VideoFreq,
I'm a pro audio guy (28 years experience), and definietly NOT a pro video guy! :)
Yes, your audio card is important.
You asked for recommendations...

For Mac, check out Apogee and MOTU.
For PC, look into M-Audio, RME, Presonus, Focusrite.
This stuff can get very subjective, and for every brand I tell you to look at, there'll be some guy with a comment that reads like "I had one of them and it was sh!t".
YMMV.
:)

If you want to be able to mix in sourround formats (5.1 et al), you will need an audio card which has at least that many outputs (ie the number of audio channels you want in your exported mix). Want to mix in 7.1? You'll need a soundcard with at least 8 outputs. You'll also need that many channels of amplification and the requisite speakers.
So, what toothbrush does a dentist use?
I'm currently using an M-Audio Fasttrack Ultra 8R. That's a discontinued product, so don't bother looking.
But I'm very happy with it.

These days, almost all soundcards will do 96kHz sample rates, and some will offer 192kHz.
Do you NEED 192kHz? No, not really. Nice to have, but not essential.
Unless you're planning on preparing content for SACD or one of those other funky formats that support 192Khz audio.

As stated, I don't do video professionally. I bought VP13 for ediing my home movies together.
My home studio is NOT set up for surround, and I have no plans to go down that route (at least at this stage, anyway).
If I can be of further assistance, sing out.

Oh, and set your templates to use audio (inside your timeline) at 24 bit resolution.
But when you render, render to 16 bit.
And if I can condense 28 years of audio knowledge into one power tip for video guys, it's this:

DON'T mix your audio with a compressor across your master output.
A compressor will hide a multitude of sins and bad habits.
If you have a narrator/narration vo, use this as your starting point. Set it to playback so you're seeing peaks around -12 to -10dBFS. Yes, there'll be the occasional blip that's louder, and some parts that will be softer, but that's ok.
At least we have a starting point.
Don't ever change you narration level again.
Now, here's the kicker...
From here on, IGNORE your audio meters! Yep, you heard me. Ignore 'em.
From this point forth, mix all OTHER (non-narration) pieces of DIALOGUE around your narration. Using. Your. Ears. Seriously, use your ears to gauge the level that these other pieces of dialogue should be playing back at. Not your audio meters.... your ears.
Once you've done this for your entire timeline, go back and repeat the process for any sound effects or foley (not music... yet).
Mix them AROUND your dialogue. If they are drowning your voices, drop them in level. If the voices seem too loud, just bring the sfx or foley UP in the mix and turn your speakers down! :)
Finished that?
Now go back and do the same thing for your music.
Ride your music levels up and down along your entire timeline so that the music never drowns anything else out.
At this point, you should have a VERY dynamic mix, but one where EVERYTHING is audible.
And now, you may look at your audio meters again.
Now, and ONLY now, can you go ahead and strap a compressor across your master output.
Understanding compressor ratio is very complicated, but if your mix is WILDLY dynamic, start with a ratio of say 5:1.
Set the compressor attack time to around 15ms and the release time to around 100ms.
Sure, that's a subjective call, but with those settings, you won't get any nasty surprises.
If the compressor offers a detection circuit choice (usually 'peak' or 'rms'), use peak.
Now, set the threshold of the compressor so that you're seeing around 6-8dB of gain reduction on the compressor's gain reduction meter.
Adjust the compressor's output so that your narration is peaking around -3dBFS on your master output.
Now, your mix should really be coming together.
The final plugin on your master output should be a peak limiter.
Set it to limit output at -1dBFS.
Now, your mix should be totally rocking, and no peaks will exceed -1dBFS.
If you're mixing commercially (for broadcast), you may have to abide by the rules which use LKFS metering. Sorry, don't have time to keep writing this... people hounding me to do work.
Sing out for more info....
Later.

I'd posit that the sound card is much more important than the video card, at least as far as video editing is concerned. A $1000 video card isn't going to produce results that look any better than a $10 card. However, a $100 sound card will produce results that can sound much better than a $25 card.

Audio is subject to the "weakest link in the chain".

