Another Question About "Tube Pre-amp"

Tom Olson · Post by **Tom Olson** » 15 Aug 2004 9:08 am

Let's say you're running a guitar into a digital signal processing device such as a "POD" or the like, and you've got the POD set to emulate some sort of tube amp with some sort of speaker cabinet effect on top of that. Then, from the POD you're running the signal into a digital recording device and digitally recording it.

I've heard that tube preamps are good for "warming up" the signal. However, in the scenario above, would it really be able to warm it up? I admit that I have no knowlege in this area, but my gut instinct tells me that since you've already got a signal coming out of the POD that's an emulation of a tube amp signal, that would not be much different than recording the tube pre-amp'd signal into the digital recorder (which would end up being a digital emulation of a tube amp signal).

Am I wrong? If so, can anyone explain how the digitally recorded signal from the tube pre-amp will be significantly different from the tube amp emulated signal from the POD? Thanks for your help.

David L. Donald · Post by **David L. Donald** » 15 Aug 2004 12:50 pm

For one thing, tube emulation is just that, not the real thing. It is a decent aproximation, but not quite.

Same with cabinet emulators.
Pretty good, but you can't change it by moving a mic or aiming the cab differently in a room.

I do use Line 6 plug in Amp Farm a lot, because it is flexibl AFTER a recording,
but I do the guitar through a tube preamp BEFORE the recording chain most of the time, before the plug in.

I think this address many of your questions.
But I am sure more will pipe up here shortly.

Tom Olson · Post by **Tom Olson** » 15 Aug 2004 2:37 pm

Hi David -- thanks for the post. I never thought of putting the pre-amp before the POD.

But in any case, let me explain my question a bit further. Like I said, I don't know a whole lot about the subject so this is educated guessing at best on my part. But -- if you take a signal coming out of a tube amp or pre-amp and you record it DIGITALLY (on a digital recording device), you end up with a digital representation of a signal produced by a tube amp or pre-amp.

But, this is essentially the same thing that you get from a POD, right? At least that's my understanding of things.

So, regardless of whether you have a real tube amp and/or a tube emulation from a POD or something similar -- if you're recording the signal with a digital recorder it seems to me that the sound quality of the recording won't be that different because the recorded signal is digital.

Or, stated another way, if you take a signal produced by a digital tube emulation device and also run it through a tube pre-amp, BUT you record it with an analog recorder, you WILL be able to tell a difference because the recorded signal is analog.

However, maybe my "logic" is flawed.

Jon Light (deceased) · Post by **Jon Light (deceased)** » 15 Aug 2004 2:51 pm

Digital recording is not emulation. It is as representative of the original sound as digital photography is of the original image. Digital emulation as in tube emulators, speaker emulators, reverb emulators, etc., is just that--a clever fabrication. And depending on the product it can indeed be pretty damn good. But good digital recording will differ from analog only in that analog provides some inaccuracies and distortions that happen to be very pleasing to the ears--described as fatness, warmth, etc. Good digital recording of warm tube sound will give you warm tube sound. IOW, it will accurately represent the source signal. If it doesn't, there is something wrong in the recording chain.
So your premise is flawed (said without any intended attitude).

David Doggett · Post by **David Doggett** » 15 Aug 2004 4:21 pm

To add to the complexity of this problem, capturing the digital emulation of the tube sound requires a very high fidelity system all the way through, including the speakers. Therefore, if you run a good digital tube emulation through a good board and good monitors, it will sound pretty good, likewise if you run it through a good PA with good full spectrum speakers including tweeters. Paradoxically, if you run it through a guitar amp and guitar speaker, it may not sound so good, because it is not a hi-fi full spectrum system. For example, I found my POD XT Twin emulation sounded pretty good through my Nashville 400 at low volume, where the system is fairly accurate. But the louder I played the amp, the more it sounded like a stressed out solid state amp and not a tube amp.

If you look at the Line-6 guitar amps that are designed to reproduce the amp emulations, they are very high-powered clean hi-fi systems compared to other guitar amps. If you want to have good amp emulations, you might do better to forget about guitar amps and either run through the PA or a powerful hi-fi acoustic amp. [This message was edited by David Doggett on 15 August 2004 at 05:24 PM.]

