When does resistivity start to matter with ICs?

With voltage transmission into high input impedance, at what point will increased resistance of the conductor(s) start to measurably affect performance detrimentally? As in not just attenuation of the whole band but significant roll-off?

I don’t believe an aluminium interconnect sounds intrinsically different from a silver one (with same geometry) because of the difference in resistance. Well, negligibly.
It’s just silly that it’s so common to promote silver as the best conductor for every purpose just because it’s least resistive. So many other parameters…

So yes, I know resistance is not a problem with good design, but would like to know how much it’d actually take for it to become problematic.
This is strictly about voltage transmission.

IC cable looks at a theoretical “infinity” load, so ALL of the signal drops across the load and nose across the cable.

OK, that’s the theory. We really use 10K ohm to 47K ohm in that ball park as a load. The resistance of a IC conductor would need to go really high to not be a good vokltage transfer function into a high impedance termination.

So we have the ratio of the conductor and the load to consider. Even with a high 0.5 ohm RCA resistance, that would be no sweat into an even 10K load impedance (20,000:1). The load will wrestle most of the signal away from the cable!

The “attenuation” in such a short length isn’t too material, either, and that’s good. We don’t have current squared times resistance losses in IC cable, so we are left with voltage attenuation only. The current is “zero” theoretically. We don’t need current in IC cables!

On longer XLR, you can use several INDIVIDUALLY INSULATED wires to lower the loop DCR of the cable relative to the load. RCA can us that trick, too, and use heavy outer braid design to lower the voltage drop on the return shield.


1 Like

Isn’t this quite exactly why material choice has a far bigger impact in interconnects than speaker cables?
Since there is essentially no current, am I wrong in thinking that the signal (EM wave) travelling along an interconnect is “essentially” missing the magnetic part and has primarily the geometrically radial characteristics of an electric field?
Could you explain why the star quad configuration is beneficial in an interconnect when the magnetic counterpart of the conductors’ surrounding fields is negligible? Or is it? Still enough current to have non-negligible magnetic fields at the very small distances we have in a star quad design?

I’m confused. Yes, should go learn some physics… But while I’m lazy at it, you could help Galen!

1 Like

Over on the Iconoclast thread some of this has been discussed and there’s a few white papers posted as well. It’s a really long thread, but I believe that some of what you are interested in has been discussed starting around Jan of this year.


The star quad does lower inductance, but it ALSO allows four 30 AWG wires to be used instead of 25 AWG wire. This much improves the Vp with respect to frequency as we can see in the chart. We have to not forget that we have a lot of moving parts and to not pick favorites.

Inductance is ALWAYS there, and is lowered from 0.15 to 0.11 uH/foot. So no, the magnetic field is still in full force relative to the E field. OK, we changed the inductance, but the capacitance is opposite goes UP from 12.5 pF/foot to 17.5 pF/foot, still very low and why a good dielectric design is important. Improvements to the Vp linearity outweigh the total capacitance in listening tests and that was the primary reason to change the “conductor” to four 30 AWG wires.

On attenuation, the four 30 AWG have a lower total CMA than the single 25 AWG so it is a win-win there, too.

There you have it, and why the series II IC cable were designed. We move the performance up and also keep costs in check as we use a lot of the same BOM materials and processes.


1 Like

I wonder if van den Hul’s pure Carbon Nano Tube interconnects at a stated 65 ohms per meter are the leading edge in the field of high resistance cabling…
Now those would be interesting to try. Only about 7 kilodollars apparently…

A material without the calculations and measurements to show WHAT and HOW it is improving anything except their pocketbook is useless. You’re adding more “error” to already inadequate and poor designs and how’s that all of a sudden become “correct”? Oh, I forgot, raise the price.

A cable is a balance or R, L and C. Anyone variable that gets too high unbalances the cable such that it is far from ideal.The real issue is analog cable does not act in a linear fashion with frequency. We have a changing target who’s variables influences are different as frequency rises.


1 Like

How come? You are offering us three varieties of material for Iconoclast and you have no clue (so certainly no data either) WHY they sound different. A material is useful when it’s applicable to a given design and subjective evaluation of it tells us that it makes a difference in sound (a difference that can only be lateral in absolute performance, because we are now assuming the material choices available for a design measure the same!)
Yes, forget about comparing different materials in different designs, sure!
Let’s say a Teresonic 24k gold interconnect sounds “like this and that”, ok, but we shouldn’t in absolute terms compare it to either the ICONOCLAST ICs as they are NOR should we compare it to hypothetical 24k gold ICONOCLAST ICs. We could instead compare (in absolutes) the Teresonic copper interconnects with the 24k gold ones, because they share the same structure. Right.


