🛑 9850x3D VS 13900K, 2026 MAX TUNE Benchmark... 😱👑 — Transcript

Full transcript

1. 0:00AMD versus Intel in 2026. Have things
2. 0:04changed? Well, a little bit, but I got
3. 0:07some to show you in this video.
4. 0:15So, Raptor Lake, aka the 13900K,
5. 0:19released in 2022. And basically, I've
6. 0:23been using this for my main gaming CPU
7. 0:26for 4 years now. And in that time frame,
8. 0:29we've had the 5800 X3D, 7800 X3D, 9800
9. 0:34X3D, 9850 X3D, and 9950X3D2.
10. 0:42And of all those AMD CPUs that have
11. 0:44released, I'm still going to be going
12. 0:46with Raptor Lake for my main gaming rig.
13. 0:50Why is that? And the short answer of it
14. 0:53is the AM DIP. Now, for those of you
15. 0:56that don't know what that is, I will
16. 0:57show you an example of it later on in
17. 0:59the video, but basically what the AM DIP
18. 1:02is is an architectural flaw in Ryzen
19. 1:06CPUs because basically the IO die and
20. 1:10the core chiplet have a physical
21. 1:13separation between them limiting the
22. 1:16memory latency. And it doesn't seem like
23. 1:19no matter what AMD does with this
24. 1:21architecture, they cannot fix that
25. 1:24problem. Now, before the AMD fanboys go
26. 1:26in the comments and say it works for me,
27. 1:29or there is no AM dip, a little bit of
28. 1:32backstory. So, when Call of Duty War
29. 1:35Zone 1 came out, this was back in the
30. 1:386900 XT days, I was the only guy saying
31. 1:43the 6T900 XT did not work in that game.
32. 1:482 years after the fact AMD came out and
33. 1:52said, "Yeah, the sixtine XT is broken in
34. 1:56War Zone. We now have fixed the driver."
35. 1:58And that first scenario when that
36. 2:00happened, it was a little odd to me
37. 2:02because I was like, "How did nobody else
38. 2:04pick this up?" So, okay, that's a little
39. 2:06bit of a wash. Whatever. So the second
40. 2:08example of kind of this situation
41. 2:10happening was the Intel degradation meta
42. 2:13where I was the only tech tuber that
43. 2:16diagnosed and fixed the problem almost 6
44. 2:20months ahead of everybody else. So then
45. 2:22I was like what is really going on here
46. 2:25with hardware bros and maybe kicking and
47. 2:28screaming. Next example, AM DIP. Only
48. 2:32now, three or four years later, AMD
49. 2:35finally acknowledged that there are
50. 2:38micro stutters with Ryzen CPUs and that
51. 2:41it will be fixed with Zen 6. So then I'm
52. 2:45like, okay, how is it that I am the only
53. 2:48guy that always seems to diagnose and
54. 2:51see these problems? And why is it
55. 2:53everybody in the comment section says it
56. 2:56works fine for me or I don't see it? So,
57. 3:00I actually experienced this myself. Uh,
58. 3:03basically what the air conditioning
59. 3:05effect is is the brain's ability to
60. 3:09normalize
61. 3:10the current environment that it's in for
62. 3:13survival. The easiest way to explain it
63. 3:16is basically you're stuck in a car in
64. 3:19the middle of a desert and you've never
65. 3:21had air conditioning before. So you
66. 3:23don't know what it feels like. Then when
67. 3:26you actually have air conditioning and
68. 3:28you take it away, well now your brain
69. 3:31has a concept of it and it and it knows
70. 3:33what it feels like. Now, for the last
71. 3:35couple of months, I've been playing on
72. 3:37my AMD test bench, the 9850X3D,
73. 3:41just because my my PC is torn apart,
74. 3:44right? And you know, it was an
75. 3:45interesting phenomena because like I
76. 3:48first started playing Expedition 33 a
77. 3:50little while back uh last Marchish, May
78. 3:54kind of time and I noticed the AM dip in
79. 4:00this game very slightly, but I figured
80. 4:04it was more because of like I was
81. 4:06running Optains on my Intel system and
82. 4:10not so much the SSD speed on the AMD
83. 4:13one. So, I just chocked it up to that,
84. 4:16right? And then when I was playing
85. 4:18Gunfire Reborn with my buddies, there
86. 4:22was actually no issue at all. Not a
87. 4:24single hiccup, not uh no problem
88. 4:27whatsoever, right? And then and then we
89. 4:29finished that game obviously. And then
90. 4:30Dead Zone Rogue is the game that we're
91. 4:32playing right now currently. And
92. 4:35[snorts] again, no problems. I was like,
93. 4:36"Okay." So, I'm like, "Maybe maybe the
94. 4:39AM DIP isn't as bad as uh you know what
95. 4:43I was talking about." And then what had
96. 4:45happened was I actually built this PC
97. 4:48here on my shop uh frameers.org/shop.
