Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated:
0
00:00:00,780 --> 00:00:05,960
foreign
1
00:00:02,570 --> 00:00:08,400
[Music]
2
00:00:05,960 --> 00:00:10,200
books today we're continuing our CPU and
3
00:00:08,400 --> 00:00:12,179
GPU scaling Series this time with a
4
00:00:10,200 --> 00:00:16,340
comparison between Intel's new Core i5
5
00:00:12,179 --> 00:00:19,440
13400 And the tried and true AMD ryzen 7
6
00:00:16,340 --> 00:00:21,660
5700x now as it stands the ryzen 7
7
00:00:19,440 --> 00:00:25,980
processor is actually a good bit cheaper
8
00:00:21,660 --> 00:00:29,279
at around 190 us whereas the 13400 is
9
00:00:25,980 --> 00:00:31,500
currently fetching 240 U.S which is the
10
00:00:29,279 --> 00:00:34,020
MSRP for this part as the most
11
00:00:31,500 --> 00:00:36,300
affordable 13th gen Core i5 part the
12
00:00:34,020 --> 00:00:38,460
13400 has quite a lot of you very
13
00:00:36,300 --> 00:00:40,739
excited and while it looks quite good on
14
00:00:38,460 --> 00:00:43,800
paper it might also look very familiar
15
00:00:40,739 --> 00:00:46,620
to some of you with 6p cores four equals
16
00:00:43,800 --> 00:00:49,920
and a 20 megabyte L3 cache you might
17
00:00:46,620 --> 00:00:52,440
have mistaken as the core i5 12 600k
18
00:00:49,920 --> 00:00:56,180
because well that's basically what it is
19
00:00:52,440 --> 00:00:59,100
in fact when compared to the 12600k the
20
00:00:56,180 --> 00:01:01,500
13400p cores a clock just 300 megahertz
21
00:00:59,100 --> 00:01:03,899
lower at 4.6 gigahertz so a mere six
22
00:01:01,500 --> 00:01:06,060
percent reduction there while the e-cors
23
00:01:03,899 --> 00:01:08,580
have been closed Lockdown from 3.6
24
00:01:06,060 --> 00:01:10,380
gigahertz to 3.3 gigahertz and that's a
25
00:01:08,580 --> 00:01:12,600
slightly larger eight percent reduction
26
00:01:10,380 --> 00:01:14,640
other than that they are basically the
27
00:01:12,600 --> 00:01:17,280
same CPU with the only other notable
28
00:01:14,640 --> 00:01:21,780
change being to the integrated Graphics
29
00:01:17,280 --> 00:01:25,740
the 12 600k use the UHD 770 while the
30
00:01:21,780 --> 00:01:29,280
13400 uses the UHD 730 sounds trivial
31
00:01:25,740 --> 00:01:31,200
but the 770 does pack 33 more execution
32
00:01:29,280 --> 00:01:33,360
units so that it is quite a significant
33
00:01:31,200 --> 00:01:35,280
difference though at the end of the day
34
00:01:33,360 --> 00:01:37,380
you won't be gaming with either one of
35
00:01:35,280 --> 00:01:40,079
these Solutions so at least in that
36
00:01:37,380 --> 00:01:42,180
sense it is a trivial difference now
37
00:01:40,079 --> 00:01:45,240
when compared to the 12400 this newer
38
00:01:42,180 --> 00:01:47,579
13th gen version and I do say 13th gen
39
00:01:45,240 --> 00:01:50,460
with air quotes it clocks five percent
40
00:01:47,579 --> 00:01:53,159
higher for the P cores while the 12 400
41
00:01:50,460 --> 00:01:56,159
didn't offer any e-cors at all the 12
42
00:01:53,159 --> 00:01:59,159
400 was also limited to an 18 megabyte
43
00:01:56,159 --> 00:02:01,380
L3 cache and used the UHD 730 integrated
44
00:01:59,159 --> 00:02:04,920
Graphics so that means when compared to
45
00:02:01,380 --> 00:02:06,180
the 12 400 the 13 400 it is a big Gap
46
00:02:04,920 --> 00:02:08,280
grade for those of you looking at
47
00:02:06,180 --> 00:02:11,580
productivity performance but at today's
48
00:02:08,280 --> 00:02:13,620
prices it does also cost 30 percent more
49
00:02:11,580 --> 00:02:15,540
but as usual for this content we will be
50
00:02:13,620 --> 00:02:16,860
focusing on gaming performance so I
51
00:02:15,540 --> 00:02:20,160
thought it would be interesting to not
52
00:02:16,860 --> 00:02:22,560
only compare the 13 400 with the 5700x
53
00:02:20,160 --> 00:02:25,560
but also the previous generation value
54
00:02:22,560 --> 00:02:28,440
champ the core i5 12400 along with the
55
00:02:25,560 --> 00:02:29,700
ryzen 5 5600 so we have four CPUs in
56
00:02:28,440 --> 00:02:33,000
total all of which can still be
57
00:02:29,700 --> 00:02:34,260
purchased today for between 130 and 240
58
00:02:33,000 --> 00:02:36,239
U.S
59
00:02:34,260 --> 00:02:37,980
now given this is a GPU scaling
60
00:02:36,239 --> 00:02:39,900
Benchmark we are obviously focusing on
61
00:02:37,980 --> 00:02:41,879
