I believe you. I am a science guy, so I am always a skeptic and verify everything I can. Trust, but verified. I am assuming your measurements are correct which is why I'd like @Precogvision to give us his take on what is going on once confirmed. My theory is that the scoop out is happening in the presence region of the curve and where sibilance is most common. So, What the scoop out might be doing is allowing us to listen to higher volumes than normal. If our ears are missing presence, we simply raise the volume until we get the transparency our brains we need. That simultaneously raises the bass and treble giving it more excitement. I noticed that I listen to higher volumes with the Bird than I do with other IEMs. And, with some sources, when you turn the volume down, the Bird does not maintain an even response across the curve. That's my theory.
So I understand your theory and it makes sense, however, I think it is different. I think you are raising the volume more than normal because you are probably not used to the way the Traillii's sound presentation is. With that I mean how diffused and laid back the sound is, it does not image close to your forehead or front of head like some other IEMs I have heard. Erlkonig is a prime example of an IEM that does do that. With the Erlkonig I always had the perception I could listen at much lower volumes because the image was literally like, in my eyeballs. The Traillii puts music very wide and behind you almost, when I first got mine I was cranking it up too as it was something I was not used to.
Thanks to a generous reader for sending all of these IEMs in for evaluation! I'm kind of de-sensitized to flagship IEM's at this point, but I gotta admit...I was pretty giddy unboxing all this at once haha. Top left is qdc Anole VX, top right is Sony IER-Z1R, bottom left is VE8, bottom right is qdc Anole V14. I've heard most of these before, but with some of them it's been so long (eg. with the VE8, it's been going on a year) that I've forgotten what they sound like.
Now, here's the game plan. I'll be doing a shootout between the VX, Z1R, VE8, and U12t by request. Some will recognize that these are Crin's S-tier IEMs. When I asked, he told me he's still got a "World's Best IEMs" write-up in the pipeline which obviously entails these IEMs. His will be a round-up; mine will be a shootout. Still, I don't want to steal his thunder, so I'll probably write this on the side and drop it on Head-Fi/Reddit/HP.com forums first. Then, if he's OK with it, I'll publish it via official channels under something like "reviewing Crinacle's top IEMs" or something like that. The Anole V14 will get a full review, VE8 will get a re-visit; I think my thoughts are still pretty consistent with VX and IER-Z1R.
All measurements are up on the graph tool. By request, I've graphed the Anole V14 using the recommended sound signatures from qdc.
Sure, have not heard either of those but Traillii treble goes for pretty natural and smooth rather than airy and super extended. Does not seem to be the way it’s treble tuning is intended.
I believe you. I am a science guy, so I am always a skeptic and verify everything I can. Trust, but verified. I am assuming your measurements are correct which is why I'd like @Precogvision to give us his take on what is going on once confirmed. My theory is that the scoop out is happening in the presence region of the curve and where sibilance is most common. So, What the scoop out might be doing is allowing us to listen to higher volumes than normal. If our ears are missing presence, we simply raise the volume until we get the transparency our brains we need. That simultaneously raises the bass and treble giving it more excitement. I noticed that I listen to higher volumes with the Bird than I do with other IEMs. And, with some sources, when you turn the volume down, the Bird does not maintain an even response across the curve. That's my theory.
Do you know any other sets that using the same trick? Dunu sa6 has a scoop at 6khz too. I kind of like that trick when i heard VX in terms of killing the sibilance trick.
Recently, Apple Music announced the release of lossless music which is big news for us audiophiles. I know I myself was pretty excited when the feature showed up on my iPhone a few days ago! But lossless music sucks up lots of space and bandwidth. And let's be honest: Can you actually tell the difference between lossy and lossless? Of course, it's easy to say you hear a difference between track file A & B when it's sighted. But placebo is very real. So what better way to find out for certain than taking on a few nasty ABX tests? You know, just the thing that would have most audiophiles tails tucked, running away? Well, that's exactly what I decided to do for fun.
