HM-1 Headphone Monitor Correction

[Rason]

Hey
I am trying the plugin by DLD Technology. It surprisingly seems to be working very well. It is designed to be used as final filter placed right before the audio output in order to mix properly on headphones. Suitable for people who are travelling or mixng late in the night. My room is not treated at all and it is basically the worst combination of dimensions, windows, etc. I do not see any improvement possible in near future. My mixes- although sounding ok on my monitors-sound worse when played somewhere else. They always seem to share the same problems... sounding strange, boxy and with harsh highs. My room is reflected in them. I was not able to solve it by even proper referrencing with pro tracks previously. Now, when mixing to this plugin, it seems to be making wonders. The plugin emulates speakers inside the headphones and it makes the existing mix sound baad, it makes it sound like my mixes on other speakers in other environments. When I manage to make it sound at least ok and balanced with the plugin, then it sounds just soo great when it is turned of again and it translates much better on other systems. I think I would really recommend it to anyone who is sentenced to bad environemnt or headphone mixing. There is one thing I am not sure about. People should calibrate the plugin to their type of head/ears. There are 50 types of heads. Not quite sure I understand the calibration well after reading the manual. The head type presets seem to differ so much one from another. If anyone has any experience here, would you please give some tips or try to describe in detail how to properly calibrate it. Thanks a lot!

[tibah]

I can't be much of help here, but I did share the same problem trialing HM-1. I had at least 5 models that could work. The rest seemed like a mess of phase issues to my ears, but then again, this might be due to the model not fitting my, eh, head?

I would still rather try to learn my headphones, without such a tool. There are more good sounding mixes made on headphones than you would guess, but it's down to, again, knowing your tools and in this case, your room won't be an issue, with or without HM-1.

[XysteR]

I use the VRM built into the DSP on my focusrite audio interface with my KRK 8400 headphones. I swear by it. it's really good and saves me having to burn off CD's to play in the car, lounge etc. And because The VRM runs on the DPS of the interface it has no overhead on the CPU. Brilliant stuff

[Libraquaricorn]

The calibration process for that thing is ridiculous. I think I remember the manual saying you should pick the one that sounds best. Yeah, right...which one of these 50 models do you think sounds the most accurate to you? Gee, I don't know. I don't even have studio monitors so I just bought it. It would be nice if one could measure ones head and pick the closest match though, or one of the five closest matches.

It's in my rack, saved in my regular template but I rarely turn it on. Maybe I should, since I do everything with headphones. Not that I spend much time mixing...

I'm happy it works for you though.

I have a great pair of headphones - the AKG K701. They take forever to break in but I've had them for years and they sound better an better. i'm usually content when I play back my finished product on my stereo or in my car or on my portable and cheap Koss set, but I'm not a pro. I'm also into lo-fi (I love Crapre2) so mixing and mastering is mostly about making shure things look ok before I just press go and I'm done.

I would love to see a decent alternative to HM-1 in the shop though. I'd be willing to pay 6 times more for it if it was any good.

[DLDTech]

Hello - DLD here, I'll try to explain a few inner workings of HM-1 (and Revolution)...

HM-1 uses HRTF models, which is basically a way of recording how stuff 'sounds' with a certain speaker setup and head shape. It is actually done by placing small microphones in a volunteers ears, and moving the speakers around them whilst recording test tones.. The time and cost of doing this is very high, but fortunately MIT released the recordings from around 50 different people, which is where the 50 models comes from in HM-1.

Revolution (and many other 3D audio software) requires a 'generic' model that works with everyone. This is done by using a complex algorithm to take an average of the 50 models - this unfortunately 'dulls' the 3D, but is good enough for effects rather than mastering.

With HM-1, I wanted to give the best possible 3D effect AND frequency response, so included all the available original models. Unfortunately there's no way to know which will work best for a certain person, so you just have to try them all and pick the most believable. Even if you measured your head, differences in the thickness and texture of skin all make a difference (especially with frequency response). The only way to be 100% accurate is to be one of the original volunteers!

Think of HM-1 as a pair of NS10s - if you can make it sound good with HM-1 then it will generally sound good on anything

The main alternative to HM-1 is the HAAS effect which uses calculated delays rather than real world recorded data. There is a RE in the shop that does HAAS, but I haven't personally tried it. I've not heard the VRM either, but suspect they spent a lot of time perfecting their models, and sold a lot more units than HM-1!

[8cros]

Hello - DLD here, I'll try to explain a few inner workings of HM-1 (and Revolution)...

HM-1 uses HRTF models, which is basically a way of recording how stuff 'sounds' with a certain speaker setup and head shape. It is actually done by placing small microphones in a volunteers ears, and moving the speakers around them whilst recording test tones.. The time and cost of doing this is very high, but fortunately MIT released the recordings from around 50 different people, which is where the 50 models comes from in HM-1.

Revolution (and many other 3D audio software) requires a 'generic' model that works with everyone. This is done by using a complex algorithm to take an average of the 50 models - this unfortunately 'dulls' the 3D, but is good enough for effects rather than mastering.

With HM-1, I wanted to give the best possible 3D effect AND frequency response, so included all the available original models. Unfortunately there's no way to know which will work best for a certain person, so you just have to try them all and pick the most believable. Even if you measured your head, differences in the thickness and texture of skin all make a difference (especially with frequency response). The only way to be 100% accurate is to be one of the original volunteers!

Think of HM-1 as a pair of NS10s - if you can make it sound good with HM-1 then it will generally sound good on anything

The main alternative to HM-1 is the HAAS effect which uses calculated delays rather than real world recorded data. There is a RE in the shop that does HAAS, but I haven't personally tried it. I've not heard the VRM either, but suspect they spent a lot of time perfecting their models, and sold a lot more units than HM-1!

I suspect that really fit any model, after all, the main difference in the localization of the sound source. I use two devices. To the left and right separately. Of course with different models of the head to them.
I only wish to place the speaker as the behind and in front of the virtual listener. It would be great to have such a setting.

[Aquila]

I've just tried the HM-1 and also noticed the phasing artefacts.

For what it's worth, the stereo field sounds great. Hearing an instrument panned left or right does feel like it's coming from a speaker in my room however anything close to the centre, ie mono audio, has a very prominent comb filter on it regardless of which head shape I pick.

I'm left wondering if there was a way for this unit to process mid/side audio independently, and if it would overcome the comb effect?

[normen]

I've just tried the HM-1 and also noticed the phasing artefacts.

For what it's worth, the stereo field sounds great. Hearing an instrument panned left or right does feel like it's coming from a speaker in my room however anything close to the centre, ie mono audio, has a very prominent comb filter on it regardless of which head shape I pick.

I'm left wondering if there was a way for this unit to process mid/side audio independently, and if it would overcome the comb effect?

The "funny mid" thing is kind of typical for binaural/HRTF type stereo imaging. The Neumann Kunstkopf has the same issue.

[Aquila]

The "funny mid" thing is kind of typical for binaural/HRTF type stereo imaging. The Neumann Kunstkopf has the same issue.

Well it's good to know it's not a bug then. Still, I find it distracting to use myself. I may just have to put this back on the "not for me" shelf.

[Rason]

Thanks all for the feedback! So in order to properly calibrate i need to pan some uneffected rather simple sound hard right or left and find head type that makes it sound as clear as possible and as properly panned as posible for me, correct?