RTX 4090 & 53 Games: Core i9-13900K vs Ryzen 7 5800X3D

[80-watt Hamster]

Gamers normally do not run heavy MT workloads, besides 13600K is nearly as fast in games as 13900K, and as an all-rounder it's as good as 5800X3D if not better. Both are just fine power consumption-wise in gaming. I've never talked about undervolting Intel CPUs. No idea where from you drew that conclusion. .

If you like the Ryzen 5800X3D great, just stop parroting that RPL is a power use hog. It's only 13900K/13700K and only in heavy MT tasks.

In order to show that the 13900K is superior to the 5800X3D in frames/W, you posted a chart where the 13900K alone was tuned with an undervolt and underclock. The X3D still came in second place, and about as far behind the tuned i9 as the 90W i9 is behind the X3D.

I can post lots more reviews which show that 13900K is a fine CPU while gaming

Of course you can. Is that even in dispute?

 

[Space Lynx]

Well, I kind of regret getting raptor lake now after watching this video... looks like 7600x beats it easily across the board as long as you slot in some ddr5 6000 cas 30 ram. i think even lows are better by 20% with the 7600x over the 5800x3d and 13600k... turns out 7600x was the sleeper winner all along you just have to use really high end ram.

fuck. i don't know know what to do now. i should have never betrayed my love for AMD after all...

7600x beats x3d and raptor lake according to this video, in highs, in lows, everything. as long as you pair with ddr5 6000 cas 30... how come I didn't realize this when I bought my 13600k on launch day... im so fucking confused right now. ugh my head hurts.

 

[EatingDirt]

I think you just didn't check the 13900k review: they did not test Ace Combat 7 (+30%), Detroit become human (+18% @1080p), Dos (+20.8% @1080p, +19% @1440p), Guardian of the Galaxy (+12.3% @1080p), SpiderMan remastered (+17.6% @1080p, +14.6% @1440p).

5 titles with that guarantee +20% at 1080p not tested in the 13900k review which shows a better scaling compared to this comparison. Don't you think that's strange?

If you add 41 games to the original 12 to the test suite, and only 5 of of those new games have 25% better performance, and the rest have ~0%, those outliers aren't going to make a big difference in the final percentage. The ~0% will make up 77% of the score, the ~25% will make up 23% of the score.

Weighted and simplified that turns into 5.75%.

 

[nomadka670]

Well, I kind of regret getting raptor lake now after watching this video... looks like 7600x beats it easily across the board as long as you slot in some ddr5 6000 cas 30 ram. i think even lows are better by 20% with the 7600x over the 5800x3d and 13600k... turns out 7600x was the sleeper winner all along you just have to use really high end ram.

fuck. i don't know know what to do now. i should have never betrayed my love for AMD after all...

7600x beats x3d and raptor lake according to this video, in highs, in lows, everything. as long as you pair with ddr5 6000 cas 30... how come I didn't realize this when I bought my 13600k on launch day... im so fucking confused right now. ugh my head hurts.

Thats why I said test again the 7700x vs 13700k with the Right RAM and rtx 4090 . People like to manipulate benchmarks

 

[AlainCh2]

Somewhere in USA, may be. Worldwide it's still $430, excluding taxes.

Taxes & Shipping included - Amazon prices in the Union

 

[gffermari]

Well, I kind of regret getting raptor lake now after watching this video... looks like 7600x beats it easily across the board as long as you slot in some ddr5 6000 cas 30 ram. i think even lows are better by 20% with the 7600x over the 5800x3d and 13600k... turns out 7600x was the sleeper winner all along you just have to use really high end ram.

fuck. i don't know know what to do now. i should have never betrayed my love for AMD after all...

7600x beats x3d and raptor lake according to this video, in highs, in lows, everything. as long as you pair with ddr5 6000 cas 30... how come I didn't realize this when I bought my 13600k on launch day... im so fucking confused right now. ugh my head hurts.

The 13600K/KF with DDR5 is the best overall combination.
The 7600X performs the same in gaming (the differences between all three cpus are a joke), worse in productivity but offers better upgrade path.

So....apart from the fact that you use DDR4 with the 13600KF, and may lose some frames, your choise was the best possible for an Intel build.

Just enjoy your build.

 

[Space Lynx]

The 13600K/KF with DDR5 is the best overall combination.
The 7600X performs the same in gaming (the differences between all three cpus are a joke), worse in productivity but offers better upgrade path.

So....apart from the fact that you use DDR4 with the 13600KF, and may lose some frames, your choise was the best possible for an Intel build.

Just enjoy your build.

I already bought my mobo though, and I did not realize how much DDR4 ram was holding me back. Sucks ass.

Also, if I am going to spend a shit ton of money on DDR5 ram I might as well go with 7600x then I can slot in 7800X3D someday down the road or 8800X3D a few years from now, etc.

I thought I was getting the best bang for my buck, but eh I guess not. I still saved about $300 overall by not going with AM5 premium prices, but in the longer term I didn't, and I get worse performance out the gate...

 

[N/A]

I already bought my mobo though, and I did not realize how much DDR4 ram was holding me back.

To avoid buyers remorse wait or don't buy. Just sell it. I think they're all the same, with 58X3 / 7600X 10-15% faster than 13600K in 1440p, but that will not be the case with 7900 as it tends to have less driver overhead.

 

[vmarv]

The most important subset of these 53 games to me, is the one of "low frame rate games". I would like to see the bottom 25 percent of games taken out of the group of 53 (the bottom 13 games in FPS for example), so we can see what difference the CPU is making in games that run slowly. I don't need a CPU upgrade to get from 400 to 600fps. I would like to bring 100 fps titles up to 150fps etc.
Very funny that out of all the games that have large differences at 4k, DMC5 and Civilization 6 are the only ones I play and they are both heavily AMD favored. Upgrading to Intel would slow me down. Intel has to get fixes for those 2 games out! The scheduler is still not working right?

If I'm not wrong, the hybrid Intels work better with Windows 11. I'm not speaking of gaming in particular, but overall. I saw so many reviews these days and unfortunally I can't remember where I read it, but the scheduler seems to works better in the new Windows.
Is it correct guys?

 

[Asni]

If you add 41 games to the original 12 to the test suite, and only 5 of of those new games have 25% better performance, and the rest have ~0%, those outliers aren't going to make a big difference in the final percentage. The ~0% will make up 77% of the score, the ~25% will make up 23% of the score.

Weighted and simplified that turns into 5.75%.

Yep, exactly: 5.75% in addition to the initial 13% difference (keeping 13% as an approximation: you should evaluate the number of games, the scaling on the rtx 4090 in these initial 12 games and the percentage in the the other 36 games which isn't 0%). That's the point.
While this comparison talks about a 6.2% total!

Spoiler

Spoiler

 

[Imouto]

Somewhere in USA, may be. Worldwide it's still $430, excluding taxes.

In Europe it's 360~380€ taxes included.

 

[Space Lynx]

To avoid buyers remorse wait or don't buy. Just sell it. I think they're all the same, with 58X3 / 7600X 10-15% faster than 13600K in 1440p, but that will not be the case with 7900 as it tends to have less driver overhead.

I'm just going to say fuck it and keep my current setup. this is my 10 yr build and im retiring from the hardware hobby. i used to love this shit, but i don't these days. haven't for quite some time really, even this build was more just out of necessity cause my laptop is really quite outdated.

so yeah fuck it, im still going to get great performance and i did save $300 or so by not doing AM5, so eh its all good. this rig will do me good for next ten years. fuck all the nonsense, time to game boys

 

[b1k3rdude]

On top of the charts I had a notice that the titles were wrong, but I'm traveling, so I can only fix them when back home. Looks like many people don't read text. Back home and the chart title are fixed now

Nice one Wiz, I thought it might be something simple.

Well I have my 5900x and a 3080-12gb, so am happy to wait for the 7800X3D and by then the 4080-16GB/RX7900XT & DDR5-6000 will have dropped to more reasobale prices. At which point if finances allow, will then do a complete upgrade and sell the old rig etc.

 

[bastordd]

Intel 13 gen sweet spot for ram is 6400...
6000 with cl 36 is to slow

 

[regs]

Taxes & Shipping included - Amazon prices in the Union

View attachment 268691

China - $400. And so as most of the world.

https://www.aliexpress.us/item/3256804012905041.html

Still doesn't change anything. 12400F cost 150 euro, excluding VAT. 350-150 is still 200 euro difference, which you can spend on 6750 XT instead of 6600 XT. That would give you far greater experience. Except if you playing strategies, of course.

 

[ShiningSapphire]

Well, I kind of regret getting raptor lake now after watching this video... looks like 7600x beats it easily across the board as long as you slot in some ddr5 6000 cas 30 ram. i think even lows are better by 20% with the 7600x over the 5800x3d and 13600k... turns out 7600x was the sleeper winner all along you just have to use really high end ram.

fuck. i don't know know what to do now. i should have never betrayed my love for AMD after all...

7600x beats x3d and raptor lake according to this video, in highs, in lows, everything. as long as you pair with ddr5 6000 cas 30... how come I didn't realize this when I bought my 13600k on launch day... im so fucking confused right now. ugh my head hurts.

You made no mistake with this choice. 13600K is much better all-rounder beating 7600X by a lot in multi core workloads and providing great performance in games. Here you can read a review in which the testing methodology in games is based on finding the most cpu heavy scenarios (not built-in benchmarks like so many reviewers like to do) to show differences between different cpus and as you will see 13600K performs really solid better than 7600X in games not far away behind 13900K. It's in Polish but you'll get by with a translator. https://www.purepc.pl/test-procesor...tna-wydajnosc-porownanie-w-grach-i-programach

 

[EatingDirt]

Yep, exactly: 5.75% in addition to the initial 13% difference (keeping 13% as an approximation: you should evaluate the number of games, the scaling on the rtx 4090 in these initial 12 games and the percentage in the the other 36 games which isn't 0%). That's the point.
While this comparison talks about a 6.2% total!

Spoiler

Spoiler

The initial difference was 11.4%. The CPU games suite was a lot more in favor of the 13900k when paired with the 3080. Which had CS:GO @32%, AoE5 @28%, Borderlands 3 @6% & FC6 @27% all in favor of the 13900k.

It may be that the 4090 is just better with the 5800X3D than the 3080 was. Driver updates, AGESA updates and game patches could also account for the increase in performance for the 5800X3D. There are so many factors at play when benchmarking that you're going to get anomaly's and differences from month-to-month, year-to-year, etc. If we bench this a month from now, Devil May Cry 5 might be in favor of the 13900k by 10%, which would put us closer to the initial 11.4%.

 

[gffermari]

im still going to get great performance

You're getting 99% of the absolute performance. I think you can live with that, taking into account that you're not gonna get a 4090 now and a 4090Ti the day it's released etc...

For the money you spent....you actually stole them.

See you in 4-5 years. We'll both need to upgrade some time then.

 

[Asni]

The initial difference was 11.4%. The CPU games suite was a lot more in favor of the 13900k when paired with the 3080. Which had CS:GO @32%, AoE5 @28%, Borderlands 3 @6% & FC6 @27% all in favor of the 13900k.

It may be that the 4090 is just better with the 5800X3D than the 3080 was. Driver updates, AGESA updates and game patches could also account for the increase in performance for the 5800X3D. There are so many factors at play when benchmarking that you're going to get anomaly's and differences from month-to-month, year-to-year, etc. If we bench this a month from now, Devil May Cry 5 might be in favor of the 13900k by 10%, which would put us closer to the initial 11.4%.

In that review, the 13900k performs 13% better than the 5800x3d (100/88.6). The 13900k review is 15 days old, nothing changed.
I'd just like to know how we went from 13% with a rtx 3080 to 6.2% with a rtx 4090 in a completely cpu bound scenario.

 

[nomadka670]

You made no mistake with this choice. 13600K is much better all-rounder beating 7600X by a lot in multi core workloads and providing great performance in games. Here you can read a review in which the testing methodology in games is based on finding the most cpu heavy scenarios (not built-in benchmarks like so many reviewers like to do) to show differences between different cpus and as you will see 13600K performs really solid better than 7600X in games not far away behind 13900K. It's in Polish but you'll get by with a translator. https://www.purepc.pl/test-procesor...tna-wydajnosc-porownanie-w-grach-i-programach
View attachment 268696

Then what about the hardware unboxed video where the ryzen 7700x beats the 13700k . And they not bullshittin.

 

[Space Lynx]

You're getting 99% of the absolute performance. I think you can live with that, taking into account that you're not gonna get a 4090 now and a 4090Ti the day it's released etc...

For the money you spent....you actually stole them.

See you in 4-5 years. We'll both need to upgrade some time then.

It's true, its more of just the confusion of the variances being so great between review methodologies this round that made me a little frustrated. Ram never used to matter much, especially not on Intel's side of things. I just assumed nothing had changed cause the initial reviews showed 13600k dominating.

It's all good though, I actually just did the math, and by going with my current setup I saved exactly $320 by avoiding high end ddr5 ram and a $330 AM5 mobo (if I went AM5 I would have gone with the Riptide mobo). total savings were $320 by going with a high end clearance sale mobo for z690 and budget dual ranked ddr4 ram, for you are correct, even if i build again in 5 years AM6 socket will be out.

I might do a new build in 5-6 years, depends where I am at in life. I do intend to try to make this a 10 year build though, with maybe one gpu upgrade in-between, since i do have a gen5 slot for gpu on my mobo.

 

[Mistral]

What's the deal with DMC at 1080p and Civ at 4k?

 

[Scrizz]

the Civ VI result at 4k seems odd. It also seems most games are GPU not CPU limited.

 

[DeadmanFatboy]

In that review, the 13900k performs 13% better than the 5800x3d (100/88.6). The 13900k review is 15 days old, nothing changed.
I'd just like to know how we went from 13% with a rtx 3080 to 6.2% with a rtx 4090 in a completely cpu bound scenario.

I'm not sure what is so confusing. The average got dragged down when more games were tested. Averages can be misleading. That's why most reviewers show individual game benchmarks.

Conclusion is clear, though: 13900K and 5800X3D are evenly matched in most games, but a few favor one (usually Intel) with double digits relative performance gains.

 

[wheresmycar]

I'm really impressed with these results.. the less power munching DDR4 X3D competing head-to-head with a faster stock clocked and DDR5 pushed 13900K gaming overkill. In some of the titles benched which are currently in my play-library the difference in perf going either way is only -/+2.5%. The only 2 games which i'm interested in where intel offers a nice performance uplift (13-16%) is death stranding and far cry 6. Havent touched these titles yet but definitely looking to secure a third title with similar performance gains, spiderman remastered, for my nephews and nieces.

Bottom line, not interested in RPL/ZEN 4 anymore (a growing sentiment). Unless Zen 4 drops an X3D for a reasonable asking price + affordable boards, its looking more likely i'm gonna grab a 5800X3D