Just How Much Faster are Intel CPUs for Gaming?

Thanks for this. So, overall, Intel is faster, but not by much, and it is noticeable only for a very narrow group of users, in some extreme cases. I think the conclusion is fair: if you need every last drop of FPS (for whatever reason), you still need to go Intel; however, if you just want to enjoy games to their full, AMD is a perfectly viable and solid option. No suprises there, but good to see it tried & tested :)
 
Thats completely true, if you ignore results like these:
https://www.techspot.com/review/1871-amd-ryzen-3600/

Battlefield 5: 8700K faster by 11FPS/18FPS
8700K:114/167
3600: 103/149

Tomb Raider: 8700K faster by 10FPS/19FPS
8700K: 78/114
3600: 68/95

Far Cry: 8700K faster by 10FPS/7/FPS
8700K: 84/110
3600: 74/103

World War Z: 8700K faster by 34FPS/25FPS
8700K: 169/201
3600: 135/176

Rage 2: 8700K faster by 4FPS/8FPS
8700K: 122/168
3600: 118/160

Hitman 2: 8700K faster by 12 FPS/13FPS
8700K: 94/118
3600: 82/105
Perhaps you gave a wrong example here. Because this example only proves the value of a 3600. How much do these CPUs sell for? Can you upgrade your coffee lake CPU to a newer gen without a new motherboard?
 
Honestly can't recommend anyone buy Intel anymore. Zen 3 is not too far away and will bring massive gains when it comes to gaming. The only area Intel is able to maintain a slight lead.

Being slightly faster in a few games that don't scale much past 4 cores is only good today. Not the future. Zen 2 already does great in games and Honestly is the CPU to get today. But with Zen3 so close and Intel having nothing to compete with it. I don't see why anyone would go Intel.

Even the XT versions of the Zen2 CPU’s will bring very worthwhile gains. Alas there won’t be a 3700XT, only the more expensive 3800XT, but the 3600XT should be a good match for the 10600K if you’re into this low res e-sports rubbish.

I would get a 3600XT if it were around now to repalce my ancient 3570K, but I also want a better GPU than my 1070, and don’t want current gen GPUS, so will wait for next gen Ampere and RDNA2 and if I do that I’ll wait for Zen 3.
 
Wow, whether or not its noticeable, 10-30FPS is pretty substantial.
I knew Intel was better for gaming but when you compare them like this, I didn't think it was THAT much of a difference.
Intel for gaming all the way, which is why, gamers haven't had a reason to switch since Ryzen released. In fact, if building a gaming rig in 2020, Intel still makes alot of sense. Glad to finally see a review that purely focuses on this aspect, seeing a 20-30FPS difference in some cases is really eye opening.
10 year old Intel architecture with a paint job STILL putting a beatdown on AMD's best, yikes, that will drive a stake through any AMD fanboy's heart.

If gaming @ 1080p with low settings is your thing, especially with esport titles. Considering that when we start looking at 1440p with High settings AMD vs Intel looks nearly the same. And will only get worse with Intel over time as more and more games start scaling past 4-6 cores.

Zen 3 has some big changes that will video games will vary much favor. And with Intel still giving us rebranded Skylake chips, future doesnt look so bright.

Intel clearly is the way to go if you only want to play world of tanks.
 
It would be interesting to see a performance .comparison between real world builds within specific budgets... $1k build and a $1.5k build or whatever budgets make sense.
 
Doesn't sound like you read my comment.
The VERY FIRST thing I said was "whether or not its noticeable".
So I guess I have to repeat myself? Ok....

Wow, whether or not its noticeable, 10-30FPS is pretty substantial.
Thats a beatdown.

It depends on what it's being compared to.

Example 1: in testing, with the same amount of RAM & the same GPU/resolution settings, CPU 1 gets 30 FPS while CPU 2 gets 45 FPS. Not only is there a +15 FPS margin between their performance, but CPU 2 provides +50% more performance over CPU 1.

Example 2: as above, but while CPU 2 had 45 FPS, CPU 3 had 60 FPS. Again, the margin between the 2 CPUs is still +15 FPS, but CPU 3 is now only providing +33% performance over CPU 2...still impressive, but not as impressive as the difference between CPU 1 & CPU 2.

Example 3: same situation, only this time we're comparing CPU 9 & CPU 10. CPU 9 provided 300 FPS...by itself a very impressive feat, as the performance will max out any gaming monitor on the market. CPU 10 managed to outdo it with 315 FPS...but while it has the same +15FPS margin over its competition as the prior 2 examples, it's only a +5% improvement over CPU 9's performance. That's a fairly small margin -- so small, in fact, that in some cases that could be attributed to the margin of error in the testing itself.

The results they're showing in this article? They're analogous to Example 3, not Examples 1 or 2. Which is why the article's conclusion was "Yes, the 10600K is faster, but you probably won't notice the difference in gaming, & we still think the 3700X is better for all-around performance".
 
Wow, whether or not its noticeable, 10-30FPS is pretty substantial.
I knew Intel was better for gaming but when you compare them like this, I didn't think it was THAT much of a difference.
Intel for gaming all the way, which is why, gamers haven't had a reason to switch since Ryzen released. In fact, if building a gaming rig in 2020, Intel still makes alot of sense. Glad to finally see a review that purely focuses on this aspect, seeing a 20-30FPS difference in some cases is really eye opening.
10 year old Intel architecture with a paint job STILL putting a beatdown on AMD's best, yikes, that will drive a stake through any AMD fanboy's heart.

