1
00:00:00,160 --> 00:00:06,640
so the general wisdom these days seems to be that if you want to save a buck

2
00:00:04,080 --> 00:00:10,240
and build a gaming rig AMD is the way to go but

3
00:00:07,759 --> 00:00:14,000
what does that mean exactly like AMD's for poor people who can't afford

4
00:00:12,400 --> 00:00:18,400
Intel well at a very high price point you

5
00:00:16,560 --> 00:00:23,519
could make that argument i mean remember we're talking about gaming here where

6
00:00:20,160 --> 00:00:24,800
Intel's top spec chips do outperform

7
00:00:23,519 --> 00:00:29,519
AMD's but most people aren't buying top of the

8
00:00:27,599 --> 00:00:33,680
line computers and that doesn't mean that they're poor

9
00:00:30,960 --> 00:00:39,680
it just means they want a good value so we designed today's challenge to find

10
00:00:36,239 --> 00:00:43,360
out if AMD's performance deficit can be

11
00:00:39,680 --> 00:00:46,879
overcome by its advantages which include

12
00:00:43,360 --> 00:00:48,960
less expensive cpus and motherboards and

13
00:00:46,879 --> 00:00:53,280
included overclocking support all the way up and down the ryzen lineup and i

14
00:00:51,600 --> 00:00:59,039
think you guys are going to find our results really surprising no matter what you

15
00:00:57,039 --> 00:01:02,079
were expecting i certainly did

16
00:01:09,680 --> 00:01:16,960
11 hardware sponsors are giving away a slew of awesome prizes at random to

17
00:01:14,240 --> 00:01:21,680
people voting in the builds.gg 10k challenge so head over to builds.gg

18
00:01:19,119 --> 00:01:25,439
today for your chance to win so we're gonna start things off right

19
00:01:23,520 --> 00:01:32,000
with a head-to-head off basically here's how it's going to go

20
00:01:28,240 --> 00:01:34,320
down using pc part picker Anthony and i

21
00:01:32,000 --> 00:01:41,280
are each going to put together the best build we can one for each side with a

22
00:01:37,280 --> 00:01:44,799
budget of 750 us dollars before taxes

23
00:01:41,280 --> 00:01:46,479
and shipping and a big list of

24
00:01:44,799 --> 00:01:51,040
requirements we both want to be able to overclock if

25
00:01:49,040 --> 00:01:54,799
not now then later on we're both going to need at least one

26
00:01:52,799 --> 00:02:00,240
terabyte of storage and eight gigs of RAM and we both want at least a 500 watt

27
00:01:58,079 --> 00:02:02,799
power supply with 80 plus braun certification

28
00:02:01,840 --> 00:02:06,799
so here we go then

29
00:02:05,200 --> 00:02:10,800
first order of business is going to be the CPU both of us know that we need

30
00:02:08,879 --> 00:02:14,640
something reasonably powerful but because of the overclocking stipulation

31
00:02:12,720 --> 00:02:19,360
Anthony's going to get a big discount by going with AMD because all desktop ryzen

32
00:02:17,440 --> 00:02:23,840
processors are unlocked thankfully the core i3 8350k isn't too

33
00:02:22,640 --> 00:02:28,640
expensive also because AMD's mid-tier chipset

34
00:02:26,160 --> 00:02:31,680
enables overclocking i can save money by going with a b series motherboard not

35
00:02:30,400 --> 00:02:35,840
the cheapest one mind you because overclocking is a consideration i'll

36
00:02:33,440 --> 00:02:40,000
want a board with decent vrms of course my k-series chip also doesn't

37
00:02:38,319 --> 00:02:43,519
come with a cooler which means i need to drop yet more of my precious budget on

38
00:02:42,239 --> 00:02:46,720
one of those wow this is starting to add up really

39
00:02:45,840 --> 00:02:52,480
quickly knowing Anthony's going for higher speed memory i can claw back a bit of my

40
00:02:50,640 --> 00:02:57,360
budget by going with a less expensive kit unfortunately that does mean that the

41
00:02:55,519 --> 00:03:01,920
per thread performance gap is going to close between us which is something i

42
00:02:59,360 --> 00:03:06,480
can't really afford pardon the pun for storage this hitachi one terabyte drive

43
00:03:04,400 --> 00:03:11,760
is a no-brainer there are cheaper one terabyte drives but none are 7200 RPM