Brad Sarno · Post by **Brad Sarno** » 15 Aug 2004 5:18 pm

Well for the sake of arguing, I'd argue that much digital recording is actually less true than an analog recording. Analog may have it's noise and other issues, but when compared to your typical computer digital recording setup, it will introduce fewer artifacts into the sound. Digital has a wide range of quality. 16bit soundcards and older cd players have sloppy clocks (jitter) and poor quality anti-alias filters. Even cheap 24bit digital will introduce some very non-musical artifacts into the audio.

Digital may give you low noise floor and "flat" frequency response up to about 18kHz, but unless it's GOOD digital, the nasty digital artifacts can often overshadow the music. Even to this day, the high budgets will still often track and mix to analog tape and save the digitizing till the end in mastering.

In digital, since you're taking a snapshot of the audio waveform periodically, typically 44,100 times per second, the timing of WHEN those many snapshots are taken is critical. If the clocking is sloppy then you are taking the snapshots at inexact times thus losing the integrity of the original sound forever. Tape won't do this. There is wow-and-flutter, but the sonic effects of jitter can be much worse than tape motor speed fluctuations. Good, low-jitter digital conversion is the way to make digital sound great.

Also, the big gripe that so many digital haters have is the issue of anti-alias filtering. If you take the CD digital standard of 44.1kHz sampling rate and divide it by two you get 22.05kHz which is your theoretical highest frequency possible because it takes two samples to define a complete wave. BUT, you have to very steeply filter out any sound above that frequency or you'll have some very unpleasant overtones intermodulating with the music. The problem is that to make a filter that steep so as to allow the music pass thru up to about 20kHz, but not pass anything above that requires the kind of filtering that creates ringing overtones. The sound of these ringing tones is what many call the "CD Sound". It's that crispy sound where a "s" sounds like a stick hitting a cymbal and sounds like a pick hitting a string and sounds like any other bright sound. CD's played thru average gear will have this crispy "haze" all over it.

So my argument is that digital is not necessarily as true or as good as analog unless it's good digital. 96kHz sampling rates at 24bits with low jitter sounds real nice though. Cheap digital sounds awful and can take a nice warm steel tone and ruin it.

Brad Sarno
Blue Jade Audio Mastering

Tom Olson · Post by **Tom Olson** » 15 Aug 2004 9:52 pm

Digital recording is not emulation. It is as representative of the original sound as digital photography is of the original image.

So, because digital data is digital data, at least theoretically, you can generate a digital representation of a photograph that's not really a photograph at all, but is simply artificially generated from a computer software "model" that tells the digital display what to display in order to look like an actual photograph -- such as in some computer games and flight simulators, and other computer animation, etc. etc.

That is, it doesn't really matter what the source of the signal or data is. If you have some sort of "software" to digitally model the signal, you can (at least theoretically) produce a signal (or data) that will be substantially the same (or at least so similar as to be practically the same) as an analog signal (or analog data) of the actual source, provided that the modeling software is accurate enough.

Good digital recording of warm tube sound will give you warm tube sound.

So, based on that, it seem like you could, at least theoretically, digitally generate a signal in a digital processing device that is substantially the same as an analog signal of a guitar played through a tube amp, AS CAPTURED ON DIGITAL RECORDING MEDIA. My understanding is that some of the DSP's (like the POD) are extremely accurate as far as what they are modeling.

I guess the question now becomes, "how accurate is the modeling of the DSP's (like the POD), and what is the "density" or "resolution" (for lack of a better term) of the modeling.

If the answer is "very accurate" and/or that the modeling "resolution" is greater than that of a home CD audio system, then I still don't understand what a pre-amp will add when recording digitally.

However, if the answer is "not very accurate" then I can certainly understand what the tube pre-amp will add.

T. C. Furlong · Post by **T. C. Furlong** » 16 Aug 2004 3:54 am

Tom,

I think Brad hit the nail on the head. No matter how accurate the emulation is, the bi-products of digital will always be present. As I understand it, there are filters in digital signal processing used in the audible spectrum that are very steep and will intoduce phase shift. All filters cause phase shift. There is also processing time involved in all DSP's. Add it all together and in the real world, it can't possibly produce a faithful representation of a guitar performance when compared to an analog path. Not to say that the digital emulation is not valid, just not the same.