A 5fps difference at 60fps is more substantlal than 30fps at 400fps. That's how substantial it is. Even less so because of diminishing returns.
 
A 5fps difference at 60fps is more substantlal than 30fps at 400fps. That's how substantial it is. Even less so because of diminishing returns.
Exactly, thats why those results I posted between the 8700K and 3600 are so important, the folks saying it only happens at 400FPS are just doing selective reading.
 
And why wasn't a Ryzen 9 3950X used?

Because in the real world when you upgrade/buy a computer you compare things that cost the same. As with anything else cars/plans/rockets/etc..
Price is always the #1 factor unless you are Bill gates or Amazon CEO and have a billion to spare.

The 10600k and the 3700X are priced almost exactly the same, pretty obvious.
This is also why people settle for a "slower" R5 3600 because given the choice to spend almost double that for an intel counterpart that offers 10% more performance, its a no brainer.
 
Wow, whether or not its noticeable, 10-30FPS is pretty substantial.
I knew Intel was better for gaming but when you compare them like this, I didn't think it was THAT much of a difference.
Intel for gaming all the way, which is why, gamers haven't had a reason to switch since Ryzen released. In fact, if building a gaming rig in 2020, Intel still makes alot of sense. Glad to finally see a review that purely focuses on this aspect, seeing a 20-30FPS difference in some cases is really eye opening.
10 year old Intel architecture with a paint job STILL putting a beatdown on AMD's best, yikes, that will drive a stake through any AMD fanboy's heart.

And as you increase the resolution and settings the difference will decrease as well. It's purely a theoretical test not real world. The one area where Intel has a slight advantage is in the 1% measurement. But even then I really doubt the low is unplayable for either processor.
 
People who are gaming a lot buy 10700k at the least, so I don't know what the 10600k does here. Sure, price wise they are a match, but people are willing to pay for 10700k and OC it on top of that. So redo these tests with 10700k oc-ed to 5Ghz+ and you'll see then why Intel is indeed better for gaming.
Not trying to be a hater or something, but I don't understand why it is so hard to accept the fact that Intel parts are better for gaming? They have a low latency internal architecture which makes the difference.
 
Exactly, thats why those results I posted between the 8700K and 3600 are so important, the folks saying it only happens at 400FPS are just doing selective reading.

The example where the 9 game average is 141 on the 8700K and 134 on the 3700X? That's even more insubstantial.
 
As stated above, winning is winning. Intel fans are happy with a 5% edge in games even though it costs more and is much slower for core-heavy workloads. Many people don't do core heavy workloads and the differences between Intel and AMD in light core workloads is small enough not to be noticeable unless you're benchmarking.

Just like while gaming.
 
As stated above, winning is winning. Intel fans are happy with a 5% edge in games even though it costs more and is much slower for core-heavy workloads.
Its more then 5%, its about 10% overall, and in certain games, 15-25%.
Those 8700K vs 3600 bench's show its not just high end gaming when your seeing a 7-15FPS difference.
Most importantly no one cares about core heavy workloads when building a gaming rig.
Any new i5, i7 or i9 performs very well in synthetic benchmarks/multitasking, especially when you compare them core to core, thread to thread, they are almost even, and Intel actually wins a benchmark here and there.
Nobody cares how fast your PC encodes an MP4 or zips a file.


People who are gaming a lot buy 10700k at the least, so I don't know what the 10600k does here. Sure, price wise they are a match, but people are willing to pay for 10700k and OC it on top of that. So redo these tests with 10700k oc-ed to 5Ghz+ and you'll see then why Intel is indeed better for gaming.
Not trying to be a hater or something, but I don't understand why it is so hard to accept the fact that Intel parts are better for gaming? They have a low latency internal architecture which makes the difference.
I'd love to see this too.
 
Its more then 5%, its about 10% overall, and in certain games, 15-25%.
Those 8700K vs 3600 bench's show its not just high end gaming when your seeing a 7-15FPS difference.
Most importantly no one cares about core heavy workloads when building a gaming rig.
Any new i5, i7 or i9 performs very well in synthetic benchmarks/multitasking, especially when you compare them core to core, thread to thread, they are almost even, and Intel actually wins a benchmark here and there.
Nobody cares how fast your PC encodes an MP4 or zips a file.



I'd love to see this too.

The difference you posted are not playable vs non playable performance though so at the end of the day the numbers are academic and don't mean much.

But it seems to me 100% of your workload is just playing games do you actually do anything else on your pc? if not why didn't you buy a console ?
 
Its more then 5%, its about 10% overall, and in certain games, 15-25%.
Those 8700K vs 3600 bench's show its not just high end gaming when your seeing a 7-15FPS difference.
Most importantly no one cares about core heavy workloads when building a gaming rig.
Where do you live? You keep bringing up 3600 vs 8700k and the difference in gaming performance, but you don't mention the price difference. A sane person in today's world would ask: "why does intel's 6 core cost almost twice the price?". 6 core vs 6 core, according to your own metrics? One costs 200 and the other 350 (according to the techspot article you shared). People don't care about core heavy workloads, and they also don't care about price of a product?
 
Back