gaming performance and in total we have
62
00:02:39,900 --> 00:02:44,840
a dozen games all of which have been
63
00:02:41,879 --> 00:02:49,500
tested at 1080p using the Radeon RX
64
00:02:44,840 --> 00:02:51,239
6650xt 6950 XT and RTX 4090 now the
65
00:02:49,500 --> 00:02:53,400
reason for testing at 1080p is quite
66
00:02:51,239 --> 00:02:55,260
simple we want to minimize the GPU
67
00:02:53,400 --> 00:02:57,180
bottleneck as we're looking at CPU
68
00:02:55,260 --> 00:02:59,580
performance but by including a number of
69
00:02:57,180 --> 00:03:01,800
different tier gpus we still get to view
70
00:02:59,580 --> 00:03:04,260
the effects of GPU bottlenecks on the
71
00:03:01,800 --> 00:03:06,360
CPU as for the test system specs we're
72
00:03:04,260 --> 00:03:08,160
using ddr4 memory exclusive here as this
73
00:03:06,360 --> 00:03:10,860
older but much cheaper memory standard
74
00:03:08,160 --> 00:03:13,080
makes sense when talking about sub 250
75
00:03:10,860 --> 00:03:14,940
us processors though there are certainly
76
00:03:13,080 --> 00:03:17,519
scenarios where you would opt for ddr5
77
00:03:14,940 --> 00:03:19,140
but for simplicity's sake we've gone for
78
00:03:17,519 --> 00:03:20,459
an Apples to Apples comparison using
79
00:03:19,140 --> 00:03:22,560
ddr4
80
00:03:20,459 --> 00:03:25,860
I've also decided to use G skills rip
81
00:03:22,560 --> 00:03:28,860
Jaws V series 32 gigabyte ddr4 3600cl 16
82
00:03:25,860 --> 00:03:30,659
memory because at 115 us it is a good
83
00:03:28,860 --> 00:03:33,000
quality kit that offers a nice balance
84
00:03:30,659 --> 00:03:34,739
of price to Performance the am4
85
00:03:33,000 --> 00:03:37,700
processors are able to run this kit at
86
00:03:34,739 --> 00:03:41,040
the 3600 spec using the standard CL 16
87
00:03:37,700 --> 00:03:44,580
1936 timings the locked Intel processors
88
00:03:41,040 --> 00:03:46,200
though can't run above 34.66 if you want
89
00:03:44,580 --> 00:03:48,180
to use the gear one mode and ideally you
90
00:03:46,200 --> 00:03:50,940
do want to use the gear one mode as the
91
00:03:48,180 --> 00:03:53,340
memory has to run well above 4000 before
92
00:03:50,940 --> 00:03:55,200
Gear 2 can match the performance of gear
93
00:03:53,340 --> 00:03:57,060
one and the reason for this limitation
94
00:03:55,200 --> 00:04:00,299
is due to the fact that the system agent
95
00:03:57,060 --> 00:04:01,980
or sa voltage is locked on non-k sku
96
00:04:00,299 --> 00:04:03,959
processors and this limits the memory
97
00:04:01,980 --> 00:04:06,000
controller's ability to support high
98
00:04:03,959 --> 00:04:08,099
frequency memory using the gear one mode
99
00:04:06,000 --> 00:04:10,200
and I should note that all overclocking
100
00:04:08,099 --> 00:04:12,180
is locked down on these 13th gen Parts
101
00:04:10,200 --> 00:04:14,459
there are no special motherboards that
102
00:04:12,180 --> 00:04:16,199
can alade overclock all of them are
103
00:04:14,459 --> 00:04:18,419
locked down there is no non-k
104
00:04:16,199 --> 00:04:20,220
overclocking for 13th gen parts
105
00:04:18,419 --> 00:04:22,979
so that being the case I've tested the
106
00:04:20,220 --> 00:04:26,040
12 400 and 13 400 using the rip Jaws
107
00:04:22,979 --> 00:04:27,780
memory adjusted to 3466 as this is the
108
00:04:26,040 --> 00:04:29,940
most optimal configuration for these
109
00:04:27,780 --> 00:04:31,380
processors short of going and manually
110
00:04:29,940 --> 00:04:34,259
tuning the subtimes which we're not
111
00:04:31,380 --> 00:04:36,120
doing for this content so let's get into
112
00:04:34,259 --> 00:04:37,979
the benchmarks
113
00:04:36,120 --> 00:04:39,780
starting with watchdog's Legion we find
114
00:04:37,979 --> 00:04:42,360
very competitive results across the
115
00:04:39,780 --> 00:04:44,100
board regardless of the GPU used it's
116
00:04:42,360 --> 00:04:47,400
really crazy to see that we're actually
117
00:04:44,100 --> 00:04:50,520
not that GPU limited when using the 6650
118
00:04:47,400 --> 00:04:52,919
XT sure the 13400 was 21 faster with the
119
00:04:50,520 --> 00:04:55,199
6950 XT but that's actually quite a
120
00:04:52,919 --> 00:04:57,419
small margin given the difference in GPU
121
00:04:55,199 --> 00:05:00,660
power there the GeForce driver I've had
122
00:04:57,419 --> 00:05:03,479