Coincidentally, today I am working with the man that developed "the" high resolution measurement platform for digital audio. He has confirmed that the new "high resolution" 96K sampling rate found in modern digital recording actually reduces signal to noise ratio to about 72dB. (sounds nasty to me)The point is that there are trade-offs and by-products in any technology.

To support the audio/photography analogy. What if Ansel Adams had a digital camera?

Tom Olson · Post by **Tom Olson** » 16 Aug 2004 9:32 am

T.C. -- thanks for the post. I'd have to say that I actually agree with everyone who's posted so far to the extent that I understand what is being said. But, I think my problem is that I haven't been able to phrase my question in a way that makes it understandable. And, of course, it may be that I'm not understanding the points being made by the posters above.

I don't mean to beat a dead horse, but one more try:

I'll try this analogy --

Let's say you take the same photograph of something with two cameras: 1)a digital camera; and, 2) a 35mm camera.

Then you make poster-sized prints from the photographs taken by each of the cameras.

Now, you use a second digital camera to take a photograph of each of the poster-sized prints. Finally, you print a hardcopy of each of the two photographs taken by the second digital camera.

At least the way I see it, you can analogize this scenario to an audio signal path as follows:

The original thing that you took the photographs of is the raw signal from the instrument. The 35mm camera is an analog device such as a tube pre-amp. The first digital camera is a digital modeling device such as a POD. The second camera is the digital recording device. The printing of the last two photographs taken by the second digital camera is the play-back process (home CD player).

Now (at least as I see it), if the resolution of the FIRST digital camera is GREATER than the resolution of the SECOND digital camera -- OR -- if the resolution of the PRINTER used to print the last two photographs is LESS then that of either the first or second digital camera, THEN -- you won't be able to tell any difference between the photograph that passed through the first digital camera and the photograph which passed through the 35mm camera (they will look exactly the same).

My understanding is that the digital playback equipment (home CD player) and the CD itself are sort of the "weak link" in the chain as far as the audio resolution. If that is true, and the modeling of the POD is sufficiently accurate, then it seems like the pre-amp signal and POD signal would not be distinguishable when played through an average home CD player system.

However, this would not be the case if you were using analog recording equipment, analog recording media and playback equipment.

David L. Donald · Post by **David L. Donald** » 16 Aug 2004 11:58 am

What Jon, Brad, DDogg, said.

As far as the camera analogy. It is not to bad.

A silver based 35mm photo will be as accurate as each grain of silver and it's dyes for each color.
If you scan it on a Scitex or Hell etc, at a pixel density where the pixeles are 3-4 per silver grain,
then it would look essentially the same.

IF the printer has the same density of DPI....

You can scan a superfine grain film like, Technical Pan, with a pixel bigger than the grain and it will sort of average off, but you lose resolution and accuracy.

Same with a digital audio sample
If you sample at 24 bit 192khz, then it will be darned close to analog, but at the price off extreme disk usage and lower track count.

But most ears will not tell the difference...
until it is down sampled to say DVD 96khz audio
Which will be aceptable to all but the most neurotic ears.

But down sample it to 24 bit, 44.1 CD audio, and you have again lost a lot of the signal.
So the modeling of the down sampling convert er becomes critical in how it averages the more complete signal for a smaller bandwidth.

This is like the bigger than grain pixel. If the averging model is right then it gets the best aproximation.

As with all prcess, Garbage In = Garbage Out.
You system is only as good as it's weakest link.
If that weak link is early in the chain, there is no going back and replacing that audio signal lost.

So if you have your guitar into a tube preamp, like Brad's BlackBox, then a good non loss pedal, and then a normal use tube preamp, before sampling,
you get the interaction of the tube with the pick up and then sample that sound.

This would be preferable to plugging in solidstate and then sampling it.

If you put the post Bbox signal into a POD for example,
you still have the sampling to digital, but with a warmer interaction like a tube amp, before sampling.
If you then choose an amp emulator, it will be seeing a signal based on a more amp like waveform.

