electro-music.com   Dedicated to experimental electro-acoustic
and electronic music
 
    Front Page  |  Radio
 |  Media  |  Forum  |  Wiki  |  Links
Forum with support of Syndicator RSS
 FAQFAQ   CalendarCalendar   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   LinksLinks
 RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in  Chat RoomChat Room 
go to the radio page Live at electro-music.com radio 1 Please visit the chat
poster
 Forum index » DIY Hardware and Software » Developers' Corner
Tube Amp Models
Post new topic   Reply to topic Moderators: DrJustice
Page 3 of 4 [87 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
Goto page: Previous 1, 2, 3, 4 Next
Author Message
FreeElectrons



Joined: Apr 30, 2008
Posts: 9
Location: Denmark

PostPosted: Fri May 02, 2008 5:02 pm    Post subject: Reply with quote  Mark this post and the followings unread

Most tubes, save for those few types with known design issues, are way more reliable than what people tend to 'remember'. Many of the tube problems back in the day were actually due to problems located elsewhere in el cheapo consumer electronics. In particular bad passive components indirectly led to premature tube failure due to massive overload. As tubes are generally fairly resilient to power overload, they will frequently last a fair while even when run wildly out of specs.

The classic problem with consumer electronics from the tube age are bad capacitors with paper insulation, and power resistors to a lesser degree. The paper in the capacitors deteriorates over time due to moisture in the air, causing a 'leaky' capacitor, and eventually indirect tube failure. Even today many people try to 'restore' old radios by buying new 'tube sets' on ePrey and elsewhere, believing that the problems are fixed by replacing the 'unreliable' tubes. Sure enough, the old radio may even play for a while, while the new set of tubes silently scream out their pain over a very short lifespan. Silly semiconductors would just have fried instantly if given the same treatment.

My daily driver FM receiver is a Grundig of Germany from 1956. When I restored it, I *did* replace one of the original tubes with another used one, though it would have worked without replacement. But all the rest, including the full class A audio output power tube are the factory original ones. Of course I replaced about 25 faulty passive components before putting the radio back into service...

It just so happens that I have another 13 of the same audio output tube as used in the above radio. The particular type, EL84 / 6BQ5, was extremely common in consumer electronics in the fifties and sixties. The type was almost exclusively used in class A audio output stages, running at full dissipation 100% of the time. All my 13 used EL84s works perfectly, as tested on my transconductance tester, which also tests for the presence of gas leaks and heater/cathode shorts/low resistance.

Operated within spec and in circuitry designed for the peculiarities of tubes, they last thousands of hours, usually much longer.

When was the last time you saw the picture tube in a TV or a computer monitor die and needing replacement?

Which leads to my next comment: If you don't have any experience with tubes, then your guitar amp simulation is in trouble IMHO. I don't believe the answer to your chosen problem lies in a few neat formulas you might find on the 'net, but more in an understanding on how tubes actually work and were (properly) used by the big manufacturers like Mullard, Philips, GEC, Telefunken, RCA, GE, Raytheon and their likes.

Frank.
Back to top
View user's profile Send private message
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Fri May 02, 2008 6:15 pm    Post subject: Reply with quote  Mark this post and the followings unread

FreeElectrons wrote:
Which leads to my next comment: If you don't have any experience with tubes, then your guitar amp simulation is in trouble IMHO. I don't believe the answer to your chosen problem lies in a few neat formulas you might find on the 'net, but more in an understanding on how tubes actually work and were (properly) used by the big manufacturers like Mullard, Philips, GEC, Telefunken, RCA, GE, Raytheon and their likes.

Frank.


That's very frank, Frank! But joking aside, I enjoyed your history lesson above, it's always a good read when someone entertains us with tales of yesteryear. And I think you're right, if i don't have any experience with tubes then my guitar amp simulation is in trouble... So I will get experience with tubes! I can read about them and learn different aspects of them from the web and from books. I can design and simulate them. I can continue to model them better and better also. It's a gradual learning process and I'm sure it will be enjoyable for the most part. That's the plan anyway, wish me luck!

_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Fri May 02, 2008 6:25 pm    Post subject: Reply with quote  Mark this post and the followings unread

I also have some better news about that Fourier signal warmer thingie that I made. I had to really tweak it and track down a few glaring errors in my coding, but eventually I got it working. It produces a sinusoid as shown in the right half of the photo, ramped up to a pointy thing in the left half. You can hear the 10-second ramp three times in the audio file.