Here's the first test I ran. I could definitely discern an audible difference here; scoring a 9/10 has a p-value of less than 1% which is well within the realm of statistical significance. In order to pass this test, I focused on only one section of the song; specifically, the part where it explodes into loudness at about a minute in. I ignored pretty much everything else when I was doing the test.
Here's the third test I ran using MP3 320 vs. FLAC 24/48. I actually failed the second test using these same files, getting only 6/10 correct. Even getting 7/10 correct only corresponds to a p-value of 0.11 (using a straight binomial pdf calculation) which most would not deem statistically significant. In the future, I might be able to score higher (as I was suffering from listener fatigue at this point), but it stands that I was mostly acting on gut instinct for these comparisons.
Finally, after taking a break, here's another track I tried playing with. The files are 24-bit FLAC and MP3 320 respectively. Once upon a time - unblinded - I could have sworn I heard a noticeable difference between the two tracks. But as you can see, I might as well have been guessing at the beginning! Up until test #7, I actually got more wrong (way more LOL) than I did correct. Why is that? Simply put, my approach was wrong. I was trying to discern a difference in clarity using Taeyeon's vocals. Unfortunately, it's actually very difficult to pick up a difference in sheer note clarity, particularly in the midrange. Recognizing that I couldn't discern a difference using her vocals, I switched up my approach. You can see that my score steadily picks up starting from #8 - in fact, I get 7/8 correct after this point! So what changed? Well, like the Sawano Hiroyuki track, I started listening for loudness. An MP3 file's dynamic range is more compressed, so - listening carefully - it came off just a hint louder on the MP3 in the beginning.
The Verdict
You can see that the differences I used to pick up on differences between the files mainly came down to one thing: dynamic range. If you want to pass one of these tests, it's mostly going to come down to listening for stuff like noise floor, peak loudness, and reverb trails. Honestly, I don't think (the vast majority of) people can even tell the difference between MP3 320 and FLAC based off of sheer note clarity. I'm certainly not able to. And you know what's even crazier? These are cherry-picked songs, songs that I have listened to hundreds, if not thousands, of times. If I was taking a test like this under duress, or using tracks I was not familiar with, I highly doubt I would pass. For better or worse, the audible differences between lossy and lossless are teeny-tiny, to the extent of which they're pretty much negligible for most people. Don't even get me started on lossless vs. 24-bit or MQA, oh boy.
Now, am I going to stop using lossless files? Hell no. Just like anyone else, I'm a fat sucker for FOMO. I don't want to know that I'm getting something lesser if I could have better. And hey, if I do hear a difference sighted - even if it's just that juicy placebo - I still heard a difference, right? That said, I still wanted to put it out there that ABX tests like this are a great way of keeping things in check the next time you think you hear a "huge" difference. They're way more difficult to pass than you'd think, and even if many of us don't like them, they could probably help us save a ton of money.
Recently, Apple Music announced the release of lossless music which is big news for us audiophiles. I know I myself was pretty excited when the feature showed up on my iPhone a few days ago! But lossless music sucks up lots of space and bandwidth. And let's be honest: Can you actually tell the difference between lossy and lossless? Of course, it's easy to say you hear a difference between track file A & B when it's sighted. But placebo is very real. So what better way to find out for certain than taking on a few nasty ABX tests? You know, just the thing that would have most audiophiles tails tucked, running away? Well, that's exactly what I decided to do for fun.
Here's the first test I ran. I could definitely discern an audible difference here; scoring a 9/10 has a p-value of less than 1% which is well within the realm of statistical significance. In order to pass this test, I focused on only one section of the song; specifically, the part where it explodes into loudness at about a minute in. I ignored pretty much everything else when I was doing the test.