44
00:03:09,840 --> 00:03:16,080
i'm also going to pick up this diy pc case and a rosewill arc 650 watt power

45
00:03:13,840 --> 00:03:20,400
supply it's not the best quality unit but for 40 dollars it's 80 plus bronze

46
00:03:18,080 --> 00:03:25,599
and has a three year warranty on to the graphics card i really don't have much

47
00:03:23,120 --> 00:03:29,840
money left to spend here and this is probably where i'm gonna get killed

48
00:03:27,680 --> 00:03:34,400
i can only hope that Anthony isn't very good at putting together value builds

49
00:03:32,640 --> 00:03:38,319
okay you just about ready there Linus it's been a while yeah i think i'm done

50
00:03:36,239 --> 00:03:41,920
you good yep and with that

51
00:03:39,760 --> 00:03:45,680
our builds are complete so pencils down

52
00:03:43,519 --> 00:03:49,120
time to compare ooh

53
00:03:47,360 --> 00:03:53,120
yeah i think i win well it's okay it's not about whether

54
00:03:50,720 --> 00:03:57,200
you win or lose it's is it more about how you play the game oh okay well i

55
00:03:55,280 --> 00:04:01,519
think i played better gaming um

56
00:03:59,360 --> 00:04:05,040
so we've got a lot in common we're both only spending 20 25 bucks on our cases

57
00:04:04,159 --> 00:04:10,000
yeah um roseville power supply yeah you got

58
00:04:07,599 --> 00:04:12,799
the same hitachi hard drive as well the graphics card's gonna kill me

59
00:04:11,519 --> 00:04:17,840
honestly i could almost afford an rtx except i

60
00:04:16,400 --> 00:04:21,440
couldn't right and i've only got a quad core processor

61
00:04:19,919 --> 00:04:25,199
because your CPU is only five bucks more even though it's only a quad core but

62
00:04:23,199 --> 00:04:30,000
like it's a faster quad core than it is than the ryzen cores so it's the 45

63
00:04:28,720 --> 00:04:34,000
on the board and the cooler that are costing me that

64
00:04:32,000 --> 00:04:37,520
GPU upgrade yeah and your RAM is cheaper too yeah i probably should have gone

65
00:04:36,000 --> 00:04:44,240
dual channel it didn't make that much of a difference so here's the thing

66
00:04:40,479 --> 00:04:45,600
requiring overclocking capability

67
00:04:44,240 --> 00:04:51,040
isn't really fair in the sense that Intel can deliver

68
00:04:48,800 --> 00:04:54,720
much better performance per dollar outside of their k series overclockable

69
00:04:53,840 --> 00:05:00,880
chips even if those limitations are entirely

70
00:04:57,919 --> 00:05:05,680
their own doing so let's actually work together then and

71
00:05:03,120 --> 00:05:11,199
modify the blue team build to try and get a little bit closer to that bang for

72
00:05:08,639 --> 00:05:16,560
the buck shall we sure basically CPU's got to go yeah and

73
00:05:14,160 --> 00:05:20,880
i think you can get like an 8 400 out of that

74
00:05:17,440 --> 00:05:23,600
8400 is such an underrated CPU 64

75
00:05:20,880 --> 00:05:27,440
reviews is all it's got the value is entirely there yeah it's outstanding so

76
00:05:25,680 --> 00:05:31,600
that means i don't need a cooler anymore you can kill that i don't need a z

77
00:05:29,440 --> 00:05:37,440
series board anymore yep you can go with a i would say b360 so now you should be

78
00:05:34,000 --> 00:05:39,520
able to afford the 2060. okay so

79
00:05:37,440 --> 00:05:44,639
with the Intel build now running a more reasonable b series chipset because we

80
00:05:42,160 --> 00:05:52,160
can't overclock anyway we're able to get both a more powerful CPU and GPU so

81
00:05:48,400 --> 00:05:54,800
that's six cores now and an rtx 2060. so

82
00:05:52,160 --> 00:06:00,479
with this modification then our specs actually even out significantly

83
00:05:57,440 --> 00:06:02,639
but it's worth noting that any edge the

84
00:06:00,479 --> 00:06:07,199
Intel system gets in its stock performance in games

85
00:06:04,479 --> 00:06:12,560
may be offset at least partially by AMD's unlocked core multiplier and

