 LCD2.
 
P = (V^2) / R

P is power in watts (not milliwatts). V is voltage. R is impedance of the headphone. The larger R is, the smaller P is. It's simple math. (I talk about the voltage changing in the next section.) If we look at the impedance rated for a headphone vs measured impedance, the rated impedance by the manufacturer tends to be the lowest impedance for a headphone at all audible frequencies (in other words, shows you the worst case scenario). HD800 has lowest impedance of about 330ish but Sennheiser rounds it down to 300 just to be safe. This is why companies call it "nominal impedance". The listed specs are already pretty much worst case scenario from the get-go.
 
Sensitivity Ratings @ 1khz... A problem?
So, I've already shown why a higher impedance by itself actually makes a load easier to drive. But that's assuming that voltage required stays constant. (Hint: It does.) Sensitivity ratings are typically taken at 1khz. Sensitivity is measured in dB/Vrms. It's not as complicated as it looks: You give the headphone 1Vrms and watch how loud it gets for a given frequency. Why 1khz? Why don't we measure for other frequencies? If we took a sensitivity reading for every frequency from 20hz to 20khz, took the data, plotted it on a graph, and connected the dots, we would actually be making a frequency response graph. That's the key: We are trying to pick a spot on the frequency response that represents the "average volume". And so, many industry leaders choose 1khz. (Please go complain to Sennheiser, for example.)

No, you can't look at the quietest parts of the frequency response graph and say they are "the hardest to drive". The DT990 has a trough of -15dB and a peak of +5dB. When the DT990 plays 120dB at its -15dB trough (13khz), it will play 100hz notes 20dB louder (120+20 = 140dB). You're not buffering for anything by trying to strive for 120dB at 13khz. Given the same amount of voltage, a headphone is not going to play as loudly at every frequency, yes, but that's called a frequency response. DT990's sound signature is more V shaped. That's a characteristic of the headphone, not the amp. If your amp is changing the frequency response of your headphone, it's not transparent.
 
You want to pick a frequency that represents the average loudness of a headphone, not its peaks or troughs. Many headphones screw with things outside of 1khz. Note how in the graph, I picked four different headphones from four different companies, yet they all have non-screwy frequency response at 1khz. The Grado PS1000 has stupid amounts of treble. The old Beats might have ridiculously loud bass. Some headphones are V shaped in their frequency response. 1khz is a fine point to pick as far as points go. If your headphone happens to be ridiculously hard to drive AND have a big 1khz peak in its frequency response according to a frequency response graph, you might worry. But if you actually have a high quality headphone with a frequency response that's remotely flat, you should be just fine.
 
So when we look at the equation, P = (V^2)/R, we now know a lower R increases the amount of power required, and the voltage variation required to get a given loudness is only useful if we pick one reading that represents the average loudness of a headphone across its entire frequency response. I've just accounted for every single variable in the power equation.
 
 
 
TL;DR:
O2 can power LCD 2.
Quit complaining and enjoy the music.

You shouldn't be having any problems with those headphones. The only thing I can think of was that you used high gain with it and pushed too much power.
 
 
We are working on it...prototype is in our shop.

This post has gotten out of hand.... i just ordered a5+ and the audioengine subwoofer. i did my research and im content with my purchase. please stop the dick measuring contest about audiophile gear.

A newer reviewer named Lachlanlikesathing made an excellent video recently that discusses the issues with head-fi forums.
 
https://www.youtube.com/watch?v=AqC0gYL9wFk
 
The audio industry is poisoned with places such as head-fi it seems. The moderators apparently care more about their sponsors than the truth, than actually helping their forum members. Sad.
 
Funnily enough, that hifiguy528 clown is commenting all over this video. Probably to protect what he's built, which is a relationship with audio companies that give him so much free shit that he says nothing but positive things about their products and lies to everyone.

I know this will draw hate as is common on criticism, keep in mind this is my opinion and I'm sure Linus will do fine when shifting his content on to a different demographic altogether. Afterall, smartphones and general consumer gadgets are popular watching and I'm sure many of Linus' new audience is here and ready to defend him. But this is constructive criticism, so I would appreciate the same gesture in return.
 
 
Having been watching LTT for a long time, it has gone increasingly downhill over a long period of time for me:
 
Linus prioritizing boutique video quality sacrificing all quantity and depth of insight. We're down to less than a video a day, and the ones we get are usually shorter and less opinionated/ insightful. Forgetting who the audience is with an increasing number of consumer gadgets, (seriously, a wi-fi plant humidifier??) mobile phones, and other borderline tech of fringe interest to PC enthusiasts, even further lowering the quantity of content desired. Getting on the same release bandwagon as all. the. other. tech. channels. Every video with a product I want to watch either doesn't exist at all or is covered by OC3D, Linus, TekSyndicate, TastyPC, SWEclockers and HardwareCanucks ALL AT ONCE. Random WAN show guests that I very often find to be irrelevant and annoying. Getting that weird uncomfortable YouTube "buddy club" vibe that just kind of smells bad to me. Josh from Fractal being an exception. Why go to PAX? There are hundreds of YouTube channels doing PAX. Just why? Where are all the hidden gems in PC hardware? Have Linus completely abandoned reaching out to manufacturers for samples of interesting new coolers, cases, NAS boxes, you name it? It just seems Linus is more interested in autonomy, getting all the same stuff as everyone else, and losing touch with the market. It used to be about consumer enlightenment, showcasing stuff you might not have heard of. Remember the Thermaltake Tt eSports BAG? That kind of thing. I bought one because of Linus. What has Linus shown me in the past 6 months that I wanted to buy afterwards? Absolutely nothing, I'm afraid.  
But the final step that is probably going to make me unsubscribe, is not going to Computex. Seriously, that is gutting. That is a stone-cold knife in the back to the original audience of the channel, to whom Computex is probably the highlight of the entire year, after the huge thrill of covering it last year. I could forgive anything from Linus but not that.
 
It makes me sad, as Linus Tech Tips has been my very favourite channel for years, but now it is my least favourite of all my subscriptions.
 
My 2 rubel, anyway.

Headphones are extremely subjective about what one person will like and one person won't. I personally like my ATH-M50s despite all the hate they get around these forums they are a solid pair of cans. If you've tried your friends pair of them and liked them who cares if someone else doesn't like em? 
 
I've had my pair of ATH-M50s for damn near 5 years at this point & they've stood up to constant use and abuse. They fold up pretty compact for carrying which is nice because I bring em to work every day & aside from the pleather earpads getting worn out (which is gonna happen, its why they make replacements) they are still as good as the day I bought em.
 
And even if you do somehow manage to break them Audio-Technica has pretty damn good tech support I've found out. I had a friend blow out a driver on his pair after 4 years & I was able to get a replacement driver from Audio-Technica for $25 shipped & do the repair for him. 

I'll just leave this here...
 

 

Success! Well...sort of. 
 
Finally got V 0.1 up and running after a little bit of debugging. Had made an error when copying my schematic over from multisim to eagle pcb & switched up my V+ & V- pins on my LME49990s. 
 
Nothing a little bit of swearing & judicious application of an x-acto knife couldn't sort out though. This is why you prototype things
 
Unfortunately I didn't take pictures after completing assembly & initial debugging because my phone battery died but I'll take it back apart tomorrow & take some.
 
 
Here's the pic showing where I screwed up in the schematic:
 

 
 
Edit: Took a pic of the top side at least. Blue wire is the jumper fixing the swapped power rails.