is also visible here with the RTX 4090
123
00:05:00,660 --> 00:05:05,940
coming in slightly slower than the 6950
124
00:05:03,479 --> 00:05:08,400
XT that means we saw the highest frame
125
00:05:05,940 --> 00:05:11,580
rates using the Radeon GPU but even so
126
00:05:08,400 --> 00:05:14,100
the 13 400 and 5700x delivered basically
127
00:05:11,580 --> 00:05:16,440
identical performance the Intel CPU is
128
00:05:14,100 --> 00:05:18,419
just one and a half percent faster even
129
00:05:16,440 --> 00:05:21,720
when compared to the much cheaper ryzen
130
00:05:18,419 --> 00:05:24,060
5 5600 the 13400 doesn't look that
131
00:05:21,720 --> 00:05:26,039
impressive winning that matchup by a
132
00:05:24,060 --> 00:05:28,020
mere five percent margin
133
00:05:26,039 --> 00:05:30,780
moving on to Warhammer 3 we see when
134
00:05:28,020 --> 00:05:34,680
using the 6650 XT that all CPUs are
135
00:05:30,780 --> 00:05:37,680
limited by the GPU at exactly 81 FPS and
136
00:05:34,680 --> 00:05:39,660
even with the much faster 6950 XT we're
137
00:05:37,680 --> 00:05:42,240
still quite heavily GPU limited this
138
00:05:39,660 --> 00:05:45,240
time to around 180 FPS in fact it's not
139
00:05:42,240 --> 00:05:46,800
until we employ the RTX 1490 that we
140
00:05:45,240 --> 00:05:50,460
find the performance limits of these
141
00:05:46,800 --> 00:05:52,320
CPUs the ryzen 5 5600 and core i5 12400
142
00:05:50,460 --> 00:05:55,380
for example both capped out at around
143
00:05:52,320 --> 00:05:58,979
200 FPS so very respectable performance
144
00:05:55,380 --> 00:06:02,759
there the 5700x was however 12 fast in
145
00:05:58,979 --> 00:06:06,060
the 5600 while the 13400 was 9 faster
146
00:06:02,759 --> 00:06:07,620
than the 5700x so ultimately Intel does
147
00:06:06,060 --> 00:06:09,900
win in this game though that doesn't
148
00:06:07,620 --> 00:06:11,880
mean the core i5 is necessarily the
149
00:06:09,900 --> 00:06:13,500
better value choice but of course we
150
00:06:11,880 --> 00:06:14,400
will look at that a bit later on in the
151
00:06:13,500 --> 00:06:18,240
video
152
00:06:14,400 --> 00:06:20,580
next up we have Hitman 3 and no the 6950
153
00:06:18,240 --> 00:06:22,800
XT and 40 90 results aren't mixed up
154
00:06:20,580 --> 00:06:24,720
here for whatever reason the 40 series
155
00:06:22,800 --> 00:06:26,280
currently suffers from really weak one
156
00:06:24,720 --> 00:06:28,139
percent lows in this title and
157
00:06:26,280 --> 00:06:30,600
presumably this is some kind of driver
158
00:06:28,139 --> 00:06:33,539
related issue in any case the average
159
00:06:30,600 --> 00:06:36,060
frame rate of the 6950x10 RTX 4090 are
160
00:06:33,539 --> 00:06:38,880
quite similar at 1080p due to the
161
00:06:36,060 --> 00:06:40,500
performance being CPU limited if we were
162
00:06:38,880 --> 00:06:43,380
to increase the resolution therefore
163
00:06:40,500 --> 00:06:47,220
reduce the CPU limitation we would start
164
00:06:43,380 --> 00:06:48,780
to see the 6950 XT Trail the RTX 4090 by
165
00:06:47,220 --> 00:06:51,300
a larger margin of course if we were to
166
00:06:48,780 --> 00:06:53,819
use much faster CPUs then it wouldn't
167
00:06:51,300 --> 00:06:56,220
require a 4K resolution but just want to
168
00:06:53,819 --> 00:06:59,160
make it clear that the 6950 XT is in no
169
00:06:56,220 --> 00:07:01,259
way a competitor to the RTX 4090 we are
170
00:06:59,160 --> 00:07:04,199
heavily limiting the RTX 4090 here at
171
00:07:01,259 --> 00:07:07,319
1080p with these lower end CPUs or at
172
00:07:04,199 --> 00:07:09,660
least lower end by today's standards now
173
00:07:07,319 --> 00:07:12,180
the more limited and heavily utilized 12
174
00:07:09,660 --> 00:07:15,300
400 and 5600 did perform quite a bit
175
00:07:12,180 --> 00:07:17,280
better with the Radeon GPU again due to
176
00:07:15,300 --> 00:07:19,620
nvidia's driver overhead which sends
177
00:07:17,280 --> 00:07:21,840
performance backwards in high CPU
178
00:07:19,620 --> 00:07:23,340
utilization scenarios for the most part
179
00:07:21,840 --> 00:07:24,660
though there was little to no
180
00:07:23,340 --> 00:07:26,099
performance difference between these
181
00:07:24,660 --> 00:07:28,620
four CPUs
182
00:07:26,099 --> 00:07:30,720
the a plague tail Requiem results are
183
00:07:28,620 --> 00:07:34,680