86
00:06:10,319 --> 00:06:17,120
higher thread count depending on the workload

87
00:06:14,479 --> 00:06:20,319
so now what well why don't i throw a couple of benches together and see if i

88
00:06:19,199 --> 00:06:24,880
can't figure out what the performance difference is between these oh so we're not just going to do this theoretical

89
00:06:22,880 --> 00:06:30,000
we're going to build these rigs and see what happens yeah i like it

90
00:06:28,080 --> 00:06:34,000
after an afternoon of prep our performance results came out to be

91
00:06:31,600 --> 00:06:39,039
pretty interesting as you can see our first builds are really different but

92
00:06:36,800 --> 00:06:44,960
while our AMD machine is way out ahead thanks to its powerful rtx 2060 graphics

93
00:06:41,680 --> 00:06:48,400
card Intel does pull up close in some

94
00:06:44,960 --> 00:06:50,479
less GPU bound cases and in cs go

95
00:06:48,400 --> 00:06:54,400
even manages to pull ahead though it should be noted that this game is

96
00:06:52,160 --> 00:06:57,280
notorious for performing much better on Intel

97
00:06:55,360 --> 00:07:02,319
when we overclocked our first builds all those extra threads on AMD's processor

98
00:06:59,440 --> 00:07:06,560
added up to a lot of oomph and since our graphics card was a little bottlenecked

99
00:07:04,160 --> 00:07:11,039
at stock this translated to a nice increase in gaming performance so AMD

100
00:07:09,120 --> 00:07:16,800
beat out Intel by about six percent overall with that said our overclocked

101
00:07:14,000 --> 00:07:21,840
Intel machine on average beat our stock AMD system's rtx numbers not bad for a

102
00:07:19,840 --> 00:07:26,720
gtx 1660. moving along we've got our adjusted

103
00:07:23,759 --> 00:07:32,800
builds so remember our AMD machine saved a few bucks on the motherboard and our

104
00:07:28,720 --> 00:07:36,560
Intel one jumps up to a core i5 8400 and

105
00:07:32,800 --> 00:07:37,840
rtx 2060 but loses its overclocking

106
00:07:36,560 --> 00:07:43,360
capability so here Intel is top dog for performance

107
00:07:40,639 --> 00:07:49,039
with neither machine overclocked but then gets edged out by our overclocked

108
00:07:46,080 --> 00:07:54,240
AMD machine this is proving to be a much closer race than we could have possibly

109
00:07:51,120 --> 00:07:56,720
hoped and it gets even more interesting

110
00:07:54,240 --> 00:08:01,840
when we look at the value proposition for our original configurations despite

111
00:07:59,520 --> 00:08:07,360
very decent performance levels and a lower price tag the Intel build ends up

112
00:08:04,560 --> 00:08:13,280
delivering poorer performance per dollar than our AMD one both at stock speeds

113
00:08:10,479 --> 00:08:17,599
and when both of them are overclocked now our Intel build ended up a little

114
00:08:15,280 --> 00:08:20,800
bit cheaper though because we had some budget left over

115
00:08:19,120 --> 00:08:26,879
just not enough for anything that would impact performance like an rtx upgrade

116
00:08:24,000 --> 00:08:32,479
so if you were building this box at home and didn't have a hard and fast 750

117
00:08:30,240 --> 00:08:35,039
limit the story might be a little bit different

118
00:08:33,519 --> 00:08:39,200
but we are operating within the constraints that we set and AMD came out

119
00:08:37,200 --> 00:08:44,399
ahead at this price point as for the adjusted builds this puts

120
00:08:41,680 --> 00:08:48,480
Intel's value at the closest to parity out of all of our comparisons

121
00:08:46,640 --> 00:08:53,360
but then when we compare it against our AMD builds overclocked numbers it again

122
00:08:51,680 --> 00:08:58,399
gets edged out and i guess it's that that word that

123
00:08:55,839 --> 00:09:02,560
keeps coming up edged that is kind of the most interesting

124
00:09:00,480 --> 00:09:08,399
takeaway here so as it turns out go figure neither AMD

125
00:09:06,560 --> 00:09:15,680
nor Intel's product management departments are staffed by sea sponges

