Snake Oil Department, Top This

[Sal1950]

Lowell George was (and I am, with some incredulity, paraphrasing & quoting snippets of google's AI summary) "linked to" Fraternity of Man "through his earlier group, the Factory".
Most of us (I reckon) know Don't Bogart that Joint, though, from Little Feat's* very fine live album Waiting for Columbus.

______________
* Lowell George's later group.

One of my favorite all time bands.
I saw Little Feat sometime in the early 80s (I think) at Chicago's House of Blues.
A great time was had by all !!!

I don't remember Feat' doing the song though, I got high for many years on the original Fraternity version, I can remember rolling a few in Nam during my Southeast Asian vacation.

 

[Labhound]

One of my favorite all time bands.
I saw Little Feat sometime in the early 80s (I think) at Chicago's House of Blues.
A great time was had by all !!!

I don't remember Feat' doing the song though, I got high for many years on the original Fraternity version, I can remember rolling a few in Nam during my Southeast Asian vacation.

I saw Little Feat in 1978 at the Royal Oak Music Theater (Detroit area). Lowell George white bib overalls era.

 

[Mnyb]

After watching Danny's latest videos, I officially nominate GR Research as the holy snake oil emperor.

Thank you for taking one for the team I would never watch a Danny video

 

[Audiofire]

What the audiophool arguments boil down to:

 

[ErVikingo]

Most of these "tweaks" are outright fraud or technical changes that fall well below/above the audibility threshold. In the case of the ones that actually have technical merit (there might be some), there's the issue of measurable vs audible.

It’s the audio equivalent of putting a race tune on a commuter car: there’s a technical change, but it’s in a performance range you’ll never actually utilize. One of our engineers calls these "spec-sheet engineering". We had a guy that brought his factory racecar to which he added all sorts of tweaks (aero, exhaust and tuning), measured a higher peak HP and torque (yet not a flatter curve). We had 4 drivers a/b that car against my "bone stock" factory race car (3 serial numbers away from his), lap times were materially similar for every driver. Does it look cooler, yes; does it sound wicked, yes; is it faster, nope. And, I won the championship and he came a distant 5th!

At the end of the day, this is a hobby, and those "upgrades" sell regardless of whether they move the needle in the real world. Like they say in MX, the drug problem in the USA is not our doing, you created the market, we just supply it. As long as audiophiles demand a fix for problems they can't actually hear, the industry will be happy to sell it to them.

 

[MatrixS2000]

Yeah I got that.

I’m not sure if the eye charts were to show that the data was the same.
But any noise jiggling the ground or power rails would be removed with the balans.

The eye diagrams are showing that the noise was not an issue on the ethernet cable in all of the tests conducted. If there are no bit errors, the data would be the same for all tests (assuming the same file was used in each test).

Any noise on the ethernet cable gets attenuated by the switch port transformers which typically provide 30–50dB CMRR. That’s why you don't see ferrite cores in data centres or enterprise wiring closets.

 

[antcollinet]

The eye diagrams are showing that the noise was not an issue on the ethernet cable in all of the tests conducted. If there are no bit errors, the data would be the same for all tests (assuming the same file was used in each test).

Any noise on the ethernet cable gets attenuated by the switch port transformers which typically provide 30–50dB CMRR. That’s why you don't see ferrite cores in data centres or enterprise wiring closets.

In any case bit errors don't mattter. Ethernet implements full error connection. Otherwise it couldn't be used for all the other applications (including banking) for which accurate data is mandatory.

 

[ErVikingo]

The only area I could think anyone could argue that an ethernet could cause issues would be electrical noise BUT, ethernet uses isolation transformers so NOPE.

 

[pablolie]

In any case bit errors don't mattter. Ethernet implements full error connection. Otherwise it couldn't be used for all the other applications (including banking) for which accurate data is mandatory.

Exactly... anyone that thinks that digital requires perfect looking 0-1 "square states" doesn't understand digital transmission at all, and it's a waste of time to try to explain it to them.

 

[MatrixS2000]

In any case bit errors don't mattter. Ethernet implements full error connection. Otherwise it couldn't be used for all the other applications (including banking) for which accurate data is mandatory.

This is incorrect. Ethernet uses error detection via CRC-32. That means it detects the bit error and tosses the frame - it has no retransmission capabilities. For retransmission, ethernet relies on a higher layer protocol (L4) - TCP to request the retransmission. Relating this back to audio. Qobuz uses TCP and buffering to hide retransmission for smooth playback. UDP is another upper layer protocol but it does not have a mechanism for retransmission. Airplay uses this protocol, you will hear click/pops/dropouts when bit errors happen.