98. 4:52I built a 13900K PC out of spare parts
99. 4:56that I had. And then when I put a PC up
100. 5:00for sale on the shop, I obviously play
101. 5:02on it, stress test it, make sure it's
102. 5:05100% good before I ship the PC, right?
103. 5:09Then I played Dead Zone Rogue on this PC
104. 5:13with my buddies and then it hit me. I
105. 5:18could not believe how much smoother the
106. 5:22gaming experience was. I was like, I com
107. 5:25I completely forgot
108. 5:28how good the experience was on an Intel
109. 5:31platform because I had been playing on
110. 5:34the AMD one for so long and my brain had
111. 5:38actually normalized the stutters and the
112. 5:40micro dips and all that as like I
113. 5:43couldn't see it, man. It was wild. So,
114. 5:46my theory is that unless you actually
115. 5:49have both systems right in front of you,
116. 5:52uh, the longer you play on AMD systems,
117. 5:56the more your brain will normalize it.
118. 5:59And that's the only thing that I can
119. 6:00actually think of that explains all the
120. 6:03comments saying it works fine for me or
121. 6:07you're crazy, juofus, because they don't
122. 6:09they can't AB test it in the moment. So,
123. 6:13what we're going to do then, I'm going
124. 6:14to run Dead Zone Rogue on both PCs and
125. 6:18I'm going to show you the difference. Or
126. 6:20at the very least, I'm going to try to
127. 6:22show you the difference because I can
128. 6:24only capture the game in 120 fps, right?
129. 6:27But I'll have uh 1% lows up so you can
130. 6:30kind of see it. Now, I put the memory at
131. 6:346,200 instead of 6400.
132. 6:38And the reason why I did that is because
133. 6:40every time I do an AM dip video
134. 6:44and it's at 6400, someone in the
135. 6:47comments is gonna be like, "It's because
136. 6:48your memory speed and fabric are too
137. 6:50high, bro." It's like, dude, it does it.
138. 6:53It does it no matter what you do with
139. 6:56the fabric speed. It's a It's a flaw
140. 6:58with chiplet architectures.
141. 7:01But let me show you what happens with
142. 7:03Dead Zone Rogue on this PC. And this has
143. 7:06a 5090 in it. Now, in this game
144. 7:09specifically with the AM dip, when I
145. 7:12noticed it, it's later on in the stages
146. 7:15when there's like hundreds of units
147. 7:17spawning. So, I actually got to cuz
148. 7:19right now, see, it's fine,
149. 7:20>> but I'm in this lobby here.
150. 7:23>> 300. I have the frames locked at 390 and
151. 7:26the 1% lows are basically 390, right?
152. 7:29So, like no no dip whip, a mouse whip,
153. 7:32nothing, right? So, I got to go and I
154. 7:34got to play the actual game for a little
155. 7:36bit. Get to level 20 in one of these
156. 7:38like heavy ones. Then I'll show you what
157. 7:40happens.
158. 7:44So, you can see there I did I played
159. 7:46that entire thing without a single blip
160. 7:48in the lows, right? But we're only on
161. 7:50round five of 20.
162. 7:58So [snorts] yeah, that was round nine of
163. 8:0020 and it's starting to get worse with
164. 8:03the dip, right? You saw on the 1% lows.
165. 8:06Not bad enough to where it actually
166. 8:09affects my game play though, right?
167. 8:11That's the key. That's the key. Round 10
168. 8:13there. The dips were getting slightly
169. 8:16worse down to like 350ish. Right.
