Bedini Base

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

tonwerker

Member
Joined
Sep 24, 2004
Messages
13
Bedini Base
Hello,
do anybody know how this unit is working?
So far I understand it puts the monosignal out of a stereo source material
and left the source Stereo "untouched"....

Any idea about the concept behind or if it is a "DIY try" worth?
This unit was recommend me as something truthly amazing - price range is 6000 $....
 
i wouldn't spend my time diying this one. The spacial widening circuitry can be found duplicated in cheapie pieces of crap like this.

Cool effect though. I'm wondering if anyone makes an AU plugin.
 
I've used a Waves stereo-widening plugin which was great. Very powerful.

It came in handy with a mix which I recorded direct to stereo 2-track.
 
I got the tip for the bedini from a friend - and he knows that it works a bit better then the "noisinger" stuff -
I think it works also a bit different then some delay plugins like Waves
(not that it is doing a bad job, but with a simple sample delay you can create an equal effect)...

Ok - this is the patent description:
realistic pleasant sounding binaural output in which optical coupling means, having a nonlinear transfer characteristic, creates an unbalanced output to a pair of audio outputs. The system provides for a high or low level input, which may be selected by a switch, to an emitter follower which drives an optical coupling circuit. The optical coupling circuit is comprised of a light emitting diode and a phototransistor connected in a phase splitting network. The phase splitting network provides an unbalanced output to terminals for connecting to preamplifiers, amplifiers, etc. An alternative embodiment employs two optical coupling circuits in two separate signal processing channels, including two phase splitting networks.
 
[quote author="tonwerker"]Ok - this is the patent description:
realistic pleasant sounding binaural output in which optical coupling means, having a nonlinear transfer characteristic, creates an unbalanced output to a pair of audio outputs. The system provides for a high or low level input, which may be selected by a switch, to an emitter follower which drives an optical coupling circuit. The optical coupling circuit is comprised of a light emitting diode and a phototransistor connected in a phase splitting network. The phase splitting network provides an unbalanced output to terminals for connecting to preamplifiers, amplifiers, etc. An alternative embodiment employs two optical coupling circuits in two separate signal processing channels, including two phase splitting networks.[/quote]

Yeah, yeah heah... blah blah blah...

I've used one. We have one round here somewhere, It's not used any more... -Ever.

it's not actually crap as such, but it's not very good, and the manufacturer's description of how it works and what it does is basically psycho-babble mumbo jumbo. It's based on three principles: Marketing marketing & marketing.

Here's the real truth.

You get nothing for nothing. For every improvement there is a diminishment. It makes some stuff wider, but then that same stuff collapses in mono. There's a mono-feed-in control, but then the stereo image collapses again.

It's fundamentally a blumlein 'phase shuffler'. Nothing more. Useful at times of course, but far more useful is an M/S encoder, then a couple of mono EQs, then an M/S decoder. You can do a LOT more, with more control.

Forget the Bedini. I used one whe it came out, I was intrigued until I found out what a simple, overpriced boat-anchor it really is.

Over-hyped, marketing-driven, psychobabble. Nobody wants them, that's why they never sold.

Keith
 
Hi Keith,

Do  you (or does anyone else) know how the Center Distance on the BASE works?
The process behind it is a complete mystery.

I've studied a patent assigned by Bedini, from 1984. The description does not appear to match that of the 4 dial BASE = Some of the processes may well have been carried across though, as there was some 3 other variations of the 84 BASE.

It seems BASE has also been re incarnated as products by other companies since.
Circa 1989 Behringer worked on an analog replica, the Edison EX-1. This has been re-released over the years but the essence still the same. (4 dials, LMF and Stereo Spread, Center Pan and Distance).
The same processes were later reworked digitally and made available on the Behringer  X32 mixer and DEQ249 products.
There is a more recent  VST by AntiPop called TeslaST as well. Identical controls.

The manuals for ALL of these products are dilute, with very little description other than "it brings sound closer" or something along those lines.

