AMD Ryzen 7 5800X 3D Cache Eight Core 4.5GHz (Socket AM4) Processor - Retail - Go Go Go xD

[Richdog]

Surprisingly Anandtech did a CPU review of the 5800X3D!
(Long running joke especially on the AT forums as they seldom do any reviews anymore and tend to be very very late.)

The AMD Ryzen 7 5800X3D Review: 96 MB of L3 3D V-Cache Designed For Gamers

www.anandtech.com

And they had mins in their 4K part. The biggest gain was in GTA V but they tested that at 4K Low:

GTA has some similarities with Skyrim/FO4 in that it can be extremely CPU dependent but that is probably a bigger gain than we saw in CAT's FO4 benching thread (which is at 480P). While Ryzen 5800X3D does lead in the FO4 benches there (but nobody has submitted a 12900K yet), even with the 3D memory timings still seem to matter. Pity as the real appeal of 5800X3D vs 12700K/12900K is that it can be largely plug and play and mostly works quite well with even cheap PC3200 memory. (Well that, and Alder Lake can get really power hungry; still seems that Intel P cores are very much brute force - not that throwing transistors at a problem is necessarily bad until space and heat becomes a problem - which is mainly a problem for Intel with servers.)

Thanks, seems that there are no tests showing maxed out 4k settings but I guess this is pretty niche as far as overall gamers go and I doubt would show much improvement in CPU benchmarking. I will keep hunting Youtube as that's where the occasional amazing video can be found of someone playing games and testing minimum and average frames in real-time.

 

[KompuKare]

Yes if Intel could rival the efficiency of the 5800X3D they would have a beast with their high clocks. Seems like at the moment you can't have everything. I do like the low power draw and therefore heat output. I'm too old to be sweating playing PC games, I've got a rowing machine for that and trying to keep up with the dog

It's almost like Intel are caught in a bad space.
While there are lots of strategies to CPU design

1. Be super efficient in transistors and power. Non-custom ARMs are the leaders in that but Zen1/2/3 to some extend do this. The previous Intel CPUs until about Skylake had some of that (they had a motto similar to "only can only go up by x% if it gains *x% in performance").
2. Be extra wide but still be super power efficient. Apple's cores follow this idea
3. Be extra wide with power not that relevant. Some of IBM's big server cores do this (as it happens these are POWER architecture!).

Now I'd say 1 and 2 are the most popular trends.
With Alder Lake it seems that Intel went a bit with 2, but because they couldn't compete with their big cores in certain metrics (great IPC, but a core size which really isn't that efficient at higher clocks) and they were force to clock a bit like 3.
For servers I guess Intel intend to rely on the E cores but that is dangerous. Some loads love lots of slower cores (but ARM server chips are able to that far better), but many workloads do not.

AMD's 3D stacking is really more a 1. thing as it leads to chips which are power efficient but still fast (at loads where cache helps).

 

[Vulture]

Interesting review, 5800x-3d is excellent, beating or tied with 12900ks while consuming much much less power, and it works on 2017 boards, also you get max performance with cheap RAM, imagine what will Zen 4 with second generation of 3dcache do to Intel: https://www.youtube.com/watch?v=Md6YxfnhpAU&feature=emb_imp_woyt

 

[Troezar]

Interesting review, 5800x-3d is excellent, beating or tied with 12900ks while consuming much much less power, and it works on 2017 boards, also you get max performance with cheap RAM, imagine what will Zen 4 with second generation of 3dcache do to Intel: https://www.youtube.com/watch?v=Md6YxfnhpAU&feature=emb_imp_woyt

Will be interesting as I fully expect Intel to up their cache too.

 

[Cantankerous]

I'm also very interested with what amd do with the pricing of new chips considering how cheap some of the older ones are. Such as the normal 5800x.

 

[Troezar]

I'm also very interested with what amd do with the pricing of new chips considering how cheap some of the older ones are. Such as the normal 5800x.

If say the fastest 8 core will be similar to current 5800X3D pricing.

 

[kmax86]

What is the UK MSRP for the 5800X3D ? Is it £410 or £420 ?
Are there any hints going sub £400 in the next few months?
Looking to change from 3600 but the freezer 34 Duo that I have may not be enough to cool 5800X3D

Thanks

 

[Aegis]

Looking to change from 3600 but the freezer 34 Duo that I have may not be enough to cool 5800X3D

I'm using the Freezer 34 eSports (single fan) on mine and it's fine (low 80's running Cinebench R23, low-mid 60's during gaming) - I'm using the PBO2 Tuner tweaks linked in this thread.

 

[Richdog]

Just found this HWU comparison of the 5800x vs the 5800X3D, as I expected at 4k or higher the differences even to minimum 1% are negligible due to GPU limitations (below link goes directly to 4k summary).

 

[Troezar]

Just found this HWU comparison of the 5800x vs the 5800X3D, as I expected at 4k or higher the differences even to minimum 1% are negligible due to GPU limitations (below link goes directly to 4k summary).

Good just we're about to get new GPUs potentially twice as fast

 

[Poneros]

Just found this HWU comparison of the 5800x vs the 5800X3D, as I expected at 4k or higher the differences even to minimum 1% are negligible due to GPU limitations (below link goes directly to 4k summary).

HWU CPU testing is almost entirely worthless. They rely too much on built-in benchmarks which means in some cases it's actually really light on CPU compared to GPU (WD:L is the most egregious example), they don't do 720p to eliminate GPU bottlenecks, and they lower even CPU-stressing settings and they ignore RT (which hits CPU hard), so what's the point? If you want real results look at the germans, they're (sadly) the only ones who understand how to test.

