Cat5e vs Audio Snake, Hear Signal Loss of 100 meter cables 

Now people can advertise their vintage multicore analog snakes as sounding more “warm” than twisted pair snakes haha.

Wooo!!! Great to see a new video out Dave, hopefully this means you'll be treating us with more in the near future :)

Dave, as a 30+ year audio veteran its easy to overlook the subtle details that DO make a difference. Thanks more making us think about those items! Looking forward to more info!!!

Hey Dave! Been a fan of yours for years ever since I first started doing music around 2011. Your information surpasses that entry level business and gets into the deep brass tacks and I really appreciate that about you, and I've been a fan of the RHCP since I can remember and I've seen them twice, once in Omaha in 2011 with Thundercat as the opener, and I suspect you were the man behind the console and it was one of the most full and crisp sounding rock shows I've seen, and I've seen over 100 shows since then. Just wanted to say I'm a big fan, and since you've uploaded again, quarantine must have you bored as F 😉 Lol

THANK YOU. This makes it really easy to explain to others!

Cool test! Glad to see a new video!

A very clear and comprehencive approach to a real issue in cable construction and layout. Thank you.

Dave. I love you man! You are like the NEW Bob Carver for my generation. One day, you got to tell us what you do for car audio in your ride. I have been using Cat5e and cat6 STP for years for audio. And I love it. I had a lot of people tell me they could not afford good speaker wires or RCA's, and I made them some for free when I heard that. Dude. Thank you for making this video. I could tell often that some of the RCA's I bought from the store just did not measure up when I used them in car audio compared to some of the stuff I used at home. And I'm glad you made this video. It confirmed something I had known for really long cable runs. You rock Dave!

As always, Awesome video Dave! Many thanks.

Had to double check to see how long it has been since you uploaded a video. Love applying your techniques to what I do and seeing the improvements instantly. I've probably watched every one of your videos at least 3 times, some 5+. Can't wait for more.

So glad I found this channel, love it!!!

I can't express the gratitude I have for everything you share

good to have u back

Very interesting indeed. Unrelated but interesting for me was the demonstration of the effect of reflective surfaces (your arm and your body) on what the mic was picking up as you started the demo.

About 10-15 years ago, I made what I call a Mix Rack. It was basically a DAW that sat on the stage with a remote session and control surface out front. This only ever got used when I would show up and there was something not right about the production supplied, a mix position that was in a booth and those sorts of situations. One thing that I had noticed right off the bat, is how clear my top end was when I eliminated the FOH snake - the mix rack would be sitting right next to the amp rack so there were no sends or returns going to FOH. Not such a big deal these days, but at the time it was an eye opener.

Thankyou for the video . Pleas let this be the gateway to more !

Very interesting, we switched over to data cable and stage box for our theater audio. It’s good to know we are ok for signal to stage

It would be interesting to do a test on the cross talk between the pairs for the audio snake and the data cable. At the end of your videos you should tell everyone to remember to click the like button so we don't forget. Thanks.

love your videos they are great for us sound nerds

Interesting video and showing how these cables perform side by side.

Thanks a lot sir I follow your strategy. It helps me lot.

I’ve did more then a few festivals for XM pre-merger where we’d pull back the audio from several stages to one broadcast center. We used good cat5 cable to run the feeds back to the booth. It was relatively cheap, small, light weight, and we usually just abandoned it in place on load out and saved time trying to get the hell out of there in the middle of the night. Sounded as good as any reasonably priced audio snake and you could just cut to the needed length as you set up. I imagine it would sound even better on better cable today. As an IT geek, I can tell you that the old stuff was good to 100mhz and the new stuff up to 350. 20 to 20khz is no problem. It’s got a shitload of twists so the noise rejection is pretty good as long as you stay away from power feeds. For live sound mixes that were often not pristine to begin with, it was plenty good enough.

...aways pushing audio to next step evolution!!! Thanks Dave you inspire us aways!

Dave’s a true audiophile. Been watching his videos, reminds of my favourite Sound geek, my husband, you sound guys are all so similar lol - even tho he’s a re-recording mixer.. great channel!

Thank you Dave!.

I missed the part where you explained the fact that we were listening to the balance differential and I was confused by the outcomes the whole time. Cool way to test!

We are blessed with a new video :D

Have never clicked on anything so fast as a fresh Dave Rat test video! I think the confusion in some of the comments stems from how you modded the cat box to monitor the difference between pin2 send vs return. some kind of polarity reversing / summing inside the cat box? Also, Dave while you’re there could I pls get your recommendation for current production reference cans? Both my D2000s are dead :(

No idea i was even subscribed to you but that was really f*ing cool.

Where have you been man? Glad to see you making videos again. I really enjoyed this one. Your older pro audio mixing videos were really great as well. Will you be doing any more of those?

DAVE RAT I salute you my guy. Glad to see you still active. I owe my career to many of your teachings. I am a software engineer that is now building tools for sound guys. But you helped me with more knowledge about sound than a university. Who woulda thunk you guys need computer techs and apps in 2020. This video perfect example. If I ever see you lunch on me. In the immortal words of BLACK FLAG. WE GONNA RISE ABOVE! (With Dave's help :-) ) #SALUTE! 🙏🙏🙏

Thanks Dave

Finally, a good RU-vid algorithm recommendation! Really glad I found your channel. I love your scientific:subjective balance, and this video was fascinating. P.S. I'm staying because of the rat. :)

For those that are not understanding why louder and brighter = more loss, let me see if I can explain it to you. If I am not mistaken, Dave is running a null test. In a null test, the more equal/similar the two compared signals are, the less you will hear ( theoretically nothing if the signals are identical ). Which then means the more dislike the two compared signals are, the less nulling you will have. Data cables are made to a much higher standard than your typical analog copper snake. The loss in highs is a result of capacitance between the leads. More capacitance, more loss in highs. Longer cable lengths and lower grade cable are more prone to signal loss because of this. Data cables are designed to have less loss so that data doesn't get lost ( quite literally ). With a null test, you have two signals that are out of polarity with each other, in this case, we are reading the difference between each end of the cable. If the two signals are identical, you will have 100% nulling ( no sound ). He accentuated the highs on one of the signals so that there is no chance for there to be 100% nulling. What this allows is for you to hear the nulling that does or does not occur. So less volume and less highs mean less signal loss ( the two signals are more alike than different ). With the analog snake, you had more volume and more highs because the signal was not nulling as much, meaning that the signal at each end of the cable was more dissimilar. Does that make sense?

20 Million thumbs UP..Sir

Was totally surprised by this result. With higher amount of signal carrying material, the thicker wire gauge, I expected the audio snake to do better than the data cable. Wonder if it's the extra twists in the data cable that help improve signal stability... Also didn't realize I had missed this one. Thanks, now back to wrapping my head around DI's and transformers in your latest videos...

Really good vid thank you.

Cool, love the tech talk. Thanks

Great vid. Thanks for the schooling

I designed, 24 years ago, the first and only (to my knowledge) ACTIVE INTERCONNECT. It had excellent reviews in the Audio Press at the time. It could compensate any capacitance, which is the chief culprit, responsible for the high frequency loss across the cable. Other factors for the poor performance are the usually inadequate drive capabilities of the pre-amp output stage to drive a capacitive load which a cable presents to the pre-amp. Audio cables usually are of a larger diameter so as to minimize the capacitance per unit length. I used a little known circuit topology to completely compensate the cables capacitance, very simple circuit, very effective. The user could even use his already high quality cables which would be even more beneficial. And it was possible, by the user to tailor the cable driver to his exact preference by simply plugging in a different value of resistor. I was active in the High End Audio game for only a few short years. Designed the first Digital Pre-amp which had the Golden Ears Brigade buzzing. Left that game because I found the people involved in it to be very uneducated, even downright unethical, besides, my Industrial Electronics side of the company took of exponentially, so I had very little time anyhow. But I often wondered how far it could have gone. The main opposition even offered me to become their technical director as they were unable to compete. Data Cables are not very suitable for audio. It is misleading to say the have data rates of 10 to 100MHz and more. True, but that's only possible with very special drive and receive circuitry. These drivers and receivers use Pre and De- Emphasis for instance, something that cant be done with Audio. Its a fascinating subject, but it requires a great deal of specialized know-how. Happy listening.............

Thank you

Wow, I had forgotten that I have a compressor on my computer monitors and I wasn't hearing the difference and I was scratching my head for a minute cause I couldn't hear the difference right away... just the attack and release.. Ha Ha :D ... Us nerdy people and our gadgets... Thanks Dave for all you do for us... Been doing sound for a long time but I always learn something valuable every time I watch your videos... I need to definitely (Turn off the Compressor and Limiter) when you are giving sound examples...

25yrs ago I was using cat5 in vehicles due the the issue of coaxial cable being more prone to the magnetic induction of AC currents on the chassis. The only other problem that was a significant factor at the time was, most car audio gear was single ended and not balanced. So people needed equipment with low impedance signal systems as to mitigate the imbalance and common mode noise

Awesome!

