What's the best way forward? to a better recording sound?

[Stuie]

Ive got a M-audio Blackbox, going through a 16bit soundcard.sample rate of 44mhz(i think) Sequencer im using is the Live Lite 5.2. My p.c specs is 1.3ghz processor,512mb ram memory, 40gigb HD,. I'm wanting to improve the sound quality., to make a more polished proffesional sound. Should I upgrade to a 24bit or 32bit sound card? or up the sample rate on the 16bit card to make the difference? Sorry I'm new to Computer Recording, and its taken me 3 days to get my music equipment working, I'm not too impressed with the sound I'm getting at the moment. I write mostly Heavy Metal tunes and want that heavy in your face sound. Any Ideas for me that would be grand! Stu_C.

[M. Bréqs]

Ive got a M-audio Blackbox, going through a 16bit soundcard.sample rate of 44mhz(i think) Sequencer im using is the Live Lite 5.2. My p.c specs is 1.3ghz processor,512mb ram memory, 40gigb HD,. I'm wanting to improve the sound quality., to make a more polished proffesional sound. Should I upgrade to a 24bit or 32bit sound card? or up the sample rate on the 16bit card to make the difference? Sorry I'm new to Computer Recording, and its taken me 3 days to get my music equipment working, I'm not too impressed with the sound I'm getting at the moment. I write mostly Heavy Metal tunes and want that heavy in your face sound. Any Ideas for me that would be grand! Stu_C.

Well, first off, 24 bit will certainly improve your sound. It's going to cost you +50% more computing power though, so if your system is bouncing at the top of the CPU meter, you'll be in for a surprise when you switch.

Secondly, 32 bit will not help much more than 24 bit in my opinion. 32 bit sound cards are friggin expensive, as 24 bit is pretty much the regular for semi-pro and serious hobbyist cards these days. Also, I don't know if Live even supports 32 bit!

For sample rate, it's a trade off. There's advantages to upping your sample rate, but it comes at a much greater CPU power cost than going to 24 bits. Unless you have a monster computer and an above average range budget, 24 bit 44.1 KHz is your best bet. It will sound friggin sweet to everybody except perhaps the snottiest engineers and afficianados, and even then only in high quality listening environments.

Moving from 44.1 to 48 KHz won't make an appreciable difference to your sound really, and some people say it can cause some artifacts if you want to burn standard CDs. Thus, the next options are 88.2 KHz and 96 KHz.

moving to these higher sample rates will increase your maximum frequency, but will have virtually no impact on your bass or midrange. It's only going to be heard on your cymbals, air, or some synthesized sounds.

Also, in terms of CPU power, you're more than DOUBLING your computing requirements when you jump to 96 KHz. However, the other advantage comes in latency. A higher sample rate means a faster response. If you're finding that you're having problems playing virtual instruments live because of lag time, then it might be worth considering. But, 44.1 at a buffer setting of a very conservative 256 samples will be unnoticeable to all but Jerry Lee Lewis on the keys, providing you are using solid ASIO drivers. There's more latency in the time it takes your finger to travel from the top of a piano key to the contact point in all likliehood. At 128 samples, nobody will notice any delay without loading your tracks into a sample editor.

All in all, 96 KHz is not worth it in my book, but that's just me.

How many ins and outs do you need for your audio card? There are some good really compact solutions on firewire and particularly PCMCIA, but if you need a bajillion ins and outs you're gonna pay.

Also, I hate to say it, but you might want to go find a producer you respect and get them to actually listen to what you have now. I don't know how your mixing skills are, but I have met a lot of people on the gear quest who never realized that the problem was their eqing and levels, not their interface. I'm not saying it's your case, I'm just saying that it happens to people sometimes, and in fact it happened to me before I went and got a skilled producer to critique my mixing.

Depending on

[Stuie]

M. Breq's, that was awesome thanks alot. You've answered my questions, cheers.... I'm gonna get a 24 bit soundcard. Thanks again! Stu_C.