126
00:09:11,279 --> 00:09:17,920
and every rig from both teams regardless

127
00:09:15,680 --> 00:09:22,080
of overclocking ended up being within about

128
00:09:19,040 --> 00:09:25,040
five percent in terms of the performance

129
00:09:22,080 --> 00:09:29,760
per dollar like it's almost like AMD's pricing structure is designed to

130
00:09:26,880 --> 00:09:34,080
maximize their profits while slightly undercutting Intel but not so much that

131
00:09:32,160 --> 00:09:38,880
they have to react and drop their pricing and like Intel's is designed to

132
00:09:36,959 --> 00:09:43,680
demand a small premium because of their more recognizable brand

133
00:09:41,040 --> 00:09:46,800
truly fascinating isn't it i actually i honestly did not expect it to be this

134
00:09:45,360 --> 00:09:51,600
close no like when we set out for this what was

135
00:09:49,200 --> 00:09:55,920
the what was your expectation i was expecting AMD to just cream Intel at the

136
00:09:53,680 --> 00:10:00,080
low end and then when we you know adjusted them i would expect Intel to

137
00:09:58,080 --> 00:10:04,399
cream AMD because it's roughly the same price and

138
00:10:01,360 --> 00:10:06,320
and you had six cores of more gaming

139
00:10:04,399 --> 00:10:09,360
optimized single threaded performance like it yeah

140
00:10:07,920 --> 00:10:13,519
so i don't know we

141
00:10:11,680 --> 00:10:17,839
we ended up with a result that wasn't what we were going for but we certainly

142
00:10:15,600 --> 00:10:22,320
had a lot of fun with it and let us know in the comments below if you'd like to

143
00:10:19,600 --> 00:10:26,720
see any other of these kinds of value analysis pieces in the future maybe at

144
00:10:24,320 --> 00:10:31,600
different price points or looking at different components of the system

145
00:10:28,640 --> 00:10:35,680
rather than just the CPU speaking of in the future

146
00:10:33,519 --> 00:10:39,120
this message from our sponsor used to be in the future but now

147
00:10:37,839 --> 00:10:43,600
it's now get your fix on with ifixit's new line

148
00:10:41,360 --> 00:10:47,360
of marlin screwdrivers each fixed blade screwdriver features an ergonomic

149
00:10:45,360 --> 00:10:52,640
knurled rubberized handle with a swivel top plus a black oxide coated tip so it

150
00:10:49,920 --> 00:10:56,160
hangs onto tiny screws with a steel grip the set case lid is held in place

151
00:10:54,320 --> 00:11:02,000
magnetically and can be used as a sorting tray which is super handy and

152
00:10:59,279 --> 00:11:07,200
it's manufactured in germany and backed by ifixit's lifetime warranty so even in

153
00:11:05,519 --> 00:11:11,519
the event that something goes wrong you can contact the folks over there and

154
00:11:09,200 --> 00:11:14,560
they will get you fixed up in a hurry it's like it's almost like they fix

155
00:11:13,040 --> 00:11:18,480
things like for a living we've been using ifix's tools for years and they

156
00:11:16,640 --> 00:11:21,839
make building pcs and repairing electronics around the office a breeze

157
00:11:21,040 --> 00:11:28,320
or at least as good as it can be sometimes it's very frustrating so get the marlin

158
00:11:25,440 --> 00:11:31,839
screwdriver set for 24.99 at ifixit.com forward slash Linus at the link in the

159
00:11:30,320 --> 00:11:36,000
video description so thanks for watching guys if you disliked this video you can hit that

160
00:11:34,640 --> 00:11:40,160
button but if you liked it i don't know who could dislike it because like there

161
00:11:38,240 --> 00:11:44,800
wasn't really uh yeah i don't think fanboys lost yeah like Intel can't be

162
00:11:42,399 --> 00:11:48,240
like oh AMD is so stupid and AMD can't be like oh Intel's little stupid but if

163
00:11:46,720 --> 00:11:52,800
you liked it hit like get subscribed or maybe consider checking out where to buy the stuff we featured in the video

164
00:11:50,800 --> 00:11:56,800
description also link down there is our merch store which has cool shirts like

165
00:11:54,640 --> 00:12:01,200
this one and our community forum which you should totally join

166
00:11:59,279 --> 00:12:06,000
no one can be upset okay someone someone can be upset always

167
00:12:03,440 --> 00:12:06,000
somebody
