To a close approximation doing that will double the cable capacitance that is seen by the component that is driving the cable, and it will cut the resistance and inductance in half. For a line-level interconnect conducting analog signals the most significant of those changes is almost certain to be the doubled capacitance. The resulting sonic difference, if any, will depend on the output impedance of the component driving the cable (especially its impedance at high frequencies), the length of the cable, and the characteristics of the particular cable type. And if there is a difference, it may or may not be for the better, depending on those variables, the sonics of the system as a whole, and listener preference.
P.S. to my previous post: There's one more possible effect I can think of that doing this may have in some systems. If the two components that are being connected are susceptible to ground loop issues (which might take the form of high frequency noise and/or a reduction of "background blackness," as well as or instead of low frequency hum), the lowered resistance of the ground connection between the components might reduce the severity of that issue to some degree.
Have to concur, the results are quite spectacular with the use of two cable sets connected with splitters, and even more so when we terminated two cable assemblies into one plug to produce the Double Double.
The sound stage in this configuration is huge, dynamic slam increases, and while theory would predict a diminished high end response due to high capacitance the highs are actually more extended, detailed, and with no attendant harshness or listeners fatigue.
We have been playing this configuration for several weeks now and have had none of the issues that the theory quoted above seems to predict. That being said we still are being cautious because of those theoretical concerns, but so far the theory is having some serious trouble reconciling itself with the reality on the ground.
theory/hypothesis is merely an observation of an event or occurrence that covers only that observation.
even then the theory/hypothesis of the observation may only be partially correct and even occasionally, not correct at all.
when there is a variation of the original observation then the current theory/hypothesis may not work, or be relevant, at all.
FYI for those interested:
Related forum threads are here: https://forum.audiogon.com/discussions/new-teo-audio-ics-who-has-them
and here: https://forum.audiogon.com/discussions/anyone-try-teo-gc-jr-interconnects-yet
@rcprince a single cable assembly having 2 IC’s in parallel with single termination connectors at both ends of the assembly provides one solution to your concern of using 2 Y-signal splitters with 2 IC’s (i.e., avoiding adding more connectors to the signal path).
See this post by way of example:
I will share hear what I just posted in the comments section of my article which is linked above.
I wish to add to the body of knowledge about the Schroeder Method, as well as clarify some misinformation that is forming around it. I have found also in experimentation that it is efficacious in additional applications including a digital connection, an XLR connection, and connection between dedicated preamplifier and amplifier(s). The balanced (XLR) connection was the first to be tried, as a natural extension of the original RCA assembly. Next was a twin XLR assembly acting as AES/EBU output from a transport to DAC.
The final iteration relating to preamplifiers and amplifiers must be prefaced with the necessary warnings that caution should be exercised in regard to some amplifiers which are NOT suitable for using the Schroeder Method (double interconnects) as it may cause them to oscillate, making them unstable and damaging them, and or speakers. Again, as always has been said, this is a DO AT YOUR OWN RISK activity. Please consult your equipment manufacturer if you have questions or concerns about the suitability of this activity.
The final iteration was a dedicated preamp to amp setup which worked as splendidly as all the others. I discussed this further in the soon to be published review of the Belles ARIA Preamplifier and Mono Block Amplifiers.
Does multiple parallel transmission lines exist as a cable, of any kind? is there a a history in this area?
This is apparently a nordost cable, in this image.
The idea of parallel transmission lines is as old as the transmission line idea itself.
The trick is.... can it be made to work. It depends. On a lot of things.
One last point which I do not believe has been emphasized enough is that the longer the interconnects, the lower the characteristic impedance of the cable. Doubling a cable halves the impedance, and I have been avoiding longer cables (i.e. I have only worked with 1m interconnects to date) in doing the Schroeder Method, as I do not wish to encounter a situation where the output of the preamplifier (whether internal to a DAC or dedicated preamp) is not able to drive the amplifier, and potentially incur damage due to the characteristic impedance being too low.
Those with more esoteric designs in DACS, i.e. NOS DACs, and preamps should exercise caution and understand the limitations of their gear, and accept the potential for damage if they proceed. At this point in time caution is still to be exercised as regards types of systems in which this can be done safely.
Regarding the mention of digital interconnections, one thing that theory can predict with certainty is that doubling a digital cable will result in a substantial mismatch between the characteristic impedance of the cable and the 75 ohm or 110 ohm impedance (for coaxial S/PDIF or balanced AES/EBU, respectively) of the components being connected.
That in turn can be predicted with certainty to cause or increase signal reflection effects at the high frequencies that are present in digital audio signals, and consequently to degrade waveform quality to some degree, in some manner. Which can certainly be expected to result in sonic differences in many and probably most systems, in part by contributing to timing jitter at the point of D/A conversion. And while those differences might turn out to be to be preferable to some listeners with some systems and some recordings, they would constitute a reduction in how accurately the recordings are being reproduced.
An interesting excerpt from this paper by Steve Nugent of Empirical Audio:
jayctoy, yes, the effect is cumulative with successive additions. Agreed; the sense of the music being more lifelike, whether a studio or live recording, is powerful. The characteristics of a particular interconnect's properties is intensified and the entire system is changed profoundly. The experience is much more convincing as to recreation of a real event.
" Regarding the mention of digital interconnections, one thing that theory can predict with certainty is that doubling a digital cable will result in a substantial mismatch between the characteristic impedance of the cable and the 75 ohm or 110 ohm impedance (for coaxial S/PDIF or balanced AES/EBU, respectively) of the components being connected."
+1 almarg,...well said, thanks for the rest of the message as well.
In short, this method should not be used for any digital connection that expects specific impedance, S/PDIF, AES/EBU, Word-clocking or Master clocking cabling of any kind whether 50-ohm or 75-ohm are all examples.
It’s very unlikely, Ozzy.
In the case of analog outputs, some designs drive XLR and RCA outputs from separate output stages (i.e., they individually buffer them), and some don’t.
If they do, obviously connecting both would not cause either of the connections to have an effect on the other. (Aside, that is, from the possibility of introducing a ground loop and its negative effects, assuming the other ends of both cables are connected to a single component).
If the two outputs are not individually buffered, the signal provided to the RCA connector is probably the same signal that is provided to one of the two signal pins on the XLR connector. Connecting them both would do nothing for the other signal on the XLR connector, while resulting in at least a slight imbalance in the impedances of the two signals on the XLR connector, thereby degrading the noise rejection that a balanced interface can provide, at least slightly. Also, depending on the output impedance of that component and on the total of the capacitances of both cables, upper treble rolloff of whichever signal is used by the destination component might occur to some degree, and sharp transients might become more sluggish.
In the case of digital outputs, coaxial S/PDIF and balanced AES/EBU outputs would certainly be individually buffered, since their output impedances have to be different. So connecting both would not accomplish the same thing as doubling a cable, which as I’ve said I don’t recommend anyway for digital cables.
Zephyr24069, thanks for your comment.
Oddly what you are doing makes sense to a degree. I know nothing about the interactions of cables and gear. As in why it happens. Only a glossy idea of why it occurs. Ok, I'm uneducated. Now that that is clarified, I have made the following observations. I have connected some interconnects between my DAC and preamp, as well as between my Preamp and amp.
One set of interconnects is the Inakustik NF2404
One set is the Genesis by Gary Ko
The other is an inexpensive coax interconnect.
Here is what I observed.
Between the DAC and the preamp the coax and the Inakustik are hard to tell apart. The Inakustik is better, but not by much. The Genesis is an OMG that is great. The world opens up. I use the Genesis there.
Between the preamp and amp the coax, Inakustik and Genesis are hard to tell apart but they are different. The Genesis is a little thin, so I keep it where it excels. The Inakustik is more fleshed out and full. Very clear and accurate. Pretty neutral but very full. I like that so it stays between the preamp and amp. The coax is just there. Nothing special but pretty darn good for $250. It is not as clear as either the Genesis or Inakustik. There is some haze and cloudiness to it. It is not in use.
