AMD Ryzen 9 7900X3D Linux Gaming Performance

[Keats]

Game world time should not materially change between supported CPUs as that's a function of an algorithmic governor in the program. The governors are there to make games playable for humans because CPUs can make decisions far faster than either the screen can display and the human player can respond. For one known example try The Ur-Quan Masters and turn Super Melee's time governor off. Even on an ancient 486 machine turning off the time governor resulted in a game no human could play, but it was fun watching the computer fight itself.

I understand what you're getting at. Some games like the X series and Civilization, etc have a computer AI in which decisions are made either in the background (as in X4, RTSes) or at specific turns (Civilization, Gal Civ, MOO). I just don't see how you can automatically measure that kind of thing. You'd have to play the game or connect it to a test harness. Decision trees are going to vary based on the situation presented to the game's AI at any given time or turn. How do you know if a given problem is taking too much time because of a bug, or the decision tree is validly complex? Most people aren't going to have the skills or patience to map out the logic flow in those systems to know if a benchmark is a valid result or not even if the code is available. They'll just make assumptions that may or may not be valid. You may not even be benchmarking the CPU itself but some logic decision that depends on other code paths before it finishes.

Paradox grand strategy games have variable time settings, including "Go as fast as you can" setting (this is very useful especially for experienced players because a lot of the time playing is spent waiting for the next important thing to happen - when something important does happen, you can either reduce speed or even just pause the game to deal with it). But you're right in that benchmarking those is pretty tricky because Paradox doesn't include built-in benchmarking tools and there's a lot of randomness included, which can impact the results.

[microcode]

It'd be interesting to see a 3DV$ mobile part, given the power savings of not going to main memory. The reason the peak performance is sometimes lower is lower clocks, but mobile parts aren't bothering to hit those clocks anyway so it'd be moot.

Seems like mild heterogeneous multiprocessing is becoming the norm in the industry; whether it's different cores or different memory architecture on differently clocked similar cores; so I wonder who is responsible for the big ideas and execution in the scheduler to make these things really shine...

EAS (Energy Aware Scheduler) and QHMP (Qualcomm Heterogeneous MultiProcessing) have been around for a while for big.LITTLE type systems.

[Weasel]

you attack the userbase as brainless and clueless and thats right apple customers are complete garbage people.

but that has nothing to do with the tech engineers at apple.

i would say apple M1/M2 are technically speaking successfull products

they are successful in terms of tech...

They're decent products. Nothing more. Way overpriced though.

[Weasel]

...and a 7700X is probably fine for those things as well in 99% of cases unless you're just compiling or running 3d renders all the time. Faster clock beats more cores unless you're able to hammer every core and 8 cores is plenty for most multi-tasking.

Don't get me wrong. It's a nice processor, it is more power efficient (slightly), but if you're watching your dollars you're most likely better off with the cheaper option for similar performance in many / most scenarios.

Depends if you like to multi-task or not. While most games typically don't make use of more than a couple cores (this is starting to change though), you could be doing something else in the background and that changes things.

[pal666]

...and a 7700X is probably fine for those things as well in 99% of cases unless you're just compiling or running 3d renders all the time. Faster clock beats more cores unless you're able to hammer every core and 8 cores is plenty for most multi-tasking.

7950x has faster clock than 7700x. so what's wrong with 7950x in all those tests?

[qarium]

They're decent products. Nothing more. Way overpriced though.

but you have to agree that the technical enginners do not set the price right?

its the money people who want this to be overpriced not the technicians.

[yump]

The 7950X system is clearly misconfigured or mislabeled. In some tests it's even behind the 5950X:

[Xnor0]

How come 7950x is behind both 7700x and 7900x? If 7700x leads that should be the lack of inter CCD latency right? But then 7900x would suffer just as much as 7950x from the this scheduling issue

[Linuxxx]

The 7950X system is clearly misconfigured or mislabeled. In some tests it's even behind the 5950X:

Mislabeling is out of the question, because it's all done automatically by the PTS.

Misconfiguration depends on the possible root cause:

- Inadequate cooling?
- AMD-PSTATE CPU driver causing problems?
- "irqbalance" daemon screwing up the interrupt handling?

Really hard to tell with many possible variables, unfortunately...

[jaxa]

It could have been some weird dual CCD bug as has been seen before, but that doesn't explain why it doesn't affect 7900X, etc.