Super Best Audio Friends

The evolution of the original irreverent and irrelevant and non-authoritative site for headphone measurements, i.e. frequency response graphs, CSD waterfall plots, subjective gear reviews. Too objective for subjectivists; too subjective for objectivists

A not-exhaustive* guide for helping listeners get the best synergy for the HD800** I have heard various HD800 headphones. To be honest, even with extensive mods, SBAF mod, SDR mod, any combination of mod, etc, the HD800 never quite jived with my tastes. However, under good synergy and the right mods, I do find it to have very significant sonic merit and think it still does deserve its reputation and place.

Other members of this forum have commented that the HD800 does not respond to EQ well (the data definitely exists to back this line of reasoning). This is because the early reflections inside the housing, the large ribs, and mounting structure of the cups, combined with the driver's angling towards the ear and its suspended nature cause some very interesting (and measurable) effects on the FR and harmonics. I do NOT like the HD800 stock because of the FR oddities, and thus this guide, like @purr1n's original guide, will assume that a SDR mod will have been used at the bare minimum (or some similar mod). In a personal sense, @tommytakis kindly let me borrow his HD800, which had both the SBAF and SDR mod (as well as a few others), and I found this iteration of the HD800 to be the most enjoyable to listen to for me. I do not discount the other mods; just haven't heard them.
So I was on eBay shopping around for phono preamps and stumbled across something that looked like another Chi-Fi box. Upon further inspection, it appears to be made in Massachussetts by hand. There are three models right now, the
  • MM-3 - $119 (currently listed at $99 right now on sale)
  • MM-5 - $199 (currently listed at $179 right now on sale)
  • MM-6 - $329
Now all of them are 40 db of gain for moving magnet carts for 47K + 100pF loading, however 150pF, 220pF , or 330pF available upon request.Not much change between specs, although slightly lower THD, lower noise, and closer to RIAA spec. All run on a 24V AC adapter/step down transformer.

All hand matched discrete parts. Says using NOS JFET's, although does not say what exact ones (ie, 2SK170), and if is not that, I am sure it is something similar.
Vali2+ technical measurements, Vali2 is longer in production, technical measurements for comparison may be found here.


I have long had a love / hate releationship with the Schiit Vali2. It has good tonal balance and a pleasing harmonic distortion profile. Conventional wisdom for budding tube headamp listeners was to acquire this inexpensive component, then listen long, gaining insight while saving funds until a significant ascension with more knowledge could be afforded. At headphone meets the Vali2+ has been popular with listeners, myself included. However an important note about can meets is the rather high ambient noise, which may mask low level issues.
Let me start off by thanking purr1n and SBAF for the opportunity to start this thread, as I have a great passion for what I'm going to present, and being able to share it directly with the community is awesome and makes it more meaningful. I hope you'll like what you'll see and make good use of it.


I'm a headphone enthusiast, a long-time audiophile and music lover, and a hobbyist headphone designer/DIY-er from Croatia. Currently also a medical student at a university, which often limits the time I have for enjoying this hobby, but that just makes every moment I can spend doing this count even more. I'm going to present the project I've been working on for the past almost four years, and I'm going to do it in the most open and transparent way possible by giving the headphone community full access to it, including the ability to use the raw CAD files of two fully developed, completely modular headphone models, the LTA V1 and V2, for FREE. Anyone interested will be able to use the files as they are, or further modify and build upon the existing designs, in order to manufacture the parts on their own, and make their own headphones by using the project as a guide. This is something unprecedented and will allow you to skip a ton of designing, R&D, testing and prototyping, and will enable you to go straight to actually making headphones, saving a ton of headaches and money in the process.​
Read here for initial subjective impressions:


I will be providing mostly measurements with limited subjective impressions pre-measurements. In a nutshell, the Hive is a great e-stat. The presentation is closer to the Sennheiser estats (HE60 and HE90) or HE Jade than the STAX headphones. This meaning they have a more tactile attack. Bass slam actually seems stronger then even the Sennheisers. I believe there were some design considerations made here with regards to diaphragm tensioning, which is quite low. The downside is that the diaphragm has a tendency to stick to the stators if the headphones are not put on very gently. The workaround is to put the headphones on first, let the pressure equalize in a second, and then plug in the headphones in. Running a lower bias can help too. If one likes the ethereal and limp STAX sound, then it is possible that the Hive may not suite them. Personally, I love their presentation. Tonally, they are bit like the HD600s. The stock pads have an 8kHz peak that emphasize sibilance. Fostex woody like pads flatten out this peak a bit.
I have created a new measurement website with the help from Oratory1990. To goal is to have a place for the community to easily share measurements of their headphones, etc. It can be found here,31

