CPU and GPU Scaling with the Ryzen 7 5800X3D: 4 GPUs, 4 games, 3 resolutions

[Steve]

In a sea of next-gen CPUs, the Ryzen 7 5800X3D still stands tall. We benchmark it across 4 games, resolutions, and GPUs to see how it scales with today's top-tier and midrange graphics cards.

Read the full article here.

CPU and GPU Scaling with the Ryzen 7 5800X3D: 4 GPUs, 4 games, 3 resolutions

 

[Lounds]

I bought a Ryzen 7 5700X3D in Nov24. It's good to see that I can basically pair it up to a max of RTX 5080 level of performance before I'm CPU bottlenecking my GPU.
Thinking RX 9070 XT or 5070 Super (when they come out) as an upgrade over my 3060 Ti.

 

[ScottSoapbox]

This makes me so happy. I'd hoped my upgrade from 3900 to 5800x3D would hold me through my future planned 5080 upgrade and it will! (with no appreciable loss compare to the 9800x3D at my normal settings.)

Sadly, 5080s are still a ripoff so luckily it will also hold me through my future planned 5070 TI Super.

 

[scavengerspc]

I was hoping to see the old CPU\GPU articles come back for a while now. Nice to see this!

A blast from the past. Steve did this one 15 years ago:

https://www.techspot.com/review/294-singularity-performance/

 

[letsgoiowa]

Looks like I'm going to keep my 5700X3D until the 2nd or 3rd gen of Ryzen on DDR6. Using a launch day MSI X370 Gaming Pro Carbon board. Best hardware investment EVER. It'll be like 13 years old before I need to swap the motherboard.

 

[Chris Rchr]

In short, the 5800X3D, which is similar to the I7 12000 family and after, shows that all these CPUs from 3 years ago do a great job to the point of not limiting (or just a bit) large GPUs. The upgrade possibilities are excellent in my eyes.

 

[BogdanR]

“ Also, as a side note: if you're wondering why we didn't use the 9070 XT, don't worry. The RX 9070 can clearly exceed 446 fps, as seen in the 7600X and 9800X3D data. A faster Radeon GPU would not have changed the Ryzen 5 5600 results…”
…but might have gotten too close to the 5080’s results for comfort.

 

[PatchFixGaming]

Now this is epic & the kind of CPU+GPU scaling benchmarks I've always wanted,, even if we can 'extrapolate' single CPU or GPU focused benchmarks,, work like this really drives the message of CPU or GPU bound @ X res home for new users & paints a more complete picture.

Cheers & never stop doing comparisons like this! Greatly appreciated.

 

[Inthenstus]

There’s always something about these reviews that throws me off. If I’m playing an MMORPG, and am in the main city or a raid, playing at 4K, my GPU is capped at 100%, and according to reviews like this, my GPU is the bottleneck. But in these scenarios, having a faster CPU (like the 9800X3D) will almost always improve frame times, and provide a more playable experience than upgrading from a 4060 to a 5090. But going off this review alone, as a noob, I’d upgrade my GPU, and still have all the same stuttering I did before. Food for thought.

 

[b3rdm4n]

…but might have gotten too close to the 5080’s results for comfort.

Depends how anyone would personally interpret "too close". The 9070XT is some 12% faster than the 9070 at 4k, factoring that in, FPS in that 4 game average would go up from 61 to ~68.3, with the 5080 at 84 it's still 23% ahead.

No doubt the 9070XT is better value, and both can OC some extra 10% or so.

 

[BogdanR]

Depends how anyone would personally interpret "too close". The 9070XT is some 12% faster than the 9070 at 4k, factoring that in, FPS in that 4 game average would go up from 61 to ~68.3, with the 5080 at 84 it's still 23% ahead.

No doubt the 9070XT is better value, and both can OC some extra 10% or so.

