WEBVTT

00:00:00.240 --> 00:00:06.160
in this box is a pci express power

00:00:04.319 --> 00:00:12.960
filter card it claims to smooth out the dirty power

00:00:10.639 --> 00:00:18.160
from bad components and improve the audio signals coming out of your pc

00:00:16.560 --> 00:00:23.279
and what we're going to do is we're going to feed it a worst case scenario

00:00:20.800 --> 00:00:28.320
with a super cheap 1000 watt power supply along with a very coil whiny

00:00:26.000 --> 00:00:32.880
graphics card to see if it actually makes any appreciable difference to

00:00:31.039 --> 00:00:36.079
sound quality my expectations

00:00:34.960 --> 00:00:39.520
well i wasn't born yesterday so

00:00:38.399 --> 00:00:44.719
pretty low speaking of born yesterday though

00:00:41.760 --> 00:00:50.800
i wrote a children's board book and we just got our first set of mostly final

00:00:47.920 --> 00:00:55.120
samples so if you or anyone you know works for a bookseller and wants to have

00:00:52.960 --> 00:00:58.800
the abcs of gaming on your shelves this holiday season for those little gamers

00:00:56.800 --> 00:01:03.440
who are still too young for a next-gen console for christmas contact us at

00:01:00.879 --> 00:01:08.600
books lttstore.com and we'll get the info you need to place your initial

00:01:05.600 --> 00:01:08.600
order

00:01:15.040 --> 00:01:21.600
before we can get into this card and what we expect we have to talk about

00:01:19.520 --> 00:01:26.880
power and how it's converted from the mains ac power out of your wall to the

00:01:24.240 --> 00:01:31.600
dc power that your hardware works with most pcs today use what's called a

00:01:28.960 --> 00:01:36.079
switch mode or switching power supply this is mostly because compared to a

00:01:33.600 --> 00:01:40.720
linear power supply they are smaller lighter and more efficient a switching

00:01:38.560 --> 00:01:45.119
power supply can be as much as 80 smaller and can have efficiencies as

00:01:42.640 --> 00:01:48.320
high as 96 compared to the peak of around 60

00:01:46.399 --> 00:01:52.000
for a linear power supply but that doesn't mean that switching

00:01:50.640 --> 00:01:57.520
power supplies like this one don't have any downsides

00:01:55.119 --> 00:02:01.040
and one of them is that they create more electrical noise

00:01:59.200 --> 00:02:04.240
if you've ever turned on a speaker and heard a s

00:02:02.640 --> 00:02:08.720
or when no audio is being played that's

00:02:06.560 --> 00:02:13.440
exactly what i'm talking about quote-unquote dirty power can rear its

00:02:11.280 --> 00:02:18.080
ugly head from the external supply things like voltage spikes or low or

00:02:15.920 --> 00:02:23.680
unbalanced ac power like we have here in our building or even internally within

00:02:20.800 --> 00:02:28.319
the system after the conversion to dc from components like graphics cards or

00:02:26.720 --> 00:02:33.360
other chips now efficient autos of high-end audio

00:02:30.560 --> 00:02:36.560
have tried a huge variety of ways to mitigate this unwanted electrical noise

00:02:35.680 --> 00:02:41.040
with varying levels of success one of the

00:02:38.720 --> 00:02:46.000
most common methods of noise suppression is power conditioning where essentially

00:02:43.840 --> 00:02:51.040
you use a device that takes the less than desirable power that it's given

00:02:48.080 --> 00:02:55.680
then acts as a buffer outputting a nice clean ac power signal to sensitive audio

00:02:53.680 --> 00:03:01.440
hardware farther down the line if you watched our overview of the abyss

00:02:58.000 --> 00:03:04.400
diana headphones you may remember this

00:03:01.440 --> 00:03:08.800
monster of an amp tower this is the sort of thing that has had

00:03:06.879 --> 00:03:15.360
plenty of forethought given to power delivery for example where is it

00:03:11.920 --> 00:03:15.360
where is it there it is ah

00:03:17.280 --> 00:03:24.959
actually it fell pretty flat it's probably fine for example

00:03:22.640 --> 00:03:29.440
for example look at this plug here with its bi-directional noise containment