$1000 audio card connected to bad monitors or using a poor amp will result in many problems.

A MB audio card connected to a good amp and speakers will prove better - in my experience.

Then you have to consider your "space". If you have bad reflections, poor placement, etc, etc. then no combination of gear will produce great results. A good to better set of headphones may be a better solution - though there are pros and cons to mixing with headphones as well.

Making sure your amp/speaker volume and other points in the signal chain are in "harmony" make a big differences too. That distortion you may here may be from your amp/speakers. Take time in setting your monitoring levels (external to Vegas).

96, 192 kHz; 16- 24 bit. Meh. Vegas mixes internally at floating point vs fixed, so again, the weakest link when you monitor comes into play. Unless you have a really good listening space with decent speakers and amp, the results of your 'room' will affect things more than sample rate and bit depth. It is best not to re-sample too often - mixing 96 khz with 44.1 and setting your project to 192 khz wont buy you much. In some cases it can hurt.

The only time bit depth really comes into play is right before it is sent to the sound hardware ( and the render!) . That is when the internal floating point mix stream is converted to the output format. Rendering is no different than monitoring in this respect. just different destinations. There are cases where you want to be aware of this. Many will recommend a ready for mastering mix - as simple and with little final dynamics added. Get it to sound right in the mix, then take that mix and master to the final destination.

The best advice is to mix with your ears, not your eyes. Get the best mix with no processing - particularly dynamics. Compare to other material that you are familiar with and you know how it sounds in your room space.

oh, use the down mix button to see how it plays in mono. It can sometimes let you hear where things can get "lost in the mix".

There is no 'easy button". Listen, and compare is the best advise. It will overcome all of the technical sides more often than adding gear. Understand your gear and signal chain both internally and externally can produce superior results than the top of the line toys you can throw money at. You will start to hear the weakest links in your chain and know where to put your gear budget at. I always recommend amplification and monitors before a new audio device.

Peter

As a 30+ year audio pro, I concur with the above.
But there is no 'one-size-fits-all' in terms of processor, buss and order. Plug-in presets should be considered as a starting point, and 99.9% of the time will need to be tweaked. Volume envelopes are a better choice to control levels opposed to compressing the sh__ out a track.
I record location dialog at 24/48k, music and S/FX at 96k or higher.
Get the best pro-audio soundcard you can afford.
FWIW, I always liked the Echo soundcards for the great sound and rock solid drivers.

FWIW....




Part 2 is interesting, as well as the comparison of a Sound Blaster and to more expensive audio cards. Time 41:29 approximately.

Peter

Most people actually recommend a USB sound device these days, even if you have a desktop computer. There are plenty that go from 100 to several hundred dollars. As posted, Sound on Sound is great, but so is Homerecording.com, a forum for both newbies and experienced. Good place to ask questions.

I'm looking into Focusrite, because I like the type of sound they produce. But that's purely subjective.
I have midrange M-Audio monitors and an old Firewire 4-10 that's worked pretty well for me. I also have Sony MDR 7509HD headphones that do not color or add to the sound mix like consumer cans do. If you go with headphones, find ones that do not add base or midrange or high end. You want the sound to be accurate as possible when mixing no matter how bad it sounds.

audio2u,
One compressor across the master bus?
Limiting in post-mix?
Let the tracks stack without levels compensation?
Sorry, I don't get it.

I usually run 32-36 mics into 4-6 subs on the main board(s). Backing tracks are on their own sub(s). Each one has a unique dynamic loudness and peak range that must be tackled individually before getting them to work together in the mix. I plan on 2-3 days' work just for that.

We must live in separate quadrants of the universe.

As rraud said, volume envelopes are much better way to get your mix happening than to just slam everything through a compressor.
I've always said on my podcast that you should try and get your mix sounding as close to what you EXPECT fo finish up with, without emplying any compression at all.
Use your volume envelopes to get the mix "in the ballpark".
Then, and ONLY then, should you apply any compression. The reason for this is that the compression occurs on the post-fader signal level. So, if you have a really dynamic voiceover, and you try compressing without any volume automation in place, the compressor will be working WAY too hard on the loud parts of the voiceover, and barely doing anything at all on the softer parts of the voiceover.
But by automating your envelopes first, the signal going IN to the compressor is much more consistent in level. Thus, the compressor doesn't have to work as hard, you don't get the "pumping" artifacts, and the loud parts aren't sounding MORE squished than the quiet parts. In general, the whole damn thing sounds much smoother.
Now apply that philosophy to every part of the audio mix.
Your sound effects, your ADR, your foley, your music/score.

