First I have to say, that I think, this site is much better then some expensive ear-training-programmes.

What could be even better in my eyes is:

In the Inversion of 4-note-chords, you should absoloutly start only with dom7-chords, because only the dom7-inversions are nearly always part of ear-training-test at the music-conservatories.

The progression of difficulty in intervals depends not that much in the ambitus of intervals, but more in the similarity of consonant/dissonant. So its easier make out a seventh then to distinguish a fifth from a fourth.

At the Melody and the Rhytms1 tests you could look more, what's really difficult and what's really easy. Sometimes its very easy to check the right answer and sometimes its very hard. I think the wrong answers are created automatically. Would be better if could do it yourselve.
In the other tests the progression of difficulty is much better.

In the melodies some Melodies with 2-4 parts would be great (also rhythms with more parts, but that needs different sounds for each part)

The metronome is too loud.

Good training would be also too memorize a melody (or a rhytm) and you see the options to chose not until the melody ended (has to be repeatable 3-4times)

Some tests, esp. the triads and 4note-chords sound too low, then its too hard to distinguish.

Would be nice, if you can chose another sound, e.g. piano and some standing sounds.

Would be nice, when the alltime earplane list lists more then only 25, but 1000 (or at least 100). This list tells more about how good you are, then the weekly competitions and the earlympics - it would be very easy but very boring to earn every week 1000 points with intervals)

But nevertheless, this site is already very good! So far.
Lorenz

and a last thing (little bug):

If you answered the last question of a round wrong, you dont get to see, how much points you did in that round, because there comes always "Your round has been logged"

Thank you! That's quite a collection of good points, and I agree in most of them.

It's almost a little sad that I consider the tests here final. I haven't touched the source code in years, and I'm not even sure where I left it. I'm rewriting it from scratch in java, and the offline version will eventually end up online (but that won't happen for some time...).

But I'll save this posting and have it in mind as the work progress. Hopefully the offline version will be updated twice a month, but I've still got some startup problems to solve.

Best,
Knut

One thing more came to my mind (I'm a computer fool, so I have no idea about how difficult it is to realize):

As a good musician you have to make out musical phenomenons no matter what instrument is playing. People are often specialized to hear very good from the piano or from the guitar but they perceive nearly nothing, when they hear a choir for example.

[Do you know the "Vienna Sound Library (VSL)". They recorded about 10million (!!) orchestral sounds and all note-conjunctions(!!) (nearly all orchestral cinema music is produces with VSL since it exists). The student version of VSL is about 500Euro, complete version about 10000Euro.]

It would be really really great to hear all the tests with these original sounds. Especially the intonation tests would be so helpful. How often do you hear a group sounding badly? How great would it be, if you can say: "Second clarinet was about 10ct too low"? Conductors waste so much time with that issue, which could be practised so easily.

I would spend at least 200Euro (when I earn more accordingly more) for a program, where I get wrong pitched wind instrument chords and where I can try out to tune every single note of the chord up and down and to see how I did it (but that needs good knowledge about intonation: how exactly pitch does change in special chord-progressions)

Please let me know, if you are willing and have time to improve (some parts of) the tests of the program. I'd like to be some kind of assistant in developing it, if you dont mind

Cheers
Lorenz

Hi Lorenzo,

I think your hopes for this site may be a little too optimistic. My goals for writing the tests from scratch is to get my java chops going, so that I'm better suited for rewriting another application that I wrote in the nineties (but it died with Windows2000).

The sounds in the new version will be midi-based (you can enjoy the piano in the downloadable version), and there may be some other options, but I don't think you can expect any sampled sounds in the foreseeable future.

But then again, I don't know what may seem easy some six months down the road...

Hi,

I also posted this question up on http://www.ear-trainer.com

I'm having trouble distinguishing between Perfect 4th & Perfect 5th intervals. My guitar teacher says it's something to do with inversions. What's up with that? The guys at ear trainer haven't replied yet.

Mark

I'm not sure if this answers your question, but if you go from C up to F then the interval is a perfect fourth. If you go from C down to F it's a perfect fifth. The note names are the same, but the distance is different.

hth,
Knut