"DHCP sounds worse than a fixed IP address"

[Palladium]

If your IP has as many zeroes as possibly you get quieter blacker background 10.0.0.1 is much better than 192.168.1.99 the 9 are the worst’s
/s

More dynamic range can be achieved with a class A netmask than a class C!

 

[fatoldgit]

Glad to see so many network savvy people in here.

Indeed.

My background is (was, now retired) software development on mission critical, high throughput data interchange backends (typically with a database on at least the receiving end but more generally on both the send and receive), hosted on Unix then latterly Linux, all predicated on sockets and the TCP/IP stack at the lowest level. So have some understanding of this.

These kinds of stupid proclamations used to drive me mad and I would try and help the poor souls that brought into it but I don't bother any more...it's their money and if they don't have the wherewithall to do some basic research as to how ethernet and tcp/ip work then...whatever.

Do you have Black Friday specials on your routers...if so might pick one up next time around.

Peter

 

[fatoldgit]

More dynamic range can be achieved with a class A netmask than a class C!

Token Ring... that's where it's at!!!!

 

[voodooless]

According to the good people at Lejonklou, setting a fixed IP address for their Kalla streamer will lead to "higher musical performance".

https://www.lejonklou.com/tips/how-to-set-kalla-to-fixed-ip-for-improved-performance/

Can somebody reach out and ask them what mechanism makes this performance gain work ?

Oh boy, that streaming thing is truly next level! € 5700 for a product, and then provide only a 2-page manual, and a product that has basically zero connectivity options other than LAN in and RCA out. No proper specs either... It's really idiotic

 

[Multicore]

Token Ring... that's where it's at!!!!

The one that rules them all?

 

[threni]

Can somebody reach out and ask them what mechanism makes this performance gain work ?

Way ahead of you - I emailed them yesterday! I was concerned that should I splash out on their expensive knulli I'd have to forgo the convenient pinning of a MAC address to an IP on the DHCP server - which is less of a faff than manually assigning IP addresses to each device and having to make a note of the ones I'd used to avoid clashes - to avoid what I called "...the inferior sound of a connection where the IP address could differ upon each connection to the DHCP server". The reply states that I need not worry - I get wonderful sound and ease of IP address assignment:

"The musical performance in Källa achieved by a fixed IP is a very small one. So it’s rather a bonus thing that we discovered and in no way a necessity to enjoy it. I normally don’t use the fixed IP setting at the lab, but I do at home as I don’t change things there very often.

Anyway, the increased performance comes only from setting Källa to a fixed address and therefore shutting down the little process that runs in order to keep an eye out for when the DHCP address lease time is ending. When it ends, Källa gets a new (usually the same) address and lease time from the router and the timer restarts. When Källa is set to fixed IP, it kills this process.

So it doesn’t really matter what you do on the router side, the small increase in performance is obtained by going to Källa’s web interface and entering a fixed address, which stops its DHCP lease process from running."

I think this is cognitive bias, rather than a scam/ignorance - hard to be too upset about to be honest. Made me smile, anyway.

 

[voodooless]

I think this is cognitive bias, rather than a scam/ignorance - hard to be too upset about to be honest. Made me smile, anyway.

No, I'm sorry, not good enough! If a company thinks it's okay to have a customer pay € 5700 for such a product, they'd better have their shit in order.

 

[Purité Audio]

They know what they are doing, clearly an engineering focused manufacturer. From their ‘Tundra’ monos,
Quote,
‘And being less bothered with the amount of work required at each moment means that Tundra Mono delivers the emotions of the music more convincingly.’
Keith

 

[Hayabusa]

Makes perfect sense.

What you are all missing is the DHCP lease time.

With a fixed IP, the computer doesn't need to keep a energy sapping timer running.

With DHCP, the computer now needs to run said energy sapping timer so it knows when to renew the connection.

So it's obvious that the energy sapping timer causes issues with the temporal flow of co-located packets within the ethernet frame stream which reduces top end sparkle, muddies the mid range and gives you bloated bass.

My system has never sounded better since I installed a $5k 10GB network switch with $3k external reclocker, started using expensive CAT 8 cables capable of 40GB/s and fronted these all with $7K of LPS's.

I have also spent $4k on special footers for the switch, reclocker and lps's which really help reduce packet flow congestion.

Of course I need the headroom of this setup, which is capable of 5000 MBytes/second, to avoid contention/packet retries when playing back my 200 KiloByte/second FLAC tracks. You gotta have head room, just in case!!!!.

But I plan to convert my 60,000 FLAC tracks to WAV cause I am told that the time and energy needed to convert FLAC to raw PCM corrupts my sound and that WAV is the way to go.

I can also see moving to fixed IP addresses will lift my sound to a new level. I will report back soon!!!!

Peter

Also make sure you chose the fixed ip adress carefully as the serialized bit pattern can have unexpected side effects

 

[tw99]

So it doesn’t really matter what you do on the router side, the small increase in performance is obtained by going to Källa’s web interface and entering a fixed address, which stops its DHCP lease process from running."