Although I do warn that there could still be something wrong here, it does look in the waveform like the signals are believable and the sound has a quality that varies in a pleasing way. So it seems OK at first glance...


Tube_FFT_Warmer.jpg
 Description:
Audacity waveform view of Fourier Signal Warmer output signal at transition of phasor from 1 to 0.
 Filesize:  14.21 KB
 Viewed:  15855 Time(s)

Tube_FFT_Warmer.jpg



Tube_FFT_Warmer.mp3
 Description:
The sound of "warmth", ahhh! Or so I think, you decide!

Download
 Filename:  Tube_FFT_Warmer.mp3
 Filesize:  468.68 KB
 Downloaded:  1340 Time(s)


_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Sun May 04, 2008 4:18 pm    Post subject: Reply with quote  Mark this post and the followings unread

Well, I'm taking a break and asking for comments. Specifically, I'd like for my next step to be expanding that signal warmer to include arbitrary music signals, warming them up in a tube-like way. Also I'd like to entertain your comments on whether this "signal warming" approach is any good.

Naturally once that is all said and done, I'll get back to really modifying the device models. I have learned more about what a pentode does and why it does it, so I can perhaps model that next. I also have some thoughts about saturation modeling so that's all good.

Just waiting for your comments and when the mood is right, or the phase of the moon, I'll get back into it for sure! Comments anyone?

_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Mon May 05, 2008 2:53 am    Post subject: Reply with quote  Mark this post and the followings unread

I altered the tube-warming program to work on arbitrary signals and fed a guitar solo into it (from my Guitar Lab software). It does add some sound to the clean guitar signal, but honestly it sounds like the guitar is wrapped in aluminum foil which is vibrating, haha. Not so good first try so I won't bother to post the input and output files here. But do let me know what you think of the sample I provided, I'd like to get comments on that, thanks.
_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
blue hell
Site Admin


Joined: Apr 03, 2004
Posts: 24075
Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320

PostPosted: Mon May 05, 2008 12:48 pm    Post subject: Reply with quote  Mark this post and the followings unread

It sounds like noisy mains hum Shocked Laughing

But it's hard to make out a meaning from this sample, as it's isolated, just a single tone and there is no reference to the original sound.

When you say it does not work as intended on your guitar sounds I think that statement has more meaning than the sample you posted .. I would be interested to hear a guitar wrapped in alu-foil BTW as I've always liked to put tea spoons or bits of cardboard between the strings of a guitar.

_________________
Jan
also .. could someone please turn down the thermostat a bit.
Posted Image, might have been reduced in size. Click Image to view fullscreen.
Back to top
View user's profile Send private message Visit poster's website
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Mon May 05, 2008 1:05 pm    Post subject: Reply with quote  Mark this post and the followings unread

OK, Jan, below are the unmodified original synthesized guitar sound, plus the enhanced one. I see what you mean about the original sample being like ground noise or power supply noise. Maybe this technique is not so hot after all. Just exploring, and occasional failures are part of the picture!


Tube_Warmer_in.mp3
 Description:
The "clean" guitar signal recorded from Guitar Lab.

Download
 Filename:  Tube_Warmer_in.mp3
 Filesize:  466.12 KB
 Downloaded:  1342 Time(s)


Tube_Warmer_out.mp3
 Description:
The "distorted" output signal with harmonics added.

Download
 Filename:  Tube_Warmer_out.mp3
 Filesize:  468.57 KB
 Downloaded:  1269 Time(s)


_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
blue hell
Site Admin


Joined: Apr 03, 2004
Posts: 24075
Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320

PostPosted: Mon May 05, 2008 1:26 pm    Post subject: Reply with quote  Mark this post and the followings unread

The purpose of doing the experiment is to determine the nature of it's failure, as Jan Booerman used to say, except that he said it in Dutch of course.

It's not a tube warmer that I hear, and not alu-foil either, but I do like the effect!

Cooledit has a spectral noise removal tool (I think Audacity has it too), you point it to an example of noise and then it builds a filter from that to silence it. When you apply that filter to a signal in heavy reduction mode it will produce similar sounding stuff. I use that quite a lot actually.

I think it's inherent to banding the signal frequency wise?

