[yosoro]

I'm curious if Impulcifer can bring the headphones close to the sound of the speaker.
I have used software such as waves and ooyh software that does not require ear measurement, and the results are not good.

 

[castleofargh]

I'm curious if Impulcifer can bring the headphones close to the sound of the speaker.
I have used software such as waves and ooyh software that does not require ear measurement, and the results are not good.

It is a massive step up in getting more accurate audio cues. But how convincing it will be, might depend on you. Can you trick your brain into thinking you're listening to speakers, when there are no speakers to see in front of you? Can you remain tricked if you move your head and the sound turns with you? I feel like different people can be impacted differently.

But in term of acoustic, with the impulses you're only missing non linear variations(which can be seen as bad and good). This time the rest is measured at your ears instead of being measured by some dummy head. That may or may not make a big difference(how close to the human average are you?).
I happen to care for head movement A LOT. But even then, I felt that custom impulses without head tracking, were more important than head tracking with the wrong acoustic model(Waves NX was wrong for me, I can't tell about anybody else). OOYH has the benefit of having some really cool rooms/speakers, but nothing is made for our own head, so once again it's all down to genetic luck. I'm sure some people are getting a very convincing result from it. I didn't. And among the stuff I've tested, it's the one I stopped using the fastest.

All that to say, it depends, but yes it's probably a significant step up if you measure the signal of the speakers and of your headphone from inside your ears(so you need in ear mics, DIY or binaural products).

 

[musicreo]

I'm curious if Impulcifer can bring the headphones close to the sound of the speaker.

For me it sounds very close to real speakers.

Except headtracking impulcifer has functions (room correction) that even the realizer doesn't provide.

To be honest I have done about 50 measurements with 5 self build mics (two different mic types) and there are some measurements that did not sound good. I tested different speakers, different distances, different positon of the speakers and different fixing of the mics in the ear. There is a learning curve but now all of my last measurements were very good.

 

[johnn29]

Along with room correction, Impulcifer also lets you use IEM's which Smyth doesn't.

It's exactly like my real speakers when you sit in the same spot and do an A vs B. Combine it with a Bass Transducer and play some Atmos demo clips and you'll be shocked!

The biggest impact on the quality of the HRIR I made was actually cutting the hooks off my Sound Professionals mics. Since gluing the earplugs to them channel balance is perfect everytime and it just sounds great everytime.

Once you take whatever HRIR you make to other places your brain does do tricks and some sound odd. But if you take it back to the seat you measured it's going to sound the same. The way I've got around this is taking an ultra nearfield measurement that I use with my mobile devices or out of my own house. That sounds good anywhere.

I think head tracking is valuable for identifying where sounds are located, especially behind you or above you. But if you're just listening to regular movies/music I don't think it's that valuable. But I guess this varies with each user. It works well with my Mobius because the HRIR is generic - so the tracking helps the center channel pop out of your head.

 

[sander99]

Impulcifer also lets you use IEM's which Smyth doesn't.

You can use IEMs with a Smyth Realiser. You can create a HPEQ (headphone compensation) with either the so called manloud or manspeaker method.
[Edit: see my next post]

 

[johnn29]

Ah - wasn't aware of that. Thanks

 

[sander99]

You can use IEMs with a Smyth Realiser. You can create a HPEQ (headphone compensation) with either the so called manloud or manspeaker method.

Actually the manspeaker method would be not so suitable, my mistake. In this method you compare the headphones with the real speakers by ear using test signals. With IEMs that would mean taking them out and inserting them all the time.
With the manloud you compare and try to level match different frequencies with each other by ear. Then you can leave the IEMs in your ears during the entire process.

 

[lowdown]

My experience with Impulcifer is a bit different. It doesn't match the sound of my speakers, it's much better. The illusion of listening to actual speakers in front of me is stunningly convincing. After months I still sometimes have to take off my headphones to make sure the speakers aren't playing late at night disturbing my neighbors, even though given my setup I know they can't be playing. But also on several key points the Impulcifer sound is much better than my speakers. The tonal balance, clarity, resolution of detail, elimination of any "in the speaker" sound, and zero crossover or room anomalies, all make Impulcifer a major upgrade. It's a very obvious improvement, so much so that I no longer listen to my speakers for music.

 

[jaakkopasanen]

I made some performance improvements to Impulcifer. On my machine the demo measurements get processed now in 9 seconds instead of 35. Plots are still very slow though and I doubt there is a lot that can be done about those. There are 256 graphs plotted for 7.1 system afterall.

Reverberation management is coming to Impulcifer proper soon.

 

[johnn29]

Very excited about reverb management!

Edit: Liking the progress dialogues in the latest update:

 

[kalstein]

I have a question about 'room correction' part.
Is that additional information for making 'hesuvi.wav' or standalone solution for room correction like REW ?

 

[johnn29]

Room correction just amends the processed hesuvi.wav with different parameters - a specified curve/tilt, flatter frequency response and a bass boost if your actual room can't dive as low as the target.

Users/Jaakko - some more videos that might be of interest. I saw there's an open issue around mixed phase correction so the Dirac one might be useful for you. Although you probably know it already! Audyssey one is also from the software engineer

- Audyssey room correction.
- Dirac creator

 

[johnn29]

Loudness Compensation / Dynamic EQ with HeSuVi/Impulcifer use:

I found that Equalizer APO has a built in loudness compensation filter that works ideally! Just load up Editor.exe in the EAPO folder and then click the plus button.

Bass perception falls off as volume falls. So when we listen at lower volumes bass seems to be missing and sound is balanced. The target curves we shoot for with our real AVR's or Impulcifer (like 6db Harman slope) apply for reference volume. The good thing with the loudness compensation algo is that you can actually get a reference volume on headphones ala 0 on your AVR and it'll boost the bass/treble based on psychoacoustic models (Fletcher-Munson although more modern models).

The filter lets you set a reference volume and then you just leave your device volume alone and use windows to control. When you set your volume to 100 (or whatever it finds as the 75db mark) - that's reference volume. Now - it's easier to get a reference volume on an AVR because you can use a UMIK-1 with calibration file loaded and shoot for 75db with your internal test tone. Over headphones it's harder. What I did was get the tone playing at 72db on my real AVR then flip to my open back DT990 and perceptually match.

Works really well.

Some headphones have this built in - like the Bose 700. But unlike with Bose - this one actually knows what the reference volume should be so works ideally, not approximately with any content that has a reference volume like movies.

Offset is there for things that are mixed louder than movies - like music or TV. Basically if you google Dynamic EQ by Audyssey you'll get a better understanding.

 

[jaakkopasanen]

Recommended reading for those who are interested in the reverberations and early reflections. https://www.audioholics.com/room-ac...ions-in-home-theaters-a-different-perspective

The article speculates that the ideal reverberation time and first reflection delay depends on the source material. Speech requires different acoustics than say classical music. This is of course not possible in a physical room but with Impulcifer one can set different reverb times for center channel, front channels and surround channels. Center channel mainly has dialogue and the surround channels have ambiance in movie sound tracks. Additionally different room acoustics could be ideal for different genres of music and with Impulcifer one can have different BRIRs for different genres.

 

[johnn29]

Jaakko - another related one recently on speech intelligibility/reflections from Matt Poes at Audioholics: