Hi.

I've been recording a few rough demos of instrumental tracks I've made onto my computer but I don't know if I'm doing the recording in the optimal way for my equipment.

The setup I've been using is:

Guitar > Fender Frontman 15G (crappy practice amp), headphone out > Creative Sound Blaster Live! 24-bit, line-in > Audacity

I've been recording as 24-bit, 96kHz for the clean stuff. This is probably overkill; I'may reduce this to 48kHz and I probably don't need the bit-depth either. I would record at 44.1kHz but the card introduces recording delays, my guess is due to native resampling. I record as 16kHz for the overdriven parts as this is a really trebly amp and I need that lowpass!

I haven't decided whether to EQ the amp before line-in yet and so the EQ pots are set to 5 at the moment (I guess that's almost flat). I have the volume set at such a level that the headphone out provides a reasonably loud signal without clipping or analogue limiting.

I switch the computer monitor off most of the time as that does help reduce some of the noise. Overdriven sounds have some noise but once I'm playing it's not noticeable. Clean sounds have some too but I can live with it for the most part! From my experience I don't get any better results when connecting the guitar directly into the card; the level of amplification required introduces its own noise anyway.

Anyhow, I'd appreciate any opinions or suggestions from those more in the know than myself (ie anyone!) on how I could get the best quality and any hints you know. In the ideal world I'd have a better amp with a line-out and my effects pedal wouldn't be broken... but...


(Edit: added)

You already seem to know pretty well what you're doing.

Some tips:

- I would indeed record at 48 kHz instead of 96 kHz. 96 kHz is a waste of space, IMO.

- The line out of cheap guitar amps is great for a clean sound, but for distortion it sounds bad most of the times, because of the absence of a speaker simulator. Two things you could do:

* Search for a mic to record your amp through its speaker. A Shure SM57 is an example of a great mic for these kind of purposes, but quite expensive. There are cheap copies (which aren't as good, but it is a small amp anyway so you won't get a killer sound with a SM57 either).

* Search for a speaker and/or amp simulator plug-in for your recording software.

Thanks very much for your help!

I will switch to 48kHz. In case I want to transfer recorded audio to CD, am I right in thinking that the required resampling shouldn't degrade quality too much? Do you think 24-bit is worth the extra space or should I go down to 16?

Sadly my amp doesn't even have a line-out so I have to use the headphone socket at a low level. Man, that thing is trebly! Recording at 16kHz alleviates some of the annoyance but it still doesn't sound altogether great. I have tried recording with a mic previously (I have a Shure C606, any good?) and may do so again but I had to have the stuff pretty loud and it wasn't very practical for me.

I like your suggestion of software modeling solutions. I've just been using Audacity but would there be any plugins for it or equivalent free software that could do such things?



Of course, if anyone else can help I'll be very thankful.

No problem


Most recording software will resample properly back to 44.1 kHz, using the right filters and such.

If you do a lot of audio editting, 24 bit is advised over 16 bit, to avoid noise due to rounding errors (see this topic). However, since your recordings are basic and just demos 16 bit should also be sufficient if you want to save space.


I've never worked with a C606 but it looks like a nice mic for basic guitar recording. But you have to experiment with different positions for your mic (for example, try to put it close to your speaker, slightly off-center).


The UK based magazine Computer Music has a CD/DVD with each edition containing a free software package called Computer Muzys. It is a great program to get introduced in the digital recording world, *and* it supports VST plugins and VST instruments, a standard developed by Steinberg (the guys from Cubase and Nuendo). There exist a lot of guitar modeling VST plugins, I think even some free ones.

Good luck!

Hmm, I'll have to get looking. Thanks again.

What about getting a di box and reording straight/clean into your PC and then adding effects there?

Okay, now I'm embarassed... I don't know exactly what you mean by DI box. Is this just a small preamp or something? Yeah, I quite like the idea of doing effects and modelling etc in software, assuming they're reasonable realistic, but I'd prefer something free at the mo and I doubt there'll be many possibilites.

A DI (Direct Input) box basically matches the signal of your guitar to the input of your sound card (so you don't have to crank up the volume).

There are also extern devices especially designed for guitar, like the Line 6 Guitarport. That device also comes with modeling software.

If you do all your effects on your PC it is important that you have a fast computer with a low latency, otherwise a significant delay is introduced when you play your guitar.

Ah, that clears it up a bit - cheers.

Yeah, I've seen Line6 stuff before. A GuitarPort or maybe even a POD - assuming that's good for computer connection - would probably be a good investment.

Oooh yes, I think I'll pick up a GuitarPort. Thanks for the recommendation! It should give pretty good transfer quality and the amp and effects modelling seems adequate - plus I could download free presets and purchase libraries of others. Coupled with free upgrades, I could be set for a while!

Seems most of the GP advertising concerns GP Online and the ability to jam with other songs... I've always preferred playing my own stuff though. As far as I know it connects via USB and can be selected as its own recording sound device, so that should do me fine.


(Source: http://www.ramelectronics.net/html/guitar-port.html)

This should be most software, right? I use Audacity as a simple solution now but I'm sure I could find something else. I've seen Line 6's Rifftracker software on their site but I'll keep it simple and just consider the GP on its own for now.

Whoo, I need a lie down.

My friend asked me the exact same question yesterday. I would also recommend you look into getting a pre-amp it would make your life a lot easier, then you can record your mixes dry and DSP effects with Audacity. No need to use a microphone unless you are recording other acoustic instruments in a studio or the proper enviroment. It's a little bit on the pricey side, but I think it's worth it. If you want to use commercial software Pro-Tools would be great as well, but I think Audacity would be more than enough.

http://www.m-audio.com/products/en_us/Fast...kUSB-focus.html





Make sure your soundcard supports ASIO (Audio Stream Input Output) those
low-latency drivers are necessary for recording, as Bug mentioned here.



Yes, the impedance levels will be horrible that's why a pre-amp is necessary on that note as well.



Most DSP sound realistic enough to me these days I can't tell the difference when it comes to electro-acoustic instruments, for real acoustic instruments though a real recording studio is needed

Thanks for the input! Hopefully a GuitarPort coupled with its bundled modelling software Audacity can do the business - at least until I get rich and famous...

ha! no problem if you have any other questions just ask. . I want name recognition if you do (just kidding).

Haha, don't worry!

another good program is cubase sx which you can download, if you know what i mean. apologies if i broke the rules.

that and guitar rig put together gives you a bindle of effects as cubase incorporates guitar rig into is recording.

A lot of sequencers both open-source and commericial have DSP effects you can use and last time I checked Cubase wasn't free . It's great if your serious and have some money to put into it, most of the time people don't so it's in their best interest to stick with open stuff, rather than go and pirate commercial software that some developer is trying to make a living off of. I won't get into entire Philosophical discussion about, but let's just say people who take things for face value greatly annoy me.

Drenholm, you could always download some demo's and purchase the software you like. Most software comes in cheap "lite" versions.

Take a look at Cubase and Sonar, for example. ProTools comes in a free version: ProTools FREE, which unfortunately doesn't work on Windows XP.

You also could try the free "Computer Muzys" I mentioned before.

*edit*

By the way, most hardware, like GuitarPort, comes with lite and/or demo versions of some sequencers.

Thanks tons, everyone. It'd be too easy for me to get confused between all the many different options available!

bug80, I picked up a copy of cm and am having a look through it. I'll be trying Computer Muzys soon - I remember using it before but having no real need for it. At least now I do!

I've reached the conclusion at the moment that I'll probably be getting a TonePort. It seems to do and have everything the GuitarPort does and more - and it can do vocals as well so once I start properly practising singing...! Also, it comes with recording software which should hopefully get the job done.

Am I right in thinking that the TonePort includes basically all the features of the GuitarPort but with a few more effects and software features and better support for bass and vocals? If so, I don't mind the extra cash. As long as the GuitarPort doesn't have anything spectacular which the TonePort lacks, I'll go for the TP.

Any mic will work better than going direct. Guitar speakers are an essential EQ element for overdriven guitars. Sometimes you can get by without it (it has been done to great effect on some commercial recordings) but usually it's tough. As you've noticed, the direct sound is raspy.

Also, try pointing the mic at various places on the speaker and at various distances. Often close mic'ing will work OK (though you won't get any room sound). Pointing the mic at the edge of the speaker cone should further subdue highs.

If you must go direct, try using a parametric or graphic EQ to roll off as much treble as possible above 4Khz or 5Khz. If that sounds too dull, peak the response with low 'Q' (gentle slope) right under the rolloff frequency. That's exactly what most guitar speakers do.

'Speaker Emulators' often create a comb-filter effect as well. That could be simulated to some extent if you have several mid-band EQ points available. In this case you'd set the Q very high, but *notch* the response at several frequencies between, say, 300Hz and 1KHz. Make the dips very sharp but leave the midband otherwise fairly flat. The idea is to simulate what happens with multiple mics or multiple speakers in a large room.

OK let us know how it works out.

I've used Computer Muzys but since I've always worked with "normal" sequencers I couldn't get used to the way of working with Muzys. But I think it is really handy once you're used to it.

Engineer, thanks. I may try your suggestions about different mic positions and I might roll off the treble on my amp or guitar as I think they are quite a sharp combination.

I guess my method of recording at 16000Hz was a quite crude way of cutting out extra treble? If you're suggesting such lower cutoff frequencies, I'll see what I can do about setting something like that up.

Hopefully this shouldn't matter for too long as I will probably use the TonePort for most recording once I get it but you've given some really helpful advice about micing up amps. Thanks!