heavily CPU limited when using the RTX
184
00:07:30,720 --> 00:07:36,060
4090 and 6950 XT the 6650 XT on the
185
00:07:34,680 --> 00:07:39,419
other hand heavily GPU limited
186
00:07:36,060 --> 00:07:40,919
performance to 68 FPS as all four CPUs
187
00:07:39,419 --> 00:07:44,099
could easily meet that performance
188
00:07:40,919 --> 00:07:47,280
Target now with the 6950 XT we see that
189
00:07:44,099 --> 00:07:50,160
the 5700x and 3400 delivered the same
190
00:07:47,280 --> 00:07:53,039
level of performance while the 5600 and
191
00:07:50,160 --> 00:07:54,960
12 400 were also comparable however
192
00:07:53,039 --> 00:07:57,960
things look a bit different with the RTX
193
00:07:54,960 --> 00:08:00,000
4090 namely for the 5600 which appears
194
00:07:57,960 --> 00:08:02,819
overwhelmed by the additional overhead
195
00:08:00,000 --> 00:08:07,080
dropping well below the 5700x with just
196
00:08:02,819 --> 00:08:08,520
87 FPS on average the 5700x and 13400
197
00:08:07,080 --> 00:08:09,720
though did deliver comparable
198
00:08:08,520 --> 00:08:11,580
performance
199
00:08:09,720 --> 00:08:13,800
next up we have Call of Duty Modern
200
00:08:11,580 --> 00:08:15,419
Warfare 2 using the basic quality preset
201
00:08:13,800 --> 00:08:17,819
as this is a competitive multiplayer
202
00:08:15,419 --> 00:08:19,919
shooter so lower quality settings are
203
00:08:17,819 --> 00:08:21,720
typically used to not just boost frame
204
00:08:19,919 --> 00:08:24,840
rates but also make it easier to spot
205
00:08:21,720 --> 00:08:26,940
enemy players with the 6650xt we see
206
00:08:24,840 --> 00:08:28,440
that performance is largely GPU limited
207
00:08:26,940 --> 00:08:30,060
with just a seven percent performance
208
00:08:28,440 --> 00:08:33,360
difference between the fastest and
209
00:08:30,060 --> 00:08:34,740
slowest CPUs tested the 5700x for
210
00:08:33,360 --> 00:08:36,659
example was just three percent faster
211
00:08:34,740 --> 00:08:39,479
than 13 400 so in other words
212
00:08:36,659 --> 00:08:42,180
performance was basically identical
213
00:08:39,479 --> 00:08:44,880
moving to the 6950 XT largely removes
214
00:08:42,180 --> 00:08:49,200
the GPU bottleneck and now the 5700x is
215
00:08:44,880 --> 00:08:51,180
a massive 37 faster than the 13400 which
216
00:08:49,200 --> 00:08:53,580
is a surprisingly large margin very
217
00:08:51,180 --> 00:08:56,580
surprising in fact the 13 400 was just
218
00:08:53,580 --> 00:08:58,680
five percent fast and the 12 400 so a
219
00:08:56,580 --> 00:09:03,120
very Mild improvement there while the
220
00:08:58,680 --> 00:09:05,160
5700x was 20 faster than the 5600 we're
221
00:09:03,120 --> 00:09:07,560
also seeing similar margins with the RTX
222
00:09:05,160 --> 00:09:09,720
4090 though due to the overhead here the
223
00:09:07,560 --> 00:09:12,540
CPU limited results are slightly slower
224
00:09:09,720 --> 00:09:15,240
with the GeForce GPU
225
00:09:12,540 --> 00:09:17,399
Spider-Man remastered is a very CPU
226
00:09:15,240 --> 00:09:20,040
limited game especially with Ray tracing
227
00:09:17,399 --> 00:09:22,500
enabled that said we do see some
228
00:09:20,040 --> 00:09:25,980
performance improvement from the 6650 XT
229
00:09:22,500 --> 00:09:28,980
to the 6950 XT particularly when looking
230
00:09:25,980 --> 00:09:30,660
at the core i5 processors interestingly
231
00:09:28,980 --> 00:09:32,220
the ryzen processors did see a
232
00:09:30,660 --> 00:09:36,240
performance uplift when moving from the
233
00:09:32,220 --> 00:09:38,640
6950 XT to the RTX 4090 around a 9 gain
234
00:09:36,240 --> 00:09:39,839
while the 13400 saw a seven percent
235
00:09:38,640 --> 00:09:42,120
Improvement
236
00:09:39,839 --> 00:09:43,980
I suspect memory performance is the key
237
00:09:42,120 --> 00:09:45,660
limiting factor for the ryzen processors
238
00:09:43,980 --> 00:09:47,519
and namely its dram latency which will
239
00:09:45,660 --> 00:09:49,500
be responsible for why the core i5 has
240
00:09:47,519 --> 00:09:53,580
performed so much better here using the
241
00:09:49,500 --> 00:09:55,620
RTX 490 the 13400 was 13 faster than the
242
00:09:53,580 --> 00:09:57,060
5700x
243
00:09:55,620 --> 00:09:59,519
the shadow of the Tomb Raider results
244
00:09:57,060 --> 00:10:01,620
are quite interesting using the 6650 XT
