WEBVTT

00:00:00.000 --> 00:00:06.879
the eagle-eyed among you have probably seen this LTT mousepad many times over

00:00:05.839 --> 00:00:13.040
the last oh i don't know six months

00:00:10.000 --> 00:00:15.120
six months that's cute past Linus

00:00:13.040 --> 00:00:19.600
we tried to launch this video eight months ago when we got our first

00:00:16.720 --> 00:00:24.080
shipment of desk pads but you you guys bought them all then we tried to launch

00:00:21.439 --> 00:00:28.400
the video again three months later and you bought them all again we have loved

00:00:26.800 --> 00:00:34.079
all of your feedback on the mousepads you guys can actually check out some reviews on the store page and finally we

00:00:31.760 --> 00:00:37.120
have our biggest restock yet but i can't promise you that they will last long

00:00:36.160 --> 00:00:42.559
anyway back to you past Linus what you haven't

00:00:39.600 --> 00:00:47.760
seen is the absolutely painful amount of r d that went into our attempt at making

00:00:44.879 --> 00:00:51.600
the best mouse pad we possibly could we didn't actually intend to make a video

00:00:49.760 --> 00:00:56.000
about this experience but we ended up learning so much that i felt that we had

00:00:54.480 --> 00:01:00.640
to share it with you here's how it went down

00:00:58.480 --> 00:01:04.239
this video is brought to you by ubico the ub key from ubico is a cost

00:01:02.559 --> 00:01:08.159
effective and easy to use two-factor authentication solution learn more about

00:01:06.080 --> 00:01:11.159
how to get 10 off yours at the link down below

00:01:18.320 --> 00:01:25.360
feast your eyes on LTT's first peripheral well sort of it doesn't look

00:01:23.520 --> 00:01:30.799
like it should have taken 15 months but the truth is we had no idea how much was

00:01:28.479 --> 00:01:34.560
involved in doing something like this right i thought it was going to be as

00:01:32.640 --> 00:01:38.880
simple as ask one of our lovely designers lloyd in this case to create

00:01:36.960 --> 00:01:44.079
some options for artwork get nick to contact some overseas vendors to

00:01:40.960 --> 00:01:45.600
requisition samples evaluate the samples

00:01:44.079 --> 00:01:50.399
and you know the communication of the suppliers and bippity boppity we print

00:01:48.159 --> 00:01:55.200
them well for starters about half of the requests for samples we sent out never

00:01:52.320 --> 00:02:00.000
even got a response their loss i guess and out of the many that we received a

00:01:57.680 --> 00:02:04.479
couple of weeks later only four of them weren't obviously garbage even out of

00:02:02.719 --> 00:02:08.479
those four none of them were perfect which was a big wake-up call for us

00:02:06.560 --> 00:02:13.599
there's varying materials print qualities and of course costs and as

00:02:11.360 --> 00:02:18.560
part of this process we also bought a whack of competing mouse pads for

00:02:15.360 --> 00:02:21.120
comparison but how do you compare

00:02:18.560 --> 00:02:27.280
if you have no data well aesthetic appeal and overall feel are easy but

00:02:25.200 --> 00:02:32.239
those aren't definitive rankings we wanted to actually quantify the

00:02:29.520 --> 00:02:37.200
performance of each sample pad so we needed data we challenged the jank

00:02:34.879 --> 00:02:41.440
master supreme alex to rig something up in the workshop and

00:02:39.280 --> 00:02:44.560
well i'll let him speak for himself this is the basic plan for our testing

00:02:43.440 --> 00:02:48.160
here you know just toss down a mouse pad

00:02:46.560 --> 00:02:52.959
and then we're going to use the little cnc rotor here to like wrap it around in

00:02:51.280 --> 00:02:58.000
a bunch of different ways to be able to precisely track

00:02:55.680 --> 00:03:01.040
how much the mouse is moving and then you know switch out the mouse pad throw

00:02:59.440 --> 00:03:06.560
another one in and it'll do its circles or whatever again

00:03:03.599 --> 00:03:09.920
only problem is i some have to attach this mouse

00:03:08.000 --> 00:03:13.760
to the router and i'm not entirely sure how to do it one

00:03:11.920 --> 00:03:16.959
of the problems is that you know if this goes up and down

00:03:15.280 --> 00:03:21.920
i don't know exactly how much force there is

00:03:18.560 --> 00:03:24.879
but i think it's uh

00:03:21.920 --> 00:03:30.480
like 960 newton meters which is it's way too much for a mouse

00:03:28.400 --> 00:03:34.000
to handle for sure you know before we do this i'm going to put on some safety

00:03:31.760 --> 00:03:37.360
glasses just in case something to be clear

00:03:35.280 --> 00:03:42.000
the scroll wheel on this mouse actually is broken so if we blow it up it's not

00:03:40.400 --> 00:03:48.799
the end of the world but you know we still don't want to

00:03:45.200 --> 00:03:48.799
that said let's give her

00:03:58.159 --> 00:04:04.959
seems to be working because we know that our cnc router's movement is