Actually the 9070 is 12% slower than the 9070XT at 4k. And no, the reverse is not true, something which escapes a lot of people here, including the editors of this site.
Here’s why:
According to this site’s review of the 9070, at 4k the 9070XT averages 74 fps and the 9070 65 fps.
So, how much slower is the 9070 than the 9070XT?
Well, we need to use the 9070XT as a reference, so the performance of the 9070 is 65/74x100=87.838 % of the 9070XT so we can say it is 100-87.838=12.162 % slower.
Conversely to see how much faster than the 9070 the 9070XT is, the 9070 becomes the comparison reference therefore we calculate 74/65x100=113.846 %.
So we can say the 9070XT is 113.846-100=13.846 % faster.

Yes, the difference is rather small but it is there. Makes the 9070XT a slightly better value

 

[b3rdm4n]

Actually the 9070 is 12% slower than the 9070XT at 4k

I'm using this image for reference. Showing the 9070XT has a relative 112% of the performance of a 9070. This image shows that a 9070 has 89% of the relative performance of a 9070XT. Perhaps we are using different data?

I agree the distinction is important, and many people do get it wrong.

 

[Gezzer]

There’s always something about these reviews that throws me off. If I’m playing an MMORPG, and am in the main city or a raid, playing at 4K, my GPU is capped at 100%, and according to reviews like this, my GPU is the bottleneck. But in these scenarios, having a faster CPU (like the 9800X3D) will almost always improve frame times, and provide a more playable experience than upgrading from a 4060 to a 5090. But going off this review alone, as a noob, I’d upgrade my GPU, and still have all the same stuttering I did before. Food for thought.

MMORPGs are a special use case, they're worse then even your run of the mill PVP FPS. Why? Well a couple of reasons.

There is virtually no graphical exchange between a player's computer and the game server. The server is a massive database containing 3 dimensional values for every location in the game world. Then there's other server/s with location data for every moving element like NPC's and players, where they are, where they're heading, and their speed. As well servers also keep track of the interactions between well... virtually everything.

So there's a lot of data flowing between a player's computer and a the game server/s. And this puts a massive load on the CPU to begin with. But it gets worse the more players are in any one area. Because not only are the server/s keeping track of everything, they also have to supply each player with information about every other player. How well things work has a lot to do with the Net code. You might have <30ms latency, but other players might have 50, 75, or 100+ ms latency and that will affect how well the game server can refresh what you see.

So how does a better CPU help? Well there's more resources available for all the non-graphical workload and this can free up more resources for graphics. It's the main reason why the article was looking specifically at single player experiences. Even your run of the mill shooter has the same problems as a MMORPG. It's why most have a hard limit on player count. Something the MMO designation doesn't allow or it wouldn't be a MMO.

There's also one other factor that has a minor effect. There's something called host processing, which simply means that most on-board "hardware" like your sound, or network are only interfaces. All the actually processing is done by the "host" processor, your CPU. In most normal use cases your CPU doesn't even notice the extra load they add. But remember how I mentioned that a MMORPG has a really high network load? So your NIC is adding a abnormal load to the CPU and more CPU resources will help mitigate it.

A perfect example of this is Ironforge in vanilla WoW. It was the only auction house in Azeroth, and due to that there was a massive amount of players there at any one time. So much so that it created all sorts of havoc. I can remember running just outside the bank one second and finding myself in the moat between the bank and the AH the next. Blizzard did a lot of work on the net code, but what really helped was simply making more AH locations to reduce the player count. Later they upgraded to phasing, so that the load was split between servers.

 

[HallowFeind]

I still cant see these benchmarks as being right. The rainbow 6 at 1080p for example. The CPUs are all the same fps on the 9070 but when you look at the 5080 they step up, but the 5600 gets more fps with the 9070 by over 100 fps. It does not seem right that the 5600 has problems maxing out with a 5080 Seems strange. Someone explain why this happens. This does not add up.

 

[B Double]

This makes me so happy. I'd hoped my upgrade from 3900 to 5800x3D would hold me through my future planned 5080 upgrade and it will! (with no appreciable loss compare to the 9800x3D at my normal settings.)

Sadly, 5080s are still a ripoff so luckily it will also hold me through my future planned 5070 TI Super.

I'd compare 9070XT with the 5070 Ti super when it launches

 

[B Double]

I really love these comparisons between new and old. We have a variety of machines in our house and we've always tried to balance the GPU and CPU. These articles really help. Thankyou so much!