Nice demo! Cat 5e specs 1GHz (1000MHz, 1000000 kHz) which is indeed a comfortable margin above 20kHz. Given the advantages in terms of weight and durability, this is clearly a no-brainer choice. .. thanks for make it so vividly clear. Of course someone is bound to say, "I hate digital. I won't even use cables designed for digital in my signal path regardless of their numerous advantages!"

#1: Common mode noise rejection can be 10x better on star-quad or "quad" audio cable. This can be more important than high frequency losses, especially with low-output dynamic microphones like an SM58. #2: High frequency loss is a combination of output (source) impedance *and* the capacitance of the cable. an active XLR output on say a Behringer x32 mixer is 75ohms so that should be able to drive 500METERS of almost any audio cable without any noticeable frequency loss. There would be no real sonic benefit in using cat5e cable for *line-level* audio use other than cost (which is nice) but for sure you should warn people about the folly's of using cat5/6/7 wire for *microphone level* signals.

Man... The new cable is Approved.

I like this approach. It fits in well with my habit of putting a tweeter and cap across power and speaker wires to hear problem. I got that idea from someone in the car audio industry. He said he installs a tweeter with a crossover and a burnout preventer on the output of his sub installs and mounts the tweeter near the drivers head. It sounds awful if you clip the sub amp so his customers don't blow up as much gear. RU-vid didn't really convey how that worked either, but in his FLAC file it was plain as day. It would be nice if you posted some high bitrate lossless audio files of this. Frankly I can't hear anything except a slight gain difference over RU-vid in that first test. The second test showed a clear gain difference, but how does it sound with makeup gain? And how do the cables do with interference? Split a lamp cord and run one conductor through the reel and connect an old-fashioned light bulb and dimmer. That'll tell you what's got good interference rejection and what doesn't.

just clarifying. a " perfect cable " in this test would be silent and the loud frequencies would be the frequencies that have losses on real cables.

Could you do a comparison of cross talk between channels? How about induced AC line noise?

Very nice that you are doing this sir .. people don’t believe in cat6 and cat7

Nice!

I've used Cat5 cable as speaker cable for years. It may be 5e, I'm not sure, it's the purple sleeved cable... It's the solid core stuff. I read an article on the TNT Audio website a long time ago, made some up and haven't looked back. Sold all my expensive posh speaker cable. They also recommend making a plaited version which I did do a few times but the thumb blisters were a pain :\ Great video, thanks :)

Interesting video Dave. A quick question: if you were to A-B test the analogue and and digital cables in a production/live environment, do you think the audience would perceive a difference?

It's funny, moment you started talking, I'm like, "I've heard this guy's voice before". Couldn't place it......thought maybe I ran into you in the trenches somewhere (nope). But then I realized there's a RHCP FOH vid that I saw years ago. Weird to remember a voice for that long. I guess you do have a pretty distinct voice. Anyway, thanks for the video!

High frequency roll off is preferable to across the range frequency loss as you can upp the treble a bit to reform the loss , less loss in midrange and lower range from w1

That's it. I'm using my RJ45 for my RCA's instead of buying expensive audio cables 😅 Thanks Man very informative video!

Not a lot of things get me to click a notification faster than a new @Dave Rat video ..

Dave, would love to see a loss and latency test between analogue over cat cable vs digital (like Dante)

Thank you!

Excellent vid - thank you! The super cat is shielded cable, so I'd assume that any good quality shielded/grounded CAT5e cable would be acceptable, but would appreciate your thoughts on that? Is super cat larger copper than "normal" network/data CAT5e? I use Ubiquiti ToughCable carrier grade cable for all my data runs and it's far superior quality to any other grounded/shielded cable I've used professionally. Thoughts? Thanks, Dave! Great stuff.

category 5 cable is 4 twisted pairs. category 7 cable is 4 twisted pairs separated by a core insulator and each pair has a grounded shield wrap accompanied by an outer grounded shielded wrap for the whole cable data cable is designed to carry a differential signal pair while audio cable is designed to carry a differential signal with a ground reference

I would never have believed that you could run audio through a data cable. Wow. Live and learn.

Wow where have you been?! I guess quarantine gives you more time to make videos for us little people. Lol. Fyi the hair shocked me. Used to watching you older videos.

Thanks Dave may we half some more please

Very interesting, I thought there would have to be a difference, in favour of data,

Nice way to actually listen to signal loss! A question, doesn't the roll of cable act as an inductor, which may cause extra HF loss?

Great comparison of the cable types! But what about the phase shift due to propagation delay over the 100 m?! Doesn´t this add to the difference in signal between the voltage drop and the direct connection?

