00:00:00
hard dis shocking looks like a great way
00:00:01
to pick up some cheap storage options
00:00:03
either for local mass storage or to put
00:00:05
in an N so in this video I'm going to
00:00:08
look at what the good options are for
00:00:09
this but I'm also going to go a bit
00:00:10
deeper and look at external storage
00:00:12
options to avoid and perform some
00:00:14
performance testing on the discs both in
00:00:16
their enclosure and out to see what you
00:00:18
can expect when you remove that juicy
00:00:20
disc inside also I'm going to look at
00:00:23
how to open the enclosure without
00:00:25
damaging it as the process Min guid show
00:00:27
damages the enclosure in the process I
00:00:29
do want to C however that I can't
00:00:31
guarantee you won't damage it and that
00:00:33
this can impact your ability to return
00:00:34
the drive if it goes bad so you do need
00:00:36
to go with your eyes open on that but if
00:00:38
done right it's certainly possible to
00:00:40
open it without damage and I'll show you
00:00:41
how to do this in the one we're looking
00:00:43
at today which is going to be a Western
00:00:45
Digital elements enclosure so let's get
00:00:47
started and first of all which unit to
00:00:49
choose if it's hard to ensure you don't
00:00:50
get an SMR drive when buying a be drive
00:00:52
then it's even harder when you're buying
00:00:54
an external disc as the manufacturer
00:00:56
won't usually say what model disc is
00:00:58
inside and they just quote a model
00:00:59
number for the full unit So to avoid
00:01:01
this risk I'd start by not buying a
00:01:03
Seagate disc of below 10 TBT as those
00:01:06
Barracuda drives that often come in
00:01:07
these are SMR up to about 8 terab in
00:01:10
size so if you buy a seate enclosure
00:01:12
from say 1 to 8 tab it's likely going to
00:01:14
come with an SMR Barracuda compute disc
00:01:16
inside all the small caddies with 2 and
00:01:18
1/2 in drives will also typically be SMR
00:01:21
and for these reasons covered in other
00:01:22
videos avoid these for Western Digital
00:01:25
their desktop drives can also be SMR up
00:01:27
to 6 terb but as of today all of that 8
00:01:30
terb and above drives are CMR and there
00:01:32
are some notable exceptions for host
00:01:34
manag SMR drives in the highend ultra
00:01:36
stars but you're not going to get these
00:01:37
in an external caddy so you don't need
00:01:39
to worry about this so to be safe I'm
00:01:40
not going to buy an external disc load
00:01:42
on 8 terab and the element 8 terby drive
00:01:45
often comes in at a good price point and
00:01:47
it's regularly on offer so this is what
00:01:49
I'm going to look at today at times
00:01:51
discs and external caddies were
00:01:52
definitely a great way to save money but
00:01:54
the first tip today is not to jump to
00:01:56
this conclusion because it just isn't
00:01:58
always true these drives often come with
00:02:00
a fairly short warranty and when
00:02:01
shocking the unit you do run the risk of
00:02:03
causing damage to the casing that could
00:02:05
potentially result in a warranty claim
00:02:07
being denied all together so the first
00:02:09
job is to check that the price saving
00:02:11
really does make it worthwhile and
00:02:12
actually camam camo.com tells us that
00:02:15
the wdbl 8 tab Drive can come in at a
00:02:17
close price point to the WD elements
00:02:20
unit at least in the US market and it
00:02:22
turns out the discs are actually quite
00:02:23
similar spec with both being 5640 RPM
00:02:26
CMR drives you do get a cry of course in
00:02:28
the price when you buy the element unit
00:02:30
but if your plan is to extract the disc
00:02:32
from the enclosure then this may not
00:02:34
really add any benefit anyway so price
00:02:36
vary a lot by region however so in the
00:02:38
UK for example the elements package is
00:02:40
often considerably cheaper than the 8 TB
00:02:42
WD drive so it's going to be worth
00:02:44
checking and comparing so now we've
00:02:46
established the prices to compare here
00:02:47
is an elements 8 terabyte I picked up
00:02:49
for 130 UK and this is actually a good
00:02:52
discount on the WD blue 8 terabyte price
00:02:54
of £192 typically in the UK so first of
00:02:57
all we're going to Smart test the device