00:03:27.120 --> 00:03:33.760
technology thing is most people don't have

00:03:31.360 --> 00:03:37.280
thousand dollars to buy an amp stack to power their speakers what if they just

00:03:35.680 --> 00:03:42.720
wanna use a pc well that's where this

00:03:40.400 --> 00:03:50.879
at least in theory comes in meet the l fidelity axf 107 magician a

00:03:48.080 --> 00:03:55.519
pci express power purifying filter fresh out of senzen china it sports seven

00:03:53.440 --> 00:04:00.080
different capacitors seven different capacitors it supports seven different

00:03:57.840 --> 00:04:03.519
capacitors and wouldn't you know it they even threw in a little bit of RGB for

00:04:02.159 --> 00:04:07.920
good measure although i'm only going to give them half points

00:04:05.680 --> 00:04:11.680
for that because to change the colors you actually have to flip these little

00:04:09.599 --> 00:04:16.959
dip switches here with a pin now it claims to and i quote here

00:04:14.239 --> 00:04:21.680
prevent video card coil wine and improve overclocking as well as reduce noise

00:04:19.440 --> 00:04:26.000
floor for an ultimate pc audio build spec that is that is quite the claim

00:04:25.199 --> 00:04:32.240
wow you know like if one 40 dollar card can

00:04:29.199 --> 00:04:34.000
create ultimate pc audio spec i will be

00:04:32.240 --> 00:04:38.479
shocked just as shocked as it was to hear that

00:04:35.919 --> 00:04:42.720
we've got a new shirt check it out it's CPU pins but if you look really close

00:04:41.040 --> 00:04:45.880
actually if you look from far away it's got an LTT logo in it

00:04:44.639 --> 00:04:50.560
up your geek street cred on ltteststore.com today now to create the

00:04:48.080 --> 00:04:55.199
perfect storm of bad audio we grabbed Anthony's GPU test bench and

00:04:53.280 --> 00:05:00.120
replaced the perfectly good silverstone power supply with the sketchiest one we

00:04:57.520 --> 00:05:04.800
had in the entire warehouse this 25

00:05:01.840 --> 00:05:08.000
1000 watt china special you're actually going to see this again by the way

00:05:06.320 --> 00:05:11.759
assuming it survives the day in our upcoming lemon pc build hit that

00:05:09.919 --> 00:05:16.880
subscribe button so you don't miss out while we were at it we tossed in this

00:05:14.080 --> 00:05:21.120
sapphire nitro plus Radeon rx 590 special edition for maximum coil

00:05:19.199 --> 00:05:26.720
whininess even when this puppy's not doing anything

00:05:23.759 --> 00:05:30.160
hear that crackling that ticking can you hear it a little bit yeah that's

00:05:28.880 --> 00:05:35.280
it well it's not even doing anything david oh yeah all right let's start with a baseline

00:05:33.280 --> 00:05:39.360
without our power filtering card in here pre-loaded on this machine i've got a

00:05:37.199 --> 00:05:46.160
handful of lossless audio tracks ready to go and grado's 420 edition headphones

00:05:42.000 --> 00:05:46.160
as our reference cans to start with

00:05:48.080 --> 00:05:53.840
i really need to move to flack for all my stuff now what i'd really love to have is

00:05:52.560 --> 00:05:59.360
something with sort of quiet parts of the track so let me uh

00:05:57.759 --> 00:06:05.110
poke around a little bit here because that's where you're gonna be able to hear that that hiss or that higher noise

00:06:03.440 --> 00:06:08.439
floor

00:06:09.440 --> 00:06:15.520
while i'm listening i'm going to go ahead and run specuperf which is like a

00:06:13.600 --> 00:06:22.000
professional applications benchmark that loads up both our CPU and GPU to see if

00:06:18.639 --> 00:06:22.000
i hear any of that uh

00:06:22.639 --> 00:06:30.479
delightful buzzing through my headphones here can you hear that wine now that

00:06:26.960 --> 00:06:30.479
we've actually loaded it up yeah

00:06:32.319 --> 00:06:38.400
i just i can't make out any problem here to be

00:06:36.720 --> 00:06:42.240
clear these are not the most difficult headphones in the world to drive so it's