_________________
Jan
also .. could someone please turn down the thermostat a bit.
Posted Image, might have been reduced in size. Click Image to view fullscreen.
Back to top
View user's profile Send private message Visit poster's website
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Mon May 05, 2008 1:37 pm    Post subject: Reply with quote  Mark this post and the followings unread

Well, what it's doing is making reduced-amplitude harmonic copies of each signal in a frequency bin of about 10 Hz. So it reads the frequency bin of 10 Hz and puts k*amplitude into 20 Hz, k*k*amplitude into 30 Hz and so on up to six harmonics. Then it starts again with 20 Hz as the fundamental, and repeats, etc.

I know that technique is creating multiple overlaps of signal enhancement, but it was the only way i could think to enhance a non-periodic spectrum of sound... Perhaps it needs more work.

At any rate, I've thought of how to do saturation with the tube model so I'll get to that tonight or tomorrow, but please keep the comments rolling in, I appreciate them!

_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
DrJustice



Joined: Sep 13, 2004
Posts: 2114
Location: Morokulien
Audio files: 4

PostPosted: Mon May 05, 2008 3:43 pm    Post subject: Reply with quote  Mark this post and the followings unread

I would not call what I hear a 'tube warmer', but I think the effect in itself is quite interesting. There's certainly nothing wrong with a bit of alu foil once in a while Very Happy

It's an interesting choice to go to the frequency domain for this. I suppose one could do a (multiple?) fundamental estimation and work from there. The repeated harmonic copies technique, and variants on the theme, could be a new effect class in itself though!

I would imagine that tube warming is better suited for time domain processing. But that is just a hunch...

DJ
--
Back to top
View user's profile Send private message Visit poster's website
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Mon May 05, 2008 5:54 pm    Post subject: Reply with quote  Mark this post and the followings unread

OK, so I've got two clear opinions that adding in harmonics does not appear to be a promising tube-substitute. So there must be more to it than that. No problem, I can get back on the main track of things. Next I'd like to model saturation using our existing amplifier model. We have numbers for the (not so) linear region, I can expand the sweep of input voltage to cover some saturation and make a table accordingly. Then the proper use of gains and biases will position the overdriven input signal just right on the tube's transfer function and we'll get some nice saturation. I can work with the existing two files that we were using already. Maybe tonight, maybe tomorrow, I'll do this one next...
_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Tue May 06, 2008 8:57 am    Post subject: Reply with quote  Mark this post and the followings unread

OK, it took a lot of finagling around with gains and offsets and such, but I managed to get the model to properly simulate a saturated tube, or so I believe. Looking at the photo below, you see the tube-saturated sinusoid on the left half and the newly beginning zero-amplitude sinusoid on the right half. This waveform is exactly the shape you would expect from an overdriven tube amp, including the rounded corners at the maxima. Listening to the attached audio file will give you a feel for how it sounds. I'm happy with the results - are you?


Tube_Amp7.jpg
 Description:
The tube-clipped waveform at its peak amplitude.
 Filesize:  12.75 KB
 Viewed:  15772 Time(s)

Tube_Amp7.jpg



Tube_Amp_out7.mp3
 Description:
Tube Amp sound in response to overdriven sinusoid.

Download
 Filename:  Tube_Amp_out7.mp3
 Filesize:  469.79 KB
 Downloaded:  1213 Time(s)


_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
DrJustice



Joined: Sep 13, 2004
Posts: 2114
Location: Morokulien
Audio files: 4

PostPosted: Tue May 06, 2008 10:22 am    Post subject: Reply with quote  Mark this post and the followings unread

Inventor wrote:
I'm happy with the results - are you?

Best so far in this thread! But, as Jan mentioned a few posts back, it is much easier to evaluate the effect with a phrase played on an instrument. Perhaps continue to use that previous sample as a refererence? If testing for authenticity, then it might even be an idea to use an actual electric guitar sample.

DJ
--
Back to top
View user's profile Send private message Visit poster's website
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Tue May 06, 2008 2:39 pm    Post subject: Reply with quote  Mark this post and the followings unread

OK, I did the best I could. Since I don't have an actual electric guitar sample, I used a Guitar Lab sample - the same one as before. I'll have a guitar in a few weeks so then I can test with real sound.

I had to expand the amount of saturation that was modeled before getting good results, and you'll notice from the waveform that the clipping is pronounced but I can hardly hear it, strange.


Tube_Amp10.jpg
 Description:
The distorted "guitar" signal.
 Filesize:  20.02 KB
 Viewed:  15759 Time(s)

Tube_Amp10.jpg



Tube_Amp_out10.mp3
 Description:
The tube-model distorted output signal.

