Resampling and converting question, sound quality

[dbijoux]

Ok, ok, no more self-deprecation. The world sucks, that's why we have music, right? Plus, you're ahead of many people out here and obviously have focused on the performance side of things. Nothing wrong with that, in fact, congratulations.

It's not so much that digital-to-digital is the issue. Without knowing exactly what the problem is, quite possibly your computer, being a weak link, could be a part of the problem.

IMO, it's that you're doing all the production on the Fantom. Not to say that isn't possible, but there are a lot of things to consider and that is probably the least intuitive way.

If you are serious about a home studio, you should start looking at recording gear. Maybe you could hock a board for a Motu. With all your gear, you might be more satisfied with an interface with multiple inputs. A cheaper solution would be to buy a stereo input interface and use your mixer. There are quite a few right around $200 and all are pretty good. Anything will be better than your onboard soundcard.

From there, monitors would absolutely help too. Even if you know your Bose well enough to translate, Bose speakers are notoriously colored. I really bring this up because most likely your problem is somewhere in the mix and the best way to figure out where is to use a flat monitor.

As for a stand alone CD-Recorder, other than being somewhat convenient, I don't think it would help.

It really might be a pain in the arse. Every step of the way, you'll need more gear, not to mention acoustics. Also, every step you get closer to 'quality' the price will increase exponentially.

Not knowing your situation, maybe an option for you is booking some time in someone elses studio. It might not be as rewarding, but it'd certainly be quicker and you can see first hand what it might take.

[Nik_]

How do you get your mix from the fantom into your computer?

[buminbeer2]

DBIJOUX, thanks for all your comments. You are correct that all the mixing is done in the board, I just hate computers (I know the synth's are just that). You are also correct that I'm a composer and in a perfect world I would just play and record what I hear. I do have much gear, but no recording gear sadly and that's why I bought this board, thinking I could do it all onboard including audio.

I just resample the song(s) and import them into the cpu as a .Aiff file or whatever it is through USB. From there I was hoping to just convert it over to say a .Wav and drag/drop into the media player or whatever and burn. Just not too happy about the sound I hear and not worth burning for demos for others to hear.

While I really did not want to buy recording gear and I take music serious, sounds like my only choice. I wonder if a cheap recorder like the Boss BR-600 would work, but then I'd still have to burn to CD on the computer and guessing the BR would color the sound as well. I make decent money, but if I'm going to have to spend a few thousand, then it could take a few months to save. BY THE WAY, I have a Mackie 1202-VLX mixer to plug my guitar/keyboards all in, so I don't really need a mixer.

With all that said, it would be nice to use my other keyboards as setting up preformances from say the JV-80/880 and trying to midi in the X6 and sequence is a pain. I'd rather just play live and delete the track from my many keyboards if I don't like it and could skip the sequencer altogether anyway.
I just use the sequencer to keep from using the computer and software that I hate, plus I currently don't even have a good way to get the sound to the computer to start with as it's a cheap built in soundcard.

Help me out with some of the setups you guys use. I don't live anywhere near a recording studio (been in them in the past). We have more cows then people where I live. So I know I need a decent set of monitors to start with and then suggestions from there. I'm not the smartest guy in the world, I was just born with the gift to play music without much effort, so don't get too deep on me
Thanks all, yet again!!!!

[emenelton]

buminbeer2

While everything people have mentioned to you is correct, the irony of your situation is this - you want to be able to produce tracks that sound good on your mobos on board sound output into your computer speakers. It is a lot trickier than it seems.

First , I would make sure that you are not using the eq on your mackie for fantom playback. Are you?
Then I would run the track in the fantom and watch the meters. Make sure they are peaking right around -1. Also make sure the pan controls on your mackie are hard panned left and right for both the computer and the fantom channels.

When you play your fantom mix in your computer make sure that you have the playback hot enough in Audacity that the master meters there also achieve -1 or 0 vu without hearing distortion. Those controls in conjunction with the gain controls on your mackie should allow you to play back both the fantom and the same track on your computer at the same volume.