I've consulted  customer support at music-group (parent of behringer) as well as StudioPop for a break down - no luck in either case. Customer rep. had no idea where to start and tried to contact behringer Germany HQ but nothing came of that.
AntiPop did not reply I think because they're  based in Spain.

I managed to get in touch with Uli of behringer through linkedin.com (he personally designed the EX-1).
He said:
"I am sorry but this product has been obsoleted around 20 years ago" seeing that it exists in the X32 and DEQ this reply amazed me and  goes to show the attitude of some towards the integrity of his products.

So I've done all this and still nothing. Any ideas? Already stated in this thread; it may be a mono feed in - my suspicion is that it adds a phase split duplicates to the stereo feed. This may add additional decor elation if the control is turned up. But that doesn't explain what happens when the dial is turned down!!!

Any leads will be highly helpful to and appreciated by me as well as others who may have any of the products mentioned (5 in all, none of which have good documentation).
 
I should add; I am familiar with how the other controls work just not Center Dist.

However; also unclear is whether the differential Low shelf filter in LMF Spread is phase compensated or not.
There have been 2 developments in the past that use such a filter, with different target applications.

With Blumlein shuffling, group delay in the difference is desirable (90deg at approx 700hz) to re-align signals derived from spaced microphone, aka phase rotator.

A more recent, stereo-sonic shuffler was adapted for pan potted mono stereo signals . With it, linearity was the rule, rather than the exception.

One can only assume that BASE, Behringer and AntiPoP products are of the non-linear sort.
As such these products can be used in conjunction with spaced mikes, even though not their intended purpose. One would need to work out the corner frequency of the filter and go from there, because in blumlein shuffling, boost and frequency are coefficients that correspond to certain spacing; Once the fixed corner frequency present in the products has been determined the LMFSpread dial can be marked to denote corresponding spacing.

Im not all that good at maths but I'll try to explain the principle (alternative explanation here);
Imagine a sound coming from the right. Low frequency sound will arrive at both mics with a small delay between mic1 and mic2, with little or no difference in amplitude. The purpose of non linear shuffling is to add a phase shift at mid-high frequencies and enhance LF amplitude differences. The spacing between the mikes corresponds to a wavelength in meters, given speed of sound in air one can work out a frequency. It is at this frequency a 90deg phase shift is introduced between the channels by means of RC filter theory, recall; a phase shift by means of filter in the difference . The left and right channels are phase shifted but in a frequency selective way, with that shift rising above hitting at the target frequency. Once played back through speakers the sound is more correct.

However it would be questionable to add such shift to recordings that haven't been recorded in that way - With BASE & Co. One would get an illusion of space but of the gimmick variety considering no attention has been payed to  the underlying theory its based on and depends hugely on the source material,i.e.  if its mono panned / multi track or not.

In the original design, blumlein used a 6db/oct LPF in the difference, and attenuated the mono accordingly, a low shelf as seen in BASE etc does more or less the same thing, with the mono attenuation being made redundant.

As for the Waves S1 plugin, it has a shuffler of the Linear sort - as such its more appropriate for general use although its effect not as pro found as using the non-linear method via spaced microphones. Lexicon had/has a similar filter in their effects units as well.
 

Attachments

  • BlumleinShuffler.jpg
    BlumleinShuffler.jpg
    135.5 KB · Views: 9
Interesting, didn't know the EX-1 was based on this unit.

> a patent assigned by Bedini, from 1984

I guess this one  ?  US 4555795

And here's a related one from 1998,  US 5,748,745, using the old Philips TDA3810.

Interested in any schematic of this Bedini or the EX-1, still too lazy to trace it myself...

Thanks

 
Hi,

Keith seems to be knowledgeable I was hoping he could chime in on how that center distance works. Its the ultimate mystery in my personal research en-devours at the moment.