If I had my druthers....
I would have a Black box, with a tuned resonance cable,
the exact distance to be most sonorus with a Truetone pick up.

And then the Hilton in and out of that Bbox,
One output to my amp with 2 mics on tube
preamps in a sweet room.

One output to my tuner
and one output via XLR out to my Aphex 1100 Thermionics tube preamp / 96k converter.

The 3 signals then sampled and proccesed.

But with the benifit of the best posible REAL tube interaction with the pick up. Before any other electronics or digital conversion.

There are a number of engineers who send their
digital mixes out to stereo tube sand back into the workstation before a final mixdown.

William Peters · Post by **William Peters** » 16 Aug 2004 12:53 pm

You will probably think I am nuts, but I don't like the sound of tubes... I much prefer clean, clear solid state. I fail to hear "warmth"... all I hear is distortion.

The tubes in my Tubefex, for example, create way too much intermodulation distortion, even at low drives. Play single notes, and it sounds pretty good. Play an interval or a chord, and in comes the distortion. So I never use the tube effect. I use to have a Peavey transtube processor which also was too distorted.

In fact one of the reasons I don't like my V-amp Pro is that the digital modeling of the supposedly clean tube amps has too much intermodulation distortion for my tastes, just like the amps it is modeling.

Long live solid state!

------------------
Bill

http://www.wgpeters.com
Mullen RP SD12U, ART-SLA, Tubefex, Vamp-Pro

Ron Randall · Post by **Ron Randall** » 16 Aug 2004 1:17 pm

Question for David Donald.

If I had my druthers....I would have a Black box, with a tuned resonance cable, the exact distance to be most sonorus with a Truetone pick up.

David, I learn so much from reading discussions like this. I use the SGBB, a Trutone, a Hilton, to a tube amp. What is a tuned resonance cable? How do I make one?

Ron

Brad Sarno · Post by **Brad Sarno** » 16 Aug 2004 4:22 pm

William, you need to listen to some "good" tube gear before you leap to that conclusion. Most of the albums you listen to were likely mastered thru tubes, and many of the studio microphones you hear your favorite singers thru were also tube mic's. Tubes can sound much cleaner to the ear than solid state actually. The V-Amp2 is full of horrible amounts of IM distortion that doesn't at all exist with the amps they're supposedly modelling. Overdriven tubes sound distorted, but cleanly run tubes can sparkle with high fidelity and with greater depth and detail than many transistors can offer. It's just that "good" tube gear is often very expensive.

Brad Sarno

William Peters · Post by **William Peters** » 16 Aug 2004 5:28 pm

Brad,

Would you expect that there is something wrong with my tubefex then? Should I replace the tubes? Really, it isn't very clean at all... I think it has Sovteks in it now.

Bill

Brad Sarno · Post by **Brad Sarno** » 16 Aug 2004 5:43 pm

William, the Tubefex is really only partially a tube unit. It's mostly solid state digital. I dont have a lot of experience with it so I don't know exactly how clean it can run. Brand of tube shouldn't be the issue. Gain structure (how your inputs and gain are set) is usually how you clean or dirty up a signal path. Mike Brown could probably help you. The TubeFex also digitizes the signal so you've got your signal being converted and re-converted as well as sending it thru numerous IC chips and tube stages, so by the end your guitar signal has undergone a LOT of processing. I'd ask other TubeFex users if they can get a truly clean signal thru that thing.

Brad Sarno

William Peters · Post by **William Peters** » 16 Aug 2004 5:56 pm

Brad,

Fair enough. I bought the Tubefex used from a fellow forumite, so I don't really know what it sounded like new. I sure like the rest of it though. As I've said before, I think the V-amp is pretty much crap.... just a bad, cheap design.

------------------
Bill

http://www.wgpeters.com
Mullen RP SD12U, ART-SLA, Tubefex, Vamp-Pro

Jan Dunn · Post by **Jan Dunn** » 17 Aug 2004 8:33 am

In a similar vain, how would the blackbox compare to a tube based clean boast pedal? I have a FatDrive (with a 127AU in it)that I'm happy with though I find it a tad noisey at times.

