I need some serious advice on warmth --and--digital recording

D

Deadboyy

New member
Ok after reading this message board and realizing that im not the only clueless person out there trying to record quality music on a budget.......

Im in the process of building a studio and for the low cost of it im going the PC ,,,,interface,,,soundcard,,,software route........

What I do know is that when you convert analog to digital it tends to sound kind of flat and flakey,,,im going with the omni studio which comes with the delta 66 card and I heard that the A/D converter is good on that one..
(I hope) but anyway in order to avoid getting my sound to flat and hollow sounding is to make it as warm as possible before it hits that A/D converter on the sound card....so im thinking to get tube mics,,,tube pre amps,,maybe 2 or 3 tube preamps,,and just go tube crazy,,so when my songs that sound great coming out of my amp doesnt get dittered down to digital flatness...

So I guess my question is ,,,,am I right in doing this or am I just a stupid newbie just guessing at all this

Thanks and i hope you reply cause im clueless
 
Deadboyy said:
What I do know is that when you convert analog to digital it tends to sound kind of flat and flakey
Click to expand...
Complete and utter nonsense.........

...doesnt get dittered down to digital flatness...
Click to expand...
More complete and utter nonsense.........

So I guess my question is ,,,,am I right in doing this or am I just a stupid newbie just guessing at all this
Click to expand...
I don't know about you "being stupid" - I doubt it, if you were really stupid, you probably wouldn't have even been able to turn on the computer, log-in to the net, and get to this forum to type your question! ;)

But it certainly sounds like you got some poor information about how good digital recording actually is, and you simply ARE guessing about the rest of it!

If you get a good sound coming out of your sound sources, and you put the right mic in the right position to capture it... if it sounds like shit or in your words "digitally flat", I'll tell you right now it isn't 'cos of digital recording! Recording technique plays into it quite a bit........

Bruce
 
kool

best response iv gotten yet about this , I know I sounded really misinformed,, but all these things I have heard ,, key phrases like,,,digital and flat...

And thanks alot Bear for lifting my worries about home digital ,, its makes it alot easier to fork over all that money:)
 
Tubes, Warmth...

Quality equip seems to make a big difference (whether it is dig or analog, solid state or tube)

Tubes do seem to add a certain desireable flavor to some things...but it is actually harmonic distortion...but hey, fuzzy guitars (and sometimes vocals) do sound good !

I think a lot depends on what you are recording and what you want it to sound like.

Here is my own "warmth" formula:
Modern digital equipment simply has a lot better high frequency response than old "vintage" equipment. Much of the older stuff rolled off dramatically above 15 kHz. Also, in a real-life concert situation, you are typically sitting some distance away from the sound source. You don't have your ears sitting 4-6 inches in front of the guitar or amp like most people set their mikes. So try just rolling off the highs to make the recording sound warmer: Set a parametric EQ to Q = 1, Freq = 20 kHz, Gain = -6 dB. Adjust to taste. The trick is to make it "warm" but not "dull".

Report back on how you like this !
Thanks,
Rick
 
I'd like to chime in on Rick's comments.Those Edison recordings of Caruso are certainly "vintage",but we rightly don't consider that level of lack of fidelity desireable.
On the other hand,the modern equipment is so startlingly accurate that it also is considered not so desireable in terms of our expectations.
Where is the middle ground?The equipment used in the late jazz era and early rock era produced a body of musical culture that still dominates public background music (ballpark PAs,elevators,restraunts etc.) and THAT stuff (despite whatever sonic faults and technical defects) is the RIGHT sound in our collective unconsious simply because it is what we are used to.
So...back to "warmth".How do you sound warm?A couple of the charactistic byproducts of older tube gear are compression as well as a harmonic distortion that emphasizes the even order harmonics.There weren't a huge body of effects available then so it was mostly just reverb and delay,not nearly as slickly done as a Roger Nichols,say,and the latest Steely Dan CD(amazing job IMHO)
You can buy old gear but really you can do the same thing with new stuff,and it doesn't necessarily have to be tube.Study the techniques of the old engineers by listening to Sinatra on Capitol or George Martin's work on the Beatles.

Tom
 
Oh boy oh boy I could write a book.., so I'll keep it as brief as poss.

First of all, digital can sound good, as a matter of fact, digital can sound VERY good.

Second, Axis, you'll have to re-think your warmth formula :), nothing of it holds true.

Digital sound stands of falls with the quality of conversion and the accuracy of the digital clock. The better the converter, the better the sound quality, the more accurate the clock, the better the sound quality.

Most A/D converters on cards have a very low dB dynamic range, and there is the main problem. No wonder really, they are made at a price. All that stuff including A/D and D/A converters for a couple of hunfred bucks maximum, while a good 2 ch A/D or D/A starts at about 7 to 900.
That does NOT mean there are no good cheaper ones, they are just not that good.

If you read the above carefully, it should tell you something. Your quality of sound is dependent on your signal path and your conversion. If it is bad .... it will be bad, no matter what you do. I could hang a $12.000 Telefunken U47, on a $4.000 pre amp, going into a low dynamic range, low quality converter - and it will sound like crap. Therefore, "warming up" digital sound by means of using tube gear is a complete waste of money, unless you address the priorities first.
On the other hand, if you get quality conversion, even a cheap mic will sound OK.

So, If you start of in recording, and look for a cheap way to do so, perhaps start with a sytem which allows you the flexibility to improve things along the way. Fortunately that option is getting cheaper and cheaper. You have got a computer already, perhaps if you post what you can affort to spend, we can come up with a solution that will get you started and will keep you going with warm and happy ears :)
 
sjoko2:
Second, Axis, you'll have to re-think your warmth formula , nothing of it holds true.
Click to expand...
If you try it, and don't like it...fine.
But it actually originally came from a very well-respected professional recording engineer. I didn't post the reference, because I simply can't remember the location. If you do like it, he deserves the credit. If you don't like it, I accept the blame. :)
I rechecked my personal settings, and i usually have the Q at about 2-2.5 rather than 1 though.
Luckily, it is very easy to just turn it off.

Peace,
Rick
 
I have listened to them, and I have had stuff in the studio recorded through them. I'm not very impressed, but them - I'm very critical. I think 20 bit ADAT recordings sound better
 
Hey Axis,
Its often misconcieved that "warmth" just comes from the lower frequencies. It doesn't, its an overall smoothness that creates the feeling of a track being "warm".
On top of that, you have to consider that whatever you do, and whatever it is, every track has to be part of a complete picture. Very often people try and "warm-up" a vocal track by boosting the lower regions of a vocal. As the level of a vocal (unless you have a bass opera singer:) is rather low in this region, the boost increases noise levels, and the result will start to fight with other instruments which have a predominance in those regions.
Also, the often heard statement that digital has a better high frequency response as compared to analogue is, to be honest, total bullshit (don't take that as an attack on you - its often heard, as a result people start believing the wrong things - natural result).
A lot of the vitage tube gear did roll off - because of good design. If you'd measure the frequency curves of some old mics and reach 20k - keep measuring, you'll still get a response when you come close to 40k. Whereas digital has a 20k ceiling.... bang - stop.
Digital sound is perceived to have a higher top end, mostly because its rather harsh sound caused by bad conversion, as well as its limited dynamic range. Play / record through good converters with a higher dynamic range and your sound will once again be as smooth as butter.

Hope that helped
 
Okey Doke..

Also, the often heard statement that digital has a better high frequency response as compared to analogue is, to be honest, total bullshit (don't take that as an attack on you - its often heard, as a result people start believing the wrong things - natural result).
A lot of the vitage tube gear did roll off - because of good design. If you'd measure the frequency curves of some old mics and reach 20k - keep measuring, you'll still get a response when you come close to 40k. Whereas digital has a 20k ceiling.... bang - stop.
Click to expand...

hmmm...Well, don't take it as a personal attack on you when I point out that THAT is total bullshit.

Digital is quite commonly VERY flat in frequency response right to 20 kHz. True. Now look at the frequency response curves of some well respected vintage gear: sm57 mikes for example, or the Neumann U87 at the link below. Way down above 15 k...hence, "warm". I said nothing about "boosting the lower regions", and I don't think it would help either..

In the Neumann written spec, it is said to have response "to 20 kHz.", because typically in the acoustic/electronics industry, it is common to cite up to the -3 dB point in response (one HALF power). Now subtract ANOTHER 3 dB for a "20 kHz" preamp and...pretty soon you have lost a lot of dB. Notice also that those popular mikes were designed with resonance peaks in the 10-15 K region in an attempt to keep them sound "bright enough" even though they fall off rapidly above that. The approx. 40 K response you speak of is merely a narrow-band resonant harmonic of the lower resonance. Whether it contributes to the audible tone is a highly suspect theory.

[image]
http://www.neumannusa.com/images/specs/u87/ppU87Cardiod.gif
[/image]
OOPS, I guess image codes are "off". Just look at the site.
The line furthest right represents 20 KHz. The place where it crosses 0 dB, falling is about 12-13 KHz. Notice also that in spite of the written spec, it is actually down to -12 dB at 20 kHz. !!!!! (I guess the mike industry can claim any spec they want )
PS: I design ultrasonic acoustic transducers for a living. :)

http://www.neumannusa.com/pages/products/micSpecs.asp?microphone=U87a

Peace,
Rick
 
Last edited:
You're absolutely right with your comments about old microphones and the narrow band stuff. Just plain fact.
Now compare that to "modern" equipment....... and what do you see?
 
What do I see:...

I see two parallel trends:
1. Microphones, and particularly mike preamps, bragging about their extended high frequency response, and extreme linearity and flatness at those high frequencies (because it captures transients more accurately). And it really is "more accurate".

2. People reacting against "digital harshness" because they don't realize that they don't really want to hear it with brutal accuracy, but with a little more softness and idealization...like they are sitting out in the audience about 50 feet away, not with their heads 6" in front of the speaker cone !!! So they buy 'vintage gear", or gear that emulates its designs, in spite of the fact that it is older technology in terms of measureable performance.

BTW: I am not arguing that digital converters, etc. don't vary in quality and and MAY contribute to smoothness. I am just pointing out that what a lot of people call "vintage smooth" is really "lowered high-freq. response." I would go to the highest bit length and sample rate economically feasible, because digital IS harsher. It is just a fact that 44.1 K is the MINIMUM to cover the human hearing band and gives short-shrift to transient response. 24 bit is also probably just getting to an acceptable level of volume-level discretization.

Also: Could you explain to us WHY some "more accurate" digital converters enhance smoothness. The industry min. standard for jitter is down around 20 nanoseconds. This is an immeasureable contribution to any audible "wow or flutter" type spec, so it is hard to imagine that it makes that big a difference. If you are talking about the whole conversion process, then the other components in the A/D and D/A converters may be influencing the audible differences. I really don't know and would like an explanation.

Peace,
Rick
 
Hey Axis, cool response.

I have been involved in very extensive analogue / digital audio testing over the last decade, and it is only in the last 2 years that I have made the decission to "let go" of analogue recording and make the switch permanently. This for one reason only, you can now get digital sound to a standard which equals, and in some cases surpasses, what was previously possible with analogue.
I don't think this is the place to go into detail regarding tech & spec of converters, but yes, there is a huge difference between one type of converter and the next, even with 2 brands using the same chip. It is also an area where psychoacoustics starts to become a factor.
Just as one example, you could look up the specification of a Digidesign 888, a Lucid and an Apogee. Similarities in the tech spec, yes. Listen to them and the Digidesign drops out immediately, the other 2 sound substantially better, both of them. Now you start layering tracks, one after the other. Both the Apogee and the Lucid use the same chips, both sound good, but the more tracks you stack, the more the different characteristics of the converters become apparent.
The Lucid and Apogee are, in my and most people's opinion, the best affordable option for recording digitally.
Now you come to another thing, tech specs. Look at cheap converters, they all claim an "A" test, and between something like 90 and 106 dB dynamic range.
I think most such spec writers should be politicians. Ooops, getting carried away there :)
For cost purposes these converters use one single chip. Yes you can obtain very, even extremely good results with them, no question. But it is possible to build much better converters, with a dynamic range of up to 157dB
Can you imagine that - with the cleanliness of digital?
That, in 24/48 or preferably even 24/96 is where the industry has to go, and is slowly going. Than we are back at a point where you can "hear" a tube, a pre-amp etc.

I have had many discussions with "techies" like this. Looking at it from a pure technology perspective is very different. Just consider this, according to your perspective, everything recorded analogue should have a much lowered high frequency response, and everything recorded digital should sound brutally accurate. Whereass if you take 2 such albums and put them through a spectrum analyser - you would see none.

Damn I love this, its interesting and I gotta run! Glad to meet you, I'll be back soon!
 
...according to your perspective, everything recorded analogue should have a much lowered high frequency response, and everything recorded digital should sound brutally accurate. Whereass if you take 2 such albums and put them through a spectrum analyser - you would see none.
Click to expand...

You may have assumed that that is what I am saying, but would be an oversimplified and extremized version of my perspective.