00:02:59
and this is this is what's going to tell
00:03:00
us what is inside um and then we're
00:03:02
going to perform a full surface scan of
00:03:04
the drive before I Shu it this way if
00:03:05
the disc does come up bad I can return
00:03:07
it without any risk from actually
00:03:09
opening the caddy and unlike the Seagate
00:03:11
which often have barracudas WD actually
00:03:13
often package white label drives here
00:03:15
and these are drivers with model numbers
00:03:17
that you're not going to find on a data
00:03:18
sheet that look to be built for
00:03:20
specifically this purpose and if we look
00:03:22
at the hardware spec a first look they
00:03:24
seem very much like the 8 TB WD blue or
00:03:26
the 8 TB WD red plus drives which also
00:03:29
have an PM of 5640 and if we see how the
00:03:31
data sheet performance Compares for
00:03:33
these then we see both of these claiming
00:03:35
to have 250 MB per second for the 256 MB
00:03:38
cach versions and 185 mbes per second
00:03:41
for the 128 MB cache versions so we're
00:03:44
going to performance test our disc also
00:03:45
to see how it measured up to these but
00:03:48
right now this looks like the drive is
00:03:49
probably based on the same unit as these
00:03:51
two other drives um and I'm going to
00:03:53
provide some affiliate links below both
00:03:54
to the elements drive but also what the
00:03:56
other options I'm looking at here so you
00:03:57
can go and compare okay so looking at
00:03:59
the smart data we see that is labeled as
00:04:01
part of the Western Digital Ultra star h
00:04:04
10/12 family so once inside we can
00:04:06
actually check this but this isn't
00:04:08
likely correct and it may be based on
00:04:10
incorrect data in the smart database the
00:04:12
H here specifically signifies it's a
00:04:14
helium fill Drive which for a 5640 RPM 8
00:04:17
TB Drive is pretty unlikely I'd say so
00:04:20
we'll know for sure once it's opened
00:04:21
it's also a 512e drive which means it's
00:04:24
got 4K physical sector size and 512
00:04:26
logical sector size and it reports that
00:04:28
it doesn't support trim which makes it
00:04:30
unlikely it's SMR though some mly
00:04:33
Toshiba drives do appear to be SMR
00:04:35
without trim support So this isn't a
00:04:36
slam dunk next I would recommend a full
00:04:39
surface scam using something like Omi
00:04:41
petition assistant or just using check
00:04:43
dis SLR like this I'm going to run a
00:04:45
full performance test however in its
00:04:47
caddy to see how this performs out of
00:04:48
the box and this test will involve
00:04:50
filling the entire disc with 10 GB files
00:04:53
and then reading them off again and then
00:04:54
filling the disc again with 10 GB
00:04:56
directories containing about 5 1/ 12,000
00:04:58
mixed file SI es and then again reading
00:05:01
them off and this is going to perform
00:05:02
the same job which is to fully test the
00:05:04
entire disc surface and find any areas
00:05:06
that are suffering errors and what we
00:05:08
see actually when we do these tests is
00:05:10
that for sequential rights the dis in
00:05:12
its cad actually performs really well
00:05:13
it's getting about 226 megabytes per
00:05:16
second for large right performance of
00:05:17
the outer edge dropping to about 100
00:05:19
megabytes per second for the inner
00:05:20
tracks and this exceeds the stated specs
00:05:23
with those blue and red plus drives I
00:05:24
mentioned read performance is also
00:05:26
between 230 mb per second and 100 MB
00:05:29
also so very similar read and WR
00:05:31
performance for large files performing a
00:05:33
sequential mix file test we get about
00:05:36
185 mb/ second right performance at the
00:05:38
outer edge down to around 85 megab per
00:05:40
second at the inner tracks and between
00:05:43
200 mbes per second and 94 MBT per
00:05:45
second for read performance on these
00:05:47
files and if we perform a 20%
00:05:50
nonsequential rewrite test this is where
00:05:52
we rewrite 20% of all of these mix files
00:05:54
we get about 100 megabits per second and
00:05:56
67 mb per second rewrite performance and
00:05:59
this is consistent with what we'd
00:06:00
normally see from a CMR disc and what
00:06:03
this really shows is that this doesn't
00:06:04