The Genesis is designed around 2 wires on the positive and 1 on the negative. Kind of like your doubling up of interconnects. Different but similar. The Inakustik is 2 wires in an air dielectric spread very far apart. About an inch plus. Totally different geometry. I am sure their inductance, reactance and capacitance and radically different. I assume that is why the results based upon where they were used if quite different. If I were to do a review on either the Genesis or Inakustik and I based the finding on one topology alone, I might find either cable inferior. If I used the cable in another area, I might find it heavens sent. Interconnects are tricky business. It's hard to tell if you are going to get the same results as someone else who raved about them. From what you are all saying and the links included say, it seems cables are very system dependent.
Thanks for validating this idea for me. I got in a fight with one vendor over this very issue and feelings were hurt. Now I can't get any more cables at discounted pricing. I'm just another chump having to pay list price. For that brand at least. Owe Well. I really don't need anymore unless I purchase a tape machine.
@kingrex Your description of the design of the Genesis IC is indeed “different but similar.” The Canare Starquad cable design is even more similar to “Schroeder Method configuration.” The Starquad design has 2 hot, 2 returns and 1 shield. In the Schroeder Method configuration, however, everything (including the shield) is doubled and configured in parallel.
All: I’m adding to my last prior post here to provide more context to the Canare Starquad story and their wiring scheme. As one can see from their wiring diagram, their 2 hot conductors are sliced together in parallel. Likewise, their 2 return conductors are spliced together in parallel. Note one common shield element, however.
@kingrex I don’t think anything is being “filtered.”
Some speculate that the improved sound qualities perceived during even casual listening sessions is attributed to the parallel configuration having nearly twice the bandwidth of their single-run counterparts, with the attendant reduction in internal reflections for electrical signals transmitted through the parallel configuration relative to their single-run counterparts.
I’m not aware of other ways of achieving this effect. If the “increased bandwidth” speculation is the correct attribution to the audible improvements in reproduced sound, then interconnect designers should devote their energies to designing cables having the greatest bandwidth possible. But even then, implementing the Schroeder Method to those cable designs should yield further improved sonic benefits over their best single-run counterparts.
@elizabeth A single cable assembly is certainly preferable to using 2 single-run cables with terminal Y-splitters at both ends. But using a set of Y-splitters (4 total for a stereo run with 4 IC’s) is inexpensive enough for one to tinker with the method in the short term.
Teo Audio makes a Double-Double version of their liquid metal GC-II Jr as a set of paralleled assemblies terminated with single RCA connectors at each end. (Alas, their A’gon listing expired earlier today, but I’m sure they’ll relist it.). Taras of Teo Audio (@taras22 here) shared listening impressions of that product compared against some of their pricier single-run cables. See one of the Teo Audio threads in the “Cables” forum for that post.
@ozzy I think @douglas_schroeder tested his method with XLR-terminated balanced IC’s. Check the longer Teo Audio thread for details. BTW, that thread—like most here—goes down several rabbit holes unrelated to the thread topic. I think Doug tested the method using XLR-terminated balanced cables from manufacturers other than Teo Audio cables.
Celander, How do you get extra bandwidth. You lowered the resistance with parallel conductors and upped the capacitance and induction. I don't see any reason for the bandwidth to change just because those changed.. I get the DAC and preamp are reacting to the change in current, but I don't see that raising bandwidth.
Unless the conclusion was changing those three electrical components resulted in a cleaner signal with less noise allowing a component to operate more efficiently and more readily reproduce higher frequencies. But that is filtering out noise to allow better operating parameters.
What is it that you see in changing the induction capacitance and resistance that would make frequency change.
@kingrex See Bob Smith’s comment at the end of this link:
@ozzy See comment by @douglas_schroeder at the end of the above link with respect to his findings with the method using a parallel run of balanced XLR-terminated IC’s.
Ok, I re-read it. I gather the resistance went up, not down. And a warning that amps may not like this. I still see nothing talking about frequency extension or any engineering theories on how changing the electrical values passing the signal would benefit frequency extension.
