Yes, cables do make a difference -- regardless of price...

Good post from @benanders but I really appreciate that the OP recorded voice tracks using different cables so that they could go back and compare the tracks repeatedly. If you are comparing different recorded tracks back to back it removes a lot of the opportunity for bias. My issue with cable enthusiasts is where they disconnect one set of cables, plug in a new much more expensive pair, listen for weeks and claim that once the new cables "broke in" they sounded much better than the original cables.

I had a home recording studio in the 90's and I used this recording technique to compare preamps and microphones (I used a close-miced acoustic guitar which is very revealing). I even used this process to compare the sound of different tubes in a couple of my mic preamps as well as an opamp upgrade on one of my solid state preams. I posted many of these files on a recording forum and invited folks to listen and submit their reactions. Most listeners were surprised that the differences were so obvious. I used Mogami cables exclusively and it never occurred to me to compare XLR cables this way.

FWIW, I've never been able to hear the difference in interconnects. I realize it woulc be a hassle but it would be interesting if @crozbo could post these files so us skeptics could hear for ourselves.

The OP touched on something I have wondered about for years. If cables really make a difference then we should be able to hear the difference between two kinds of cable on  the same recorded track. My skeptical nature tells me that if there really were differences in cables audiophiles would have done this and posted the files to prove their point. Sure, cables are "system dependent" but, I'm sorry, that sounds like a cop out to me.

Does anyone know of a well controlled A/B/X test where cable differences were demonstrated? I've never heard of one. I know of dozens of tests where listeners failed to hear a difference but no test in which they were successful.

Lastly, my position is that if audiophiles hear a difference in cables then, for them, the difference exists. Cables seem to give a many of our creed a lot of pleasure and if that's where they want to spend their money then more power to them. I don't know if I'm lucky or cursed that I can't seem to hear the difference.

OK there are 3 major problems that audiophile people have that no audio engineer who records or mixes music concerning cables has and that is 
#1- Of course a cheap cable is not going to sound very good there are many reasons why poor quality cables don't sound good from being to thin to oxidation to having crimped ends. Audiophiles who compare inadequate cables to well built cables are comparing apples and oranges.
#2- I started my sound career 40 years ago live mixing and editing radio shows, the other editors always listened at 2.5x speed it was hard for me to understand the dialogue at first but after my brain "burned in" I was able to edit at fast speed. That to say your brain is variable it compensates for everything in the sound world and when you pay 10k for your AC cable you will hear a difference, there are definitely many emotional reasons but zero technical reasons.
#3- The description of the way cables make a difference include words like warm, airy, tighter low end, and imaging. All these things can be measured and measured very easily and accurately with way more resolution than your ears so those descriptions are proven incorrect. If the description were less technical like "musical", "beautiful", or "pastoral" then you could say there is no measurement for that, and ok who can argue with that, but that is not the case in most of your descriptions in this group. 

@kraftwerkturbo
The company I was working for was building a new set of studios from the ground up and money was not a problem custom everything including famous speaker designer who came in and designed the speakers and wall they were mounted it (he didn't care about the speaker cables) so the engineers did lots of tests and ended up using Romex as the speaker cable for whatever reason they said it sounded the best. I've heard of people trying RG59 coax and had the same results you did. These examples are preferences (I like strawberry or vanilla ice cream better) this is what happens when there is so much confusion. Audiophile cables of over a few hundred dollars each should always have unmeasurable differences or else there is a problem with the cable.

I like how if you don't hear a difference you're not "open minded" or don't have a "highly resolving" system. 😐

The hardware part of the system is simple:  Audio Technica Cardioid Condenser Microphone AT2035 connected with a XLR cable to the Focusrite Scarlett 2i2 preamp.

@crozbo I've said this many times on line; if your equipment supports AES48 then the interconnect cables will hardly make any difference. I don't think that mic supports it, so you'll have to audition the cables to find the one that works best for you.

Its pretty amazing to me to see how mics have changed over the decades; that you can buy a large element condenser mic now for only $150.00! I'm sure that's given a lot of people access to making music that were shut out before.

But to make a mic at that price you have to cut some corners somewhere, and I suspect that it does so with the tiny preamp that's built into the mic. If it references ground, then cable immunity is out the window. IOW it might be a balanced output, but it doesn't support AES48, the balanced standard. The Audio Technica site doesn't say, so this is theory on my part. 

We use Neumann U67s in my studio, running into custom vacuum tube mic preamps, which drive the recorder directly, bypassing our mixer. You can put whatever cable you want in that setup and you don't hear a difference. But the Neumanns have a proper floating output which is the only way AES48 can be supported. They also have no problem driving 150 Ohms, which is a typical microphone level input impedance (quite different from the 3KOhm input of the Focusrite).

These sort of things have to be taken into account when making statements about cables. When you have the lower impedances of the pro equipment, cable differences vanish.