00:04:02.319 --> 00:04:09.360
sufficiently accurate and repeatable we tested it and we mostly trust the

00:04:07.439 --> 00:04:13.519
engineers at logitech who designed their hero sensor we can assume that on a

00:04:11.680 --> 00:04:18.560
mouse pad with perfect tracking characteristics the input path should

00:04:16.160 --> 00:04:22.960
match the output plot of our mouse's movement within experimental margin of

00:04:20.320 --> 00:04:28.080
error so the idea here is we tell the router to draw a perfect circle over and

00:04:25.840 --> 00:04:31.759
over again and we expect to see basically one circle in the software at

00:04:30.479 --> 00:04:37.120
the end we expect unfortunately the data was

00:04:35.199 --> 00:04:41.120
all over the place which is why we were moments from placing our bulk order only

00:04:39.440 --> 00:04:46.320
to end up delaying this product for another six months at the time we

00:04:44.080 --> 00:04:51.360
weren't sure what was to blame we had some theories that maybe our pressure on

00:04:48.320 --> 00:04:54.000
the mouse was too high or too low or our

00:04:51.360 --> 00:04:58.160
dpi settings were bad and that is something we ultimately had to fine-tune

00:04:56.080 --> 00:05:02.800
but what ultimately ended up solving the problem was a stiffer mount for the

00:05:00.400 --> 00:05:08.639
mouse how did we know that our testing rig was bad ah i'm glad you asked

00:05:06.080 --> 00:05:12.960
truthfully until fnatic contacted us claiming to have made the world's most

00:05:10.639 --> 00:05:16.800
accurate mouse pad i hadn't even given it a lot of thought i mean

00:05:14.800 --> 00:05:22.479
yeah isn't it the mouse that determines how well it tracks well sort of

00:05:19.680 --> 00:05:26.880
here's the thing an optical mouse works essentially by taking pictures

00:05:24.800 --> 00:05:30.240
many times per second of the surface that it's moving over

00:05:28.320 --> 00:05:35.280
some fancy math happens and it determines how far it's moved and in

00:05:32.400 --> 00:05:38.880
which direction by comparing the images which is great as long as the mouse

00:05:37.280 --> 00:05:42.960
exists in a world where the surface is perfect the pictures are perfect and all

00:05:40.960 --> 00:05:46.400
bodies are completely rigid but they're not

00:05:44.320 --> 00:05:50.960
so credit to fanatic they really did make a great mouse pad and studying it

00:05:48.880 --> 00:05:54.479
has taught us a lot about how to make ours better

00:05:52.720 --> 00:05:58.960
one thing we figured out even before using theirs as a point of comparison

00:05:56.400 --> 00:06:02.800
was that using a stretchy material could affect the tracking of the mouse

00:06:00.960 --> 00:06:08.479
think about it take this pad for instance it stretches in one direction

00:06:05.759 --> 00:06:13.759
but not in the other direction so when you move this way it kind of

00:06:11.520 --> 00:06:17.199
forms a bow wave ripple if you press hard enough into the pad

00:06:15.520 --> 00:06:21.600
not that you would do that but it's interesting nonetheless

00:06:19.280 --> 00:06:26.080
and while we're at it what if the bottom surface of the pad doesn't grip the desk

00:06:23.840 --> 00:06:31.440
properly i mean that's a ridiculous example we're looking at here but on a

00:06:28.080 --> 00:06:34.319
smaller scale that is a real factor

00:06:31.440 --> 00:06:39.199
armed then with our stiffer mount and a mouse pad that the manufacturer claims

00:06:36.880 --> 00:06:43.600
can produce this perfectly consistent performance we set to work trying to

00:06:41.520 --> 00:06:48.560
replicate their results from this point on creator warehouse's very own engineer

00:06:46.000 --> 00:06:52.720
kyle took the reins and in short order built us a new test rig using our prusa

00:06:50.880 --> 00:06:56.479
print farm for all the custom bits aluminum extrusion for the structure and

00:06:54.639 --> 00:07:00.880
a bunch of 3d printer hardware for the electronics it's pretty bare bones as

00:06:58.400 --> 00:07:05.039
far as a cnc machine goes it uses a ramps board for the electronics

00:07:02.639 --> 00:07:08.639
essentially an arduino mega some cheap nema 17 size stepper motors and gt2

00:07:07.680 --> 00:07:12.720
belts we couldn't know until it was built if

00:07:10.880 --> 00:07:17.280
it would be rigid enough for our testing but the name of the game here is if

00:07:14.639 --> 00:07:20.160
you're going to fail do it fast and do it cheap

00:07:18.240 --> 00:07:23.840
and conveniently once it was equipped with a new mouse holder that kyle

00:07:21.840 --> 00:07:29.120
reverse engineered from fnatic's picture of their test rig it did end up doing

00:07:26.400 --> 00:07:34.319
the trick it's not perfect perfect but given that logitech only guarantees 98