behave like an SMR drive would where far
00:06:07
rewrite performance can actually be
00:06:09
really bad okay so having verified the
00:06:11
disc is good and the performance test
00:06:13
out in the caddy let's shut the drive
00:06:15
and see what we have inside physically
00:06:18
so just to orientate you on how this
00:06:19
fits together this unit comes in
00:06:21
effectively two parts there's an
00:06:23
external case that's made of shiny
00:06:24
plastic and you can see it here covered
00:06:26
in the protective film and then
00:06:28
internally there's a portion made kind
00:06:29
of this vented mat plastic and this
00:06:31
slides out along two rails on either
00:06:34
sides shown here and at the end with the
00:06:36
power and the USB connector here there's
00:06:38
four plastic tabs which hold the two
00:06:40
parts together so we're going to need to
00:06:41
release these four tabs and then slide
00:06:43
the inner portion out from the external
00:06:45
shell and I'm going to use a couple of
00:06:47
simple and cheap tools for this and I'll
00:06:49
also link down below some examples of
00:06:51
something like this and just a note here
00:06:53
right I've seen a few guys where people
00:06:54
leave along this long edge here where
00:06:56
the retaining Clips are but if you do
00:06:58
this there is a fair chance of breaking
00:07:00
these clips with this approach so the
00:07:01
approach I take is to gently lever one
00:07:03
part of the case away from the other so
00:07:05
you can see here there's a little bit of
00:07:06
pressure around the edge and then I run
00:07:08
a shim down the edge to release the clip
00:07:10
from the lips to hold it in place on the
00:07:11
internal part you don't need to leave
00:07:13
the Clips open as this is how they're
00:07:15
going to get broken but just use a tool
00:07:16
like this or even something like a
00:07:18
guitar pick and just run it along this
00:07:20
Edge once you've done one side and the
00:07:21
clips are released you can repeat on the
00:07:23
other side and you should be able to see
00:07:24
the clips are free and the unit is ready
00:07:26
to separate and once you've done this it
00:07:28
should slide out easily now again I
00:07:30
can't guarantee that you won't break a
00:07:31
clip this way but I haven't yet and I
00:07:33
find this approach um really successful
00:07:35
and straightforward now if we look
00:07:37
inside the um the outer chassis here we
00:07:39
can see how the clips work and you can
00:07:40
probably see why this approach works
00:07:42
okay there are these four lips that sit
00:07:44
inside the clips on the outer case and
00:07:46
they just need to be slid out while
00:07:47
applying a slight separation to ease
00:07:49
them past that kind of half a millimeter
00:07:51
or so lip without actually bending the
00:07:53
clips themselves and once the two parts
00:07:55
are separated you can get to the disc
00:07:57
remove this clear plastic that connects
00:07:59
the activity LED on the board to the
00:08:01
front of the case and you can pop the
00:08:02
disc out from the side opposite the
00:08:04
power connector easily CLI this plastic
00:08:06
part which is easy to lose and the four
00:08:08
rubber shock absorbers safe in the box
00:08:10
so you can put the disc back in if you
00:08:12
do need to return it under warranty
00:08:14
later maybe take a picture of how these
00:08:15
fit or you can always refer back to this
00:08:17
video if you do need to reassemble the
00:08:19
unit and then there's two screws that
00:08:20
fit the USB board to the disc so remove
00:08:22
these and again keep these parts safe
00:08:24
there's also this retaining clip here
00:08:26
that fits around the USB port once again
00:08:28
we can see that the disc is l AED as the
00:08:30
smart data showed as a
00:08:32
wazz drive also we can see actually that
00:08:35
it is an airfi Drive the giveaway being
00:08:37
that the screws holding the top place on
00:08:39
the drive and the Brea holes helium fi
00:08:41
discs are sealed more robustly and they
00:08:43
have no bavers or visible screws in the
00:08:45
top plate now let's test the drive again
00:08:47
performance when it's actually SRA
00:08:49
attached now in Windows you can enable
00:08:51
or disable the drive's right cach via
00:08:53