Here's the third test I ran using MP3 320 vs. FLAC 24/48. I actually failed the second test using these same files, getting only 6/10 correct. Even getting 7/10 correct only corresponds to a p-value of 0.11 (using a straight binomial pdf calculation) which most would not deem statistically significant. In the future, I might be able to score higher (as I was suffering from listener fatigue at this point), but it stands that I was mostly acting on gut instinct for these comparisons.
Finally, after taking a break, here's another track I tried playing with. The files are 24-bit FLAC and MP3 320 respectively. Once upon a time - unblinded - I could have sworn I heard a noticeable difference between the two tracks. But as you can see, I might as well have been guessing at the beginning! Up until test #7, I actually got more wrong (way more LOL) than I did correct. Why is that? Simply put, my approach was wrong. I was trying to discern a difference in clarity using Taeyeon's vocals. Unfortunately, it's actually very difficult to pick up a difference in sheer note clarity, particularly in the midrange. Recognizing that I couldn't discern a difference using her vocals, I switched up my approach. You can see that my score steadily picks up starting from #8 - in fact, I get 7/8 correct after this point! So what changed? Well, like the Sawano Hiroyuki track, I started listening for loudness. An MP3 file's dynamic range is more compressed, so - listening carefully - it came off just a hint louder on the MP3 in the beginning.
The Verdict
You can see that the differences I used to pick up on differences between the files mainly came down to one thing: dynamic range. If you want to pass one of these tests, it's mostly going to come down to listening for stuff like noise floor, peak loudness, and reverb trails. Honestly, I don't think (the vast majority of) people can even tell the difference between MP3 320 and FLAC based off of sheer note clarity. I'm certainly not able to. And you know what's even crazier? These are cherry-picked songs, songs that I have listened to hundreds, if not thousands, of times. If I was taking a test like this under duress, or using tracks I was not familiar with, I highly doubt I would pass. For better or worse, the audible differences between lossy and lossless are teeny-tiny, to the extent of which they're pretty much negligible for most people. Don't even get me started on lossless vs. 24-bit or MQA, oh boy.
Now, am I going to stop using lossless files? Hell no. Just like anyone else, I'm a fat sucker for FOMO. I don't want to know that I'm getting something lesser if I could have better. And hey, if I do hear a difference sighted - even if it's just that juicy placebo - I still heard a difference, right? That said, I still wanted to put it out there that ABX tests like this are a great way of keeping things in check the next time you think you hear a "huge" difference. They're way more difficult to pass than you'd think, and even if many of us don't like them, they could probably help us save a ton of money.
I've done quite a bit of CD ripping in the past year or so and I'm happy that I did that "project" now and to FLAC instead of like 15 years ago when I might have chosen something silly for the format and then been burdened with (possibly very relevant) thoughts of having to do it again. I'm also quite happy to not have sunk into the "EvEn bEtteR than LossLesS" circus so just going with a "keep it simple" approach. I certainly wouldn't discern any difference in those FLAC vs. 320kb MP3 tests, but no point in going for them with the price/capacity of MicroSD cards or external drives these days.
Music always comes first, and if you want to dig deeper then it's good to be aware of the quality of mastering (partly subjective of course).
Recently, Apple Music announced the release of lossless music which is big news for us audiophiles. I know I myself was pretty excited when the feature showed up on my iPhone a few days ago! But lossless music sucks up lots of space and bandwidth. And let's be honest: Can you actually tell the difference between lossy and lossless? Of course, it's easy to say you hear a difference between track file A & B when it's sighted. But placebo is very real. So what better way to find out for certain than taking on a few nasty ABX tests? You know, just the thing that would have most audiophiles tails tucked, running away? Well, that's exactly what I decided to do for fun.
Here's the first test I ran. I could definitely discern an audible difference here; scoring a 9/10 has a p-value of less than 1% which is well within the realm of statistical significance. In order to pass this test, I focused on only one section of the song; specifically, the part where it explodes into loudness at about a minute in. I ignored pretty much everything else when I was doing the test.
