could you post your recording on dropbox, soundcloud, or you tube? it's tough to make specific calls w/ out hearing them. Also these are well trained singers, who are alrtering there distance from the mic on purpose to keep consistence volume and tone. and they aren't kidding about acoustic treatment.

If you have some compressors/emulations, some of them exhibit exhibit a smoothing, or dulling quality, the summit audio tla-100 is the most extreme one i've used to dull bright mics and singers. mess with the threshold, and attack settings and you'll should hear most compressors start to dull the sound.

Also using eq, just crank the mid dot , or knob around as high as it goes, set the bandwidth (q) setting to medium/high, maybe around 8-12. Now take the frequency setting, and sweep (move back and forth between) the frequencies of around 1k-and 15k, and just listen for what the most obnoxious, harsh sound to you. It's gonna sound like ssssshhhhwwweeeeeewwwwuuuuuu. When you find the range you don't like, keep it there. Use the bandwidth to include/exclude what you like or not. then just turn that down to taste, maybe 6db -10db under the zero line.

what your doing w/ that is basically exaggerating the ugly (like zooming in on a zit), and turning the ugly part of the sound down. Just listen and set one control knob at a time as you tweak it to final, and a/b it w/ the raw track regularly, so you can keep being reminded what your 'fixing'. that way you don't make it worse.

really though trying different mics, and pres and stuff will improve you tenfold. i duno what ya have, so i'll guess, it's the nt into an interface. Vocal technique is more important than equipment. maybe put some quilts up around you, or a sock on the mic? find your best sounding distance.

Also, remember, they're pro singers, engineers, in a pro studio, and didn't offer comparison on processed/unprocessed vocal. So don't beat yourself up, do the best you can do and you'll keep getting better. Find some singers that have a similar voice to you, and import the songs into your session, and just compare your recording to there's when your getting sounds. Then during mixing you can work on enhancing your voice track, rather than correcting it.

it just could come down to there is a better mic out there for ya. if your a bright singer, you generally wouldn't want a bright mic. i look at mic selection as eq. you wanna tame the bad and bring out the good.

The quality of the pre, the room, the distance, the processing, the singer, all of those things have a dog in the fight.

And, no two people will sound the same on any given mic.

There are mics that you can use that will hedge your bets a little bit, generally the nicer the mic (and the more expensive) will help to translate well from vocalist to vocalist, but not in every situation.

I recall a session once with a female vocalist where we started her on a U87 - we weren't really diggin' the tones, so we moved her to a 414 - slightly better but still not what we were after, and in the end we ended up using an EV RE20, which for her, at least on that track, sounded great.

Now, not everyone has the luxury of having all those different mics to choose from, so in the end, you need to experiment with proximity, and use corrective EQ - and in most cases, a subtractive correction is best...

So, if you are hearing tones that are a bit harsh or brittle on the top end, attenuate those respective frequencies - a db at a time to start, don't go crazy and notch anything down by 4 or 5 db - and... work with your bandwidth (Q) to try to isolate the frequencies that are making it harsh. I can't give you an exact frequency because I don't know your voice, but generally, 1k to 4k can be the area of harshness on a vocal. Again, it's gonna differ from vocalist to vocalist, and mic to mic, and room to room, etc.

And, as K pointed out, you are trying to compare your sound with that of what is most certainly an ideal audio situation on the demo... a pro studio with a very nice pre, and a pro vocalist, etc., all these things come into play. You can't necessarily compare what your results are with the results that they got under what I'm sure were ideal circumstances.

fwiw
-d.

I want to thank both of you for your amazing replies!
I'm currently suffering from bad allergies and shortening of breath, but I picked a very simple song and recorded a little sample.
Here is the link:
[MEDIA=soundcloud]massimo480/obsession-sample

The little sample is broken into two little tracks, both recorded with a different loudness.
"Be still, I will not accept defeat" is the second track, and I'm sure it's noticeable that it gets quieter.

PS: As I just recorded this, I believe I've figured out my issue. I think I kept recording vocals with the gain set too high >.>
I always believed it's better to have it rather higher and lower.
I'd appreciate comments to that as well.
I usually sing things within tenor range, like We Are The Champions for instance, and I guess when you get louder with the vocals, it IS crucial to step back a bit?

THANK YOU GUYS

I always believed it's better to have it rather higher and lower.

when your working w/ tape you use higher levels 1. to get the maximum 'signal to noise ratio' which basically means minimizeing noise. 2. to use the tapes compression, or to intentionally overdrive it for an effect.

in digital, this is not good practice. noise isn't really an issue in unless there is something seriously wrong. it's better to err on the side of of lower, than louder. it's easy to make it louder after, and by the time you eq/comress, and everything else it's going to be louder anyway. You want plenty 'headroom' or room above your peak. this is there so you don't clip the signal and get nasty distortion, if you get a little overly enthusiastic during a part.

The most commonly agreed on level for tracking 24bit digital, is -18dbfs. which usually about halfway up the meter, this number translates to 0db on a tape machines vu meter. and means your leaving 18db of headroom before nasty clipping occurs. if the average level stays around 18dbfs, you'll have nice clear clean full recording w/ no nasty artifacts, and you won't be pushing the master bus into clipping as you stack tracks. providing you do everything else right.