[buminbeer2]

Emenelton - Keep in mind at this point it is 100% nothing but X6 into Bose speakers for what I hear. The Mackie mixer is a what if for the other keyboards. So, my X6 goes audio out to the Bose speakers. I record through the sequencer and only hear the sound direct. My computer also plays back through the same speakers.

Now, interesting note on the second half. When the fantom mix goes into the computer (as a resample I assume since that is all I do), how do I keep the mix hot through Audacity? The only option I can find is to "normalize" at -3 db and it sounds "louder" but still lose a lot of what I hear straight to the speakers to start with. I'm trying to use the onboard X6 for the whole process and only comparing my playback straight, versus when it is imported USB by resampling. Volume is NOT the issue (for the most part), losing sound is.

ALL, I am looking for the best answer to it sounding like what I hear and recording it the same way I hear it to burn for demo reasons. I do now understand that my system with the Bose speakers will "color" the sound. Fine I can live with that. I just need to record to CD what I hear without the cheap computer converting the sound to mud after resampling from the X6. Using Audacity to convert to .wav is very OK at best, MP3 sucks. I need studio quality stuff for the sound tracks.

So, if I need to run as hot as possible in the X6 without peaking, that is what I need to know and how to do that. I just hit record and do say 6-16 tracks and don't know how to master in the keyboard (if you can?) If that will give me a better sound, I'm game. Otherwise, I don't want to spare any expense in my sound.

I'm really hoping this is not too hard to try to do or understand. If I could record the exact sound I hear though the speakers, then I have gold

[lordelix]

So now we come to the crux of the matter!! I have said this in the past and will say it again. There is no keyboard in the world with the exception of a fully tricked out Open Labs Neko, that can produce a truely studio quality ready to burn project. Hardware workstation keyboards are great as scratch pads and sound sources. To get the awsome recorded resaults of a studio you need something close to a studio.

Alot of time and work goes into producing professional audio tracks and one of the essential things needed is track seperation for applying things such as compression and EQ. One of the main problems with resampling the sound as you are doing is that you loose all ability to change the levels or seperate parts for individule processing. The fantom does not have the ability to process all the tracks indiviualy with the needed effects for proper mastering.

Computer recording software and capture hardware is the answer to your issue. Even a stand alone multitrack recorder ( if you dont like computer mixing ) will give you far better resaults than going thru it the way you are now. At least that way you could use the Fantoms Effects on each track one by one. Those Bose speakers that have become such a hot topic are a problem also. If you cant hear the true sound of your mix then its gonna be way off on anything else but your system.

So the answer here is to get yourself some good studio or nearfield monitors ( as flat a sound as possible ). Get either an audio capture device as I had sugested or better ( plus software..... which I think you have a basic one ). Or get a stand alone hardware multitrack recorder.

[dbijoux]

Quite welcome, thanks also for the question(s). I think it's one more people than you have.

I can't speak specifically to the process you need to use on your Fantom, but I can offer an analogy. Think of all your recording gear as you might a microphone. While you would think it's recording exactly what you hear, in practice, that is far from the case. It's the same with all your gear, including the fantom, it's outputs, even resampling; each will add or take away something from the original sound.

Essentially, the quest we all share is how to get the music we hear or play into some medium. If it were easy everyone would be a recording artist and while it's getting easier everyday, to be really successful one still needs to understand what's going on.

I don't have all the answers, not even close, but I do know this, there is no magic bullet, unfortunate or not. Each case, each track, each sound is going to be handled differently.

One thing you might have noticed here is, everyone thinks you are recording with Audacity. The reason that is, is because most every production today is done on a computer. No doubt, it's a catch-22, but you really should consider trying Sonar, Cubase, or Pro Tools.

You could seriously be impressed, by just DI'ing into an interface and recording the stereo output from the Fantom. Additionally, with a DAW you could track and mix everything there.

If you're absolutely set on 'outboard' gear, traditionally, all this would be done on ADAT. You could go with the cheaper BOSS unit, but it seriously lacks in sample rate @ 44.1k/16 bit. For example, if you're aspirations include film, this wouldn't suffice. It's also only 8 tracks.

I have to digress and get back to your original question, resampling. If you take a look in the Juno-G forum, you'll see you're not alone in your concern over sound quality. There is even an answer to your woes.