 

[DZillaXx]

I bought a Ryzen 1700x what feels like a decade ago. I replaced it with a 5700X3D from Aliexpress for around $150 shipped. It is crazy the Bios Support some AM4 boards had, my x370 keeps ticking away. GTX1080ti to my current AMD 6800XT.

Hard to justify the cost of a new PC build when even the old stuff just works fine. Games have not pushed the hardware other than in RT. And I'm really not sold on RT as of yet.

 

[ScottSoapbox]

I'd compare 9070XT with the 5070 Ti super when it launches

FSR 4 isn't in enough games and is still behind DLSS 4 upscaling.

I also just learned I can get a 5070 TI at MSRP locally while the 9070 XT is the same price which kills the value argument. I'm going to think about it, but I'm tempted to just get a 5070 TI now and then I can happily game while waiting for the next Gens (or two) to launch and take half a year to come back to MSRP.

 

[BogdanR]

I'm using this image for reference. Showing the 9070XT has a relative 112% of the performance of a 9070. This image shows that a 9070 has 89% of the relative performance of a 9070XT. Perhaps we are using different data?

I agree the distinction is important, and many people do get it wrong.

Thank you.
Back to my original statement that the 9070XT would be too close to the 5080 for comfort.
Indeed the 5080 ends up 22-23% ahead in performance but it is also at least 50-60% more expensive. From a gamer’s perspective that’s too close for comfort.

From a Canadian gamer’s perspective, considering the least costly versions of the cards I could actually walk in a store and buy, the difference is around 600 Canadian $, money which would be better spent on other components in your system.

 

[Nintenboy01]

I have a 5700X3D and will keep it until AM6/DDR6 is out

 

[Skiprudder]

Great article and review - thank you!

I happen to be running a 5600x at the moment, and I found this article very useful in leading me to move to an AM5 socket CPU. This may be due to differences in the Australian and US (where I am) retail markets, but I can't find a 5800x3d for less than $410 at the moment. The supply of new chips almost totally dried up here in the Fall of 2024. The used market isn't much better either.

On the other hand, I can pick up a 7600x for $160, or a full CPU+Mobo+16GB RAM combo for $270. I'll want some more RAM but even with that, I'm going to come in well under $410.

Maybe a caveat "if you can find a 5700x3d or 5800x3d for less than $XXX it's a great drop in upgrade for your AM4 machine."

 

[vaemarr]

You've just copied all of Hardware Unboxed's tests and the results match 100% of the gpus used, CPUs, the results. Everything.

Do your own testing bro.

 

[BogdanR]

You've just copied all of their tests and the results match 100% of the gpus used, CPUs, the results. Everything.

Do your own testing bro.

So, you didn’t notice they are the exact same people?

 

[b3rdm4n]

Indeed the 5080 ends up 22-23% ahead in performance but it is also at least 50-60% more expensive

That's why the 9070XT competes with a 5070ti. I agree the 5080 not worth the price difference (over a 5070ti or 9070XT) , but it sits in a performance teir above both, and AMD doesn't have a product there.

And plenty of gamers are still buying them.

 

[Huubster]

There’s always something about these reviews that throws me off. If I’m playing an MMORPG, and am in the main city or a raid, playing at 4K, my GPU is capped at 100%, and according to reviews like this, my GPU is the bottleneck. But in these scenarios, having a faster CPU (like the 9800X3D) will almost always improve frame times, and provide a more playable experience than upgrading from a 4060 to a 5090. But going off this review alone, as a noob, I’d upgrade my GPU, and still have all the same stuttering I did before. Food for thought.

Couldn't agree more, those FPS bars are just a small part of the story, actually less important than the impact of CPU on frametime and stutter. The impact of that is massively underrated, actually completely neglected in these kind of articles.

I made a whole journey, from 5800x, to 5800x3d, to 7800x3d, to 7950x3d to currently a 9950x3d, while using higher end GPU's. I sometimes hardly noticed any improvement in fps in games, but the gaming experience got waaaay better, the better the CPU.

Please take these FPS graphs with a grain of salt, 5800x3d is a nice CPU, but it's by far not an 9800x3d, not even close.