Hi, this is Wayne again with a topic “My Thoughts on “Is 4 Cores Enough?” from LinusTechTips + Bonus 8 Core Testing @ The End”.
Hello and welcome to tech deals your eyes do not deceive you. This is a very old benchmark that i have previously published on the channel. It is primarily here to provide some background. While i talk i’m going to show you a member exclusive video published 18 months ago here in a few minutes, i’ll explain more on that, and i want to address a couple of comments to the is four cores enough.
In 2021, video that linus over on linus tech tips recently did it’s a reasonable point. That’S made in that video, because after all many game benchmarks and many esports games, casual games and older games that people like to play in fact run just fine on four cores. Cs: go and grand theft, auto 5 run lovely on four cores. Of course those are very old games assassins, creed, valhalla. Will the built-in benchmark tends to run pretty well on four cores, because the built-in benchmark is not the actual game, but if you get into certain parts of that game, i wouldn’t want to be on a four court chip. If i could all avoid it. The interesting thing about cores is that you don’t need more until you do, and it deeply depends upon what you’re doing clean test benches with nothing running in the background. Running built-in benchmarks are a world different from dirty systems, with two and a half year old installations of windows, with 14 things running in the task, tray a second monitor, listening to music or watching youtube videos and then trying to play modern games testing. For that scenario, however, is very very difficult, because it’s not repeatable, you can do it, but then, if you do it three more times, you’ll get three different results, because windows is going to run background processes differently.
It’S going to cache background processes differently. So, even if you load up the same programs with the same youtube, video playing with the same discord, audio and obs recording in the background, the results are not repeatable. It is extremely difficult to do that, and while there are some benchmark, reviewers who have attempted to do it with trace runs where basically programs are used to record the actual reads and writes to the disk or attempting to record or batch programs running with scripts.
That’S a really bad approximation because it doesn’t include the one thing that really matters when it comes to performance: the person in front of the keyboard when you watch youtube videos of benchmarks. What you will often miss is that the actual control of the game, how, when you press the w key to move forward, is there any lag between when you press the w key and your character actually moves? Sometimes, what will happen is the frame rate will be fine, but the input lag will not and when you go from a four core to a six chord to an eight chord chip, what can sometimes happen is the on-screen. Drawing is fine, but the game’s controllability is not. That varies deeply from game to game. Some games are very more forgiving of it than others, and then some games are just very sluggish.
When you run out of cpu performance, the truth is programs games. Applications are a mixture of needs, they need ram, bandwidth, storage, bandwidth, they need cpu, speed, graphics card performance, all of these things come together and there is usually always going to be something that is preventing more performance, be it your ram bandwidth or your storage bandwidth or You don’t have enough cpu cores or you don’t have fast enough cores. Maybe you don’t need more cores, but you need faster cores. A good example of this is a i7 7700k will benchmark faster than a ryzen 7 1700 in a wide variety of titles, simply because it has much faster individual cores.
That does not mean, however, it would be as good a gaming experience. You may very well get 30 to 40 percent, more stock to stock between a 7 700 k and a 1700, but if the game is jumpy and the frame times aren’t very good and i’m not talking about one percent lows either. This is also something i think a lot of people are confused about.
They go well. I saw the one percent low yeah, but the one percent low is a single number. It’S a single statistic: it is not frame times a one percent low basically takes the bottom.
One percent of frame times and deletes them and shows you what the uh frame rate was with the bottom. One percent of the frame times removed the time between frame to frame so frame times are measured in milliseconds. How many milliseconds between frames and the longest frame times are simply deleted to give you the best one percent frame rate, a two percent – a point: one percent uh frame rate is going to have a different number of frame times dropped. The other thing is that you can have a very uneven frame rate, but you can have a good one percent low and it’s still going to be a terrible experience and that’s also something that’s very hard to show without actually just showing a frame time graph. But when people like to quote numbers and say well, this computer is faster than that computer.
You can look at a number on a one percent low or an average frame rate. You say well clearly, this chip is faster than that chipper or these chips are basically the same and that’s true for those individual data points. There’S an old saying, there’s lies, damn lies and statistics, and it’s not that hard to take numbers and make them say what you want them to say.
