Is Zen 3 Worth It for Gaming? Ryzen 5600X vs. 3600 vs. Core i5-10400F

[Tantor]

maybe wait for the 5700 (non-x). who knows, maybe it will be good
I would try to get at least the same number of cores if possible.

Thanks for the response.

I'm considering getting a 3900x, it should run ok on my AB350 mobo. It's cheaper than a 5800x. It's slower in games, but that won't matter because I game at 4k. It's faster in rendering, and I do some light video rendering for youtube. And I save $100.

It's amazing that my old Gigabyte AB350 Gaming 3 board is still going strong. Everytime I look into upgrading, I end up asking myself ...why?

 

[Irata]

According to these benchmarks:
https://cpu.userbenchmark.com/Compare/AMD-Ryzen-5-3600-vs-Intel-Core-i7-4790K/4040vs2384
and
https://www.gpucheck.com/compare-ga...tel-core-i7-4790k-4-00ghz-vs-amd-ryzen-5-3600

Ryzen 3600 is only slightly better than your current CPU. It's much better in apps that use 6 or more cores, but in gaming that's no so important.

Really, you use Userbenchmark and a test with RDR2 on a 1660 ?

To quote from Techspot‘s RDR2 GPU benchmarking using a 9900k:

We get a real sense of just how demanding this game is when looking at the GTX 1080 and 1660 Ti, both struggling to push past 60 fps -- 60 fps at 1080p

Red Dead Redemption 2 PC Graphics Benchmark

Launched on PC a year after it debuted on consoles, Red Dead Redemption 2 is still a big release that deserves a detailed benchmark analysis, much like...

www.techspot.com

 

[Shadowboxer]

Really, you use Userbenchmark and a test with RDR2 on a 1660 ?

To quote from Techspot‘s RDR2 GPU benchmarking using a 9900k:

Red Dead Redemption 2 PC Graphics Benchmark

Launched on PC a year after it debuted on consoles, Red Dead Redemption 2 is still a big release that deserves a detailed benchmark analysis, much like...

www.techspot.com

Techspots findings in that test are way off. For a start they DX12 when GeForce performs much better on Vulkan. My RTX 2080 does significantly better than it does in that test so I’m guessing there was either been a big driver improvement since then or some kind of testing error when Techspot carried out those benchmarks.

 

[Irata]

Techspots findings in that test are way off. For a start they DX12 when GeForce performs much better on Vulkan. My RTX 2080 does significantly better than it does in that test so I’m guessing there was either been a big driver improvement since then or some kind of testing error when Techspot carried out those benchmarks.

The point is that he used both Userbenchmark and RDR2 that‘s GPU limited with a 1660 (plain vanilla, not Ti or Super) even at 1080p to support his statement.

Honestly, Userbenchmark is already a big red flag, but the second link just added a few more, making the statement misleading at best.

 

[Shadowboxer]

The point is that he used both Userbenchmark and RDR2 that‘s GPU limited with a 1660 (plain vanilla, not Ti or Super) even at 1080p to support his statement.

Honestly, Userbenchmark is already a big red flag, but the second link just added a few more, making the statement misleading at best.

Oh I don’t care about that. But the data from RDR2 is bad. I get better on epic settings than those graphs show I should have at high.

It’s quite misleading as it shows a 5700XT matching a 2080. But the 2080 was artificially gimped by DX12 when Radeons run better on DX12. The article also recommends DX12 but if you go on Reddit everyone will say avoid it like the plague if you’re on GeForce.

But meh, it was a while ago now I’m over it. Maybe sheds some light on Nvidias frustration with Steve?

 

[Irata]

Oh I don’t care about that. But the data from RDR2 is bad. I get better on epic settings than those graphs show I should have at high.

It’s quite misleading as it shows a 5700XT matching a 2080. But the 2080 was artificially gimped by DX12 when Radeons run better on DX12. The article also recommends DX12 but if you go on Reddit everyone will say avoid it like the plague if you’re on GeForce.

But meh, it was a while ago now I’m over it. Maybe sheds some light on Nvidias frustration with Steve?

What does any of this have to do with the original statement that a 4790k is basically equal to a 3600 in gaming ?

 

[Steve]

This article is misleading! By limiting the GPU, the benchmark doesn't show the true difference between the CPUs. Notice how so many have the same fps numbers - that's the GPU bottleneck- not the CPU. If you use a 3090 rtx, the graphs will show that the 5600x is faster than the 10900k in more games than not. Is Techspot being biased to nurf 5600x artificially? Check out the benchmarks on more reputable tech sites

(loud noises)

 

[CybaGirl]

You'd also get access to NVMe not choked over a 4 lane DMI connection and PCI-E 4.0, for example. So it would be a meaningful upgrade, but up to you if you think it's worth the dollars.

Thank you for taking the time to reply to my question. Much appreciated .

I keep forgetting about the PCI-E 4.0 and NVMe. I currently have a Sasmung EVO 860 256GB mSATA drive as my boot drive as well as other drives for storage. I mean my system is very quick and I seem to be able to play all games maxed out just that I would like that additional 16GB of RAM which seems to be difficult to find.

I also won a GALAX 3060Ti on Christmas Eve which will be replacing my iGame Vulcan X OC 1070.

So I am unsure if the 4790K will be a good match with this card either.

Thanks.

 

[CybaGirl]

Ryzen 3600 is only slightly better than your current CPU. It's much better in apps that use 6 or more cores, but in gaming that's no so important.

Thank you for taking the time to reply to my question. Much appreciated .

Thank you for the links and I see what you mean. So it's hard for me to justify spending all that money for such a little gain I guess.

Just would be nice if I could find another 16GB of RAM the same as what I already have.

Prices seem to have skyrocketed since COVID-19 and it's so expensive even thinking of upgrading let alone buying a whole new system.

 

[Bamda]

I built my GF a new computer using my old GTX 1080 Ti and a AMD Ryzen 5 3600, and she loves the new computer. I will upgrade the GPU when more product becomes available.

 

[Bamda]

I do a gaming but I also do a lot of photo processing and actually run simulations using Matlab, Mathematica, Comsol, Ansys etc, so actually the Zen3 CPU's are a better choice for me, but I can't get one or if you can it's at a 300% mark-up. I may have to settle for a Zen 2 for now. I only game at 1440p so it doesn't really matter in that regard but I tend to keep a fresh build for at least 5 years so Zen 3 would future proof me a lot more for non-gaming.

What I did was build a system using a Zen 2 CPU (Ryzen 5 3600) and used my current GPU. Then I can replace the CPU and GPU as stock becomes available. You can then sell or repurpose the used parts.

 

[Irata]

Thank you for taking the time to reply to my question. Much appreciated .

Thank you for the links and I see what you mean. So it's hard for me to justify spending all that money for such a little gain I guess.

Just would be nice if I could find another 16GB of RAM the same as what I already have.

Prices seem to have skyrocketed since COVID-19 and it's so expensive even thinking of upgrading let alone buying a whole new system.

You can basically ignore the two links that were posted.
Userbenchmark is widely discredited and the second link showed a GPU limited game, I.e. performance is limited by the graphics card. In this case a faster CPU will not make much of a difference.

Since you now have a 3060Ti being GPU limited should not be an issue.

That said, if your PC still does what you want it to do, there‘s really little reason to upgrade. So if the games you enjoy all run fine, you don‘t need to shut down other apps while gaming and am still happy with how the system performs, save your money.

In the end it comes down to your individual use case, which games you play and if / how you multitask.

-> As an example, there are games that might stutter on a 4C CPU but if they are not games you play, it‘s not a problem.

 

[mAdmAnDingo]

This article is misleading! By limiting the GPU, the benchmark doesn't show the true difference between the CPUs. Notice how so many have the same fps numbers - that's the GPU bottleneck- not the CPU. If you use a 3090 rtx, the graphs will show that the 5600x is faster than the 10900k in more games than not. Is Techspot being biased to nurf 5600x artificially? Check out the benchmarks on more reputable tech sites

You completely missed the point of this article, and you went ahead and made a fool of yourself, by insulting Steve for no reason thanks to your ignorance. Steve has already done an 5600X review (with an 3090), you were just too lazy to look for it.
https://www.techspot.com/review/2135-amd-ryzen-5600x/

So instead you call it misleading. Next time, do your own research, prior to throwing ignorant accusations around.

 

[Irata]

You completely missed the point of this article, and you went ahead and made a fool of yourself, by insulting Steve for no reason thanks to your ignorance. Steve has already done an 5600X review (with an 3090), you were just too lazy to look for it.
https://www.techspot.com/review/2135-amd-ryzen-5600x/

So instead you call it misleading. Next time, do your own research, prior to throwing ignorant accusations around.

And I‘d add how realistic is the 10400F / Ryzen 3600 + RTX 3090 scenario ?

I‘d assume that if you spend that much on a GPU (and probably PSU and mainboard), getting a better SKU CPU isn‘t an issue, I.e. those aren‘t the customers thinking of getting a cheaper CPU alternative to save money.

 

[Zooropa93]

Why not include a 10600K in the test? That's more of a competitor to a 5600X than the 10400F and is still less expensive than a 5600X.

 

[Shadowboxer]

Why not include a 10600K in the test? That's more of a competitor to a 5600X than the 10400F and is still less expensive than a 5600X.

That would defeat the point of the test. The idea is to find out if you really need 6 really fast cores like the ones on zen 3 compared to a budget 6 core like the 10400F to play games on a modern GPU. It’s not a test to see which is faster, we already know the 5600X is faster than the 10600K.

 

[Irata]

Why not include a 10600K in the test? That's more of a competitor to a 5600X than the 10400F and is still less expensive than a 5600X.

That would probably be a bit pointless - it‘s a budget vs. premium comparison and the 10600K is neither.

Once you add the cost of an HSF and a good Z490 mainboard, you might as well go for a 5600X. The 10400 and 3600 are noticeably cheaper alternatives otoh.

 

[TheBigT42]

All I saw was a bunch of apples and oranges when comparing mid 5000 CPUs with top 3000 CPUs The Intel Bananas don't even belong in the comparison if you are looking to see if it is a good idea to upgrade from Zen2 to Zen3.

Try comparing apples and apples:
3950x - 5950x
3900x - 5900x

 

[poohbear]

I don't think there's any point discussing someone upgrading from a CPU that's 2-3 old. That would be a completely pointless upgrade. Talk to the people who have CPUs from 5-6 years ago, as they're the ones that'll be upgrading and deciding between a 3600 vs 5600.

 

[flee2020]

I think that we should turn gaming reviews around - we should first select the GPU and then test it using the various CPUs. We can then determine which is the fastest CPU that need to be used before the GPU becomes the bottleneck and using a higher end CPU will not benefit the gaming. It may be better to spend more money on a better GPU than wasting it on a more expensive CPU.

 

[BBoogey]

Not need change 3600 only few game give 10+fps