I think this is cognitive bias, rather than a scam/ignorance - hard to be too upset about to be honest. Made me smile, anyway.

Oh, I'm sure the guy believes it.

However, his "audiophile logic" type explanation doesn't make any sense either. Why is having less processes running a good thing for sound quality ?

One could equally well argue that having the processor running at close to 100% all the time will be better because it will produce less switching noise from going through multiple power/idle states, and hence you will get much more inky blackness or whatever

 

[phofman]

IIUC Källa runs Asus Tinkerboard S which has an old quad-core RK3288 (32bit A-17), 2GB DDR3. Of course that extra process waiting on select for the renewal timeout has zero impact on the audio, even if the analog circuits are powered by the same supply as the computer (which is claimed to have a sonic advantage... https://www.lejonklou.com/forum/viewtopic.php?p=56234#p56234 , go figure... ) Since the project started already in 2017, its author has to have a good command of linux because using linux on non-RPi ARM at that time was not common. IMO he knows it's just a marketing... but that's everywhere...

 

[somebodyelse]

Do you mean this guy? https://www.machinadynamica.com/machina17.htm

He actually goes into very in-depth explanations as to how taping little bags of rocks to your cables could work. What I find interesting is he mixes more-or-less accurate science with total snake oil nonsense about audio. I am not sure if he ever sold pieces of paper but he's sold everything else.

Or maybe these goofballs?

https://www.synergisticresearch.com/isolation/accessories/carbon-tuning-disks/ - little stickers you put on your cables that are supposed to improve the sound. In this case the attempt at explanation is thinner:



It "interacts with EM fields... to improve sound quality", you see.

I thought it might have been the late Peter Belt who claimed to have 'treated' various things so that they could influence the sound quality via 'quantum', 'morphic resonance' and more.

 

[Multicore]

I think this is cognitive bias, rather than a scam/ignorance - hard to be too upset about to be honest. Made me smile, anyway.

No. This bafflegab was written with care.

The supplicant's address won't change while it uses the network. Even after going away and coming back it won't change unless the address pool is oversubscribed. Unlikely but even if it happened it wouldn't affect audio since the address is sable while streaming. This is basic stuff and I do not believe the author is unaware of it. Or is the author trying to suggest that the state of the supplicant process in memory makes the difference to audio?

Note also that the author is carefully trying to have it both ways. That doesn't wash. Either the difference is audible and important or it isn't and should not be mentioned at all.

Before I read this answer you quoted I thought this thread was just a good laugh. But the answer annoyed me.

 

[CuteStudio]

€ 5700 for such a product,

Wow, that's expensive.

A raspberry Pi 3B has multiple cores of GHz levels and is about £35, so what's their excuse?
Often I use a Pi + the Apple £9 DAC, but since discovering the Joy of USB Isolators (and the tweak to get native 44.1kHz from the DAC) it just gets plugged into my music laptop

So the (fanless) laptop was around £350, the DAC £9, the US Isolator + upgrade PSU chip £15, Total about €411.4, so I now have €5288.6 free for booze and hookers - a great result !!

Now where did I leave that table of fixed IPs I'd assigned? Oh - wait, only my router has a fixed IP

 

[Nopiano]

I suppose that reading a manufacturer’s recommendation and acting on it is somewhat similar to listening to cable demo where the presenter tells you what you will hear. That implanted idea becomes what you experience, as we are all highly suggestible whether we think we are or not.

Ironically, therefore, I would bet that the majority who follow the instructions will hear an improvement!

 

[kemmler3D]

I think this is cognitive bias, rather than a scam/ignorance - hard to be too upset about to be honest. Made me smile, anyway.

I agree with those saying it's "a scam" although I think that's a little strong. The clue:

The musical performance in Källa achieved by a fixed IP is a very small one. So it’s rather a bonus thing that we discovered and in no way a necessity to enjoy it.

Since they know it does nothing for the audio, they hedge by saying it's a "very small" audible benefit. This is so you don't get upset when you don't hear anything. I'm sure you've noticed that in most cases, subjective audiophile stuff is claimed to have a night-and-day effect, deeper blacks, wider soundstage, blah blah blah. They're downplaying it because you're digging and they think you might actually A/B test it.

 

[Mr. Widget]

But my home networking is perfectly happy running on 2.5GE and Wifi6. Would be perfectly fine with 1GE and Wifi5, too. :-D

Thank you!

People jonesing for 10GE home networks and 8K TVs are just compensating for something...

 

[sq225917]

If your IP has as many zeroes as possibly you get quieter blacker background 10.0.0.1 is much better than 192.168.1.99 the 9 are the worst’s
/s

We should push this as fact

 

[SKBubba]

Do you mean this guy? https://www.machinadynamica.com/machina17.htm

I PayPal'd that guy a $5 tip one time for making me laugh so hard. I told him I was working on a quantum powered Faraday cage to protect against emf and solar flares. I said it kept vanishing into another dimension and wondered if he would help me fix it and partner up to bring it to market.

 

[Palladium]

Token Ring... that's where it's at!!!!

Imagine how dark the background is with all those zeroes in IPv6.