I might add that some of the processes in these boxes based on the base (pun) appear to be signal dependent, so my previous comment about multi track can be ignored. I think the non linearity added, increased as the signal level went up (maybe the inverse?). Some form of 'follower' is described in the 84 patent, at least from my interpretation.
On another forum the designer of the EX-1 is quoted, in where he describes EX-1 as 'program dependent':
First of all, we create a mono signal (L+R) and the stereo part of the original signal (L-R). We then start to process these signals. In order to achieve greater ambience reception, we use a program-dependent delay on the stereo part and a low-mid frequency filter as well, which allows boosting or cutting of those frequencies which are normally hard to perceive as being in stereo (i.e. low-mid frequencies)
source

FWIW a similar process was definitely used in the Hughes SRS systems. the AK-500 and 100.
SRS process amount was dependant on out-of-phase component in source material, in order to avoid over 'oops' modulation.

Another post here talks about the opto couplers in the BASE:
"I had a look at the patent and they have a circuit diagram of the "Audio Processor" in Figure 3. Q2 is a conventional split-load phase splitter of the type found in many Class AB valve amps just using a transistor rather than a triode. I can't even begin to reason why it is driven as a photo-transistor rather than using any conventional drive method. My guess is that the optocoupler will exhibit soft clipping when over-driven, but that is just a guess. What comes out the other side is just two copies of the input signal 180 degrees out of phase with whatever nonlinear effect the optocoupler has introduced"

As for center distance, its not obvious.

Of all the processors, the SRS seems the most refined. Hughes made its process available for license to 3rd parties. Spatializer Audio and NuReality made alternatives, (HTMS-2510 and VHT-200). The HTMS product has a Schreiber sphere[1] display, which in itself has interesting origins, Peter Schreiber being the 'inventor' of quadrophony which set the foundations for and inspired Dolbys first surround system (Dolby and Schreiber have a long history of legal battles on elements of Pro Logic).

[1] Understanding the Scheiber Sphere by William Sommerwerck, MCS Review , Vol.4, No.3, Winter 1983
 
Center Distance could very well be Sum/Difference ratio, with Stereo width and LMF spread being filter gain and frequency...
Then again I doubt it.

A more intuitive guess would be that it works like this:
Mono or Sum Portion is split into two and subsequently shifted by +/- 90deg from 0 to 20khz (as in Pro Logic Rears but without the 7khz LPF). This is either added (+) side of the dial, or subtracted (-) of the dial, to/from the input signal.

I wouldn't be too harsh on Bedini, with the opto couplers also included, the BASE seems to address all of the phsycoacoustic enhancements available from stereo in one unit. When the couplers are over driven, this adds saturation which is known to have listening benefits. John  unfortunately passed away only there last November.

John-Bedini-318w.jpg


Seemed like a great guy.
 
Bump, I've  tried to PM SLLtech, (above) however it seems his PM inbox is full... could a mod perhaps address this?
 
I've since realised that JB may have been inspired by early Aphex aural exciter boxes in part, regarding the signal dependant bit already mentioned, a good guess seeing that they preceded the first b.a.s.e by some 10years or so (founded 1975).  Aaural exciter with optical big bottom is seen written on more recent designs,  it could be that they always were optical, recall above b.a.s.e has elements of optical processing.

"[The aphex] circuit combines a low-pass filter and dynamics processor to compress and delay incoming low-frequency information. The process is reverse amplitude dependent, meaning that more is applied as the input level drops and less as the signal gets hotter."
More probable is that both designs are based on  the teachings of fletcher-munson, the loudness curves (phons) in where as signal intensity decreases, certain frequency ranges become increasingly more difficult for the ear to perceive. Its specifically these ranges that the Aphex processes (high and/or low end, depending on model). See Attached.

BBE Sonic Maximizer SPL Vitalizer is similar. See pic. According to the patent,  it cuts mids via S-shape aka 'smiley' EQ curve, (mid-hi tune dial) distorts highs n lows by an amount (harmonics dial) both are done in sidechains . It adjusts width via another sidechain (width dial). Sidechains are blended with input (process dial). Not sure if anything is signal dependent here, never got a hint that it wasin patent. MMV.