However, in a complete production the problem actually is probably deeper. I go back to the mix again. Since you said, the mp3 was even worse, I have to think there are some real issues with your mixed down track. You see, there is only so much sonic space available. Each instrument needs to sit nicely in it's frequency range, like an orchestra. When things don't play nicely together, for a variety of reasons, they get even worse when recorded.

A recent example posted here concerned a mix with a runaway snare drum. As I recall, just bringing that down fixed a number of problems.

You have to understand, in simple terms, normalizing takes a signal and amplifies it to a point just before it clips or hits -0db. The problem is, it's not picky, it looks for the loudest peak and starts from there. So, if you have just one rather loud kick halfway through the song, it could make everything else sound relatively quiet.

Without rambling too much, it doesn't even have to be so obvious. It could be some frequency you barely hear, which is why monitors and paying attention to the levels are so important.

Hope this helps

@lordelix - Next time I won't spell check!!! lol ... That is the crux isn't it. It's really difficult this gap in understanding, but it does equate to some lively discussion.

[lordelix]

I just love to keep people on their toes with words not so commonly used!!
I cant even begin to tell you the fun I had with a friend that was visiting from England!! Up way too late with too much alchohol and defining as many native slang terms as possible!!

[dbijoux]

Bob's your uncle.

[Nik_]

A lot of you are missing the point.

The OP connects his Fantom to his speakers and playback is satisfactory.

So he does a mix down/resample insode the Fantom. Exports the AIF file via USB to the compter. When he plays the file i the computer, connected to the same speakers, it is no longer satisfactory.

So, there are only two variables:

1. The Resample function in the Fantom. We know it knocks several dB off the over all level. Normalising the file in Audacity to -1dB should sort that out. But does it also lose some fidelity?

2. The sound card outputs o the computer. It's quite possible that these sound significantly worse than the outputs of the Fantom itself. Not much you can do about that except get a decent sound card or audio interface. Cheaper than a stand-alone recorder.

[emenelton]

If his Bose are connected directly to his fantom, how can he play his computer mix through them?

[dbijoux]

@Nik - Right, there shouldn't be a problem. However...

1) I don't think Resampling looses fidelity, but seems to also include some deemphasis. Additionally, normalizing is not a best practice with a bad mix, it won't solve the problem and could make it worse.

2) I've never heard what I would call an arsed sound card on playback. I mean maybe that's the case, heck, it could be a jacked up cable. You understand more about it than me, and 'real' sound card certainly wouldn't hurt.

All in all, IMO Lordelix has it right on. It's a pretty big expectation to mix/master(premaster) using only a workstation. Not to say it isn't possible, just exceedingly difficult and constraining. Don't you agree?

@emenelton - I imagine everything route through his mixer?

[emenelton]

@emenelton - I imagine everything route through his mixer?

He said that the bose were connected directly to his fantom and the computer plays through them, which seems like conflicting statements. He said also that the fantom didn't run through his mixer.

One observation I can make to buminbeer2, if normalizing at -3db does make his mix louder, he should have room in his fantom rendering stage to boost the overall level of his mix.

To do that, remix by adjusting all your faders up will maintaing the balance in your mix.

It's a start in doing music to understand the various gain stages in your setup including the speaker icon in your computers tray. Getting a better soundcard is a start but it's not going to solve your issues. Basics such as knowing how to transfer your mix from the fantom into the computer and to play them back on each confirming that they sound the same is just the beginning.

maybe buminbeer2 could post the mix he's taking about. I'd be happy to listen to if if he offered me a beer too!

[buminbeer2]

To clarify, my Bose speakers have two sets of inputs So I can play the keyboard at the same time I'm trying to learn a song from the computer or Utube, the cpu conversion sucks compared to me playing the sequence from the keyboard. I can even play both at the same time for a really odd effect, hehe. Maybe the built in sound is causing some of the issues on the computer, but playing CD's sound normal or at least to my ears?

Thanks for all you rocking to help me. Great topic it seems.

The song is called dying (no where near a final track, just something I made to figure out how to get it to sound better)

http://www.box.net/shared/tli5esmhxl

[emenelton]

great work!