the device manager and this instructs
00:08:55
the drive whether it should use its
00:08:56
onboard cache or not and this option
00:08:58
allows you to select between optimized
00:09:00
right performance or effectively right
00:09:02
safety because data in cach can
00:09:04
potentially be lost during a power
00:09:05
outage even if the drive has reported to
00:09:08
the OS that the right was successful
00:09:09
with the element USB enclosure the OS
00:09:11
believes the cach is disabled but the
00:09:13
USB adapter appears to enable it behind
00:09:15
the scenes as the dis reports enabled
00:09:17
here in HW info are seen here so when
00:09:20
attaching the drive directly via SATA
00:09:22
we're going to test the drive with both
00:09:24
right cach enabled and disabled to see
00:09:26
what this does for performance and the
00:09:28
results here are actually pretty
00:09:29
interesting first of all without right
00:09:31
cach enabled we see a significant
00:09:33
reduction in ride performance with the
00:09:34
drive producing just 75 mbes per second
00:09:37
Max at the outer edge dropping to around
00:09:39
48 megabytes per second at the inner
00:09:41
tracks read performance isn't impacted
00:09:43
as you'd expect as it performs similarly
00:09:46
as it does in the enclosure at between
00:09:48
226 mb per second down to about 100 mbes
00:09:51
per second across the disc surface mix
00:09:53
rights vary from 95 mb per second down
00:09:55
to 58 mb/ second and that's actually
00:09:58
around 20% better than large far
00:10:00
performance again rep performance on
00:10:02
this test was similar to that of the
00:10:03
enclosure with around a peak of 200 mbes
00:10:06
per second and 92 megab per second at
00:10:08
the inner Edge so this isn't a surprise
00:10:10
but it does show that the elements USB
00:10:12
interface is doing something with the
00:10:14
drive right cache despite it being
00:10:16
disabled in the OS if we enable the
00:10:18
right caching on the S attached disc we
00:10:20
get better performance from the disc but
00:10:22
what we do find is that the disc doesn't
00:10:24
perform as well as it does when it's in
00:10:26
its caddy so the firmware on the drive
00:10:28
appears to be optimized for use with
00:10:29
this USB board and in this case for
00:10:32
large farri tests the drive initially
00:10:34
gives around 250 mbes right performance
00:10:37
but once the right cach is saturated
00:10:39
which happens very quickly it drops to
00:10:41
around 195 megabytes per second and
00:10:44
that's about 14% slower than the
00:10:46
elements enclosure that performed at
00:10:48
about 226 mbes per second and this
00:10:50
performance deficit is actually
00:10:52
consistently visible across the entire
00:10:54
disc with even the inner tracks
00:10:55
producing 92 mbes per second compared to
00:10:58
100 in the C and that's around an 8%
00:11:01
drop for mixed farri performance s
00:11:03
touched with the right cach enable
00:11:05
produces only between 120 megabytes and
00:11:07
65 mbes per second right speed and this
00:11:10
is significantly slower than in the
00:11:12
enclosure with around a 30% penalty and
00:11:15
this is lightly related again to how the
00:11:16
USB Bo manages the cache and how it
00:11:19
reports right completions to the OS so
00:11:21
for mixed rights it looks like the
00:11:23
performance is a fair bit worse once the
00:11:25
driver is removed from its closure and
00:11:26
this likly isn't an intentional attempt
00:11:29
to s shocking as some might say I think
00:11:31
it's likely that the drive is just
00:11:33
optimized to be used with the elements
00:11:34
interface board and the way that it
00:11:36
manages the drive caching and again
00:11:39
mixar re performance isn't impacted and
00:11:41
the drive performance almost identically
00:11:44
in the caddy when it's attached to SATA
00:11:46
with the right cache enabled or with it
00:11:48
disabled and comparing rewrites to the
00:11:50
files again the D performs best in the
00:11:52
caddy with around 20% penalty when the
00:11:54
SATA attached with right cach enabled
00:11:56
and about a 40% penalty when the cach is
00:11:58
disabled in the OS so let's also quickly
00:12:01
look at temperatures because one of the
00:12:03
main issues with discs in these caddies
00:12:04