Brad Sarno · Post by **Brad Sarno** » 17 Aug 2004 6:26 pm

Jan, the SIB FatDrive is a GREAT pedal. I use the SIB Echodrive for guitar, best delay pedal out there IMHO. I love that brand. The FatDrive is a MUCH busier circuit (than the Black Box) that has 3 bands of EQ and lots of gain, bright control, and whatnot. Actually the FatDrive might make a decent steel preamp, but maybe not the best pickup impedance matcher/buffer like the BB. The two might make a wonderful combination with the BB before the volume pedal and the FatDrive as a preamp at the end of the signal chain to tone shape, control gain, and drive an amp.

The Black Box uses better quality and much more expensive cap's and resistors, has a much lower noise floor, near-unity gain, has a more optimal input impedance for steel pickups, has fewer stages, less circuitry, no pot's or switches to get dirty or go bad, is handmade in-house, and has a real cool blue light on it

Putting a 12AU7 in the FatDrive is probably a good idea for cleaning it up a bit for steel, but it's still not going to be as clean and quiet as a SGBB due to the nature of the design of the FatDrive as a hi-gain, tone shaping device.

Honestly the SIB stuff is great and Rick Hamel is a brilliant designer and a super-friendly and helpful guy. The SIB built quality is actually what inspired me when building the Black Box. Solid steel, powdercoated, plated thru circuit boards, etc. If you use the FatDrive, it may let you eliminate the need for any other preamp since it has both gain and EQ in it.

Good Question!

Thanks,
Brad Sarno

David L. Donald · Post by **David L. Donald** » 17 Aug 2004 11:28 pm

I am in agreement with all Brad said.
It depends on the design and the quality of components.

I have a TubeFex and it is a bit distorty, but not so bad I don't use it for clean. I think better tubes properly biased would help it.

Ron, a tuned resonance cable is matched electronically to the preamp and impedence of the pick up, in this case, or other source unit.

I was being a bit audiophile esoteric with this comment.

Basically, for a given cable length
and type of wire metal composite,
and it's thickness,

In relation to it's foreseen frequency bandwidth, there is an optimal SET of distances to create the least resonances or ringing harmonics, and impeadance nulls in the cable.

Cables are often cut at arbitrary 2', 3', 6', 10' etc,
when 5'8.35" would sound a touch better. or 10'.245' as random number examples.
This is of course different for each usage, a guitar over 50' is not the same as a steel at 10'.

I am sure there is a way to calculate the best cable length for a Truetone to Black Box, with the optimal capacitance / impedance loading.

We're talking 1/2 mm precision here. And including the length of the lead off the pick-up in the equation too.

It is likely to be so short a cable as to be nearly irrelevant at steel guitar bandwidth, to most folks.

But I thought Brad and a few others would get a kick out of the concept.

Get a 2 or 3 foot gold tipped cable from you steel to pedal or black box and then to the pedal.
As I understand it the George L cables are properly matched for their length.

Also would want it going to a gold plug other than a guitar jack like an XLR, for better contact, or a linearness of the signal path...
Yes audiophiles can be truely obsesive.

Also at this time the other things I have added on my Black Box wish list...
don't exist, they are just that a wish list for Brad's consideration.

But I know he is thinking on all suggestions he gets. So that's cool.
I see a design for something and think what I wouild do with it, in ddition to it's original function.

But it is still very cool... just as it is.

[This message was edited by David L. Donald on 18 August 2004 at 12:44 AM.]

Brad Sarno · Post by **Brad Sarno** » 18 Aug 2004 8:09 am

It is a bit esoteric but I think it's based on transmission line theory. Back in the day impedances and loads were carefully matched, usually with transformers. The theory had to do with resonances and reactances related to the speed of electrons in a wire and the length of the transmission line (wire) and stuff. I barely understand it, but I have been told that these days since we no longer exactly match impedances, that the theory gets thrown out the window for any practical application. I know that Ed Fulawka dabbles in the idea and swears that he'll only cut guitar cords in multiples of certain lengths.

Here's a "fun" link: http://www.tmeg.com/tutorials/t_lines/t_lines.htm

Or this one: http://www.bytemark.com/products/tlttheory.htm

David, you're dangerous!

