How is Intel Beating AMD Zen 3 Ryzen in Gaming?

[menko2]

Once again ZEN3 is newer, on the better node and is faster in everything including PC Gaming over anything Intel has to offer.

That's what I want but I have both systems right now and at 4k (I don't test 1080p cause I don't use it) I get a few fps more with the old system. And the 8700k I'd quite old.

I have been trying the last few days everything because the 5900x crashes in everything the 8700k@5ghz I don't understand how in gaming is a bit worst. I set up the RTX 3090 in afterburner to be the same in both systems.

 

[Super XP]

That's what I want but I have both systems right now and at 4k (I don't test 1080p cause I don't use it) I get a few fps more with the old system. And the 8700k I'd quite old.

I have been trying the last few days everything because the 5900x crashes in everything the 8700k@5ghz I don't understand how in gaming is a bit worst. I set up the RTX 3090 in afterburner to be the same in both systems.

Yes but the difference is negligible at best.

 

[menko2]

Yes but the difference is negligible at best.

Yes i think I'm going to return everything until something better comes along.

My 8700k still rocks for a little bit more as I only game.

 

[HD64G]

Game engines make those differences now. Zen3 is at least equal or better in light-threaded games on average. CSGO is the ultimate example.

 

[Calmmo]

It's thanks to extremely agressive voltage behavior. 5900x gets actually hotter running games than it does running all core workloads. They dont at all behave like you'd expect based on past experience. Very different to the 3900x i had before.

 

[menko2]

Game engines make those differences now. Zen3 is at least equal or better in light-threaded games on average. CSGO is the ultimate example.

I have been trying different configurations and overclocks to test in different games and compare with my 8700@5ghz and z370 motherboard.

Horizon zero dawn, Assassins Creed Valhalla, someday you'll return (unreal engine), red dead redemption 2, shadow tomb raider,...

All of them have different engines but the result is the same. I get around 3-5 fps more at 4k with the 8700k and z370.

I don't know why is this because the 5900x crushes the 8700k in all benchmarks. But games results is what matters to me.

I used the same settings in the rtx 3090 in both platforms. About ram I used 3800mhz-1900mhz infinity frabric and cl 14-16-16-39 with the Asus x570 hero and 3900mhz 15-15-15-35 with the z370.

I also tried boost single core to 5.15ghz with the 5900x and another configuration of all core to 4.6ghz.

 

[Dredi]

I get a few fps more with the old system.

Which games? On 4k the processor matters very little

are your results taken with the peocessor at stock settings? Bad overclocks can harm performance a lot. Might also be something wrong with your OS.

 

[HD64G]

I have been trying different configurations and overclocks to test in different games and compare with my 8700@5ghz and z370 motherboard.

Horizon zero dawn, Assassins Creed Valhalla, someday you'll return (unreal engine), red dead redemption 2, shadow tomb raider,...

All of them have different engines but the result is the same. I get around 3-5 fps more at 4k with the 8700k and z370.

I don't know why is this because the 5900x crushes the 8700k in all benchmarks. But games results is what matters to me.

I used the same settings in the rtx 3090 in both platforms. About ram I used 3800mhz-1900mhz infinity frabric and cl 14-16-16-39 with the Asus x570 hero and 3900mhz 15-15-15-35 with the z370.

I also tried boost single core to 5.15ghz with the 5900x and another configuration of all core to 4.6ghz.

Maybe your timings on Ryzen are causing some errors or increase latency which decreases performance. I would check the stability-latency combo for that.

 

[menko2]

Which games? On 4k the processor matters very little
View attachment 185373
are your results taken with the peocessor at stock settings? Bad overclocks can harm performance a lot. Might also be something wrong with your OS.

I have done it in a fresh installed windows 10. It's actually the second time I try it. The first time was with a 5800x.

Tried with stock, overclocking the single core and also all core overclock.

It shouldn't matter much at 4k but I have the benchmarks of this games (the ones that have built in benchmark) and also savegames in the same location and moment in the game and the diference is there. I take pictures of every bench and fps in game to compare.

I tweaked the ram and latencies. I have samsung B-die kits that work very well.

Disapointed in gaming for me. At least I expected to have the same fps as with the z370 and 8700k@5ghz. Instead I got a little less.

 

[Dredi]

Disapointed in gaming for me. At least I expected to have the same fps as with the z370 and 8700k@5ghz. Instead I got a little less.

Then there is definitely something wrong with your system. In order to get a few fps (~10%) difference at 4k you would normally need to compare something like 10900k to 2200g. To get such a drastic performance difference (in any direction!) should not be possible with 8700k vs. 5900x. What is your GPU and is there a GPU usage % difference between the systems?

 

[menko2]

Then there is definitely something wrong with your system. In order to get a few fps (~10%) difference at 4k you would normally need to compare something like 10900k to 2200g. To get such a drastic performance difference (in any direction!) should not be possible with 8700k vs. 5900x. What is your GPU and is there a GPU usage % difference between the systems?