245
00:09:59,519 --> 00:10:03,839
the results are highly predictable here
246
00:10:01,620 --> 00:10:06,300
we're entirely GPU limited so all CPUs
247
00:10:03,839 --> 00:10:08,700
were capped at 110 FPS however this
248
00:10:06,300 --> 00:10:11,640
limit was removed with the 6950 XT and
249
00:10:08,700 --> 00:10:14,880
RTX 490 now the results are CPU limited
250
00:10:11,640 --> 00:10:18,660
basically the 12 400 and 5600 were
251
00:10:14,880 --> 00:10:20,760
evenly matched with the 13 400 and 5700x
252
00:10:18,660 --> 00:10:22,080
also evenly matched delivering around
253
00:10:20,760 --> 00:10:23,820
eight percent more performance than
254
00:10:22,080 --> 00:10:26,459
those cheaper models
255
00:10:23,820 --> 00:10:28,140
moving along we have Horizon zero Dawn
256
00:10:26,459 --> 00:10:30,120
and this total always seems to heavily
257
00:10:28,140 --> 00:10:32,760
favor AMD processors and we're certainly
258
00:10:30,120 --> 00:10:35,880
seeing an example of that here the 5700x
259
00:10:32,760 --> 00:10:39,000
was 17 faster than the 13400 when using
260
00:10:35,880 --> 00:10:43,019
the RTX 4090 and 9 faster when using the
261
00:10:39,000 --> 00:10:46,079
6950 XT interestingly the 5700x was
262
00:10:43,019 --> 00:10:47,880
slower using the 6950 XT while the 13400
263
00:10:46,079 --> 00:10:49,920
delivered its best results with the old
264
00:10:47,880 --> 00:10:52,200
rdna2 flagship
265
00:10:49,920 --> 00:10:54,839
moving on to the cyberpunk 2077 results
266
00:10:52,200 --> 00:10:57,360
we find when a CPU limited the game
267
00:10:54,839 --> 00:10:59,760
plays best with the Radeon GPU evident
268
00:10:57,360 --> 00:11:02,820
by the fact that the 6950xt was up to 12
269
00:10:59,760 --> 00:11:06,300
faster than the RTX 4090 seen when using
270
00:11:02,820 --> 00:11:08,760
the ryzen 5 5600 if we look at the 6950
271
00:11:06,300 --> 00:11:11,339
XT results we see the 13 400 is three
272
00:11:08,760 --> 00:11:15,480
percent faster than the 12 400 while the
273
00:11:11,339 --> 00:11:16,920
5600 and 5700x both hit a wall at 154
274
00:11:15,480 --> 00:11:18,899
FPS
275
00:11:16,920 --> 00:11:22,680
performance overall though was similar
276
00:11:18,899 --> 00:11:24,899
with the 13400 beating the 5700x by a
277
00:11:22,680 --> 00:11:27,060
six percent margin
278
00:11:24,899 --> 00:11:29,220
ACC has been tested using the medium
279
00:11:27,060 --> 00:11:31,380
quality preset and as a result the game
280
00:11:29,220 --> 00:11:34,320
is heavily CPU limited and we're seeing
281
00:11:31,380 --> 00:11:36,360
this even with the 6650 XT still the
282
00:11:34,320 --> 00:11:38,640
5700x was around five to ten percent
283
00:11:36,360 --> 00:11:39,959
fast in the 5600 which also made it
284
00:11:38,640 --> 00:11:42,300
seven to ten percent faster than the
285
00:11:39,959 --> 00:11:43,920
Thirteen four hundred the extra cores
286
00:11:42,300 --> 00:11:46,440
and small frequency Advantage the ryzen
287
00:11:43,920 --> 00:11:48,600
7 processor has over the 5600 is clearly
288
00:11:46,440 --> 00:11:51,180
helping here while the small frequency
289
00:11:48,600 --> 00:11:54,120
bumped the 13400 enjoys over the 12 400
290
00:11:51,180 --> 00:11:56,160
is virtually worthless in this example
291
00:11:54,120 --> 00:11:58,380
the rift breaker results are really
292
00:11:56,160 --> 00:12:00,480
interesting and perhaps in part explains
293
00:11:58,380 --> 00:12:02,700
why the AMD processors typically do
294
00:12:00,480 --> 00:12:04,860
poorly in this title for our CPU testing
295
00:12:02,700 --> 00:12:08,100
which normally only features the RTX
296
00:12:04,860 --> 00:12:11,220
4090 we see when using the 6950 XT the
297
00:12:08,100 --> 00:12:13,860
5700x and 13400 both delivered basically
298
00:12:11,220 --> 00:12:16,320
identical performance at around 140 FPS
299
00:12:13,860 --> 00:12:18,839
on average however when switching to the
300
00:12:16,320 --> 00:12:22,620
RTX 490 we see some interesting changes
301
00:12:18,839 --> 00:12:24,180
firstly the 5700x is slower dropping six
302
00:12:22,620 --> 00:12:27,180
percent of the performance seen with the
303
00:12:24,180 --> 00:12:28,740
Radeon GPU meanwhile the 3400 actually
304
00:12:27,180 --> 00:12:30,959
gains performance becoming eight percent
305
00:12:28,740 --> 00:12:33,720