Download
 Filename:  Tube_Amp_out10.mp3
 Filesize:  469.79 KB
 Downloaded:  1176 Time(s)


_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
jksuperstar



Joined: Aug 20, 2004
Posts: 2503
Location: Denver
Audio files: 1
G2 patch files: 18

PostPosted: Tue May 06, 2008 5:42 pm    Post subject: Reply with quote  Mark this post and the followings unread

did you have a ring mod and reverb on there as well? Unless it's just the original sample.
Back to top
View user's profile Send private message Visit poster's website
blue hell
Site Admin


Joined: Apr 03, 2004
Posts: 24075
Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320

PostPosted: Tue May 06, 2008 6:01 pm    Post subject: Reply with quote  Mark this post and the followings unread

jk: see tube warmer in - a couple of posts above.

Think this would be a nice distortion for on an acoustics based guitar .. erm .. a real one I mean Wink

_________________
Jan
also .. could someone please turn down the thermostat a bit.
Posted Image, might have been reduced in size. Click Image to view fullscreen.
Back to top
View user's profile Send private message Visit poster's website
DrJustice



Joined: Sep 13, 2004
Posts: 2114
Location: Morokulien
Audio files: 4

PostPosted: Tue May 06, 2008 7:25 pm    Post subject: Reply with quote  Mark this post and the followings unread

Inventor wrote:
I had to expand the amount of saturation that was modeled before getting good results, and you'll notice from the waveform that the clipping is pronounced but I can hardly hear it, strange.

It is louder than the unprocessed version, which must be taken into account on auditioning. It sounds like there's a bit of compression, and there's distortion as far as I can make out. Kind of what you want then... And as noted there's the reverb; perhaps it's better to do without it?

IMO you're getting somewhere now Smile Look forward to hearing that with a guitar (and the inevitable further algorithm development!).

DJ
--
Back to top
View user's profile Send private message Visit poster's website
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Tue May 06, 2008 8:10 pm    Post subject: Reply with quote  Mark this post and the followings unread

Well, that's the funny thing, you all detected the ringing but it comes from the tube amp model only for some reason. I haven't used a ring modulator and the "clean" signal has no reverb or echo in it, yet you hear something sort of like this in the tube-amped signal. Oops, I forgot to mention that this song uses the above Tube_Warmer_In.mp3 file as its source file, the "clean" guitar.

So now I'm wondering if ringing is caused by tube distortion or if it is this wrap-around effect that occurs. See, the table-lookup gizmo that I'm using is imperfect. If you exceed the table values, it wraps the signal around to the other side of the output rails. So if you put in a value of 1.10, out comes -0.9 for some reason. It puts the "peaks" on the opposite rail. I had to expand the clipping zone to prevent this, and perhaps it needs even more expansion.

Well, I will play around some more and see if I can get a "tube" sound without the ringing somehow. Thanks for checking the thread and adding your comments, they are helpful.

_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Tue May 13, 2008 4:55 am    Post subject: Reply with quote  Mark this post and the followings unread

Hi folks, just thought I'd leave a quick note to let you know I haven't dropped the tube amp model project. I've been busy doing other things, primarily making a chainmaille guitar strap and visiting with my mom and chatting online.

The ringing sound I believe is due to the coarseness of the lookup table that I'm using. It's based on line-segments, and I just have fewer than 8 or 10 segments defining the entire linear region, which makes angles and doesn't model the soft tails that transition between cutoff, linear, and saturation modes. So after thinking about it for some time, I believe I will customize that table so that there are more entries in the linear region and fewer in the cutoff and saturation regions.

That should reduce the slight ringing that people heard. In addition to that I will scale the whole thing to a wider range so that even more of an effect is produced and we can hear it better. Plus also I buy my guitar in about ten days, so we can run the model with real guitar input pretty soon. I'm *really* looking forward to that!

So, like anything else that's good, we have to be patient and we'll get there soon enough. Thanks to all who've contributed and I'll keep you posted on progress.

_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
DrJustice



Joined: Sep 13, 2004
Posts: 2114
Location: Morokulien
Audio files: 4

PostPosted: Tue May 13, 2008 5:57 am    Post subject: Reply with quote  Mark this post and the followings unread

Hm.... listening again I'm quite certain that there's reverb on the input sample. Either that or something which sounds exactly like it Smile

If using a lookup table, there can never be too many entries - well, it depends on the interpolation method. What order/method of interpolation are you using?

