Hi all ! I recently purchased a zoom H4 recorder which enabled me to fulfill an old project : to record a sample tune of each of my (playable) whistles, in the same conditions , without reverb or any sound treatment. The purpose is to give an idea as close as possible of the sound of these whistles as the guy sitting near to you ( typically a fiddle or box player ) can hear them. I also didn’t change the gain settings, so the differences in loudness can be clearly heard. I recorded 25 whistles, maily tradionnal budget ones ( Gens, Oak, Feadog, etc. ..) as I only own 3 " high end " whistles so far. And nothing lower than A for the moment, sorry.
The tune i’ve chosen is an air from Chieftains 4 called The tip of the whistle, for several reasons : I like it ( well, used to … ) , it goes from low D to high B on a D whistle, it’s got slow octave jumps, it contains some natural C’s.
Flash must be enabled on your browser. Click the small images to get more information about each whistle. Please use headphones as you’ll notice that differences are often subtile. Please let me know what you think, what are your preferred ones, and I’ll be glad to tell you about my personal preferences.
Of course, i’ll update regularly, as WHOA never stops …
A great idea and obviously a lot of work put in by yourself but I have spotted one blindingly obvious caveat about your recording procedure you have used a good recording device but recorded in 128Kb mp3 a very lossy compressed audio format then when you uploaded those separate mp3 whistle audio files, SoundCloud by default re-encoded (compressed) again to 128Kb mp3, joint stereo.
With lots of these whistles being so subtlety different I would have advised you use a lossless audio format such as .wav to retain all the frequency and timbrel data in your samples. Your zoom device could have been setup to record in an uncompressed format PCM WAV, 24 bit, 96 kHz. To allow the listeners to hear the pure *.wav files all you need to do is upload the *.wav files and enable the download audio file command. This does mean more interaction but anyone interested in these samples and understands audio will want to listen to uncompressed audio samples and will download.
Anyways thanks for all the effort and maybe you will consider recreating the test again? and one other small point an ascending ~ descending scale added to each whistle sample would be great comparison aid also.
@ Christophe well done, thats nice for comparing - thanks for sharing
when i bought my first whistles i always was looking for something like this because the few files i found in online shops where edited very much
i once did something similar with just 4 or 5 whistles but my playing those days … i should do it again with my collection … well my playing these days …
I’m interested in the samples. I understand audio. And I have no desire to listen to huge samples at double the standard CD audio quality. I think Christophe’s clips are perfectly fine for the intended purpose.
If you where truly interested you would want to be beside him while he played the whistles but this is we will agree is impossible so you would prefer to hear lossy mp3. Rather than the actual audio you are not making sense… I listened to the audio sample and they pretty much all sound the same so all these whistles sound the same.
The Clarke Sweetone sample does not sound like my Sweetone the Dixon Trad doesn’t sound like my Dixon Trad. What could be the difference? so your sure MT that that difference I’m hearing has nothing to do with the quality of the recordings.
The fact that you believe Mp3 128kb is CD quality shows with respect that you are out of touch with current recording and compression technology. If you had said 192Kb ogg, wma or aac you would have been closer. Mp3 is the least developed compression codec available today, falling behind advancements made by ogg, aac and wma devs. So I’m offended by seemingly scathing remark that these recordings are fine in light of the advice I offered, fine for you ok but not for me and I offered advice that I believed would help the op. I certainly didn’t expect to get my advice trodden on by someone who should know better. Your back ground is in audio production appears to be on par with your ability to be diplomatic.
When listening to a learner whistler playing a tune you are not necessarily looking to hear how beautiful or terrible the actual whistle sounds but rather the ability of the player to play the tune so yes small bit rates suffice. When you want to and need to hear every nuance of whistles and musical instruments like in this whistle comparison scenario I would never use a lossy low bit rate audio compression codec and specifically not mp3.
thanks Peter
i will read that when i have more time, i am very interested in that topic
its a pitty, we wouldnt have got your links without Jleo, and he is missing here now
Jleo, I don’t doubt you can hear a difference between my samples and what you hear when playing your own dixon and sweetone. But I doubt the difference lies in audio compression only. First, your dixon and mine ARE different . Second, the sound you hear when playing IS definitly different fom what one can hear when sitting 70 cm away from you. And finally, the sound of the various traditionnal brands ( gen, feadog, walton, oak, etc. …) is really close because the design of these whistles is quite similar. So it’s normal that you should hear only small difference between these sound samples.
Once again, I will not re-record the whole stuff in wav, but I can send a few files in wav to anyone who’s interested in a few specific brands of TW.
And I will not go into a fiering debate on ogg vs mp3 vs wma vs lossless, although I did some personal tests and have my personal point of view about it.
It seems one of my replies got lost in some internet blackhole, so here it is again :
I started recording in wav but soundcloud would anyway compress the files in 128 kbs, wich is a low bitrate i have to admit, but is compatible with audio streaming with an ordinary connexion speed. So I decided to record directly in 128 kbs to save a step in the process. i have some wav samples of all my whistles and can send one to anyone who’s interested . but I dont intend to re-record all of them. I have another smaller prjoect : effect of the tube with the same mouthpiece, an experimental approach of the old " nickel vs brass " debate.
