Others have more or less explained everything, but I'll sum it up for you.
For all intents and purposes, the frequency response of headphones, the specs of a DAC, the
sampling rate, sample size (
bit depth), and
bit rate of digital audio files, and the quality of an album's production process (recording, mixing, mastering, etc.) have little or nothing to do with each other.
The average
human hearing range only extends from 20 to 20,000 hertz. Most people literally can't hear anything above or below that. (Though you can feel bass frequencies below 20 Hz.)
Headphones that mention "high res" or "high definition" are referring to their resolving capability, that is, "the ability of a component to reveal the subtle information that is fundamental to high fidelity sound." Its auxiliary ability to emit sounds that you can't hear (which don't exist in all but the most exotic recordings anyway) should be the least of your concerns.
What you
should care about is the frequency response in a different sense: the relative performance of the bass, mids, and treble. You'll have to decide for yourself (through research and experience) which sound signature(s) you find most preferable. (Refer
here for an introduction to audiophile terminology.)
DACs are
digital-to-analog converters and convert the digital data into an analog signal. While you need a
transducer of some sort (be it headphones or speakers) to convert the signal into sound waves to listen to, most headphones do not have built-in DACs, per se. DAPs (portable digital audio players, such as iPods), smartphones, and computers have built-in DACs, but they're not always very good. Generally, you should use an external DAC (or amp/DAC combo) with mid-fi and high-end headphones if you want better sound quality.
It should be noted that just because a DAC "supports" 24-bit / 96 kHz and happens to sound better, that isn't the
reason it sounds better. A lot goes into electronic design...which is very technical and far beyond the scope of this thread. I'm not yet convinced that it's such a simple matter for a DAC to "fully decode" audio. Some hardcore audiophiles spend tens of thousands of dollars on high-end DACs because they believe they do a superior job and sound better. Others insist that all "properly designed" DACs sound identical. There is much debate on this topic, and I don't intend to start another one; just mentioning this, for your information.
The
Rockbox firmware enables FLAC playback on compatible Apple devices. It even improves the sound quality, at least on my iPod classic! (On the stock firmware, it sounded too bland and boring without EQ. I used the Electronic EQ setting to juice things up. With Rockbox, I don't use EQ, and it sounds better than I thought the iPod was capable of on its own!)
The reason some "HD" downloads sound different or better than their CD counterparts is because they are derived from a different master. Concurrently, the differences you heard between 24-bit and MP3 were primarily due to the fact that you were comparing two masters of the recording, or possibly even two recordings altogether. (Same goes for surround sound mixes.)
If the download is a different master, it may be worth the premium. Problem is, they usually don't tell you about all that, and don't offer refunds either, so tread cautiously to avoid wasting your money.
If you convert those 24-bit / 96 kHz (or whatever) files to 16-bit / 44.1 kHz (such as lossless CD quality or 256 kbps lossy AAC) with a program like
dBpoweramp, you shouldn't be able to hear any difference at all. The extra information (if any) in the higher resolution files is outside your hearing range, after all.
Again, the sample rate and bit rate of your files do not relate to how well the album was mastered. Modern albums are typically mastered at 24-bit, 32-bit, 64-bit, or higher. This has more to do with the intricacies of real-time computing than audio quality. Even if the most advanced technology is used to produce an album, that still doesn't guarantee it will sound good. Moreover, most recordings are
heavily compressed in their dynamic range, meaning that you have all the less reason to worry about it. After mastering, all sorts of things are done to the data extracted from the master files (or tapes and so on) before they are released to the public. More often than not, CDs are mastered differently than the studio master.
Don't be fooled by numbers. As long as your files are 16-bit / 44.1 kHz and at a sufficient bit rate, you're fine for most situations. (
DSD, on the other hand, is more complex.) If you would like to learn about why 44.1 kHz is used instead of 20 kHz, you can start with
this article.
Various studies have shown that it is unlikely or even impossible to distinguish between lossless and 256 kbps AAC. (MP3 is more prone to
compression artifacts at lower bit rates.) To save space, you should just convert your lossless music collection to AAC in order to fit more of it into portable devices. For use with a desktop audio system, you can leave everything as is, as long as you have enough hard drive space. In any case, it's a good idea to always keep lossless backups of your music.
I hate to break it to you, but if you want to get the most out of
any audio file, you'll have to be prepared to invest far more than a few hundred dollars. Most consumers aren't really looking for the best they can possibly get, but just "good enough" for their budget, which is fine. (
Diminishing returns come into play and escalate the higher you ascend the audio ladder.)
I recommend the Focal Spirit Professional if you can't spend more than $400.
Welcome to Head-Fi, and I wish you luck.