Superlux HD668B Review (headphone)

[bravomail]

Conclusions
You can predict my summary then: no good as is but recommended with equalization. And a bargain.

Thx for review, Amir!
Bargain, yes, recommended - no.
I got them long ago on many ppl raving/recommending them.
But quick test showed that bass/subbass muddiness/boominess, which you found.
Try any test track with low frequencies - it is a complete mess.
In addition, they r quite hard to drive from mobile (phone or tablet), so not suitable for portable use.
My recommendation - avoid Superlux as a brand.
I mean - they r tolerable, and sometimes fun, but not audiophile in any sense or form.

 

[Robbo99999]

Is this OK (pictures)?

View attachment 105451

View attachment 105452

On the Right channel Peak Gain is 2.5dB in Red.

That's ok.
Feel free to ignore that peak gain of 2.5dB.
We know the right channel is playing at a lower volume than the left channel (at least for the unit that Amir tested).
Equalizer APO doesnt understand that, so it's showing you "Peak Gain of +2.5dB" for the right channel.

I don't think you can just ignore the peaks in the red, (regardless of whether that's the headphone with the quieter channel), because we're talking about digital clipping here - that will still happen so the overall preamp for the whole system (covering both channels) needs to account for the highest peaks in the digitally boosted quieter channel.

 

[sweetchaos]

I don't think you can just ignore the peaks in the red, (regardless of whether that's the headphone with the quieter channel), because we're talking about digital clipping here - that will still happen so the overall preamp for the whole system (covering both channels) needs to account for the highest peaks in the digitally boosted quieter channel.

Thanks, corrected.

Updated my post:

Channel: L
Preamp: -11.0 dB
Filter 1: ON LS Fc 30 Hz Gain 4.0 dB Q 1.0
Filter 2: ON LS Fc 50 Hz Gain 4.0 dB Q 1.0
Filter 3: ON PK Fc 4000 Hz Gain 5.0 dB Q 2.0
Filter 4: ON PK Fc 5380 Hz Gain -7.0 dB Q 6.0
Filter 5: ON PK Fc 8582 Hz Gain -7.0 dB Q 5.0

Channel: R
Preamp: -8.5 dB
Filter 1: ON LS Fc 30 Hz Gain 4.0 dB Q 1.0
Filter 2: ON LS Fc 50 Hz Gain 4.0 dB Q 1.0
Filter 3: ON PK Fc 4000 Hz Gain 5.0 dB Q 2.0
Filter 4: ON PK Fc 5380 Hz Gain -7.0 dB Q 6.0
Filter 5: ON PK Fc 8582 Hz Gain -7.0 dB Q 5.0

 

[amirm]

Due to some shipping mix up, I wound up with a second sample of this headphone. So decided to compare the two:

The differences are within the range of measurement error. All the same characteristics remain down to small notch at 500 Hz.

This is good news as it means that our EQ recommendations should work with all samples.

 

[thewas]

This is good news as it means that our EQ recommendations should work with all samples.

I would rather say though "probably most samples" as with 2 samples we by far still don't have an idea of the tolerance band, especially for such a low cost product.

Still measuring a second sample gives a good information surplus and is very welcome.

 

[dasdoing]

the 681 is the better value champ https://www.rtings.com/headphones/tools/compare/superlux-hd-668b-vs-superlux-hd-681/439/471

 

[markanini]

the 681 is the better value champ https://www.rtings.com/headphones/tools/compare/superlux-hd-668b-vs-superlux-hd-681/439/471

Going by measuremnts it seems so. I owned a HD662F that measured better yet, still the peak in the upper-mids ruin the sound.

 

[johnk]

I managed to pick up a pair of these on ebay for <$15 delivered. I can confirm that I also see the balance issue between left and right. Paired with an Apple usb dongle they sound good to me using the suggested EQ. I am hoping they will give me a bit of a reference point to help with setting up my speakers. Maybe that is a bit optimistic given the cost, but they are certainly better than my other headphones.
Thanks, John

 

[abdo123]

I just purchased (and used) the headphones and they sound quite expensive with EQ, I had very little expectations, so it was a big (delightful) surprise.

 

[3125b]

they sound quite expensive

Oh that's nice, who doesn't love that expensive sound

 

[abdo123]

Oh that's nice, who doesn't love that expensive sound

Aaaay don't come at me, Amir used the same term hahaahah.

 

[bravomail]

That Superlux list is total nonsense.
The HD688B sounds nothing like the DT990. (owned both, still own DT990)

at the end - did u like Superlux?

 

[solderdude]

I found the HD688B too boring and could not get on with the wing system.
There are Superluxes that are 'better' IMO.
The stock pads are total crap, the early vinyl ones were improved over time and got softer in 'pleather' covering.

 

[Marcomarco]

it's correct?

 

[ehabheikal]

My dream is to have impedance sensing on the output of the amplifier which automatically sets an EQ preset for the transducer you're listening to, which can be loaded in via USB from a computer or bluetooth with a phone app. The app could take care of auto downloading, or the unit could ship with the most popular models matched to their impedances and allow additional presets to be added by the user. I think that would cover most people, since usually people have a phone or at the very least a computer.

Was just thinking of this

I am using an app on android called wavelet and it has a database of headphones that it will eq, you just need to enter model.
Tried it with cheap 9usd sony reviewed elsewhere here and wow

 

[meowchin]

I've been using these headphones for probably more than 7 years, the past year with AutoEQ preset. During this time I've recabled them, changed the earpads for velour from 681 EVO and replaced the rubber material on the head pads with fabric, which made them a lot more comfortable. They went through a lot, but they're still going strong. And yeah, you can't adjust the headband, but you can extend the pads.

 

[pasa soydan]

I cannot import the frequency response csv file to rew software i can only import the relative one.why is this?

 

[djigibao]

Probably because this is not a REW file.
For REW it must be .mdat file.

 

[markanini]

I cannot import the frequency response csv file to rew software i can only import the relative one.why is this?

File>Import>Import frequency responce

 

[pasa soydan]

File>Import>Import frequency responce

it gives me this error