170. 8:20Oh, that was bad. Yeah. So basically
171. 8:22every time I blow up one of those
172. 8:24spiders,
173. 8:26there's like a massive dip.
174. 8:36Still didn't stop me from winning the
175. 8:37game though, right? This is what I'm
176. 8:39talking about. It's my my my brain
177. 8:41slowly stopped the more I played the
178. 8:44game, the more I like compensated for
179. 8:47the dip.
180. 8:52felt that that was a big one. That was a
181. 8:54big one.
182. 9:02Oh, that was a big one too right there
183. 9:04at the end. Okay, so round 15. They are
184. 9:07the st the the the micro the AM dip the
185. 9:10stutter. It's it's lasting almost like
186. 9:14half a second now. Um, like like I was
187. 9:16saying, it gets worse the longer you
188. 9:18play. And the because some of the games
189. 9:20last like 30, 40, 50 rounds and by the
190. 9:24time me and my buddies would get to like
191. 9:25round 40, round 50, the dips would like
192. 9:29hang the entire game for like a second
193. 9:32and a half. It was bad, dude. Now, let's
194. 9:35see what happens with the Intel system.
195. 9:37Okay, so we're on the Intel system now.
196. 9:39And I know that the AMD boys again will
197. 9:42be like, "It's because you're using DDR5
198. 9:4510 million with liquid nitrogen and
199. 9:48that's the reason why Intel is bad."
200. 9:50It's like, "No, man. It it ain't even
201. 9:52like that, dude." Like, so on this rig
202. 9:55that I that I built, I'm using OEM green
203. 9:59Samsung sticks, not even Heinx ones, and
204. 10:02they only go to 6400. And the timings
205. 10:05are less than optimal, man. like I'm
206. 10:08talking like like this is probably the
207. 10:11worstc case scenario for an Intel
208. 10:13system. That being said, it is
209. 10:16completely tuned though, right? If you
210. 10:18want to learn how to tune your system,
211. 10:20frameers.org/education.
212. 10:23Again, my goal here is to kind of remove
213. 10:26all levels of cope. For some reason,
214. 10:29people think I pick sides or
215. 10:33you shouldn't pick a side. what you
216. 10:35should do or what we should all be doing
217. 10:37or my job at least. I'm just telling my
218. 10:41audience, the boys that support the
219. 10:43channel, I just tell them what is the
220. 10:46best way to get the best gaming
221. 10:48experience they can get. Okay, that's
222. 10:50it. You know what I just realized? Uh,
223. 10:54this literally just this literally just
224. 10:55happened. I uh I don't have the damage
225. 10:59values disabled on the Intel system, but
226. 11:01I have like the little numbers that are
227. 11:03popping up on the guys' heads. I
228. 11:05actually, now that I remember, I
229. 11:07actually disabled that on the AMD system
230. 11:11because I thought the numbers or the
231. 11:13calculations was what was causing the
232. 11:15stutters, dude.
233. 11:18[sighs]
234. 11:19[ __ ] me, dude. Oh my god.
235. 11:24Again.
236. 11:26So, what I'm looking for are
237. 11:30any dips
238. 11:33when I blow up those spiders.
239. 11:37No.
240. 11:40That's what I'm talking about.
241. 11:49Oh, come on, bro. Now, you'll see later
242. 11:51on that the Intel one actually does dip
243. 11:54a little bit too. Um, like probably on
244. 11:57like the last level. So, there there is
245. 11:59something with this game and like how it
246. 12:01handles like textures and stuff. Um,
247. 12:04anything involving textures hits AMD
248. 12:06CPUs much harder than Intel ones. Again,
249. 12:09I'm using 6400 megahertz
250. 12:12Samsung sticks on an Intel. It's like
251. 12:15there's nothing XOC here.
252. 12:21Let's go. So, round 16. So far, no dips.
253. 12:26So far, no dips. This is
254. 12:29so much better. So much brother.
255. 12:46NO. NO. [screaming]
256. 12:49NO.
257. 12:58[sighs]
258. 13:11NOW, it's not just the random AM dip
259. 13:14game that I play. That's the reason why
260. 13:16I'm I'm still with Intel after 4 years.