00:06:40.000 --> 00:06:46.720
not a huge surprise to me that the onboard audio on this motherboard is

00:06:44.479 --> 00:06:50.080
able to drive them just fine and i'm not saying that if you didn't have very

00:06:48.560 --> 00:06:53.840
high-end audiophile headphones you might not need a better audio solution it's

00:06:52.240 --> 00:06:58.080
just that that's not what we're testing for today today we just want to know if

00:06:56.400 --> 00:07:02.400
this makes a difference in a setup that someone might normally actually use so

00:07:01.120 --> 00:07:06.800
i don't know maybe i'm going to be blown away let's let's throw the power filter

00:07:04.560 --> 00:07:09.840
card in and see how we do here something to note guys that we didn't actually

00:07:08.160 --> 00:07:14.960
mention before is this can be installed either in a PCIe 1x slot or a 16x slot

00:07:12.960 --> 00:07:18.560
so you can insert your placebo however you like

00:07:16.960 --> 00:07:22.880
you just flip the bracket around and install it the other way where's my RGB

00:07:20.800 --> 00:07:26.479
at oh it's set to white yeah that's more work than i think it's

00:07:24.560 --> 00:07:31.160
worth to change i actually want to start with the simplest thing

00:07:28.800 --> 00:07:33.599
just the test tone

00:07:33.599 --> 00:07:38.160
and it's fine fire up spectrum perf i guess

00:07:36.880 --> 00:07:45.000
yeah i mean i guess that's one thing that we can determine like if it helps with the coil

00:07:41.440 --> 00:07:45.000
whine out of the GPU

00:07:45.759 --> 00:07:51.199
it's horrible uh let's listen to those same tracks

00:07:48.800 --> 00:07:51.199
again now

00:07:53.199 --> 00:07:55.919
still a good song

00:07:57.039 --> 00:08:04.000
yep wait rick astley made it onto this greatest

00:08:01.440 --> 00:08:10.160
hit to the 80s collection but not with never going to give you up

00:08:08.080 --> 00:08:14.560
what even is that i don't know so colin prepared this

00:08:12.160 --> 00:08:18.800
video and he wrote it as though there was no difference

00:08:16.240 --> 00:08:24.080
and i wanted to try real hard to find a difference that he didn't detect but

00:08:21.440 --> 00:08:28.080
unfortunately i couldn't

00:08:25.199 --> 00:08:30.319
not even in the coil whine of the system so

00:08:30.879 --> 00:08:38.959
the question then is what made l fidelity think that this was

00:08:36.159 --> 00:08:46.000
going to work is there any scientific basis whatsoever for why putting a pci

00:08:42.800 --> 00:08:48.800
express card into an entirely different

00:08:46.000 --> 00:08:53.839
slot would have any impact on the dc power quality of a graphics card that's

00:08:51.680 --> 00:08:56.800
over here in another slot or a sound card that's built right into the

00:08:55.360 --> 00:09:02.959
motherboard you can actually see it down at the bottom there well actually people have been putting

00:09:00.240 --> 00:09:07.600
capacitors and inductors and power lines to smooth out noisy signals for a long

00:09:05.040 --> 00:09:11.680
time now lc filters are a perfect example of this principle that are

00:09:09.040 --> 00:09:15.839
commonly used in drones to improve video quality which can be degraded by the

00:09:13.839 --> 00:09:19.839
electrical noise caused by the motors and the escs particularly when they're

00:09:17.839 --> 00:09:27.120
rapidly changing speed so the principle is actually not stupid

00:09:24.480 --> 00:09:31.120
the problem is that you're limited to just the things that this card can touch

00:09:29.279 --> 00:09:35.519
meaning that there's no impact on the pci express connectors that are going

00:09:33.279 --> 00:09:41.680
straight into your graphics card here and it also ignores that your sound card

00:09:38.959 --> 00:09:45.839
even an onboard one like this already has this kind of filtering built

00:09:44.480 --> 00:09:50.800
into it the thing is the youngsters out there might not

00:09:48.560 --> 00:09:56.399
remember this but there was a time when on-board audio was so bad that a sound

00:09:53.760 --> 00:10:00.720
card was a pretty integral part of any decent rig i can actually remember my