DJ
--
Back to top
View user's profile Send private message Visit poster's website
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Tue May 13, 2008 6:39 am    Post subject: Reply with quote  Mark this post and the followings unread

H Dr. J, I am using a ChucK feature called a GenX, specifically the Gen7 version of it. You specify a table like this:

Code:
[0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.000000,
0.025000, 0.035097,
0.025000, 0.151663,
0.025000, 0.292996,
0.025000, 0.447347,
0.025000, 0.611553,
0.025000, 0.786116,
0.025000, 0.965360,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000,
0.025000, 1.000000] => gen7.coefs;


In fact, that is the actual table from the latest code. Notice that I only have 7 data points, forming 8 line segments to describe the entire linear region. That's not very good. Plus I'm spending lots of computer resources defining duplicate little line segments in the cutoff and saturation regions, duh!

Oh, I forgot to mention: it's linear interpolation, which is also not the best.

But as we said it's a process of learning and improving. So what I plan to do next is add code to the equation solver that makes one big long line segment for the cutoff region and one for the saturation region, plus make little bitty steps in the active region. That should be much better.

You'll also notice that the table varies only from 0 to 1, that is a limitation of Gen7, which was originally intended for mapping HID sensor inputs. That's why I had to do a lot of tricky trial-and-error attempts to get the biasing and gains right. I have to do attenuation and offset on the input side, then gain and reverse offset on the output side. It's still not "just right", as you or someone observed that the output is louder than the input, but I'm getting there.

Now, if I could only make some progress on this chainmaille guitar strap, then I can get back to the tube model and do what I just said I would do. BTW, the chainmaille guitar strap is over two feet long now, and it looks wicked cool! A little heavy, but that's OK by me. Keep on truckin'!

_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Fri May 23, 2008 2:58 pm    Post subject: Reply with quote  Mark this post and the followings unread

I now have a real guitar to play so I created an input file by slowly playing all six strings in sequence from lowest to highest, with no fingers on the fretboard (producing EADGBE notes). I also happen to have exchanged the Line 6 amp that the salesperson talked me into for the amp I wanted in the first place, which is a Fender Frontman 15G. The Fender amp has a beautiful tube amp simulation which I tried and it sounds a lot like my simulation in the output file here but perhaps better than mine.

Anyway, you can hear the effect most prominently on the first note and to varying degrees on the other notes. If you can hear a faint ticking sound and a faint breathing sound, that is my apartment somehow making the amp tick and myself breathing because the Apple eMac mic is on maximum sensitivity and it picks up everything including my thoughts (now *that* is sensitive).

I also looked at the waveform of the output signal and it sure looks tube-like to me, with soft clipping, though the soft bending occurs in only the upper 5% and lower 5% of the signal. Perhaps a pentode simulation, which is next on the list to do, would be more "bendy" at the extremes, I don't know.

So... success or not? You decide! It seems totally tubular to me, but what do I know, I'm biased. What do you think?


Tube_Amp_in12.mp3
 Description:
The input file, six guitar notes: EADGBE

Download
 Filename:  Tube_Amp_in12.mp3
 Filesize:  423.26 KB
 Downloaded:  1224 Time(s)


Tube_Amp_out12.mp3
 Description:
The output file after tube simulator - best representation is the first note

Download
 Filename:  Tube_Amp_out12.mp3
 Filesize:  469.79 KB
 Downloaded:  1120 Time(s)


_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
DrJustice



Joined: Sep 13, 2004
Posts: 2114
Location: Morokulien
Audio files: 4

PostPosted: Sat May 24, 2008 3:54 am    Post subject: Reply with quote  Mark this post and the followings unread

Inventor wrote:
I now have a real guitar to play

Yay! Congratulations!

Your G and B strings are tuned to G# and C. Maybe it's not tuned yet? Guitar ownership will soon get your tuning habits sharpened anyway Wink

Quote:

So... success or not? You decide! It seems totally tubular to me, but what do I know, I'm biased.

Don't worry, tubes are biased too Smile

Quote:
What do you think?

As for the exact degree of tubularity, I'll leave that to someone who really knows their references. There's certainly lots of harmonics, but probably also lots of aliasing. It does sound a bit harsh in my ears. Looking at the waveform it may seem like the clipping is somewhat brutal, and an FFT reveals quite a bit of content below the fundamental (at lest the background noises are showing up. I think that at this level of squaring off, band limiting is in order. Of course any kind of non linear function should ideally be oversampled anyway.