How about creating a high impedance audio source to highlight small cable differences?

Dave, what about the comparison of OHMS per foot and also the Capacitance value of each coil?

1) What is the best speaker cable you have found? 2) Where can I buy Supercat unterminated for home hi-fi use in the UK? 3) Do you notice much roll off between speaker cable in the smaller runs like 3-10 meters? General home hi-fi? Would Supercat allow for better sound in these setups as well or just for longer runs? 4) Would you recommended shielded CAT vs speaker cable for home hi-fi?

Ironically I think Dave's mic audio could do with some HF loss. :) I too am surprised there's more HF loss in the thicker audio cable. Glad I don't have to worry about that in my cat6 audio runs.

When the "hack" is actually an upgrade.

Interesting. Could there be more capacitance in the blue cable or somewhere in the signal chain? My elec theory is very rusty.

Well, hum, never use 100m câble...this review is right for pro users who make sound in larges venues. With 30 m snake, no serious loss with high quality stuff. Thank for your pertinent and usefull advices. Cheers

Dave, would you mind mic'ing the speaker next time? I hear a tiny difference, but honestly, it is so small (especially considering how much the high frequencies were boosted) that I can't see a tangible reason to go digital of your analog snake is still passing relatively clean audio. What are your thoughts?

If you're listening to the current flow in the cable, do consider that cat5e is about 3 or more times the resistance per meter than good twisted pair, haven't done the math on what level difference that will equate to, just a thought, and what about noise immunity on the cat5e vs the snake? ... curious

Bit confused at the end there Dave. Did you mix up which ones in the up position?

Hi Dave! We observing losses especially at HF. I'm just thinking if it counts, or not, that you made an air-core coil with about 100 turns from that cables (technically a low-pass filter). I don't want to argue, with your findings, just asking.

Looks like I will be raiding my sons thousand foot spool of cat 7a cable, my son has a spool of cat 8 cable on order should I expect any improvement that is noticeable between cat 7 and cat 8 ?.

I’d like to see testing on this with balanced audio instead of just looking at one side by itself. Is capacitance between hot and cold going to have an effect? Also crosstalk as someone else mentioned.

I use öhlin token ring data cabel to my XLR that realy opend the music Up (ITs data cabel used on oilrigs) I found that Meny of the XLR Cabels cut a Lot of top and buttom ...

I'm interested in the technical cause(s) of this effect - and this leads me to think that in practice the amounts of high frequency loss may depend on the actual use of the pairs. I wonder if the increased high frequency loss you are getting in the conventional snake is mostly due to a higher capacitance between the signal pair and grounded screen (shield) than is the case with the Cat5e. If I understand the test-setup correctly, it seems reasonable to expect a higher capacitance to ground in the conventional snake, as its pairs are individually screened, i.e. are each closely surrounded by ground, while those of screened Cat5e are, I believe, usually only screened overall, around all 4 pairs together. So when using the Cat5e, the pair being used is not so closely surrounded by ground, because the other 3 pairs space it away from the screen - assuming that the unused three pairs in this test are left open-circuit - not grounded. If they were grounded (or had a low impedance connection to ground, as in some audio applications - depending on the source and destination equipment), then maybe each pair would in effect be surrounded by ground more similarly to how they are in the conventional snake, and so less difference in high frequency loss would be seen in a real application that makes use of all 4 pairs. It would perhaps be interesting to repeat the test with all pairs in use in various different ways (mics, DI, returns, etc). I realise there are also other factors that affect a cable's capacitance to ground besides spacing, e.g. conductor insulation material and gauge, wire thickness, etc. Also, the effect of any given cable capacitance value on the high frequency loss actually experienced depends on the signal source impedance - so, again, in practice the results experienced ought to vary depending on the actual use cases. Having said all that, I would be quite surprised if there are many (if any) cases where (under identical test conditions) the Cat5e had more high frequency loss than the conventional snake.

Dave, can you try the same audio cable coiled up vs. audio cable off the reel on a straight run (like from the stage to a mixer like you might find in a small room. I know sometimes we leave the audio cables coiled up (as extra length not needed). Just wondered how that affects analog signal quality.

I am also a little confused. Why is losing the high frequency in the data cable better?

The purple cable looks to be more than CAT-5e, cable and their marking not shown. Note, many data cables have a solid core and not stranded. Not deesigned to the bent and flexed all the time.

If this is true, then why are we still using XLR cables? Why don’t we have RJ-45 connectors on audio equipment? As a DJ using powered speakers, do you recommend converting my balanced cables to cat 5 and then back to balanced?