Stop being circularly illogical. BUY THE (fill in the blank) if you feel copper is better. That exactly why it is an option. Without pure evidence of a type of copper’s additional performance uplift it is up to you to ADD IT TO A PROPER DESIGN. COPPER DOES MEASURE within it’s resistivity variation THE SAME! ALL COPPER with the exception of resistivity measure the same R, L and C. Bring me the evidence of repeatable differences other than that. There are ZIP, ZERO, NONE outside of resistivity and mechanical T+E changes (cryo adjusts the mechanicals to an optimized low stress state).

Just because you want to work a deal with the devil on an exotic material doesn’t allow a designer to use what is called appeals to the ignorant; “Hey, I saw an invisible elephant in the room just now, prove I didn’t”. No, you prove you did, please. That animal could be ANYTHING at this point, why give it a name now? ICONOCLAST doesn’t prove anything except the knowns of the EM design and calculations. We provide the choice where it is purely subjective and you prove the merit(s) to yourself. We don’t matter in that effort. We can’t. There is nothing but BIAS in anybody’s answer but your own.

Once the right design is in physical fashion, ONLY then can we do as you say, substitute materials that do a like function and again, the differences are purely subjective for everyone. Most important, they didn’t make the cable WORSE!

It is easy to sit back and put a man on the moon in your mind, “they did what I would have done” but you didn’t. They did. When you start building your own rockets that REALLY need to work across the knowns, then you’ll appreciate how the unknowns are handled. Long before we can depend on pure theory and hypothesis, we need to get it 99% (probably more than 99%) based on reality and such that it can be repeated over and over. I won’t get in a rocket until that’s done.

Making cable isn’t a blind game of darts. There are solid reasons cable works like they do. Little evidence supports the majority of performance lies anywhere but within the knowns. Again, the HOW and WHY of a design isn’t supported with subjective opinion, and yes subjectivity is worthless. I don’t have invisible friends in my lab. As long as the electrical are retained to the knowns, the tertiary discussions can safely ride to the moon and back but they aren’t the reason the rocket worked as well as it did.

We turn unknown into knowns. We continue to challenge the knowns as possibly incorrect in science. That’s how it works. It doesn’t say we won’t ever figure out copper’s metallurgical effects on an EM wave, it means until we do known and repeatably the same every time what a material does, it has to remains off the table as a design element.

The series II speaker cable are not made with unknowns. They follow the current EM properties of analog cable. What is NOT “sold” is what everyone will HEAR with the revised numbers.



Galen, sorry, I didn’t understand. At all.
First time I didn’t understand nearly a thing of what you wrote.

Did you mean that since different material choices in a given exact geometry with the same dielectrics can only be evaluated subjectively at this point, they shouldn’t be given value as design elements as such, just options for those willing to try them as “flavours” of preference?
I kind of get it, from a rigorous standpoint we can’t design based on the subjectively evaluated differences of materials, they come after the whole design ordeal.

Really I was just commenting on you saying that a material without measurements to prove its merits is useless (might’ve misunderstood though)… Since there are currently no capabilities to measure what the materials do to the signal anyway. The materials still make audible differences which are important.

No, there are TONS of information that describe what materials do in a design as they alter R, L and C. But you can’t repeatably design to a flavor of the month theory on an “exotic” material. A design is provably repeatable. I can adjust the C value with ANY dielectric to a given value, always, and in EVERY design by adjusting the physical dimensions based on the dielectric constant. I know the material’s capabilities.

Simply put, there needs to be a repeatable measurement for true design. Subjectivity is worthless to the customer. You may as well sell them something that is broken and then say, “well, we meant what you said” but in the end you didn’t do it. I have to do it, so does PS Audio and everyone else. There are hard standards that we KNOW we are selling and THAT is what you are buying. The subjective (tube rolling, copper draw science) can be isolated off to the side as long as it doesn’t break the products salient measurements you ARE paying for.

Can you sell a product with no hard numbers at all? Sure, convince the customer he has clothes on and everyone imagines a different wordrobe. There is zero repeatability and there is technically no product to define against that is the same for all users.

True, all theories and hypothosis exist in the LAB, and some derive true hard numbers. Until then, they should stay in the lab. Copper draw science is well defined as to what, how and why the draw temps and speeds change the copper. The differences are not defined in “sound” and never have been in any measurable way except resistivity. The COST is there for sure as time is money. A slower process is a more expensive one. That’s not subjective!



I hope the differences are some day defined, well, not in “sound” necessarily but precisely how the crystal boundary arrangement affects the EM wave. I guess research into this would by necessity have to include theory from the quantum physics side? Maybe it could be simplified at least in terms of expression to something not ultimately very complex.
For example the concept of rectification distortion happening due to copper oxide boundaries in the lattice is very straightforward to understand, just how it happens, but dunno how intricate a white paper on this with measurements would be.