If i put two frame time graphs on the screen at the same time, there’s no number that can be drawn from that. That says. Well, i mean there’s averages, but the average frame rate is rarely what really matters as long as you’re within the realm of reasonableness. The the actual variability is only something you can visually look at and you can kind of sort of go well that one is smoother than that one, but by how much? How is that measurable? What’S the percentage of smoothness get very complicated and difficult to express in the sort of twitter reddit instant response? You know youtube comment, type of ways that people want just give me the number darn it uh just which chip is faster. Well, that’s complicated. It depends: what are you looking for? Take a look at our current benchmark on the this is ridiculous: we’re playing world of warships on a core 2 quad q 6600, and these numbers look terrible, but the game is surprisingly playable.
At least it was when i benchmarked it, it really wouldn’t be today because the game, it’s a service game, it’s evolved since then that benchmark was run a while ago, but i could talk for an hour on this topic. It’S performance is more than numbers, and it’s more than a test that you press f5 to run five times you drop the highest to the lowest result. You average the middle three, you make a chart, you put it on youtube and you say see: i tested a bunch of stuff yeah you did, but how much of it did you actually sit down and use and play? I have made a point to sit down and use and play man that is an old end screen. You can tell how old that video cuz – i just i strictly – took the video we’re gon na show you a member exclusive video here in a second i’d, love to hear your comments down below this was published about a year and a half ago to members.
I’M making it available to everybody now, if you’d like to support the channel, hit the join button down below, and i hope you guys enjoy this i’d love to hear what you think. Thank you very much hello and welcome to tech deals supporter exclusive video world of warships. This is actually the seventh in a long series of videos regarding the two thousand dollar sponsored builds from the end of 2019, the i7 9700 kf and the ryzen 7 3800x. Why is this a supporter exclusive, because it’s world of warships? It runs at ridiculous frame rates and if i publish this, i think i would get more eye rolls on the main channel than anything else. I thought about publishing it over on the gaming channel, but to be completely honest, i’m kind of miffed at war gaming at the moment due to their business practices with the puerto rico event. I made three hours of talking head content over on the gaming channel discussing this discussing the state of the game at the end of 2019. That was an hour what i think they need to fix in 2020 in order to increase the player base and improve the quality of life for gamers and be a better company. That was about 50 minutes, and then i spent an hour discussing their business practices and do i think they will listen to me.
No. I will say that those three videos have frankly combined more views than anything else. I think i’ve ever done over on the gaming channel. I was pleasantly surprised to see the response from the audience i thought about carrying on with more, but i’m kind of sort of in a holding pattern to see where i want to go with it.
I i legitimately am upset with wargaming and their responsiveness, or rather the lack of responsiveness. Why make this at all? Well, my wife and i still play world of warships, and so it’s kind of it’s sort of a compromise choice it’s like well, i want to give something back to you guys and be supportive. I gave me an excuse to play a couple of rounds in the game anyway and call it a video gives you something extra without necessarily putting up yet another video over on the gaming channel. The very last video published over there was actually at this point when i’m recording this – i’m actually voicing this over on january 16th uh, so not all of the other videos are published.
Yet it’s now been more than two weeks since i published a video over there. Given the low view counts over there, i may or may not actually publish any more over there. It’S that channel just never took hold, and i either would have to overhaul its content or do a lot more with it or change its theme, or i don’t know, do something different with it. So we’ll that’s in limbo.
That’S kind of on hold at the moment because, frankly, our main channel is what’s important notice. I haven’t talked about the numbers on the screen yet do i need to really i mean? First of all, some of you may say i play world of warships and i never get these kind of frame rates. What did you do to the game? The game actually has a built-in image engine limiter that normally limits the game to 75 frames per second and you have to go and edit the uh engine config file in order to get over 75 frames per second, which i did and in fact we are cpu Limited notice, the graphics cards, in fact, the battle is, oh, these co-op battles go so fast. What can i say, the battle actually ended on the right hand, side don’t worry, i went and saw what the number was before it finished.
We’Re going to see the end result here on the left-hand side pretty soon or not. There, we go 184 frames per second 196. It’S ridiculously fast! This both of these computers are stupid overkill.
We could be running this at 4k and it’d be doing it. You know 100 frames per second. If the limiter wasn’t on. I actually played this at 4k on a 1080 ti at home and i play it at high detail uh, although i have vsync turned on so it just runs at 60 frames per second and frankly doesn’t care. I don’t know.
I just made this for you guys as a fun little supporter benefit, and hopefully you guys appreciate it and enjoy it. Thank you so much for being members of the deal nation and i will see all of you next time you .