00:07:32.000 --> 00:07:39.840
accuracy for the hero sensor this test data from fanatics pad looks about right

00:07:38.160 --> 00:07:44.160
for our test the carriage moves the mouse around in a perfect circle 26

00:07:42.479 --> 00:07:48.560
times while we run a super simple pointer plotter on a laptop connected to

00:07:46.000 --> 00:07:53.520
the mouse which sorry in spite of this being a g305 is still a gpro wireless

00:07:51.680 --> 00:07:57.280
running the test this way tells us two things one

00:07:55.280 --> 00:08:01.840
if fnatic's test can be repeated by a third party which for all intents and

00:07:59.280 --> 00:08:04.879
purposes it can it's a great mouse pad and two

00:08:03.039 --> 00:08:08.879
whether or not we can match or beat their performance

00:08:06.479 --> 00:08:13.520
let's talk about that now in our attempt we sourced two new versions of the LTT

00:08:11.199 --> 00:08:17.759
pad from the same supplier now at the time we were mostly focused on the

00:08:15.440 --> 00:08:22.479
stretchiness of the material so we asked for one with two-way rather than one-way

00:08:20.400 --> 00:08:26.160
stretch so at least we'd have consistency to it

00:08:24.080 --> 00:08:31.120
and we also asked for one with no stretch and as it turns out

00:08:28.560 --> 00:08:35.440
funny story the first sample we had with the one-way stretch was actually far and

00:08:33.519 --> 00:08:39.680
away the most consistent when it came to tracking even better than the no stretch

00:08:38.000 --> 00:08:42.240
sample that our supplier couldn't even do full color printing on so it wouldn't

00:08:41.279 --> 00:08:46.240
even like look all cool we're still not quite on

00:08:44.560 --> 00:08:52.080
par with the data and fanatics marketing materials and i do still think that we

00:08:48.640 --> 00:08:54.640
might see an LTT pad pro in another year

00:08:52.080 --> 00:08:58.320
that manages to be absolutely perfect but

00:08:55.440 --> 00:09:02.800
we are close enough that anyone but the most professional of gamers is not going

00:09:00.720 --> 00:09:07.440
to notice the difference so all that was left at this point was to adapt our

00:09:04.640 --> 00:09:11.519
design for the 15 different sizes

00:09:09.200 --> 00:09:15.279
finalize the thickness we settled on three and a half millimeter by the way

00:09:12.959 --> 00:09:20.080
which i feel is a great balance between soft comfort and performance

00:09:17.600 --> 00:09:23.279
and await our final samples so

00:09:20.959 --> 00:09:26.959
this is it the LTT mouse pad

00:09:25.760 --> 00:09:33.680
take it in yeah it doesn't smell that good i cannot say how proud i am of our team

00:09:31.200 --> 00:09:37.839
for getting this done even for something as simple as a mousepad so many hands

00:09:36.000 --> 00:09:43.440
are involved in getting you guys the best possible swag and i hope that you

00:09:40.880 --> 00:09:46.880
enjoyed this long awaited piece of kit and you can see the whole lineup in our

00:09:45.440 --> 00:09:51.440
shameless promotion ShortCircuit unboxing video over on our other channel

00:09:49.839 --> 00:09:55.120
big thanks to ubico for sponsoring today's video ubico wants to make the

00:09:53.279 --> 00:09:59.279
internet safer for everyone so they created the yubi key it's a physical

00:09:57.519 --> 00:10:02.720
two-factor authentication key that helps you secure your online accounts it

00:10:01.040 --> 00:10:07.760
doesn't require batteries or a mobile connection and it's both crush and ip68

00:10:05.519 --> 00:10:11.760
water resistant it works out of the box with hundreds of popular business and

00:10:09.200 --> 00:10:16.079
personal apps like gmail twitter cryptocurrency exchanges and password

00:10:13.600 --> 00:10:19.760
managers and ub keys are super portable and work across tons of devices and

00:10:17.760 --> 00:10:24.000
operating systems like iOS Android Windows and mac it's time to take

00:10:22.000 --> 00:10:28.880
security seriously then so go beyond basic authentication and get 10 off your

00:10:26.399 --> 00:10:33.440
ubiq 5 series today by going to the link in the video description and using code

00:10:30.800 --> 00:10:37.760
Linus 10. all right video's over smash that link in the description and get one

00:10:35.120 --> 00:10:42.000
for yourself before they all sell out oh something to note by the way there

00:10:39.440 --> 00:10:46.160
are no limit switches to protect the cnc machine from traveling too far and once

00:10:44.320 --> 00:10:49.440
it powers on it has no idea where in space it is because it has no way to

00:10:47.760 --> 00:10:53.040
reference a home position so if you decide to use the files linked in the

00:10:51.440 --> 00:10:58.079
video description to make one for yourself just make sure you're a diligent

00:10:56.240 --> 00:11:04.760
operator because otherwise it could ruin itself pretty quickly but that won't

00:11:00.160 --> 00:11:04.760
happen to you you're a diligent operator