Tuesday, April 12th 2022

AMD Ryzen 7 5800X3D Gets Full Set of Gaming Benchmarks Ahead of Launch

XanxoGaming has finally released its complete set of benchmarks for the AMD Ryzen 7 5800X3D and it's been tested against an Intel Core i9-12900KF. This time both platforms are tested using an NVIDIA GeForce RTX 3080 Ti and four times 8 GB of 3200 MHz CL14 DDR4 memory. The only difference appears to be the OS drive, motherboard and cooling, although both systems rely on a 360 mm AIO cooler. Both systems were running Windows 10 21H2. The site has a full breakdown of the components used for those interested in the exact details.

The two platforms were tested in 11 different games at 720p and 1080p. To spoil the excitement, it's a dead race between the two CPUs in most games at 1080p, with Intel being ahead by about 1-3 FPS in the games where AMD loses out. However, in the games AMD takes the lead, it's by a good 10 FPS or more, with games like the Witcher 3 and Final Fantasy XV seeing an advantage of 40-50 FPS. AMD often has an advantage when it comes to the one percent low numbers, even when Intel is ahead when it comes to the average FPS, but this doesn't apply to all of the games. It's worth keeping in mind that the Intel CPU should gain extra performance when paired with DDR5 memory in some of these games, but we'll have to wait for more reviews to see by how much. The benchmarks displayed are mostly the games TPU normally tests with, but aren't the entirety of games tested by XanxoGaming.
As for the 720p tests, AMD only loses out in Strange Brigade, even though it's a loss of over 20 FPS on average FPS and by over 10 FPS when it comes to the one percent low frames. As for the other games, it's mostly a dead race here too, but with an advantage to AMD instead of Intel by 1-3 FPS. However, the 3D V-Cache seems to kick in here when it comes to the one percent low frames, as AMD edges out Intel by a large margin in more games here by at least 10 FPS, often by around 30 FPS or more. Take these benchmarks for what they are, an early, unconfirmed test of the Ryzen 7 5800X3D. We're just over a week away from the launch and we should be seeing a lot more benchmarks by then. Head over to XanxoGaming for the full set of tests and their conclusion, especially as they made an effort to write the test in English this time around.
Source: XanxoGaming
Add your own comment

139 Comments on AMD Ryzen 7 5800X3D Gets Full Set of Gaming Benchmarks Ahead of Launch

#126
Unregistered
chrcolukSeen this image doing the rounds.

Gonna test my 12700k while gaming later, have got farcry 6 and dying light 2 to compare.
#127
fevgatos
chrcolukSeen this image doing the rounds.

I can confirm the cyberpunk numbers, on 12900k it draws a LOT of power (ive seen it peak at 170 watts, lol). The rest seem questionable though, the peak numbers sure, but the averages seem way too high for the 12900k.

Regardless, the 3d consumption is impressive.
Posted on Reply
#128
ARF
chrcolukSeen this image doing the rounds.

The new 5800X3D shows lower numbers - I guess improved and mature N7 process, better binning and lower operating frequencies.
Posted on Reply
#129
Makaveli
ARFThe new 5800X3D shows lower numbers - I guess improved and mature N7 process, better binning and lower operating frequencies.
My guess is lower frequencies and the cache. With 96mb there is less reason to make trips out to main memory which lowers power consumption.
Posted on Reply
#130
ARF
MakaveliMy guess is lower frequencies and the cache. With 96mb there is less reason to make trips out to main main memory which lowers power consumption.
How much less are actually these "trips"?
With small additional L3 cache vs without it?
Posted on Reply
#131
Chrispy_
InVasManiI get what you're saying, but at the same time if the additional cache results in it never having the extremes of a heavy vehicle driving over it then it does in fact provide additional stability does it not!!? It's not so much a case of can you add more weight until the bridge collapses, but can you add more weight without it collapsing.
I guess potentially an unstable config will be less crashy?

IME the grey area between 'rock-stable' and 'doesn't even boot' with Zen3 RAM timings is quiet small. If it passes 60 seconds of MemTest86 I'm good with it, that's likely to result in less than one crash a month in real-world use.
Posted on Reply
#132
Aquinus
Resident Wat-man
MakaveliMy guess is lower frequencies and the cache. With 96mb there is less reason to make trips out to main main memory which lowers power consumption.
ARFHow much less are actually these "trips"?
With small additional L3 cache vs without it?
The CPU spends less time stalled due to a cache miss with a larger cache. It's basically fewer wasted CPU cycles waiting on system memory depending on the situation. How much different sized caches will help depends on the workload, the size of the dataset, the access patterns of said dataset, and the latency+bandwidth of said cache. Even that is an overly simplistic view because you have to consider any other tasks going on in a CPU at any given time.
Posted on Reply
#133
DeathtoGnomes
Hardware Unboxed does a 40 game review. 5800X3D vs 12900K

Posted on Reply
#134
GURU7OF9
DeathtoGnomesHardware Unboxed does a 40 game review. 5800X3D vs 12900K

It appears the extra cache has a big impact in a lot of games! If the price is not much over the 5800x then it would be well worth it for current ryzen owners with older cpus.
I could wear a slight performance loss in ms office and other programs etc if my games run faster . Probly want to check if most of the games you play would benefit from the extra 3d v cache first but !
Posted on Reply
#135
DeathtoGnomes
GURU7OF9It appears the extra cache has a big impact in a lot of games! If the price is not much over the 5800x then it would be well worth it for current ryzen owners with older cpus.
I could wear a slight performance loss in ms office and other programs etc if my games run faster . Probly want to check if most of the games you play would benefit from the extra 3d v cache first but !
I'm sure eventually there will be a database somewhere of affected games. More importantly, that impact will also mean gamers may hold onto the older builds longer. Of course hard core gamers will buy into Zen4 ASAP.
Posted on Reply
#136
Icon Charlie
DeathtoGnomesI'm sure eventually there will be a database somewhere of affected games. More importantly, that impact will also mean gamers may hold onto the older builds longer. Of course hard core gamers will buy into Zen4 ASAP.
Not at that price. You might as well grab a 5900X as the pricing on Newegg for today is $399.99 The CPU pricing continue to decline. AMD made a BIG mistake on launching this so late. But of course people will buy this and spend the money on it.
Posted on Reply
#137
DeathtoGnomes
Icon CharlieNot at that price. You might as well grab
Dont underestimate the size of the wallet of a hardcore gamer. If its faster, has more memory, and delivers higher 1% frames, they want it. Competitive gaming especially.
Posted on Reply
#138
GURU7OF9
DeathtoGnomesDont underestimate the size of the wallet of a hardcore gamer. If its faster, has more memory, and delivers higher 1% frames, they want it. Competitive gaming especially.
I guess they could sell a kidney if they get really desperate! Or better still, some one else's !
Unlimited supply of upgrades .....
Posted on Reply
#139
medi01
THU31But after several years the same games will become CPU-bound, when your new graphics card is twice as powerful or faster.
And it would not matter, even if it happens.
If FPS matters, it matters today too (competitive shooting).
Posted on Reply
Add your own comment
Apr 25th, 2024 20:06 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts