Well, I got the IM02's yesterday and was immediately disappointed. They didn't sound anywhere near as good as I expected. After fiddling with them for a while and changing the ear tips around to try the different ones, I have finally gotten the IM02's to where they sound acceptable. I couldn't get any seal at all with the small or medium Silicone tips and the large silicone tips don't sit well in my ear. If I move around much, they don't stay in place. I've noticed that these earphones are also very placement dependent. They sound the best when I have them in and have a little bit of pressure on them to push them ever so slightly into my ear canal. Their normal position is okay, but the bass and low mids suffer when they aren't being held in place. I think a lot of this is just the way my ear is shaped. The tubes need to extend about 1mm farther to fit comfortably in my ear canal.
That said, even with holding them in my ear, they just didn't sound right. It sounded like there was a frequency peak right around 2.5kHz and then a sharp roll off above 5-6kHz. I knew there would be a roll off in the highs somewhere because of the listed specs only going up to 16kHz. I wish the specs would list the range for the readings. Is it + or - 3dB, or + or -10dB. That's a big difference. So I did some measurements. Based on my measurements, it looks like the frequency response was made using a + or - 15dB range, though I'm sure that the manufacturer probably has some other way of calculating it that gives them 'better' numbers. Still, here are my results...
(note: it seems my account doesn't have permissions to embed images, so hopefully the links will work).
I ran uncompressed 24 bit 96kHz pink noise through them using a Focusrite Saffire Pro 40 as the headphone amp with the volume set to about 1.4. I then used an Audix TM-1 measurement mic back into a mic input on the Focusrite and into the computer to an RTA program. I set the measurement mic right against the port on the various ear tips and ran the tests multiple times to ensure I had good results for each test.
No seal:
https://onedrive.live.com/redir?resid=EE9EAEC1D420426E%2151684
Without any sort of seal between the microphone and the earphones, this was the result. Basically nothing below about 450Hz, then a sharp rise up to about 2.2kHz, followed by a plateau across to 6kHz, then a drop to 10k, a small bump again around 11.5k, then a steep dropoff ending around 16k where it goes below the chart.
This was pretty much expected...no seal = no bass.
Second, Foam:
https://onedrive.live.com/redir?resid=EE9EAEC1D420426E%2151685
Using the foam tips dramatically improved the low end, but it still isn't flat. It shows up on the chart right at 20Hz, then gradually slopes up (almost ruler straight slope) to about 1.2kHz, staying flat until about 1.75kHz, then a drop to 4kHz, and the rest of the HF content is the same as without a seal.
The foam was the most consistent. With making minor changes to microphone positioning, the response pretty much stayed the same. If there was a seal with the foam, it sounded the same.
Third and fourth, Silicone:
https://onedrive.live.com/redir?resid=EE9EAEC1D420426E%2151686
Silicone was very inconsistent. I had to take a lot more readings with the silicone to get any sort of usable readings. It was very difficult to get a good seal with the silicone. When I could get a good seal, I got readings very similar to the foam. Honestly, this surprised me a lot. Based on my own ears, I expected to see a sharp spike in high frequencies and a larger roll off of bass frequencies. My expectations though, were way off. The earphones themselves performed very consistently. The biggest changes in frequency response were mainly due to differences in positioning of the earphones, the ear tips, or the microphone.
https://onedrive.live.com/redir?resid=EE9EAEC1D420426E%2151687
What was very surprising is the graph when I could get a perfect seal with the silicone tips. The response was practically flat from 20Hz up to about 6kHz with a bit of a spike around 2-2.5kHz. Above 6kHz, the response was the same as all the others. However, I should be very clear that you will never, ever get that good of a response in your ear. This had the microphone practically touching the exit port of the earphones and there is no way you will ever get your eardrum that close to the sound port. Moving the microphone out to normal distances resulted in the previously listed response curve. (Before anyone argues that this means silicone are better, I should point out again that the mic was practically touching the sound port and the silicone really wasn't doing much at this point. I'm sure I would get the same response from foam if I could get the mic that close.)
Closing:
So in closing, there are two notable observations that I have about the ATH-IM02's.
First, the HF response above about 2kHz is very consistent. It is clear that there is a bump centered around 2-2.5kHz regardless of tips, and that the HF response above 7kHz goes down hill fast. Excepting a small bump around 12kHz, the HF response is pretty poor. Some corrective EQ on your music player could make this better, but it is never going to work for reference quality monitoring.
Second, the various ear tips don't alter the sound. It is harder to get a good seal with the silicone tips, but all the arguments about silicone sounding "better" than foam, or vice-versa are complete crap. The tips sound the same. The way the earphone sits in your ear is the biggest contributor to sound quality. The choice of tips should be made on what is most comfortable, what gives you the best seal, and what allows the earphone to sit closer to your eardrum (ear shape permitting, of course).