Here's the third test I ran using MP3 320 vs. FLAC 24/48. I actually failed the second test using these same files, getting only 6/10 correct. Even getting 7/10 correct only corresponds to a p-value of 0.11 (using a straight binomial pdf calculation) which most would not deem statistically significant. In the future, I might be able to score higher (as I was suffering from listener fatigue at this point), but it stands that I was mostly acting on gut instinct for these comparisons.
Finally, after taking a break, here's another track I tried playing with. The files are 24-bit FLAC and MP3 320 respectively. Once upon a time - unblinded - I could have sworn I heard a noticeable difference between the two tracks. But as you can see, I might as well have been guessing at the beginning! Up until test #7, I actually got more wrong (way more LOL) than I did correct. Why is that? Simply put, my approach was wrong. I was trying to discern a difference in clarity using Taeyeon's vocals. Unfortunately, it's actually very difficult to pick up a difference in sheer note clarity, particularly in the midrange. Recognizing that I couldn't discern a difference using her vocals, I switched up my approach. You can see that my score steadily picks up starting from #8 - in fact, I get 7/8 correct after this point! So what changed? Well, like the Sawano Hiroyuki track, I started listening for loudness. An MP3 file's dynamic range is more compressed, so - listening carefully - it came off just a hint louder on the MP3 in the beginning.
The Verdict
You can see that the differences I used to pick up on differences between the files mainly came down to one thing: dynamic range. If you want to pass one of these tests, it's mostly going to come down to listening for stuff like noise floor, peak loudness, and reverb trails. Honestly, I don't think (the vast majority of) people can even tell the difference between MP3 320 and FLAC based off of sheer note clarity. I'm certainly not able to. And you know what's even crazier? These are cherry-picked songs, songs that I have listened to hundreds, if not thousands, of times. If I was taking a test like this under duress, or using tracks I was not familiar with, I highly doubt I would pass. For better or worse, the audible differences between lossy and lossless are teeny-tiny, to the extent of which they're pretty much negligible for most people. Don't even get me started on lossless vs. 24-bit or MQA, oh boy.
Now, am I going to stop using lossless files? Hell no. Just like anyone else, I'm a fat sucker for FOMO. I don't want to know that I'm getting something lesser if I could have better. And hey, if I do hear a difference sighted - even if it's just that juicy placebo - I still heard a difference, right? That said, I still wanted to put it out there that ABX tests like this are a great way of keeping things in check the next time you think you hear a "huge" difference. They're way more difficult to pass than you'd think, and even if many of us don't like them, they could probably help us save a ton of money.
Although I noticed one more thing. Its even harder with rock/metal tracks, than with orchestral tracks like Two steps from hell or Hiroyuki Sawano tracks to tell them apart.
The only difference I can hear between 320 and flac is high frequency stuff, like cymbals and staging/imaging. The difference is very minor, though. I just recently have Tidal another chance, but I'd rather sacrifice the 1% loss in sound quality in exchange for the superior curated content and suggestions.
Recently, Apple Music announced the release of lossless music which is big news for us audiophiles. I know I myself was pretty excited when the feature showed up on my iPhone a few days ago! But lossless music sucks up lots of space and bandwidth. And let's be honest: Can you actually tell the difference between lossy and lossless? Of course, it's easy to say you hear a difference between track file A & B when it's sighted. But placebo is very real. So what better way to find out for certain than taking on a few nasty ABX tests? You know, just the thing that would have most audiophiles tails tucked, running away? Well, that's exactly what I decided to do for fun.
Here's the first test I ran. I could definitely discern an audible difference here; scoring a 9/10 has a p-value of less than 1% which is well within the realm of statistical significance. In order to pass this test, I focused on only one section of the song; specifically, the part where it explodes into loudness at about a minute in. I ignored pretty much everything else when I was doing the test.