OK, now to your questions....
One compressor across the master buss?

Yep, absolutely. Why would you need any more? If you have controlled your dynamics prior to signals getting to the master buss, then your master shouldn't need a whole lot of compression. If I've jumped ahead of you here, my apologies. My workflow is one whereby all narration/ADR voice tracks are subbed into one submix. All sound effects tracks are subbed into ANOTHER submix. Music tracks are subbed into ANOTHER submix.
In each case, the individual audio tracks are automated, which means that if you were to solo any one of those submix busses, you'd hear a submix which was already "in the ballpark".
I then slap a compressor across each of those submixes. But because all of the individual tracks were already automated, each of those submix compressors is not having to work too hard.
Then, all of those submixes feed into the master buss. And the compressor there is likewise not having to do any heavy lifting because the mix is so damned close to what I want, that it's just applying some glue to the mix.

Limiting in post-mix?
By this, I think you're asking "why is there a limiter on the final output?" If I've misunderstood, please say so. Otherwise, my answer is this: That limiter is there to make sure I can set an ABSOLUTE fixed peak value. Please understand I'm a 28 year veteran of commercial radio, and I mix to 0VU, not this new-fangled -24LKFS rubbish!! :) I jest, I know there's a reason for it... I'm just not in that mindset yet, as radio isn't yet adopting LKFS for metering. So, my peak limiter allows me to make sure that regardless of what signal is coming through the master buss, there will NEVER be a sample which exceeds -1dBFS. EVER.

Let the tracks stack without levels compensation?
Sorry. Not quite understanding what you're asking here. I suspect this may be related to my aforementioned workflow of making extensive use of submixes. The idea is that each submix, if solo'd, will sound awesome in its own right. But you will then go ahead and automate the submixes as well, to build the final mix you want to hear. Does that make sense? If not, sing out.

Cheers.

Holding up ignorance flag here!
Starting with narration/vo and mixing everything around that makes perfect sense but are you suggesting that this signal is NOT compressed at this stage? I thought it should be, in order to ensure it cuts through everything else.

There's no 'one-size-fits-all' workflow. If an element needs compressing, compress it., if it doesn't, do not. Though I almost always put a peak limiter on the master buss for rendering.
OT- My 'current' favorite is Vladgsound's 'Limiter No. 6', which has five modules,.. no I don't use them all.

1. Get tjhe best, quietest preamps that money can buy. Modern mics can have 135 dB dynamic range! Good sound cards have around 120dB snr.
2. Get an interface with at least eight Discrete I/O channels for recording and editing, not a 7.1 home surround card (You should use that to play back, not record).
3. Check here first to see that the one you want is working clean with Vegas.
4. When in doubt about features, refer to #1 above; The word the pros use for dsp, modeling, effects, contours, etc. is "distortion." If it claims it's 192/24 or better, make sure the maker isn't blowing smoke up your a**.

"
Important, but not soi much as having a little preemptive limiting at the record inputs (see #1 above -- I mix a lot of live music events. and a too-hot mic busting your pre's is no fun!

I've got a lot to learn about audio. All that went woooosh right over my head. I have to keep this topic in my notes.

Thanks  👂

Not much to add here, but yes your audio IS as important as your video.. (maybe even more so)

So invest in a good quality audio interface and dont forget good studio monitors too! Also, your room has a large effect on the audio you hear in many ways - so read up on room acoustics too. Maybe read something on the lines of 'setting up a home recording studio'. Thats taking you in the right direction.

You may have a pro level video card and display for monitoring, but do you have the audio setup to complement it? (a question to all readers not just you).

Trust me, audio is king. Even in the video editing world. Its vital to get good audio results in your video work.