crinacle's IEM FR measurement database
May 23, 2018 at 12:13 PM Post #826 of 1,335
Jun 10, 2018 at 10:20 AM Post #827 of 1,335
Greetings from Singapore, I'm back! Just had a quick tour around some shops and got a huge amount of data, so prepare for a huge dump maybe tomorrow.

Just some reminders, in preparation for any future controversy or questions:
  • The new measurements are based on the IEC60318-4 standard. They are raw and not compensated to any target curve.
  • The "jaggedness" in the bass frequencies is due to a low-resolution correction of roll-off by the software. I have been trying to resolve this the past few weeks but I haven't reached a good solution yet, so just live with it. It's not that big an issue.
  • Universal IEM measurements are taken at ~3mm from the reference plane. This results in an inherent 8k-ish resonance spike in all measurements that may or may not be ignored depending on your interpretation of the measurements.
  • Custom IEM measurements are taken at the reference plane sealed with acoustic putty (I use blutack) which is consistent with standard procedure. The resonance point for these cases are closer to 9k-10k.
  • Anyone who misinterprets the results of my measurements are not my responsibility. Vice versa applies. My analyses are mine alone as much as others' are theirs. They do not speak for me.
Secondly, I'll be in Zeppelin & Co. on Tuesday, 12th June. Anyone who's interested in having their exotic IEMs measured just drop by, say hi and let me have a go. Always a delight to play with some rare and unique gear.
 
Jun 11, 2018 at 4:58 AM Post #829 of 1,335
Nice one on the n8..only the u18t missing :)
 
Jun 11, 2018 at 10:09 AM Post #831 of 1,335
I have to say that I really prefer these new graphs not just due to accuracy but Y scale in particular. Feels like you can see a lot more information due to less of a difference between max-min dB. Also, R and L instead of averaging is a nice touch.

They are uglier looking though :beyersmile:
 
Jun 11, 2018 at 10:11 AM Post #832 of 1,335
EDIT: learn how to attach image... :rolling_eyes:

Figure_1.png
 
Last edited:
Jun 11, 2018 at 10:22 AM Post #834 of 1,335
@crinacle is there a reason why none of your new plots are log-linear?
Maybe the plotting software is non-cooperative on this front?

The current format compresses 90% of the information into 50% of the plot area.

AudioTools uses an octave scaling system which is flat on pink noise. I would definitely prefer decade scaling with white noise though, but you can only have so much when you go portable on a phone.
 
Jun 11, 2018 at 10:27 AM Post #835 of 1,335
AudioTools uses an octave scaling system which is flat on pink noise. I would definitely prefer decade scaling with white noise though, but you can only have so much when you go portable on a phone.

Fair enough...

I can just run your raw data through a bit of Python code (as above) if I want a different look.

Thanks for providing that data BTW.
I did not see your latest deluge of info at your Google Drive link previously posted.
Still planning to keep that resource updated too?
 
Jun 11, 2018 at 10:35 AM Post #836 of 1,335
Fair enough...

I can just run your raw data through a bit of Python code (as above) if I want a different look.

Thanks for providing that data BTW.
I did not see your latest deluge of info at your Google Drive link previously posted.
Still planning to keep that resource updated too?

That will take a while. The app is linked to my iCloud drive which is linked to my desktop back in Melbourne. Plus, I'll need to scrub out all personal details out of the raw text files before I upload them. I'll see if I can do it all soon off my laptop though.
 
Jun 11, 2018 at 3:22 PM Post #837 of 1,335
Agreed, R/L channel info is great.
Consumers should know how much care (or lack thereof) producers put into channel balance.

And gear manufacturers, into how much care (or lack thereof) they put into volume balancing the drivers on each side.
 
Last edited:
Jun 13, 2018 at 9:12 AM Post #838 of 1,335

Users who are viewing this thread

Back
Top