faster and that means with the GeForce
306
00:12:30,959 --> 00:12:37,380
GPU the Intel processor is a massive 18
307
00:12:33,720 --> 00:12:41,640
faster so it's possible how do I use say
308
00:12:37,380 --> 00:12:44,279
the 7900 XTX that the 13400 and 5700x
309
00:12:41,640 --> 00:12:46,980
would deliver comparable performance
310
00:12:44,279 --> 00:12:48,779
okay last up we have Counter-Strike
311
00:12:46,980 --> 00:12:51,720
Global Offensive and as you can probably
312
00:12:48,779 --> 00:12:53,519
tell this title is very much CPU limited
313
00:12:51,720 --> 00:12:55,440
and that's really the case regardless of
314
00:12:53,519 --> 00:12:58,860
the visual quality settings used here
315
00:12:55,440 --> 00:13:01,560
the 5700x is around 13 to 16 faster than
316
00:12:58,860 --> 00:13:03,480
the 13 400 and interestingly the core i5
317
00:13:01,560 --> 00:13:05,820
processor struggles the most with the
318
00:13:03,480 --> 00:13:07,740
RTX 4090 same we're looking at the one
319
00:13:05,820 --> 00:13:09,480
percent lows still if you happen to be
320
00:13:07,740 --> 00:13:10,680
using a Radeon GPU it's going to be very
321
00:13:09,480 --> 00:13:12,600
difficult to tell the difference between
322
00:13:10,680 --> 00:13:14,399
these two CPUs
323
00:13:12,600 --> 00:13:17,519
now here's a look at the 12-game average
324
00:13:14,399 --> 00:13:21,000
data and as you can see the 5700x and
325
00:13:17,519 --> 00:13:24,000
13400 are extremely evenly matched using
326
00:13:21,000 --> 00:13:25,980
the RTX 490 the 5700x was a mere two
327
00:13:24,000 --> 00:13:27,899
percent faster than three percent faster
328
00:13:25,980 --> 00:13:31,980
using the 6950 XT and then back to two
329
00:13:27,899 --> 00:13:33,180
percent faster with the 6650 XT so given
330
00:13:31,980 --> 00:13:34,560
those margins are well within five
331
00:13:33,180 --> 00:13:37,800
percent I think it's fair to say that
332
00:13:34,560 --> 00:13:39,600
overall the 5700x and 13400 delivered
333
00:13:37,800 --> 00:13:41,880
the exact same level of gaming
334
00:13:39,600 --> 00:13:44,459
performance it's also well worth noting
335
00:13:41,880 --> 00:13:46,260
that for gaming the 13400 was just four
336
00:13:44,459 --> 00:13:48,240
percent faster than the 12 400 which
337
00:13:46,260 --> 00:13:50,160
makes sense given it's only clocked five
338
00:13:48,240 --> 00:13:52,320
percent higher therefore when it comes
339
00:13:50,160 --> 00:13:55,019
to gaming the 13400 really isn't an
340
00:13:52,320 --> 00:13:56,880
upgrade over the 12 400 that also means
341
00:13:55,019 --> 00:13:58,920
that for the most part it's not a great
342
00:13:56,880 --> 00:14:01,980
deal faster than the ryzen 5 5600
343
00:13:58,920 --> 00:14:04,740
despite costing significantly more so
344
00:14:01,980 --> 00:14:07,680
from a value perspective the 13400 does
345
00:14:04,740 --> 00:14:09,779
appear very weak finally here's the cost
346
00:14:07,680 --> 00:14:11,700
per frame analysis and for this one I'm
347
00:14:09,779 --> 00:14:13,740
just going to compare the CPU cost given
348
00:14:11,700 --> 00:14:15,779
all CPUs use the same memory and
349
00:14:13,740 --> 00:14:18,839
motherboard prices are also very similar
350
00:14:15,779 --> 00:14:21,420
that being the case the ryzen 5 5600 is
351
00:14:18,839 --> 00:14:23,940
by far the best value part here offering
352
00:14:21,420 --> 00:14:26,760
a little over 20 more value than the 12
353
00:14:23,940 --> 00:14:30,300
400 and over 40 percent more than the 13
354
00:14:26,760 --> 00:14:32,279
400. meanwhile the 5700x is better value
355
00:14:30,300 --> 00:14:35,040
than the 12 400 and significantly better
356
00:14:32,279 --> 00:14:38,399
value than the 13400 reducing the cost
357
00:14:35,040 --> 00:14:40,139
per frame by 27 of course these margins
358
00:14:38,399 --> 00:14:42,839
are reduced once you factor in the
359
00:14:40,139 --> 00:14:45,779
motherboard and memory but the point is
360
00:14:42,839 --> 00:14:49,019
the 5700x presents as a far better
361
00:14:45,779 --> 00:14:52,800
investment for gamers at 180 than the
362
00:14:49,019 --> 00:14:55,320
13400 does at 240 dollars
363
00:14:52,800 --> 00:14:57,660
for productivity the 3400 will fare a
364
00:14:55,320 --> 00:15:00,300
lot better but even there we've seen in
365
00:14:57,660 --> 00:15:02,160