For example I attach here the Valhalla benchmark with the systems. The one that doesn't have the CPU detail is the 5900x. The 5800x gives a little better results overall than the 5900x. But neither of them gets to the 8700k@ghz.

Apart of benchmarks I use the method of load a save game and wait the same amount of time on the scene and same temps in the room to replicate (afterburner tweaked the same). Then I see the fps in that exact moment with the different configurations.

I still have to see the GPU usage but it shouldn't get more the 8700k.

 

[Dredi]

For example I attach here the Valhalla benchmark with the systems. The one that doesn't have the CPU detail is the 5900x. The 5800x gives a little better results overall than the 5900x. But neither of them gets to the 8700k@ghz.

Apart of benchmarks I use the method of load a save game and wait the same amount of time on the scene and same temps in the room to replicate (afterburner tweaked the same). Then I see the fps in that exact moment with the different configurations.

I still have to see the GPU usage but it shouldn't get more the 8700k.

The 5800x results are clearly better though? Higher min fps combined with lower variance produces a lot better playing experience. I still think that there is something wrong with your system though, unless valhalla is one of the two games that play better on intel, the other one being rdr2.

edit: run to run variance would be nice to know as well, it might be that the results are within it.

 

[menko2]

The 5800x results are clearly better though? Higher min fps combined with lower variance produces a lot better playing experience. I still think that there is something wrong with your system though, unless valhalla is one of the two games that play better on intel, the other one being rdr2.

edit: run to run variance would be nice to know as well, it might be that the results are within it.

Here it is the pictures I took from the scores with horizon.

Again the 5800x is better than 5900x but 87000k comes on top.

From all the savegames I load and test I get the same results of being 3-5 fps less.

I have been trying for days to get the same fps as with the 8700k. Memory tuning, bios, drivers, windows tweaks,..

I'm not a intel or amd fan. Just looking for the best gaming performance for a system. Im against Rocket Lake but im starting to think that it might give me same scores as the 8700k or better.

 

[Dredi]

Here it is the pictures I took from the scores with horizon.

Again the 5800x is better than 5900x but 87000k comes on top.

From all the savegames I load and test I get the same results of being 3-5 fps less.

I have been trying for days to get the same fps as with the 8700k. Memory tuning, bios, drivers, windows tweaks,..

I'm not a intel or amd fan. Just looking for the best gaming performance for a system. Im against Rocket Lake but im starting to think that it might give me same scores as the 8700k or better.

Again, what is the run to run variance? HZD is widely benchmarked and should not show that much of a difference. Anyway, swapping CPUs for 4k gaming makes zero sense. You are anyway GPU bound always.

 

[menko2]

Again, what is the run to run variance? HZD is widely benchmarked and should not show that much of a difference. Anyway, swapping CPUs for 4k gaming makes zero sense. You are anyway GPU bound always. View attachment 185386

In paper and for what I read yes. But testing yourself is the best thing you can do.

I re-run the test many times with all options and results are steady.

But there is a way with the savegames and exact locations and scenes that clearly indicates this small diference.

Might be intel better latency or I don't know but it's like this.

 

[Dredi]

In paper and for what I read yes. But testing yourself is the best thing you can do.

I re-run the test many times with all options and results are steady.

But there is a way with the savegames and exact locations and scenes that clearly indicates this small diference.

Might be intel better latency or I don't know but it's like this.

And in some other games AMD is noticeably faster. It's not like the margin is huge in any resolution anyway.

 

[menko2]

And in some other games AMD is noticeably faster. It's not like the margin is huge in any resolution anyway.

I guess it will be like this. I only haven't found an of my games where and was a bit faster.

I think I'll return the amd and get the Rocket Lake for future proof with pcie 4.

I thought I was already done getting amd who already has pcie 4 and also many more cores but...having the two systems here I'm not convinced.

Edit: thank you for the help

 

[Cobain]

Then there is definitely something wrong with your system. In order to get a few fps (~10%) difference at 4k you would normally need to compare something like 10900k to 2200g. To get such a drastic performance difference (in any direction!) should not be possible with 8700k vs. 5900x. What is your GPU and is there a GPU usage % difference between the systems?

There is nothing wrong. I followed his thread on another fórum where Many users also repórted and shown that tuned Intel beats tuned Ryzen 5000 in games, BE it 240hz scenario low settings or 4k. It is not wrong.

This is a vídeo example from that discussion. All this zen 3 hype makes a lot of readers think AMD has beaten Intel in games. In reality it doesnt, unless you compare max tuned Ryzen 5000 RAM against mild tuned Intel. As I usually Say, fanboys ruin the internet. Zen 3 is great and for multi thread workloads they completou destroyed Intel. Also with lower power usage. But for Pure gaming machine? No, AMD just closed the Gap. They might BE better on 10% of the games. Intel ring bus interactiom still makes them Faster.

 

[Dredi]

There is nothing wrong. I followed his thread on another fórum where Many users also repórted and shown that tuned Intel beats tuned Ryzen 5000 in games, BE it 240hz scenario low settings or 4k. It is not wrong.

This is a vídeo example from that discussion. All this zen 3 hype makes a lot of readers think AMD has beaten Intel in games. In reality it doesnt, unless you compare max tuned Ryzen 5000 RAM against mild tuned Intel. As I usually Say, fanboys ruin the internet. Zen 3 is great and for multi thread workloads they completou destroyed Intel. Also with lower power usage. But for Pure gaming machine? No, AMD just closed the Gap. They might BE better on 10% of the games. Intel ring bus interactiom still makes them Faster.

You got it the wrong way around, Zen3 is faster in 90% of games and intel in maybe 10%. With basic equally non-tuned XMP 3200 memory, see

I tried watchin the video you posted, didn't seem to include anything about the testing methodology or anything really. Just a few numbers on screen and that's it. They are also downclocking(!!!) the 5950x from typical 5050MHz boost to just mere 4650MHz, while overclocking the 10700k well beyond it being stable (silicon lottery states that just 2% can sustain those clocks and none with AVX offset disabled as in the test.

 

[NesteaZen]

just posting bc i speak it.

@Dredi
ttps://www.youtube.com/watch?v=GkpPnXKbXNU&lc=UgxUgUd663IiZb2gQ8V4AaABAg

"Why is Ryzen choked to 4.65? I have a very similar set and in Auto in games it boosts up to 5+ Ghz. And after the OC of all the cores, the fps results drop. "

"If you have such results, record the material and show us how PBO gives better results than all core, because I checked and I know it is the opposite.

Why?
-because the games opposite to the application (or benchmarks) require inter-core communication
-because hopping increases the latency
-because most modern games prefer to have equal core clocking rather than randomization
-bo system scheduler runs more smoothly

Of course, I checked PBO vs manual and there are 20-80% less frames, it depends on which one to look at (especially LOW FPS drop dramatically in the car). Of course, AMD owners are pleased with these temporary increases in MHz, who would not want to have a 5GHz Ryzenek, but unfortunately these are apparent values that do not translate into real gaming performance. It is difficult to play such AC Competizione with PBO, because low ones will drop so much that it is difficult to hit apexes, so what if AVG FPS are similar, since there is no question of a stable render.

An example from just one game

"

 

[Dredi]

Of course, I checked PBO vs manual and there are 20-80% less frames, it depends on which one to look at (especially LOW FPS drop dramatically in the car). Of course, AMD owners are pleased with these temporary increases in MHz, who would not want to have a 5GHz Ryzenek, but unfortunately these are apparent values that do not translate into real gaming performance. It is difficult to play such AC Competizione with PBO, because low ones will drop so much that it is difficult to hit apexes, so what if AVG FPS are similar, since there is no question of a stable render.

Sounds more like your memory overclock or something else is not stable. In most review outlets the lows on AMD are better than on intel, and most don't bother with all core downclocking. 80% less frames is a BOLD claim.

 

[NesteaZen]

>Sounds more like your memory overclock
his*

just posting bc i speak it.

yea I too think there's something up. the funny thing is he says "we're not building €2000-5000 gear to run it on stock. that's why I'm showing you what you can get out of your systems with easy overclocking."
proves his own statement wrong @ easy

at 9:00 he says: "it seems codemasters' ego engine seems to prefer intel. but it's weird seeing these differences bc testing zen2 3700x, 3900x vs 9900k in f1 2020 I didn't see these differences"

I love it when ppl paste random channels while disregarding the established ones. no disrespect just an observation.

 

[Dredi]

his*

Ah, my mistake. Sorry!

 

[Cobain]

You got it the wrong way around, Zen3 is faster in 90% of games and intel in maybe 10%. With basic equally non-tuned XMP 3200 memory, see

I tried watchin the video you posted, didn't seem to include anything about the testing methodology or anything really. Just a few numbers on screen and that's it. They are also downclocking(!!!) the 5950x from typical 5050MHz boost to just mere 4650MHz, while overclocking the 10700k well beyond it being stable (silicon lottery states that just 2% can sustain those clocks and none with AVX offset disabled as in the test.

Yeah right, let's test 5 games with dual rank RAM (4x8gb) wich highly favours AmD and ignore the fact most users Will use 2x8 at stock! Plus, I prefer to trust on reviews where they test 20 games, not F1 or civilization -_-

And wait, isnt GamersNexus the guy that Said a gtx 1060 can compete with the ps5 gpu? Sorry I dont rate those youtubers

Every Intel vs AMD test with max tuned settings for both, puts Intel on top for gaming. Unless you compare AMD dual rank b die against Intel single rank and no overclock

 

[NesteaZen]

Yeah right, let's test 5 games with dual rank RAM (4x8gb) wich highly favours

you seem to be new to the game.

You are the one that is referencing people that can't setup their machines. -50% less performance. like in what world. At first you talk about stock, then you talk about overclock. You are all over the place.
TPU average is only 3% difference. Most review outlets have slightly higher 5950x, 5900x averages.
It only comes down to price (with motherboard included).

so far you haven't linked anything that back up your claims.

jfyi no review outlet is reviewing at non-stock bc overclocks are not guaranteed.