is temperature and we see here three
00:12:06
measures with an ambient of 23° C we see
00:12:09
that the driver in the caddy is running
00:12:10
at about 46° C peaking at 48 and if you
00:12:14
run in an open case with ample air flow
00:12:16
all around the drive Peaks at 32° C
00:12:18
which is about 16° cooler when testing
00:12:21
inside a 5 Bay called dish chassis it
00:12:23
maxes out about 37° around halfway
00:12:26
between the two values so although the
00:12:28
elements enclosure is actually fairly
00:12:30
spacious and air can flow into and out
00:12:32
of it it will still run the drives a
00:12:34
fair bit warmer than in a typical Naz or
00:12:36
even in an internal drive bay given
00:12:38
enough space or active cooling okay so
00:12:41
we're done with the shocking and
00:12:42
comparing the performance results so
00:12:44
what are the takeways about if you
00:12:46
should shuck this drive but before I get
00:12:48
there do quickly please like the video
00:12:50
if it was useful or interesting and do
00:12:52
also consider a sub to the channel I
00:12:54
cover a lot of more in-depth topics on
00:12:56
storage and home compute and I focus on
00:12:58
bringing data that reveal was more about
00:12:59
what's really going on so I do really
00:13:01
appreciate the support in growing my
00:13:03
little corner of YouTube so thank you
00:13:05
ever so much for everything and the
00:13:06
support you give me so first of all
00:13:08
buying a drive in an enclosure can be
00:13:10
cheaper but it depends on the location
00:13:12
and the time and you cannot take it for
00:13:14
granted you also need to consider the
00:13:15
possible risk around warranty as it's
00:13:17
far clearer if you buy a be drive and
00:13:19
you don't extract it from a caddy if you
00:13:21
do this carefully you're probably going
00:13:22
to be fine to replace a disc in theal if
00:13:25
you do need to make a warranty claim
00:13:26
mine had no clear signs of removal with
00:13:28
no seals broken or damaged during the
00:13:30
process there are also no signs that the
00:13:32
drive records that this anywhere in the
00:13:34
smart data but I can't completely rule
00:13:37
out some trick WD play here so test it
00:13:39
well before removing the drive to
00:13:41
minimize this risk and although WD don't
00:13:43
make it easy to open there isn't the
00:13:45
sign to me that they're actively trying
00:13:46
to prevent it or catch people out here I
00:13:48
think the caddy just isn't designed to
00:13:50
be user accessible now secondly pick
00:13:52
your external drive carefully as
00:13:54
depending on the drive size you could
00:13:56
pick up an SMR drive and in this case I
00:13:58
avoided this by capacity where a wd
00:14:00
appear not to make SMR drives and this
00:14:02
is at 8 terab four or 6 terabytes may
00:14:05
not be the same WD don't seem to make
00:14:08
strong claims about performance on these
00:14:09
external drives so really they can swap
00:14:11
out drives without warning and they
00:14:13
don't have any specific obligation to
00:14:15
include a specific Drive of type or M
00:14:17
model thirdly and I think most
00:14:20
importantly and something I've not seen
00:14:21
called out before is that this model at
00:14:23
least seems to be optimized for use in
00:14:25
its caddy and its performance outside
00:14:28
its caddy is definely reduced I don't
00:14:30
see any any evidence of intentional
00:14:32
configuration penalty to penalize
00:14:34
shocking but the drive does seem to be
00:14:36
optimized to perform best in its caddy
00:14:38
so consider this if you're buying it for
00:14:40
another purpose and this doesn't mean
00:14:42
it's a bad drive and certainly it
00:14:44
performs fully for read performance and
00:14:46
actually produces really solid results
00:14:48
here for a very quiet and cool drive so
00:14:50
for read optimize use it's a great drive
00:14:53
for mix RS its performance is not great
00:14:55
if using directly on a SATA connection
00:14:57
but if it just comes down to a
00:15:00
significant discount and it serves your
00:15:02
use case then this could still be a good
00:15:04
option so as always thank you again for
00:15:06
taking the time and watching my video to
00:15:08
the end and I look forward to seeing you
00:15:11
in the next
