that was a pcie-e card, so not much use for retro builds ^^

can't hurt to try something like that though as old systems probably have a lot more noise on the bus.

imi wrote on 2020-09-04, 01:10:

that was a pcie-e card, so not much use for retro builds ^^

can't hurt to try something like that though as old systems probably have a lot more noise on the bus.

I honestly cannot imagine how such a card could be of any use in terms of noise reduction . I am also pretty convinced that whoever "designed" the thing is probably even more clueless as to how it is supposed to work than whoever actually ends up buying it. If this "filter" sat between the PSU and the motherboard, it might have a small claim to credibility . As it stands, I would not touch it with a ten foot pole, lest it actually damage the PC it might get installed in .

EDIT : Please feel free to prove me wrong by explaining how it could possibly do anything useful, if only in theory .

mbliss11 wrote on 2020-09-04, 00:43:

Ultimately they determined that there wasn't any distinguishable difference in their test bench which made sense but it got me thinking about older builds. Looking at a few socket 7 boards I had lying around I was curious if this thing could actually do anything for retro builds...

I'm going to make this VERY simple.

*AHEM*

NO.

It really depends on how poor the power filtering the motherboard has AND how poor the power supply power is.

For older builds with ISA slots, I can see a filter card making a difference.

With a thin client such as the HP T5720 which needs the caps added to the PCI riser for some cards to even work, then definitely a filtering card should help.

Newer PCI and PCI-e motherboards... not a lick of difference should be seen unless you have a really crappy motherboard and/or power supply in the first place.

cyclone3d wrote on 2020-09-04, 03:53:

It really depends on how poor the power filtering the motherboard has AND how poor the power supply power is. […]
Show full quote

It really depends on how poor the power filtering the motherboard has AND how poor the power supply power is.

For older builds with ISA slots, I can see a filter card making a difference.

With a thin client such as the HP T5720 which needs the caps added to the PCI riser for some cards to even work, then definitely a filtering card should help.

Newer PCI and PCI-e motherboards... not a lick of difference should be seen unless you have a really crappy motherboard and/or power supply in the first place.

If that were the case (speaking of a crappy PCI or PCIE motherboard and/or PSU), wouldn't the noise likely cause operating issues (system instability) well before it became an audible issue ? Also wouldn't there be potential issues if an arbitrarily large filtering capacitance is applied to PCI Express slot power ?

I was always under the impression that sound card noise (excluding ground loops) was mainly due to RF noise pickup due to improperly shielded components, not noisy supply voltage (hence my impression of the uselessness of this product). I would like to be educated on this matter .

Sounds like another snake oil for gullible audiophiles. Although they are gullible by default =P

OMFG this whole damn thread is going to drive me bananas...

cyclone3d wrote:

It really depends on how poor the power filtering the motherboard has AND how poor the power supply power is.

No, it does not. if your motherboard is THAT bad, it wont post.

For older builds with ISA slots, I can see a filter card making a difference.

no. it will not.

With a thin client such as the HP T5720 which needs the caps added to the PCI riser for some cards to even work, then definitely a filtering card should help.

Those caps are NOT there for filtering.

not a lick of difference should be seen unless you have a really crappy motherboard and/or power supply in the first place.

Fixed. And if this is the case, this card isn't going to help.

Without going too deep into the weeds and teaching you all the basics of electrical engineering. Take this one statement to heart:
filter caps only work when they are directly attached to the device and/or transmission line. The audio signal is not going over the ISA/PCI/MCA bus, this card cannot affect it.

yes, the board is "right next to" your sound card, and you might say that is close. But from an EE point of view, it might as well be in the next city over. Look at the small decoupling caps that are usually next to most logic chips. Ever notice each one gets its own cap and it's almost literally touching the IC? THAT'S INTENTIONAL.

This card is like having your neighbor install earthquake protection in his/her house and expecting to also work for your house. if you want a cleaner sounding SB16 card, then you are going to have to replace the caps on the card itself. you will still hear hiss and noise, because the card wasn't designed for shit (in terms of audio quality)

The Serpent Rider wrote on 2020-09-04, 04:19:

Sounds like another snake oil for gullible audiophiles. Although they are gullible by default =P

Additionally, won't any half-decently designed audio solution have power input filtering, among other things, already present ?

darry wrote:

If that were the case (speaking of a crappy PCI or PCIE motherboard and/or PSU), wouldn't the noise likely cause operating issues (system instability) well before it became an audible issue ? Also wouldn't there be potential issues if an arbitrarily large filtering capacitance is applied to PCI Express slot power ?

This is true.

I was always under the impression that sound card noise (excluding ground loops) was mainly due to RF noise pickup due to improperly shielded components, not noisy supply voltage (hence my impression of the uselessness of this product). I would like to be educated on this matter .

This is also true. CuriousMarc actually did a video about debugging a noise source in a digital clock: https://www.youtube.com/watch?v=ad8azNovsF8 While the thought was the power supply directly, it was actually electromagnetic interference.

The Serpent Rider wrote on 2020-09-04, 04:19:

Sounds like another snake oil for gullible audiophiles. Although they are gullible by default =P

The proper term is ignorant. To most people, electronics is no different than magic. But even a few basic college classes reveals the man behind the curtain.

Ok... "filtering" is the wrong term for what I was talking about then. What I am talking about is cleaning up the output for the power lines.

Say I've got one motherboard where the "bus noise" is very evident over the output from the soundcard. Take another motherboard with the same exact sound card and the output doesn't have all that "bus noise".

Wouldn't a card that has caps to clean up the power being fed to the cards help in that instance? Looking it up, it looks like that would only end up being half the equation. Looks like you would want to decouple whatever circuit that is causing the noise (power fluctuations) and that would have to be done on the other side of the circuit as well.

I understand bus noise. I've had several motherboards that I can hear the data my usb mouse makes. There isn't much you can do about it. It is a badly designed unit. If you were to re-cap a SB16 with "audiofool" grade caps, you'd still have noisy audio as the circuit was poorly designed. It just was not a concern. By comparison look at the Soundblaster ZXR.

Long story short, you can't just add caps to remove noise. https://www.youtube.com/watch?v=wopmEyZKnYo

Luckybob has covered most there is to know. Inductance of the trace and socket etc. makes the filter caps on the card wholly useless except at very lowest freqs which wouldn't be a problem anyway... There's a reason why all the capacitors are in line and as close as possible to whatever they are trying to care for, every cm from target makes them less uselful and after some point inductance has made them useless.

So in the end the thing simply is a decoration in the computer with all its colorful parts on it and it probably appeals to the modern gamers with all the fancy lights and whatnot 🤣

luckybob wrote on 2020-09-04, 03:05:

I'm going to make this VERY simple. […]
Show full quote

mbliss11 wrote on 2020-09-04, 00:43:

Ultimately they determined that there wasn't any distinguishable difference in their test bench which made sense but it got me thinking about older builds. Looking at a few socket 7 boards I had lying around I was curious if this thing could actually do anything for retro builds...

I'm going to make this VERY simple.

*AHEM*

NO.

The first thing I thought of after reading your reply 😀

Thanks for responding! I just saw the video and it reminded me of all of the noise I will sometimes get on some of my cards. I figured it was bogus. I also chuckled at their RGB implementation haha

The Linus' video was a total joke anyways.

If you want to measure something like that, then first thing you do is take out your oscilloscope!