One test you could do is to run a slow sine oscillator sweep from 20Hz - 20kHz through the transfer function and listen for artefacts (mainly aliasing).

DJ
--
Back to top
View user's profile Send private message Visit poster's website
blue hell
Site Admin


Joined: Apr 03, 2004
Posts: 24075
Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320

PostPosted: Sat May 24, 2008 4:09 am    Post subject: Reply with quote  Mark this post and the followings unread

http://rudhar.com/musica/tuneguit/tuneguit.htm Wink

I did like the rough distortion, sounds usable to me.

@DJ, you have good ears, I just heard something "strange" with the untreated sample.

_________________
Jan
also .. could someone please turn down the thermostat a bit.
Posted Image, might have been reduced in size. Click Image to view fullscreen.
Back to top
View user's profile Send private message Visit poster's website
Inventor
Stream Operator


Joined: Oct 13, 2007
Posts: 6221
Location: near Austin, Tx, USA
Audio files: 267

PostPosted: Sat May 24, 2008 4:44 am    Post subject: Reply with quote  Mark this post and the followings unread

DrJustice wrote:
Your G and B strings are tuned to G# and C. Maybe it's not tuned yet? Guitar ownership will soon get your tuning habits sharpened anyway Wink

Yes, well, I did a crude, rough and tumble tuning job with a ChucK tuner that I cooked up really quickly by taking the FFT and looking for the first harmonic, but it doesn't work very well at all! An FFT approach is not the best for a tuner, as I have learned. Perhaps I will figure out how tuners work and code that up in ChucK. After spending all I had and then some on the guitar and amp, I just couldn't get a tuner just yet. So I'll ChucK it for now.

DrJustice wrote:
As for the exact degree of tubularity, I'll leave that to someone who really knows their references. There's certainly lots of harmonics, but probably also lots of aliasing. It does sound a bit harsh in my ears. Looking at the waveform it may seem like the clipping is somewhat brutal, and an FFT reveals quite a bit of content below the fundamental (at lest the background noises are showing up. I think that at this level of squaring off, band limiting is in order. Of course any kind of non linear function should ideally be oversampled anyway.


Thanks for studying the signals that way, Dr. J! You know, I have some background noise coming from my apartment somehow. It occurred on both amplifiers, the one they sold me which was a Line 6 Spyder III, and the one I exchanged it for, which is a Fender Frontman 15G. There is a tickitty-tick sound that I recorded and attached to this post so y'all could hear it.

The ticking is not as loud as you hear in the recording, I boosted it 12 dB with Audacity, but it is quite noticeable while the guitar is not playing. I have tried shutting off every electric device I have except the refrigerator and the ticking does not subside (computer, arcade game, scope, tv, radio, clocks, printer, cell phone, lights). The Guitar Center expert suggested it could be from a neighbor with a shortwave radio, I don't know. When I unplug the guitar but leave the cable connected to the amp, the ticking gets louder and when I unplug the cable from the amp, the ticking goes away. Also with guitar plugged into amp, when I lower the guitar volume (the Strat has a volume knob), the ticking goes away. Vewwy stwange, mistwer Bugs, vewwy stwange indeed!


DrJustice wrote:

One test you could do is to run a slow sine oscillator sweep from 20Hz - 20kHz through the transfer function and listen for artefacts (mainly aliasing).

DJ
--


OK, I'll run that test sometime soon. Good Idea!


Guitar_Noise.mp3
 Description:
11 seconds of the bizarre ticking noise that my amp picks up from my apartment.

Download
 Filename:  Guitar_Noise.mp3
 Filesize:  178.9 KB
 Downloaded:  1172 Time(s)


_________________
"Let's make noise for peace." - Kijjaz
Back to top
View user's profile Send private message Send e-mail
Display posts from previous:   
Post new topic   Reply to topic Moderators: DrJustice
Page 3 of 4 [87 Posts]
View unread posts
View new posts in the last week
Goto page: Previous 1, 2, 3, 4 Next
Mark the topic unread :: View previous topic :: View next topic
 Forum index » DIY Hardware and Software » Developers' Corner
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum


Forum with support of Syndicator RSS
Powered by phpBB © 2001, 2005 phpBB Group
Copyright © 2003 through 2009 by electro-music.com - Conditions Of Use