the past that the 5700x and 12 600k do
366
00:15:00,300 --> 00:15:04,680
trade blows depending on the workload
367
00:15:02,160 --> 00:15:06,480
and I'd say best case the 12 600k is
368
00:15:04,680 --> 00:15:09,120
faster in an application like cinebench
369
00:15:06,480 --> 00:15:12,180
but again it costs 33 more so it's not
370
00:15:09,120 --> 00:15:15,360
exactly a clear win here either okay so
371
00:15:12,180 --> 00:15:17,040
like the core I3 13 100 the core i5
372
00:15:15,360 --> 00:15:20,220
13400
373
00:15:17,040 --> 00:15:22,380
it's pretty much Dead on Arrival at
374
00:15:20,220 --> 00:15:23,820
least at the current asking price if
375
00:15:22,380 --> 00:15:25,740
you're watching this video well into the
376
00:15:23,820 --> 00:15:28,440
future and this CPU is now heavily
377
00:15:25,740 --> 00:15:30,120
discounted then possibly a good buy
378
00:15:28,440 --> 00:15:32,399
that's something you'll have to evaluate
379
00:15:30,120 --> 00:15:35,699
but essentially for Intel to be
380
00:15:32,399 --> 00:15:38,820
competitive here today right now these
381
00:15:35,699 --> 00:15:40,920
new locked 13th gen Parts which I say
382
00:15:38,820 --> 00:15:43,380
new because they are essentially just
383
00:15:40,920 --> 00:15:45,779
12th gen processors with a new name but
384
00:15:43,380 --> 00:15:48,240
whatever the case they need to launch at
385
00:15:45,779 --> 00:15:50,339
the same price as the 12th gen parts
386
00:15:48,240 --> 00:15:51,420
that they are effectively replacing so
387
00:15:50,339 --> 00:15:54,480
in other words
388
00:15:51,420 --> 00:15:57,959
the 13400 should be coming in at the
389
00:15:54,480 --> 00:16:00,420
current asking price of the 12 400. so
390
00:15:57,959 --> 00:16:02,820
in the case of the 13 100 rather than
391
00:16:00,420 --> 00:16:06,060
being priced at 150 U.S it should have
392
00:16:02,820 --> 00:16:09,240
launched 130 us at the most and really
393
00:16:06,060 --> 00:16:12,600
today should be at most a 100 part like
394
00:16:09,240 --> 00:16:15,300
the 12 100 F the 13 400 it just suffers
395
00:16:12,600 --> 00:16:17,880
the exact same issue 240 for this
396
00:16:15,300 --> 00:16:21,240
processor makes no sense at all how it
397
00:16:17,880 --> 00:16:22,800
costs more than the ryzen 5 7 600 but
398
00:16:21,240 --> 00:16:25,980
worse than that is the fact that the 12
399
00:16:22,800 --> 00:16:28,320
400 can be had for 180 and while it
400
00:16:25,980 --> 00:16:30,240
doesn't have the 4E cores Gamers really
401
00:16:28,320 --> 00:16:32,339
aren't going to care about that also
402
00:16:30,240 --> 00:16:34,320
Heavens forbid a new generation of
403
00:16:32,339 --> 00:16:37,320
product actually offers something extra
404
00:16:34,320 --> 00:16:38,699
at the same price point sadly this no
405
00:16:37,320 --> 00:16:40,920
longer seems to be a thing in the PC
406
00:16:38,699 --> 00:16:43,560
space or at the very least it's becoming
407
00:16:40,920 --> 00:16:46,199
increasingly rare then of course there's
408
00:16:43,560 --> 00:16:49,860
the ryzen 7 5700x which can be had for
409
00:16:46,199 --> 00:16:52,259
as little as 180 us and the am4 platform
410
00:16:49,860 --> 00:16:55,199
has almost endless affordable quality
411
00:16:52,259 --> 00:16:58,259
motherboard options as well as cheap CPU
412
00:16:55,199 --> 00:17:01,500
upgrades you'd really only buy the 3400
413
00:16:58,259 --> 00:17:02,820
over the 5700x or 12 400 because you
414
00:17:01,500 --> 00:17:04,980
were interested in productivity
415
00:17:02,820 --> 00:17:07,799
performance but in that situation the
416
00:17:04,980 --> 00:17:10,079
13500 makes significantly more sense as
417
00:17:07,799 --> 00:17:13,079
it costs just ten dollars more comes
418
00:17:10,079 --> 00:17:15,120
with twice as many equals 20 more L3
419
00:17:13,079 --> 00:17:18,299
cash and clocks four to six percent
420
00:17:15,120 --> 00:17:19,860
higher not only that but the 13 400 will
421
00:17:18,299 --> 00:17:22,799
be around five percent slower than the
422
00:17:19,860 --> 00:17:25,620
12 600k and with the 12 600k now
423
00:17:22,799 --> 00:17:27,900
available for 250 dollars what's exactly
424
00:17:25,620 --> 00:17:30,360
the point of a 3400 it's a locked part
425
00:17:27,900 --> 00:17:33,299
which costs basically the same amount as
426
00:17:30,360 --> 00:17:34,679