Brad Sarno

Jeff Hogsten · Post by **Jeff Hogsten** » 18 Aug 2004 10:19 am

some of what has been said is true but basically digital just doesnt color the sound as much, I have a very good digital system, hammerfall sound card nuendo good mics and preamps and cut a great sound. I remember once when cds first came out everyone talked about how much better they sounded than cassettes, so one of the audio mags picked a pannel of experts to do a blindfold test.Not one could consistently tell the difference when the were using new cassettes. In fact they almost always picked the one that was played a little louder. I tend to think if we sit down and listened to some of the songs on the charts today if anyone could tell which ones were recorded digitally and which ones were recorded analog. I would say no not on a regular basis.How many great and I mean great sounding records have been cut on pro tools. Some people have just become partial to analog and I think it is from hype in the beginning of digital, consider the following comment Brad made

<So my argument is that digital is not necessarily as true or as good as analog unless it's good digital. 96kHz sampling rates at 24bits with low jitter sounds real nice though. Cheap digital sounds awful and can take a nice warm steel tone and ruin it>

so in other words digital is as good if it is good digital but bad digital sounds bad well what about bad analog I guarantee you I can find some tube amps that will make anything sound bad steel or whatever.Sample rate doesnt always mean it is going to sound better, I can take a great player and a great instrument and cut it at 44 and get a great sound.I have heard pure garbage cut on analog but no one said it sound like crap because analog stinks. It sounded like crap because the people cant sing and play and the engineer didnt know what he was doing. I have a cd of the last road group I played for the Perry Sisters that was done in nashville on black faced adats dumped to nuendo and mixed and mastered on sound forge that sound as warm and musical as you can get, maybe Ill post a mp3 and you can judge for yourself, of course the fact that they had great singers and Gary Prim on Keys and Kelley Back on guitar and Scott Sanders on steel and Jerry Kroon on drums didnt hurt anything. What Im tring to say is you can cut great warm sounding records on digtal so why is it getting bashed because someone likes their steel amp better than a pod, that doesnt mean digital stinks it means line 6 hasnt got their technology to where you cant tell the difference yet but I think they will I honestly believe with the advances being made with products like the variax being introduced that the day is coming where anything will be able to be modeled to where you cant tell the difference. Read the report this month in sound on sound mag that reviews the new focusrite pre amp that models all of the top of the line vintage pre amps the reviewer is one of the most respected engineers in the business and had the originals to compare it with and in a lot of instances he could tell no difference in the original and the simulated version at all. wouldnt it be awsome in a few years to sit down at you steel and call up a pushpull or a shobud or a rick lap steel, or a fender 1000 I believe the day is just around the corner so digital is here to stay and like I said I doubt if very few if any people on here can consistently tell if a song was recorded on digital or analog, just my opinion

Jeff

Jim Peters · Post by **Jim Peters** » 18 Aug 2004 12:21 pm

Well said Jeff, Brad, you are truly dangerous! After reading on the subject for a long while, I put a copy of JHE "Up from the skies" on the turntable, and AB'd it with a CD version, fully expecting to be blown away by the analog, but wasn't. I think it's great that there are some that live on the edge of techno, and can hear those differences, and I think I have pretty good ears, but some of this stuff is too anal for me. BTW, Brad, your non vintage Strat sounded pretty damn good thru your "vintage" Pod at Venice the other week! I think my Tele sounds pretty good thru my solid state, non vintage,cheap component tube screamer. I guess everything has its place. As they say, its only IMHO. JimP

Donny Hinson · Post by **Donny Hinson** » 18 Aug 2004 6:18 pm

Tom.

Two facts of life:

A solid state amp will never sound just like a tube amp, or vice versa.

A digital signal or recording will never sound just like an analog one, or vice versa.

You can add any kind of fancy device anywhere in the line you want to, and it still won't change those two facts.

"Stretching a duck's neck won't make him a goose."

Tom Olson · Post by **Tom Olson** » 18 Aug 2004 6:53 pm

Donny -- I agree.

Jim Palenscar · Post by **Jim Palenscar** » 18 Aug 2004 9:12 pm

"In fact they almost always picked the one that was played a little louder"
~ Fletcher-Munson rises once again