Edit, note I confused SPL Vitalizer and the BBE Maximizer, the irrelevant one is striked through and description updated..
I've not looked the BBE patent though a glance at the description implies some sort of dummy speaker load to undo lagging treble, which they call 'transient distortion, thats otherwise present in a 'phase aligned' speaker system. It should be noted this  can be already present in speakers, by means of carefully designed passive/active XO circuits for e.g.

Just for fun I'll add that Creative have a range of enhancements in their sound card products (SBX Chrystalizer, formerly X-Fi) which does more or less the same process, but digitally just prior to d/a conversion. As with any modern d/a converter, signals are upsampled as part of the conversion process, however creative adds the s-curve harmonizer in as an intermediate stage before conversion but after upsampling, which enables 'lowly' discrete-time signals to interact with  subsequent processes at a higher speed/resolution, this allows for the effect to more resemble analog processing, imparting less digital distortion and in this case undistorted-distortion it being a harmonizer (the irony).

SRS TruBass (formerly SRS WOW which can be found in Windows media player, under Enhancements) is the same, with less emphasis on the upsampling which isn't necessary for digital bass (its better resolved than treble).

Yes I'm unemployed.

Still uncertain about how Center Distance on the b.a.s.e works. Anyone willing to chime in on my effort here, at all? 


Dear Group Diy community,
 

Attachments

  • x.png
    x.png
    30.5 KB · Views: 9
Following from side line, interested - but nothing to add, this is the first I hear about the Bedini

Regarding the fletcher-munson dynamic interpretation, I'm currently working on a compressor like that - can report that it works surprisingly well even in small doses..

Jakob E.
 
Thanks and sounds cool!

You should make it a tube implementation... or using OTA current mirrors instead of transistor modeling or optical means.
In fact it seems SPL already have a tube implementation.

I'm confident that sub-harmonic enhancers and exciters in general are just plate saturation effects modeled with transistors , ignoring level dependency, delays and frequency division.. (see attachment).
In the DBX-120 thread (here) the developer of Lowender (DBX-120 VST software equivalent) explains how Aphex differs from the dbx. Essentially, the aphex cuts out entire low portion of the spectrum and saturates wheras the dbx cuts two regions in the LF, divides redult by two, then saturates, then mixes with input. Another difference is the Aphex applies 90phase shift to the effect with phase shift network. Because the 120 lacks this delay, its prone to overloading amplifiers and speakers, (but IMO can be avoided with care).

the [aphex] process is low-pass, delay, compress, mix back in. [...]the bass side-chain is delayed using only a phase network..

It seems all these products are based on valve plate saturation. OTAs or DOAs can be and are used instead.

One last thing. The dynamic processes in the Aphex/EX-1/B.a.s.e and the Hughes SRS Box are not the same.
The SRS was to be used as a playback enhancement device at home - according to the Klayman patent, the follower tracks the L-R component and only expands ambience if the recording lacks it.
Aphex on the other hand, being a recording-side enhancement trick, tracks overall signal level.
The B.a.s.e only enhances the low frequency information, whereas the aphex enhances both extremes of 20-20kHz. I could be wrong about the base though.
 

Attachments

  • s-l1600.jpg
    s-l1600.jpg
    286.6 KB · Views: 15
Very relevant DIY Project, this time with OTA amps instead of transistor modeling or valves. Its Non-signal dependent.
www.spontis.se/pdf/tantek_enhancer.pdf

 
Naa, the Tantek is signal dependent - both it's compressor and it's "harmonic generator". This is the aphex-exiter-tradition, that works by distorting and compressing a highpassed part of the signal and blending it in. Not a stereo-irizer like the BBE tradition  - and really not exiting at all (pun intended), just horrible.

Jakob E.

 

Latest posts

Back
Top