If such a simple concept as increasing mass would increased frequency then more designers would use it. I know they alter mass with speaker cables. Almost every manufacturer charges more for the better speaker cables which all seem to have heavier gauge wire. In my own personal conversations with cable manufacturers they say purity of material spacing from one another as well as shielding, construction insulation etc are the critical component to obtaining optimum sound. I doubt they would all miss added mass and not focus on that if it was relevant.
I am in no way saying Schroeder did not obtain positive results. It seems he and others have experimented with a topology and antidotal evidence is showing some benefits in their systems. I am just wondering what the real reason for the results is. Can Improvement of the signal resulting in higher frequencies being reproduce by the amp be done in a more efficient and or cost-effective manner. Doubling up interconnect which are already expensive is pretty cost prohibitive. Especially when you need to do it between a phono, a DAC, a CD a tape machine excetera. That's a whole lot of cable and additional Hydra head behind your gear.
@kingrex I’ll quote the relevant passages from Bob Smith’s comment:
“Noteworthy in the above is the fact that the Capacitance doubles and the upper Cutoff Frequency (where above which the signal begins to be attenuated) almost doubles. So in the event that we “double-up” our audio cables, we actually extend their bandwidth – albeit we are talking in the region of radio frequencies so there is no real benefit there with respect to extending the fundamental audio bandwidth.
“What IS significant though is the fact that all of the above leads to a potential reduction in reflected energy and/or standing waves within the cable, and that is because of two primary factors. By lowering the Character Impedance and consequently, raising the Cutoff Frequency, we “push” or force any potential reflections up to twice the frequency at which they would otherwise occur. That then leads to two other outcomes. First, higher frequencies find it more difficult to propagate down the length of any conductor due to the “skin effect,” and are therefore usually attenuated more with respect to those that occur at lower frequencies. As a result, lower magnitude levels of reflected energy translates into less interaction with the Source and Load circuits. That means less potential for the formation of any associated Phase Distortion artifacts as outlined above.”
Please note that this is the opinion of Bob Smith. I’m not an EE to assess his opinion.
I doubt folks are going to go “all in” with modding all of their IC’s this way, regardless of the cost or the myriad of interconnections contemplated. So the fear of HYDRA is avoided.
The questions for many are a bit different:
(1) What audible improvements could I achieve for those 1 or 2 critical sources (that is, those sources representing ~80%+ of my listening time) when those sources are connected to my receiver, integrated amp or preamp using paralleled IC’s?
(2) Could I realize a significant improvement using the Schroeder Method with IC’s of a more *modest* cost that rivals or exceeds the cost of my expensive IC’s?
(3) And if I didn’t want to engage the “enticing offer” from my favorite cable manufacturer to buy his/her best IC set due to the law of diminishing returns, then could I nevertheless achieve even better quality by buying another run of my present cable using this method?
A couple thoughts interjected...
No one as of yet knows why the Schroeder Method works, least of all me! I claim no pedigree, only curiosity leading to the discovery and implementation. Speculation would be expected until such time as a cable manufacturer would chase down the explanation, if measurable.
Yes, I have done it with both RCA and XLR with analogue signals, as well as AES/EBU (XLR) in one setup from transport to DAC. I also have successfully done one RCA setup of Schroeder Method between preamp and amp. NOTE: ALL PREVIOUS WARNINGS AND CAUTIONS APPLY; this is a "do at your own risk activity". It should be avoided with certain amps, as has been forewarned, as well as some DACs that output from the chip (ie. No opamps) etc. Again, check with your equipment mfg. Or designer as regards safety.
Imho, the Schroeder Method is a system cost saver, not a money waster. Double the price of the materials and 4 splitters is a gift given the result in every instance I have done. Others who have tried thus far seem to agree. I certainly don't mind a perceptual $10K component upgrade for the cost of one more pair of interconnects, or doubled pair. :)
Finally, funny how everyone, even designers find their pet reasons for cable performance. My most consistent benefit in cables across prices and brands has been, aside from conductor material, heavier gauge. Consequently, the idea for the Schroeder Method grew out of many years of pushing heavy gauge cabling in rigs and getting what is to my ears better sound. Ymmv