AMD Ryzen 7 5800X3D im Gaming-Benchmark: Gaming-Benchmarks mit RTX 3090 Ti, Taktraten und Leistungsaufnahme

Gaming-Benchmarks: Gaming-Benchmarks mit RTX 3090 Ti, Taktraten und Leistungsaufnahme / Der Einfluss des Testparcours

www.computerbase.de

AMD Ryzen 7 5800X3D im Test: Mit 3D-V-Cache und hoher Effizienz gegen Intel

Auf dieser Seite widmen wir uns den Benchmarks zum Ryzen 7 5800X3D. Die CPU durchläuft dabei unseren kompletten Testparcours mit 14 Spielen und 7 Anwendungsbenchmarks.

www.pcgameshardware.de

 

[Richdog]

HWU CPU testing is almost entirely worthless. They rely too much on built-in benchmarks which means in some cases it's actually really light on CPU compared to GPU (WD:L is the most egregious example), they don't do 720p to eliminate GPU bottlenecks, and they lower even CPU-stressing settings and they ignore RT (which hits CPU hard), so what's the point? If you want real results look at the germans, they're (sadly) the only ones who understand how to test.

AMD Ryzen 7 5800X3D im Gaming-Benchmark: Gaming-Benchmarks mit RTX 3090 Ti, Taktraten und Leistungsaufnahme

Gaming-Benchmarks: Gaming-Benchmarks mit RTX 3090 Ti, Taktraten und Leistungsaufnahme / Der Einfluss des Testparcours

www.computerbase.de

AMD Ryzen 7 5800X3D im Test: Mit 3D-V-Cache und hoher Effizienz gegen Intel

Auf dieser Seite widmen wir uns den Benchmarks zum Ryzen 7 5800X3D. Die CPU durchläuft dabei unseren kompletten Testparcours mit 14 Spielen und 7 Anwendungsbenchmarks.

www.pcgameshardware.de

Computerbase and PCgameshardware.de tested games at 720p, sorry if I missing something but how do I translate that to how it would practically perform at 4k?

 

[Troezar]

Computerbase and PCgameshardware.de tested games at 720p, sorry if I missing something but how do I translate that to how it would practically perform at 4k?

One way of testing will show what it will do right now at 4K the other shows the total potential of the CPU, very useful if you run multiple generations of GPU on the same CPU. I bought the 5800X3D as it showed the potential to scale with future GPU's meaning I shouldn't have to upgrade now for a while, just drop in a new GPU.

 

[Poneros]

Computerbase and PCgameshardware.de tested games at 720p, sorry if I missing something but how do I translate that to how it would practically perform at 4k?

You would look at GPU data, not CPU. The point of a CPU review is to find out how the CPU performs. But you can find UHD results with ComputerBase as well, it just wouldn't be very useful.

AMD Ryzen 7 5800X3D im Test: Benchmarks in Anwendungen und Spielen

AMD Ryzen 7 5800X3D im Test: Benchmarks in Anwendungen und Spielen / Der Einfluss des Testparcours / Das Leistungsrating lässt sich anpassen

www.computerbase.de

 

[humbug]

One way of testing will show what it will do right now at 4K the other shows the total potential of the CPU, very useful if you run multiple generations of GPU on the same CPU. I bought the 5800X3D as it showed the potential to scale with future GPU's meaning I shouldn't have to upgrade now for a while, just drop in a new GPU.

No no no..... you don't need to know about any of that, you're supposed to buy a new CPU every generation.

 

[humbug]

You see what you're supposed to do is but an i5 or Ryzen 5 now, preferably the cheapest one, when you upgrade your GPU you buy another one, and another one if you upgrade again.

You see, to buy a good CPU now so you can put the money you saved by not buying a new CPU every generation towards a better GPU next time round is a really stupid idea, so you don't need to know if the CPU can handle more than the GPU we are testing it with, just keep watching our content, we will tell you when the CPU we advised you to get needs replacing, we got you fam! Like and Subscribe.

 

[Richdog]

You would look at GPU data, not CPU. The point of a CPU review is to find out how the CPU performs.

Yes thanks for clarifying that a CPU review is about CPU performance, that is incredibly helpful. However, the info I am looking for is what, if any, impact that a 5800X3D has on gaming at UHD resolutions and perhaps unsurprisingly I already know what impact my 3090 GPU has on gaming on UHD resolutions as I used it regularly.

I assume that HWU tested the 5800X3D at all resolutions and not just 720p because gamers who game at 1080p, 1440p and 4k are also interested in seeing how a 5800X3D affects min and max framerate at the practical resolutions they will actually be playing at.

 

[COD4000]

Problem is they test a lot of Single Player games. I want to see the difference in minimums in CPU heavy multiplayer games like Planetside 2, BF2042 etc.

 

[Yaayuh!]

Problem is they test a lot of Single Player games. I want to see the difference in minimums in CPU heavy multiplayer games like Planetside 2, BF2042 etc.

Is Planetside 2 still popular? It's like 10 years old. I would have thought people would have ditched it for the modern f2p **** box FPS games.

 

[humbug]

Is Planetside 2 still popular? It's like 10 years old. I would have thought people would have ditched it for the modern f2p **** box FPS games.

Its popular with kids who have access to daddies credit card, all it is now is a bunch of invisible children running round shooting at unsuspecting free to play cannon fodder.

A once brilliant game utterly ruined by the pursuit of $$$$$$ from 12 years old's who want to feel like they have skill. Basically they are selling the worst kinds of cheats as upgrades.