Oh Boy, is this the end of all the voodoo-stuff around cables? Will there be no more noise about the noise. And who will dare to disturb the sound of silence... Good job, Dave ;-))

I have a bunch of shielded procab cat5e cable, but the shield isn't connected to the jacket of the ethercon plug. Can I still use cat boxes with this to run balanced audio over?

The treasure hual you have must be amazing

From 6 mins, I got a little confused with the brighter W1 cable being brighter and with more gain, but the losses are more in the high frq'y of the audio W! cable. Are you suggesting the brighter sound of the W! cable is due to lower frequency loss, causing the brighter sound? I can clearly hear there is a difference, but I am not in the room with you. Would you be looking at using cat 7 in the future over the W1 audio cable?

It's exciting. But I am confused. Does that mean I have to rewire my studio with standard cat 7 cables and substitute my expensive mogami audio cables for a better frequency Range?!

The audio cable, higher capacitance / ft? vs linear attenuation of the twisted pair data

I've been wondering for a little while now why current signalling isn't used more in pro audio. Most of the problems of voltage signalling (induced noise, cable losses and colouration etc) which has been the norm for at least 80 years, probably since the beginning of electrical audio, disappear when using current signalling. I've been using it for some time to pass audio, digital and analog signals between devices in my entirely non-isolated DC off-grid home, to get rid of the ground differential noise that occurs when there are devices using varying amounts of power connected to the same DC power source (the same issue that DI boxes were invented for). Sending is simply a transistor or two to convert from a voltage signal to a current signal, and then receiving is just a resistor to convert the current back to a voltage, with a capacitor to block the DC part of the signal (it needs a 1 or 2 mA DC bias). As long as all the equipment in the system shares a common earth, even if there are quite large AC voltage differentials between earth points, only one conductor per signal is required: a cat5e could carry 8 channels, without any interference between channels or from outside, with no signal loss at any frequency, over any distance. I'm just wondering if you have any idea why this isn't done? or maybe it is, and I just don't know about it?

UTP data cables are superior analog (and digital!) audio cables. By design. Good to see someone doing a video on this. Speakers and Amps should have RJ45 connectors instead of banana plugs!

I've watched it three times. Can you please explain how high frequency "loss" results in greater amplitude of the high s at the return?

Would the results be different if the wires are not in a coil?

If you invested in a TDR (Time Domain Reflectometer) a few things become clear. STP CAT 5 is 110 to 120 ohms impedance. To eliminate the issues of cable capacitance, etc except delay, go with a digital stage box. Sampling as close to the source gives the cleanest signal. Unfortunately all digital DSP, A/D, D/A conversions take time so there is latency. Running one CAT 5 to a 48 Channel stage box with 16 returns is much easier for the roadie. The big importance with data cable is shielding. It is UTP or STP. Don't skimp here. Next in line is loss. Smaller conductors have higher end to end resistance than larger cable, but most snakes are about the same conductor size. Capacitance per foot attenuates high frequency when the load is higher in impedance than the cable. Lower capacitance results in higher impedance and less loss at high frequencies.. which is what you want for a cable passing 350MHZ or less as eg a CAT-5E. Yes less is fine such as DC to 50KHZ. Watch for higher current phantom power mics. They may have too much voltage drop on a long cable (500 ft +) to function properly. FYI, CAT5 cable makes excellent Data Cable such as DMX. It is the perfect impedance for it. Many audio snakes are built with anti-static insulation so the cable is not microphonic. google cable rustle for this effect. Digital network cable will have mixed results in this realm. Many audio snakes have larger conductors with more capacitance per foot for durability in handling. This results in lower impedance and higher capacitance load per foot. The Dielectric loss in audio cable is pretty high in some snakes as materials are made for flexibility, durability, and take road abuse. Google dielectric heating and loss. This is energy absorbed by the insulation. To prevent loss of high frequency data, data cable is built with very low dielectric loss insulation. Most microphone cables are about 60 ohms impedance. CAT5 is 110 to 120 ohms. This is closer to the 200 to 600 ohm microphone input impedance of mixers.

One of the most important things to bear in mind with data cabling is that attenuation loss is not particularly critical, it's signal to noise that matters. Higher quality data cables such as Cat7 are properly shielded, but audio cables are typically well shielded, certainly when compared to Cat5, which is not shielded at all. Cat5 is designed to be *cheap* and costs a hell of a lot less than decent audio cabling. Should also test some of the high quality Aerospace multicore cables which were used for HF analogue display signals such as VGA, some of those are well-qualified for audio use.

whats your opinion on fiber optic to xlr extenders?