Here's the third test I ran using MP3 320 vs. FLAC 24/48. I actually failed the second test using these same files, getting only 6/10 correct. Even getting 7/10 correct only corresponds to a p-value of 0.11 (using a straight binomial pdf calculation) which most would not deem statistically significant. In the future, I might be able to score higher (as I was suffering from listener fatigue at this point), but it stands that I was mostly acting on gut instinct for these comparisons.
Finally, after taking a break, here's another track I tried playing with. The files are 24-bit FLAC and MP3 320 respectively. Once upon a time - unblinded - I could have sworn I heard a noticeable difference between the two tracks. But as you can see, I might as well have been guessing at the beginning! Up until test #7, I actually got more wrong (way more LOL) than I did correct. Why is that? Simply put, my approach was wrong. I was trying to discern a difference in clarity using Taeyeon's vocals. Unfortunately, it's actually very difficult to pick up a difference in sheer note clarity, particularly in the midrange. Recognizing that I couldn't discern a difference using her vocals, I switched up my approach. You can see that my score steadily picks up starting from #8 - in fact, I get 7/8 correct after this point! So what changed? Well, like the Sawano Hiroyuki track, I started listening for loudness. An MP3 file's dynamic range is more compressed, so - listening carefully - it came off just a hint louder on the MP3 in the beginning.
The Verdict
You can see that the differences I used to pick up on differences between the files mainly came down to one thing: dynamic range. If you want to pass one of these tests, it's mostly going to come down to listening for stuff like noise floor, peak loudness, and reverb trails. Honestly, I don't think (the vast majority of) people can even tell the difference between MP3 320 and FLAC based off of sheer note clarity. I'm certainly not able to. And you know what's even crazier? These are cherry-picked songs, songs that I have listened to hundreds, if not thousands, of times. If I was taking a test like this under duress, or using tracks I was not familiar with, I highly doubt I would pass. For better or worse, the audible differences between lossy and lossless are teeny-tiny, to the extent of which they're pretty much negligible for most people. Don't even get me started on lossless vs. 24-bit or MQA, oh boy.
Now, am I going to stop using lossless files? Hell no. Just like anyone else, I'm a fat sucker for FOMO. I don't want to know that I'm getting something lesser if I could have better. And hey, if I do hear a difference sighted - even if it's just that juicy placebo - I still heard a difference, right? That said, I still wanted to put it out there that ABX tests like this are a great way of keeping things in check the next time you think you hear a "huge" difference. They're way more difficult to pass than you'd think, and even if many of us don't like them, they could probably help us save a ton of money.
Agreed. I recently did a few shootouts with 320 and flac of various rips. Well recorded/ripped 320 i could not discern any difference. I always focus in on cymbals and drums. Was impossible. Super letdown as there are some flacs that sound way better than the mp3 but it really(for the most part) does come down to comparing the exact rips and producer as ive heard 5 different flavors from one band with certain rips focusing on mids, others on lows etc.
I don't think I would be able to notice if I was listening to a higher bitrate MP3 if I was not told. Even knowingly trying to tell the difference in an active AB test is difficult enough.
Recently, Apple Music announced the release of lossless music which is big news for us audiophiles. I know I myself was pretty excited when the feature showed up on my iPhone a few days ago! But lossless music sucks up lots of space and bandwidth. And let's be honest: Can you actually tell the difference between lossy and lossless? Of course, it's easy to say you hear a difference between track file A & B when it's sighted. But placebo is very real. So what better way to find out for certain than taking on a few nasty ABX tests? You know, just the thing that would have most audiophiles tails tucked, running away? Well, that's exactly what I decided to do for fun.
Here's the first test I ran. I could definitely discern an audible difference here; scoring a 9/10 has a p-value of less than 1% which is well within the realm of statistical significance. In order to pass this test, I focused on only one section of the song; specifically, the part where it explodes into loudness at about a minute in. I ignored pretty much everything else when I was doing the test.