I've talked to Purr1n, and he has given me the go ahead to upload his measurements from My site takes in raw measurements from EARS rigs (can also take in data from other rigs), and uses the SN from the EARS rig to fetch the HPN compensation directly from MiniDSP. On top of the HPN compensation, we have developed an estimate at a Harman target response, as well as what we call "Optimum", which is our desired target response.

While I do know there is already a 560S thread containing a bunch of Marv's measurements, I am personally more adhesive to subforum definition. Particularly thinking measurement posts should be placed on the measurement subforum as I've been often lost in finding specific measurements and ended up googling. I'll put a link to the previous 560S measurements at the end of this OP.


As you see in the plot, from 3khz downward, it's very difficult to bitch about 560's tonal balance. Maybe subtly uptiling. But it doesn't sound thin or lean at least for bass and midrange. I am actually kinda liking it. They do have some issues in this region, but fr is not one of them imho.,,
The latest brainchild of Lampizator founder Lukasz Fikus, is the new "mid-range" Baltic-3. This model sits below the Golden Gate and sells for about 1/3 price. A design goal by Lukasz Fikus was to offer performance comparable to the Golden Gate for a notably lower price.


Just fired it up today, so it is still burning in, so just first impressions at this point,'s very impressive. Very quiet, very transparent, especially so for a tube component. Tonally accurate with superb resolution, definition, and microdetail, but has the presentation qualities of the best analog sources like my Michell Gyro SE TT w/ SME V arm and Koetsu Urushi Vermilion M/C cartridge, with the exemplary neutrality of the Rega Planar 10 turntable.
Special thanks to @rhythmdevils for taking us back to memory lane. The AMB Gamma 2 DAC was a wonderful sounding DAC during the infancy of headphone audio. It was available as DIY only; however assembled models could easily be found for sale. The DAC uses the old Wolfson chip, which is a good sounding part, but tonally it could sound all other the place. The AMB Gamma 2 DAC implementation has a bold sound with a minimum of digital nasties. The tone was its primary selling point. It was not the most resolving or nuanced sounding DAC however.

Gamma 2.jpg

I wanted to showcase this DAC because this is an example of how marketing practices prevailed over engineering. For those who actually thought NWAVGUY was a master engineer, looking to save us all from bad measuring gear by developing the ODAC, there was already one DAC design readily available from his arch-nemesis AMB or Ti Kan with arguably better performance. Those of us who have met Ti Kan know what an outstanding and humble dude he is. If there was any DAC that should have been the ODAC, the AMB Gamma 2 should have been it.
Measures x1000 times worse than Magni 3+. Sounds cool, where do I sign up? The Vali 2+ is different from the Vali 2 in these respects.
View attachment 23784
  • More present highs - less dark
  • More lively, more resolving, more engaging similar to the Vali 1 (in retrospect, Vali 2 was a step and a half forward, but a step back).
  • I think the Vali 1 still has a small edge in immediacy, but the Vali 1 also has some issues with the super tube microphonics and highish distortion which is audible, but in a pattern that indicates euphonic distortion.
I'm not going to wax poetic on the Vali 2+. However, I do think it would be fun to take measurements using different tubes. Why not? Folks rolled all sorts in tubes in the Vali 2s of old, including rare exotic tubes that cost x4 the amp.
I have so much gear in the queue that I decided that I would do a stream of consciousness review to get you the goods on the DAC2541 as I go along instead of making everyone wait. I know that there is a lot of interest here.


After a day of warm-up, the DAC2541 is ungodly good. Those who want an oversampling DAC with a modicum of inherent warmth, but no outright warmpoo; an honest, but smooth and not-in-your-face top-end; and microdynamics and PRaT that grabs your very soul by the balls, then this is it. I had some idea of how this DAC may have sounded like given my experience with Soekris' prior generation of DACs, but my expectations have been exceeded.

1100 EUR. Wow, just wow.
Per @spwath, we shall conduct a poll on the Harman Curve. This one is public.

Read this for more technical explanation from @Vtory: