AT3035 Sounds Muddy? Recommend an Alternative?

W

Words

New member
Hey guys. So after a great suggestion from this forum I got a nice, new, CLEAR audio interface (Ionix u22), thank god. My problem now is that I've suddenly noticed how UNclear my AT3035 mic is.

Here's a basic dry recording: http://tinyurl.com/2atodkb

Basically, the sounds of n's and m's are indistinguishable, like my breath is hitting the mic causing (what I think is called) the proximity effect...except it happens whether I'm facing the microphone or not even near it, so it's not breath. Also some of the sibilance can be rough. Is this the opposite of what people refer to as warmth? This is the only microphone I've ever used, so I have nothing to compare it to. Naturally I can dramatically improve the recordings of this mic by adding high-end to the EQ, but even when I drop out most of the lows and mids, it still has this breath/proximity effect muddiness.

Or at least, I think it's the microphone. Worrying that it might be bad acoustics coloring it, I even climbed through the window out onto the roof, mic stand and all, and tried recording out there, lol. No dice. Same problem. I've listened to a few clips from other people using the at3035 and it sounds guilty.

Two questions here:

1. It IS the microphone...isn't it?
2. And if so, what am I looking for in a replacement?

According to the AT site, it says this about the AT3035: "Extended low-frequency response for rich, full sound."

I have no idea how to quantify that, or if it actually means anything at all. I liked some things I heard with what I think was the Blue Mouse, but according to the Blue site, the Mouse also emphasizes the low end for bass and baritones. Is this what's causing my problem in the first place? Do I want a mic that does the opposite? Or does it not even matter if they do the same thing considering the sole fact that the AT3035 is a $150 mic and the Blue Mouse is a $950 one?

I know that I'm not going to get my answer for the perfect aural tool by just reading text and that I have to actually try some mics and see what sounds good, but I'm hoping to get a good idea of where to start.

If anybody has any recordings with the AT3035, please show me so I can compare. It's not that this is a bad microphone, but I need a complement. I do like how it sounds up close for whisper-like stuff, but I'm looking for something that works better for grand, soaring vocals. But aren't we all.
 
- Try a HPF to lower the lows.
- Use a shock mount, it should have come with one.
- Use a pop filter (not included)
- Isolate the mic from odd reflections (relocate away from the wall)
- That is what you sound like. Changing mics wont change that.

I have an AT3035 and it's a bit too bright and brittle for my tastes. But a good mic. You should be within 12" of the mic. Which will likely require a pop filter. Get too close and you get more low end in the sound. Get to far and it sounds dark and muddy. Sounds like you're just too far from the mic IMO.

That's an odd site. Encodes the WAV as a hex dump in the html. Easily edited out, but still if you wanted help (analyze your sample), not the way to go. Anyway, after converting the hexdump back to a binary...

Source dump (trimmed of wrapping code) = 10,675,974 bytes
Resulting binary = 1,788,544 bytes

Basically 11MB to transmit 2MB of data. So much for going green.

It looks like you have an unexpected dip in EQ near 5kHz. In audacity Effect -> 181 -> Multiband EQ do a +1dB bump at 5kHz and +0.5dB on the two neighboring bands. Then amp up 12dB so you can even hear the track. And you might notice your M issue isn't so bad. The default 440Hz HPF one pole seems to help with clarity as well. Depending on IF / WHAT you intend to mix it with.

Lessons on Diction would probably help too. Sustain on vowels, not the closed mouth consonants. Or I could be wrong.
 
In case you were curious, the edited version. (for a limited time)



As far as my brittle comment, that's with high SPL sources. Like trombone, trumpet, ..... At talking volumes, it's a really nice mic. And you do want to back off a little if you're going to be belting things out. i.e. further from the mic.
 
Shadow, thanks so much for the insight. (And sorry about the .wav file; I didn't get the problem you described so I wasn't even aware.)

The 5k dip you mentioned is very interesting. This is basically what you're talking about, right?

eq.jpg


I'd looked at that graph but wasn't sure what to make of it. I think you're right and that this has more to do with eq than I thought (which is probably good).

At this point I've tried all of your suggestions and adjusting every other variable I could imagine, and I've learned a lot just in the past few days. Believe me, I don't want to blame the mic, but since we're talking about it, which others do you like for vocals?
 
I'm a trombone player.

Vocals get weird because you don't want a what you hear is what you get type mic in a lot of cases. Vocalists don't want an honest mic, they want a flattering mic. If I'm looking for a mic, vocal samples are the least informative about a mics abilities, unless you somehow know what said vocalist sounds like acoustically in person. Not that instruments don't vary to noticeable degrees, but vocalists are much more diverse.

For that sample. I noticed that you were male. And therefor assumed that you should have content < 6kHz (Hz of ladies and screaming babies). But when I looked at the graph, you were peaking above 6kHz. This could just be your natural voice and nothing to do with the mic. I just felt that it needed more man-ness in it, so I bumped 5kHz and friends. Just one of those simple stupid things. And since the source was so high specific I dropped below 440Hz (HPF) which is where most noise (mud) resides.

This is what I was seeing in audacity / Analyze -> Plot spectrum:

ss_3035.gif
 
You must log in or register to reply here.
Back
Top