[M. Bréqs]

Oh, and read this thread too for a primer on the advantages of 24 bit;

http://www.ableton.com/forum/viewtopic.php?t=39518

[Stuie]

Thanks for the threads.
I've just been looking at the preferences in options on my Live Lite 5.2 and its set at 24bit depth? automatically. I've been recording in it and never even realized it!...but I don't think my soundcard is 24bit, its 16bit I believe? I'm new to computer recording..but what would happen if it was recording in the wrong depth? The soundcard is built in its a VIA AC'97 Audio Controller (WDM)? I couldn't find anything on the web that could tell me what bit rate it is.
I've been getting a crackling noise, maybe thats the problem, the fact its set wrong... I don't know?

[j0shu@]

while the bit depth is a factor there are many other factors involved as well.

if you are doing heavy metal you are recording guitars and drums by miking them correct?

in that case i think a good mic, pre-amp, and compressor on the front end will make way more difference than the bit settings.

i dont know about the bit rate being different than the audio card.

the crackling may be due to latency since you are using onboard sound. try easing up on the buffers.

***edit***

actually i now see the whole m-audio blackbox thing. how is its sound on front end? it should sound about the same comig back out. IMO emulators never nail distortion quite right.

[Stuie]

Joshu, the USmetal sound on the Blackbox is pretty good, sounds like a dual rectifer sound, but not as good as the dual rectifer sound on my Flextone 3 Line 6 digital amplifier.
I'll try easing up on the buffers as you have said, and was reading the thread that was talking about a good mic, pre amp etc..I'm learning things everyday thanks for the advice. Stu_C.

[Michael-SW]

Well, first off, 24 bit will certainly improve your sound. It's going to cost you +50% more computing power though, so if your system is bouncing at the top of the CPU meter, you'll be in for a surprise when you switch.

24 bits shouldn't cost your 50% more CPU since Live uses 32 bit floating point internally. All audio is converted to that. Converting a 24 bit sample to 32 bit FP will be slightly more expensive CPU-wise but not much. After the conversion, Live will handle all audio equal, regardless of their original bit depth, so it shouldn't matter at all for effects, mixing etc.

It will be 50% more expensive in regards to your disc access though.

[M. Bréqs]

Well, first off, 24 bit will certainly improve your sound. It's going to cost you +50% more computing power though, so if your system is bouncing at the top of the CPU meter, you'll be in for a surprise when you switch.

24 bits shouldn't cost your 50% more CPU since Live uses 32 bit floating point internally. All audio is converted to that. Converting a 24 bit sample to 32 bit FP will be slightly more expensive CPU-wise but not much. After the conversion, Live will handle all audio equal, regardless of their original bit depth, so it shouldn't matter at all for effects, mixing etc.

It will be 50% more expensive in regards to your disc access though.

Thanks for the correction!

In fact, since the computer wouldn't have to dither down to a 16 bit output, it might save a teensy weensy percentage?

[jerry123]

I want to say something about monitors, mics, instruments, the rooms you are recording in...
I think that bit rates are great to play with, interfacing is very important and software settings need to be correct, but for a simply better recording, there are things you can improve on that will have a greater overall affect on what you final product turns out like.
If you can give a rundown of your environment and process, you may get some hints that will get you further faster.

[Michael-SW]

In fact, since the computer wouldn't have to dither down to a 16 bit output, it might save a teensy weensy percentage?

I'm not even sure Live dithers. There isn't anything about it in the manual AFAIK?

Shuffling 24 bits over the internal busses to the soundcard instead of 16 will be slightly more expensive though so lets call it even.

[Ribo]

moving to these higher sample rates will increase your maximum frequency, but will have virtually no impact on your bass or midrange. It's only going to be heard on your cymbals, air, or some synthesized sounds.

that's not true. while a higher sampling frequency allows higher frequencies to be stored, it also enhances the rest of the spectrum. just take a 100 Hz sine wave and sample it with 1 kHz and 10 kHz for instance and u'll see the difference. the more sampling points per time frame, the better the quality. same goes for the bit depths.

[dj superflat]

i haven't used the mbox, so take this with a grain of salt (though i have used the adrenalin and guitar rig), but if you're trying to record guitar based music, the best way to make it sound good is to mic a really good amp. i think emulators are fine when used in a mix where guitar doesn't dominate or doesn't need to sound like a real guitar. otherwise, micing a great old amp, even if recorded on a mediocore 8 track or somesuch, sounds better than an emulator. there are also some pedals (mesa makes one) that are effectively amps, they have tubes in them, can sound great going direct.

[The Phat Conductor]

don't ever cross zero dB on a digital eq.