We have to remember that Engineers and Artists are constantly adjusting the available technology to produce a "good" sound. We used to think it was a great thing to get a tape deck with high frequency response to 18..19 20 kHz !!! High frequencies and transients were difficult to capture, so thats where the emphasis turned...to increasing high frequency response and acuracy. Example: Creed Taylor produced jazz albums of the early 70's. Distinguished by their incredibly crisp (yet smooth, warm) sound. Mikes that had a big boost in the "presence" region were considered "great".
Then digital...suddenly high frequencies are no longer hard to get, but the digitization process does exaggerate the increase in harshness that accrues from the increase in high-freq. response by itself. Suddenly everyone is asking "Why does my guitar sounds so trebley and harsh ?" (not so noticeably on vocals, since they have lower high-freq. content to begin with)
The savvy artists and producers continue to tweak the equipment to get a good sound, by mike placement, selection, EQ adjustment, AND using high-quality converters. The difference is that now they CAN get the highs if they want. Before, they just weren't there. Listen to some of your favorite 60-70 albums and notice how much high-freq is missing (in spite of the fact that the music was actually much better ! )

Peace,
Rick
 
call me a nut but art makes a good tube a/d converter and i also use a tube compressor after my pod going into the mixer. the levelar has been a great tool at only 79.00 at musicians friend. maybe the digital problems are from a sound card that cant give you a high enough bit resolution . as for the idiot question we all had to learn somehow and its how much time you spend doing things the wrong way by doing the same thing over and over again and expecting the right results. without asking an experts advice. you took a step in the right direction coming here.
 
Quote: but the digitization process does exaggerate the increase in harshness that accrues from the increase in high-freq. response by itself.

Now we're getting there! Thats the whole thing see - if you use 24/48 or 24/96 with an accurate clock and a good converter - its gone, you're back to the "normal" sound our mind is used to.

What we did in the trials, which where done in the UK on behalf of commercial as well as institutional interests in the audio industry, was run test tones, a guitar, a male and a female vocal track and a piano through a Neve onto a Studer 2" running at high speed. We did this with a number of different calibrations to the machine, ranging from -3 to +9. Using splits we ran digital recording equipment in-sync.
We started of with generic "home" equipment CD reproduction quality at 16/44.1. After that we tried to improve the quality with stand alone converters, and various other means to improve the sound reproduction and recording quality. After we were finished with 16/44.1 we went up a step, did the same thing, and up and up. Final stage was a test with an especially build "box" running at 32/192 (I tried my damndest to make that one disappear:)) Every stage of this process was a/b'd with the analogue counterpart.
This process lasted for 15 days, full time, until we reached concencus on a point where all in agreement and everything was tested, measured and documented to the hilt.
In those tests we found a point where digital equaled the best of analogue technology - where it sounded and felt the same to a human. I won't talk about whats possible beyond that, lets just say it sounds nice.

Now the interesting thing there is that you run one signal to 2 recording media, one direct to magnetic tape, which has a limitation associated with it. The other through a conversion to a digital media which also has a limitation associated with it, namely its frequency ceiling. Here is the thing though:
Lower quality digital, measured in every way possible, from graphs to spectrum analysers, sounds like sandpaper, hard. This hardness is mainly perceived to be in the midrange area. Increase the quality of digital, and this effect gets less and less, and at a point in 24/48 with clock accuracy and an accurate conversion with a high dynamic range, the measurable effect is gone. It will still sound a little bit brighter, but this is not in the mid / high's, even though it sounds like it. Measured it is throughout the spectrum, and it is just a much greater level of transparency all round, from low to high.

Why am I typing all this? Is anyone really interested in it?
 
Very Interesting..

Final stage was a test with an especially build "box" running at 32/192 (I tried my damndest to make that one disappear
Click to expand...
We will all be there in a few more years !
I am really interested in your study. Was it ever documented and published anywhere ?
I do believe that, even though 16/44 can sound "very good", there is an inherent limit that sensitive ears can clearly distinguish. 32/192 would probably finally put those limits to bed (with accurate master clocks, of course ! :)

Rick
 
No. Nothing of it, which really sucks, because it was very comprihensive and discovered some really good stuff. I was "the recording engineer" and by far the least technical person there. It was paid for by some of the big consumer electronics manufacturers, it showed a lot of flaws in consumer goods, I think that's why.
It certainly taught me a lot.

I think we are a long way off anything 32 / 192. 32 is ideal, especialy for processing, but most of that is floating point, naturally. The problem would be storage, and format. Storage as it would eat disk space, format as none of the current standards of connectors facilitate this level.
 
You must log in or register to reply here.
Back
Top