PS. I was the Lead Systems Engineer or Architect for each of the 5 largest Canadian banks and the 3 largest Canadian Insurance companies over the course of my career with 2 very large networking companies (I am now retired).

 

[Sal1950]

Exactly... anyone that thinks that digital requires perfect looking 0-1 "square states" doesn't understand digital transmission at all, and it's a waste of time to try to explain it to them.

Oh, but it can get all jittery

 

[Redacted]

I think this is a Miele.. I'm telling you as an engineer, Miele make by far the the best laundry washers and dish washers commonly available.

But now! A new product! This looks like an audiophile version of a Miele (three trays?) that has an entirely reasonable 2000% markup, owing to the sticker that says "audiophile" on it. You can buy one from me! As usual, cash in brown envelope only. Not snake oil in this case. N.B. the very top tray (unshown) is for the dirty stylii from cartridges.

.
.

 

[MatrixS2000]

I think this is a Miele.. I'm telling you as an engineer, Miele make by far the the best laundry washers and dish washers commonly available.

But now! A new product! This looks like an audiophile version of a Miele (three trays?) that has an entirely reasonable 2000% markup, owing to the sticker that says "audiophile" on it. You can buy one from me! As usual, cash in brown envelope only. Not snake oil in this case. N.B. the very top tray (unshown) is for the dirty stylii from cartridges.

.
.


View attachment 532277

You need to post that pic in the humour thread!

 

[Redacted]

You need to post that pic in the humour thread!

Nonsense. I already have 3 firm orders. If I post in 'humour', my business model will be shot through.

 

[Sal1950]

I think this is a Miele.. I'm telling you as an engineer, Miele make by far the the best laundry washers and dish washers commonly available.

But now! A new product! This looks like an audiophile version of a Miele (three trays?) that has an entirely reasonable 2000% markup, owing to the sticker that says "audiophile" on it. You can buy one from me! As usual, cash in brown envelope only. Not snake oil in this case. N.B. the very top tray (unshown) is for the dirty stylii from cartridges.

.
.


View attachment 532277

Be sure to set the drying temp to Sahara

 

[MatrixS2000]

Nonsense. I already have 3 firm orders. If I post in 'humour', my business model will be shot through.

That too!

 

[Redacted]

That too!

Please cease and desist - I do not wish to pay punitive "success and freedom" tax for my global sales.
I'm company registered in Luxembourg and Delaware. Beijing told me to f-off as they have a cheaper clone - fools! I have the heritage and branding.

 

[antcollinet]

This is incorrect. Ethernet uses error detection via CRC-32. That means it detects the bit error and tosses the frame - it has no retransmission capabilities. For retransmission, ethernet relies on a higher layer protocol (L4) - TCP to request the retransmission. Relating this back to audio. Qobuz uses TCP and buffering to hide retransmission for smooth playback. UDP is another upper layer protocol but it does not have a mechanism for retransmission. Airplay uses this protocol, you will hear click/pops/dropouts when bit errors happen.

PS. I was the Lead Systems Engineer or Architect for each of the 5 largest Canadian banks and the 3 largest Canadian Insurance companies over the course of my career with 2 very large networking companies (I am now retired).

I will bow to your greater precision in terms. I apologise for stating ethernet instead of TCP.


I will however give you this. Error count on my mac networking over the last 64 million packets and 23 days of uptime.
***my terminal prompt*** % netstat -I en0 -b
Name Mtu Network Address Ipkts Ierrs Ibytes Opkts Oerrs Obytes Coll
en0 1500 <Link#15> 5e:2e:68:d7:86:3d 48261743 0 31980750124 15984311 0 5977551339 0

So I am essentially getting zero errors between my Mac and the switch it connects to, over about 15m of the cheapest UTP I could buy. Which does go some way to illustrating the complete redundancy of fancy ethernet cables.


EDIT : correction - that is my wifi connection, not my Ethernet - see my next post for ethernet stats.

 

[MatrixS2000]

So I am essentially getting zero errors between my Mac and the switch it connects to, over about 15m of the cheapest UTP I could buy. Which does go some way to illustrating the complete reduncancy of fancy ethernet cables.

And that is exactly how it should be. I have seen switches that ran for years with zero bit errors - no ferrite cores needed...

 

[Redacted]

16 million packets is the blink of an eye.

Packet retransmission rarely happens on a wired LAN at home, and if it does it's fixed automatically for anything you care about in the blink of an eye.