Crinacle bought the B&K 5128/4620

[Biblob]

He is going to update his database using the new B&K equipment.

Edit: It surprises me a little bit this thread doesn't get any traction. I thought more people would find this interesting to discuss

 

[julian_hughes]

I'm surprised too. The results from this new measurement set up are significantly different than before. Some *very* big differences. If you autoeq.app then you're very likely to get the new measurements provided. Apps which import autoeq measurements may not necessarily specify which set they are using. You could easily end up with very different EQ across different applications and devices, despite all the EQ's being apparently sourced from the same place and measured by the same person.

I watched Crinacle's video. One thing which stood out for me was that the old method is suspected of assuming that humans get much the same excellent seal as measuring rigs manage, and that they don't and that this causes problems. I can understand this with arbitrary size silicone tips provided by the vendor but I use foam tips and always get an excellent seal unless the IEM is such an odd shape and/or shallow fit that nothing can help. I think the old method was fine. Having a new measurement setup with very different target and some different assumptions, with all the measurements dumped in the mix with years and years of the old 711 results just seems like a recipe for confusion and for making comparisons much more difficult.

 

[Matias]

Resolve's B&K 5128 Headphone Target - you can try the EQ's.....

Hello everyone, Resolve has created a headphone target for his B&K 5128 measurement rig (which is an alternative measuring rig to the GRAS that we know that Amir uses here on ASR). Now this new measurement rig (B&K 5128) is supposed to be more anatomically accurate than the GRAS rig, but the...

www.audiosciencereview.com

 

[Dazerdoreal]

My experiences still align better with the 711, and I dont even mean the different target. The 5128 measures treble peaks which I definetely cannot hear.

 

[markanini]

My experiences still align better with the 711, and I dont even mean the different target. The 5128 measures treble peaks which I definetely cannot hear.

I find the measurements hard to read due to the extra "noise". This is the delta vs RA0045 using csglinux data, and it's noisier on individual measurements:

 

[Bow_Wazoo]

For me personally, the new measurements with the 4620 are nothing less than a milestone!

The presets that were created based on the measurements with the 711 ,never worked properly for me. They were too v-shaped.
The ones that were created based on the 4620 are a revelation. The tonality is finally right.
Less sub-bass, less treble.

I have compared 3 presets so far:
Zero 2, Blessing 3, and FH3. For all these IEMs, the 4620 preset sounds much more authentic/realistic to me. Through the measurements with the Bruel & Kjaer system, I am rediscovering the world of IEMs.
(FH5s wird morgen bei mir ankommen)

@crinacle
Thank you so much!

 

[markanini]

For me personally, the new measurements with the 4620 are nothing less than a milestone!

The presets that were created based on the measurements with the 711 ,never worked properly for me. They were too v-shaped.
The ones that were created based on the 4620 are a revelation. The tonality is finally right.
Less sub-bass, less treble.

I have compared 3 presets so far:
Zero 2 and Blessing 3, and FH3. For all these IEMs, the 4620 preset sounds much more authentic/realistic to me. Through the measurements with the Bruel & Kjaer system, I am rediscovering the world of IEMs.

@crinacle
Thank you so much!

Which presets are those? Toward which target?

 

[Bow_Wazoo]

Check this out

AutoEq/results/crinacle/Bruel & Kjaer 4620 in-ear at master · jaakkopasanen/AutoEq

Automatic headphone equalization from frequency responses - jaakkopasanen/AutoEq

github.com

Harman research targets the Bruel & Kjaer 5128

For the first time, the Harman IE 2019 headphone response target gets tested on the leading test fixture for headphone measurements.

www.soundguys.com

 

[markanini]

AutoEQ now uses a modified JM-1 with Harman filters for a target with BK5128-measured IEMs. This taget was not included in the CanJam 2024 presentation and has not undergone formal testing. The closest target that has is SoundGuys target, which was tied with Harman 2019 converted for 5128. SoundGuys target did win by a large margin among the segment of trained listeners. I think it's enough to say that these corrections will improve the sound to a lot of people. Keep in mind until recently there was only a DF target for BK5128 and that sounds bad to most people, so don't forget to give credit to those that develop and test new targets, at the same time there's still work ahead for target validation and formalization to the same level as the last generation rigs have. There will remain challenges with the new rig regarding increased operational difficulty and kinks between 200-500 Hz which are not meant to be corrected, it's unknown how it will impact AutoEQ results.

 

[Bow_Wazoo]

What a beauty.

The FH3 blew me away with the 4620 preset.
I'm really looking forward to the FH5S.

 

[InfiniteJester]

For me personally, the new measurements with the 4620 are nothing less than a milestone!

The presets that were created based on the measurements with the 711 ,never worked properly for me. They were too v-shaped.
The ones that were created based on the 4620 are a revelation. The tonality is finally right.
Less sub-bass, less treble.

I have compared 3 presets so far:
Zero 2, Blessing 3, and FH3. For all these IEMs, the 4620 preset sounds much more authentic/realistic to me. Through the measurements with the Bruel & Kjaer system, I am rediscovering the world of IEMs.
(FH5s wird morgen bei mir ankommen)

@crinacle
Thank you so much!

I agree. With Poweramp EQ, the Crinacle B&K presets are far superior.

The HD599 one is simply incredible.

 

[markanini]

I don't think any of the current proposed BK5128 targets improve the stock response of my Truthear Red + my manual EQ adjustment to the treble. To be fair it's already kind of close to one of them:

Note the JM-1 variant in this graph tool has 2dB more treble than the one in AutoEQ, confusingly they're named the same.