261. 13:19It's it's about the consistency of the
262. 13:22frames amongst every single game. For
263. 13:25example, in Dead Zone Rogue here, it has
264. 13:28the AM dip. Right now, on the other
265. 13:31hand, that Gunfire Reborn game, that one
266. 13:34performed better with the AMD CPU. The
267. 13:38thing [snorts] is though that there is
268. 13:41no Intel dip, not with Raptor Lake
269. 13:43anyway. So with Raptor Lake, what you're
270. 13:46going to get is at like let's say this
271. 13:51is probably the worstc case scenario of
272. 13:53what you're going to see with Intel,
273. 13:55right? And what you're going to see with
274. 13:58the worst case with AMD is this.
275. 14:02So the fra the the the the
276. 14:05lines being frame rate, right? So the
277. 14:07frames get super high and then the
278. 14:10frames get super low, right? Whereas
279. 14:14like the worst case scenario for Intel
280. 14:16is it doesn't get that low in the first
281. 14:19place. So amongst if you play a hundred
282. 14:22games all 100 of them are going to
283. 14:25perform well with Intel. So, for me
284. 14:28personally, I would rather take a CPU
285. 14:31where I know it's never going to perform
286. 14:33like dog [ __ ] in any game rather than
287. 14:36gambling on a CPU where I don't know if
288. 14:39I want to play a game with my buddies
289. 14:41like this game, I'm not going to have a
290. 14:44good time. It's the exact same thing
291. 14:46with Radeon graphics cards as well. It's
292. 14:49always like a lottery or a crapshoot
293. 14:51whether or not it's going to perform
294. 14:53well. I don't want any of that [ __ ]
295. 14:55man. I just want it to work. It needs to
296. 14:58just work. If you have an AMD CPU, I
297. 15:02encourage you download this game and
298. 15:04give it a shot, man. Like, it's it's I
299. 15:06think it's [clears throat] like 10
300. 15:07bucks. Like, go try it out. People also
301. 15:10forget that I own all the platforms. So,
302. 15:12I was gaming on that 9850X3D for 2
303. 15:16months. I played on it and I determined
304. 15:18that it's still worse for me than Raptor
305. 15:22Lake. I don't use the 270K or the 285K
306. 15:26either because I have the benefit of
307. 15:28being able to test everything myself.
308. 15:31It's very easy for me to not have the
309. 15:34air conditioning effect and I can do
310. 15:37direct AB tests and be like, "Yeah, this
311. 15:39one is better for the majority of
312. 15:41games." But I couldn't care less. You
313. 15:43buy whatever you want to buy, man. If
314. 15:44you play Call of Duty only and that's
315. 15:47all you play, then maybe AMD is better
316. 15:50for you. and that one specific game. I
317. 15:53sell both PCs, man. I don't care. I
318. 15:56don't care what the customer actually
319. 15:58wants to buy. I care that they're buying
320. 16:01the right thing, the right tool for the
321. 16:04job. If the tool is workstation, the
322. 16:07270K can't be beat. If the tool is Call
323. 16:11of Duty, AMD can't be beat. If the tool
324. 16:15is having the most consistent
325. 16:18performance in any game you play and you
326. 16:21never want to gamble, then the Raptor
327. 16:23Lake is the best. So, what I am building
328. 16:26is a gaming PC for me. And that means I
329. 16:30need to go Raptor Lake again. And this
330. 16:33is I don't know what's going to happen
331. 16:35with Nova Lake, but for the foreseeable
332. 16:37future, 13900K,
333. 16:4014900K,
334. 16:42it's still the best gaming CPU, and it
335. 16:45became even more apparent having just
336. 16:49tested it with playing the 9850X3D for 2
337. 16:52months. That's all I wanted to show you
338. 16:54guys. That's the reason that my personal
339. 16:57build is still going to be Raptor Lake.
340. 17:00I hope you found this content useful,
341. 17:03though. Dead Zone Rogue is another game
342. 17:05that has the AM dip. I always seem to
343. 17:07find these niche ass games that have
344. 17:09them, but it's my job to report my
345. 17:12findings to you guys. If you like this
346. 17:13content, scroll down, hit that subscribe
347. 17:15button, join the Discord, support the
348. 17:18channel any way you can. Hit up the
349. 17:20shop, pick up the PC optimization class,
350. 17:22and I will see you guys in the next one.
351. 17:25Talk to you later.
352. 17:27[applause]