the unlocked version but then you can
427
00:17:33,299 --> 00:17:37,080
just spend ten dollars more and get the
428
00:17:34,679 --> 00:17:40,260
3500 which has twice as many equals and
429
00:17:37,080 --> 00:17:42,480
more cash as I've said so yeah this one
430
00:17:40,260 --> 00:17:45,120
really is Dead on Arrival it's up
431
00:17:42,480 --> 00:17:47,700
pointless rebranding that really just
432
00:17:45,120 --> 00:17:51,419
didn't need to happen so in my opinion
433
00:17:47,700 --> 00:17:53,940
for between 200 and 250 U.S your best
434
00:17:51,419 --> 00:18:00,240
CPU options right now include the ryzen
435
00:17:53,940 --> 00:18:01,980
5 7 600 ryzen 7 5700x or core i5 13500 I
436
00:18:00,240 --> 00:18:03,720
can very easily make an argument for
437
00:18:01,980 --> 00:18:05,820
each of these options they're all valid
438
00:18:03,720 --> 00:18:09,000
depending on your preferences but in
439
00:18:05,820 --> 00:18:10,679
short the 5700x is the value King it can
440
00:18:09,000 --> 00:18:13,620
take advantage of affordable b550
441
00:18:10,679 --> 00:18:16,140
motherboards and ddr4 memory meanwhile
442
00:18:13,620 --> 00:18:18,299
the 13500 it's a budget productivity
443
00:18:16,140 --> 00:18:21,660
King it can be paired with either ddr4
444
00:18:18,299 --> 00:18:23,820
or ddr5 memory and it's supported by a
445
00:18:21,660 --> 00:18:27,179
range of Intel 600 and 700 series
446
00:18:23,820 --> 00:18:29,580
motherboards then the ryzen 5 7600 it's
447
00:18:27,179 --> 00:18:31,620
the least affordable of the three but it
448
00:18:29,580 --> 00:18:33,480
offers by far the best gaming
449
00:18:31,620 --> 00:18:36,539
performance and is without question
450
00:18:33,480 --> 00:18:39,900
backed by a superior platform that will
451
00:18:36,539 --> 00:18:42,419
support future CPU Generations the 13400
452
00:18:39,900 --> 00:18:45,660
though unless it drops to at least 180
453
00:18:42,419 --> 00:18:47,700
us you can just ignore it completely in
454
00:18:45,660 --> 00:18:49,860
favor of one of the three processes just
455
00:18:47,700 --> 00:18:52,140
mentioned and that is going to do it for
456
00:18:49,860 --> 00:18:54,660
this video If you enjoyed it do one of
457
00:18:52,140 --> 00:18:58,020
those subscribe there will be probably
458
00:18:54,660 --> 00:19:00,960
more GPU scaling benchmarks to come and
459
00:18:58,020 --> 00:19:03,120
well many more or GPU and CPU scaling
460
00:19:00,960 --> 00:19:05,400
benchmarks I should say but there will
461
00:19:03,120 --> 00:19:06,840
be more GPU content and CPU content
462
00:19:05,400 --> 00:19:08,640
coming in the not too distant future
463
00:19:06,840 --> 00:19:09,539
some exciting stuff coming up on the
464
00:19:08,640 --> 00:19:11,700
channel
465
00:19:09,539 --> 00:19:13,679
so yeah just make sure you subscribe
466
00:19:11,700 --> 00:19:16,320
that's all I ask just just hit the
467
00:19:13,679 --> 00:19:18,299
Subscribe button and do that
468
00:19:16,320 --> 00:19:19,559
I don't know also we do have float
469
00:19:18,299 --> 00:19:21,780
Planet patreon so you can do those
470
00:19:19,559 --> 00:19:23,160
things if you want to as well uh they do
471
00:19:21,780 --> 00:19:24,780
cost money they're not free like the
472
00:19:23,160 --> 00:19:27,240
Subscribe but you get some pretty cool
473
00:19:24,780 --> 00:19:28,620
perks in return monthly live streams
474
00:19:27,240 --> 00:19:31,200
with Tim myself we can ask questions
475
00:19:28,620 --> 00:19:33,179
live uh we do well we have a Discord
476
00:19:31,200 --> 00:19:35,100
server for members only so that's a cool
477
00:19:33,179 --> 00:19:36,240
place to hang out talk Tech and all that
478
00:19:35,100 --> 00:19:38,520
sort of stuff
479
00:19:36,240 --> 00:19:40,020
behind the scenes content and Q a so
480
00:19:38,520 --> 00:19:41,760
some cool stuff there if you're
481
00:19:40,020 --> 00:19:44,039
interested check that out if not
482
00:19:41,760 --> 00:19:46,080
perfectly fine and I would like to thank
483
00:19:44,039 --> 00:19:49,400
you for watching this video I'm yo Steve
484
00:19:46,080 --> 00:19:49,400
see you next time
485
00:19:50,400 --> 00:20:02,000
[Music]
486
00:19:59,000 --> 00:20:02,000
thank you
487
00:20:02,360 --> 00:20:05,410
[Music]
37004
Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.