Here's the third test I ran using MP3 320 vs. FLAC 24/48. I actually failed the second test using these same files, getting only 6/10 correct. Even getting 7/10 correct only corresponds to a p-value of 0.11 (using a straight binomial pdf calculation) which most would not deem statistically significant. In the future, I might be able to score higher (as I was suffering from listener fatigue at this point), but it stands that I was mostly acting on gut instinct for these comparisons.
Finally, after taking a break, here's another track I tried playing with. The files are 24-bit FLAC and MP3 320 respectively. Once upon a time - unblinded - I could have sworn I heard a noticeable difference between the two tracks. But as you can see, I might as well have been guessing at the beginning! Up until test #7, I actually got more wrong (way more LOL) than I did correct. Why is that? Simply put, my approach was wrong. I was trying to discern a difference in clarity using Taeyeon's vocals. Unfortunately, it's actually very difficult to pick up a difference in sheer note clarity, particularly in the midrange. Recognizing that I couldn't discern a difference using her vocals, I switched up my approach. You can see that my score steadily picks up starting from #8 - in fact, I get 7/8 correct after this point! So what changed? Well, like the Sawano Hiroyuki track, I started listening for loudness. An MP3 file's dynamic range is more compressed, so - listening carefully - it came off just a hint louder on the MP3 in the beginning.
The Verdict
You can see that the differences I used to pick up on differences between the files mainly came down to one thing: dynamic range. If you want to pass one of these tests, it's mostly going to come down to listening for stuff like noise floor, peak loudness, and reverb trails. Honestly, I don't think (the vast majority of) people can even tell the difference between MP3 320 and FLAC based off of sheer note clarity. I'm certainly not able to. And you know what's even crazier? These are cherry-picked songs, songs that I have listened to hundreds, if not thousands, of times. If I was taking a test like this under duress, or using tracks I was not familiar with, I highly doubt I would pass. For better or worse, the audible differences between lossy and lossless are teeny-tiny, to the extent of which they're pretty much negligible for most people. Don't even get me started on lossless vs. 24-bit or MQA, oh boy.
Now, am I going to stop using lossless files? Hell no. Just like anyone else, I'm a fat sucker for FOMO. I don't want to know that I'm getting something lesser if I could have better. And hey, if I do hear a difference sighted - even if it's just that juicy placebo - I still heard a difference, right? That said, I still wanted to put it out there that ABX tests like this are a great way of keeping things in check the next time you think you hear a "huge" difference. They're way more difficult to pass than you'd think, and even if many of us don't like them, they could probably help us save a ton of money.
I never conducted a proper blind test, but when I switched to a new laptop and downloaded Tidal I forgot to change the setting to HiFi/MQA, so I listened on 320 for a month without so much as realising... Ended up downgrading my subscription lol
Personally I want the best possible quality, as long as I can afford it and it doesn't impact anyone that depends on me negatively.
We spend so much time fine tuning our gear choices, eeking out the last few percentage points of performance, investing irrational amounts of money on copper wires, and then what, knowingly listen to music files with half the information missing because we don't trust our ears to hear the difference?
And it's not like 10 years ago when data was a premium and bandwidth was in short supply. We'll stream hi-def movies without any issues but balk at streaming lossless music because the file size is a little bigger?
There's absolutely nothing wrong with listening to MP3s and Bluetooth all day long. But personally I wouldn't bother investing a fraction of what I have on the best possible audio gear if that's the diet I'm going to feed it.
I did some testing with Apple Music , some Michael Jackson tracks on their typical AAC 256 codec vs high resolution lossless , I could tell the differences but they were tiny. Felt like the high res had slightly better textures around instruments and they seemed to have “more room to breathe” compared to the regular lossy codec.
It wasn’t something that’ll have me jumping ship from Spotify that’s for sure, especially considering their Hifi will be out eventually too.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