00:09:58.880 --> 00:10:05.600
first sound blaster like it was yesterday it was an auto g2 zs and it

00:10:03.360 --> 00:10:09.440
was it was night and day the difference was enough to put tears in your eyes

00:10:07.279 --> 00:10:13.200
like listening to the same songs it was huge

00:10:10.720 --> 00:10:16.480
so i'm not poo pooing the fundamental idea here

00:10:14.560 --> 00:10:21.839
it's just that we're looking at yesterday's solution tomorrow

00:10:19.760 --> 00:10:27.040
because for the most part those days are over and on board audio is good enough

00:10:24.320 --> 00:10:32.320
for the vast majority of people but wait Linus you might say what about

00:10:29.600 --> 00:10:34.959
coil wine if it doesn't oh oh that's funny well whatever we did this out of

00:10:33.839 --> 00:10:39.440
order we already checked that it doesn't help with that it's horrible so

00:10:37.519 --> 00:10:43.360
let's just conclude with this given that we used the noisiest parts

00:10:41.519 --> 00:10:47.760
that we could find here in the studio and didn't perceive any difference

00:10:45.440 --> 00:10:51.920
i wouldn't buy one of these instead i would suggest either investing in a

00:10:49.600 --> 00:10:55.839
decent power filter or power bar for your entire system to eliminate ground

00:10:54.000 --> 00:10:59.760
loops and other noise not to mention improve the longevity of your power

00:10:57.360 --> 00:11:03.760
supply or if that isn't in the cards maybe go pick up a digital analog

00:11:01.440 --> 00:11:08.720
converter that sits outside of the pc itself like one of these units from fio

00:11:06.320 --> 00:11:13.040
by moving the signal processing out of the pc's case to an external dac and amp

00:11:11.360 --> 00:11:18.000
you're isolating it from the internal noise of your system and yeah yes i know

00:11:15.839 --> 00:11:21.680
some of it can still come through usb but there are options out there if

00:11:19.680 --> 00:11:26.240
you're that paranoid to further reduce that potential crosstalk at the end of

00:11:23.600 --> 00:11:29.279
the day bottom line is a power filter isn't the first thing you should try

00:11:27.839 --> 00:11:33.519
when it comes to improving the quality of your sound and if you're running such

00:11:31.680 --> 00:11:36.880
a bad motherboard that you're getting hissing and crackling from your

00:11:34.959 --> 00:11:41.920
headphones or speakers i think your forty dollars would probably be better

00:11:38.560 --> 00:11:43.920
spent on a better motherboard that also

00:11:41.920 --> 00:11:47.279
improves the rest of your computing experience

00:11:44.959 --> 00:11:51.360
at the same time just like squarespace can improve your

00:11:49.279 --> 00:11:55.519
website building experience do you need to create a beautiful website without

00:11:53.200 --> 00:11:59.440
the hassle check out squarespace their all-in-one platform makes it easy to get

00:11:57.600 --> 00:12:02.800
up and running quickly they have award-winning templates that you can use

00:12:01.040 --> 00:12:06.079
as a starting point for a wide variety of projects and if you ever need

00:12:04.480 --> 00:12:10.480
additional help squarespace offers webinars a full series of help guides

00:12:08.399 --> 00:12:14.240
and you can even contact their 24 7 customer support via live chat and email

00:12:12.639 --> 00:12:17.839
if you already have a third party domain you don't have to give it up just

00:12:15.839 --> 00:12:21.680
transfer it over to squarespace plus you get e-commerce features to help you sell

00:12:19.519 --> 00:12:25.760
merch or services online and easily manage your inventory and orders so head

00:12:23.600 --> 00:12:30.959
over to squarespace.com forward slash LTT and get 10 off your first purchase

00:12:29.600 --> 00:12:35.200
by the way if you guys enjoyed this video and you want to learn a little bit

00:12:32.800 --> 00:12:39.680
more about this setup here we actually collabed on this video with dms he's a

00:12:37.519 --> 00:12:42.959
smaller youtuber but he's super into this stuff and very knowledgeable so you

00:12:41.200 --> 00:12:46.519
can check out the video we did with him and then maybe his channel