Are FPS benchmarks for Games,mostly GPUs false?

[DismissedFetus]

I have:

RTX 2060

i7-8700

16 GB RAM DDR4

 

The issue is most benchmarks done by third-parties specify that I'll get 30-40 FPS on the title Metro:Exodus at extreme graphics on 1080p resolution.

I downloaded Afterburner to check that for myself.I turned ALL the settings including ray-tracing to ultra and got stable 80 FPS.I was honestly was scared to crank to extreme before as thought it would crash.

The source I took the benchmark from used i7-5820K while I don't think this matters as most benchmarks display the same amount (being lower than what the RTX 2060 is capable of)

Could someone shed some light to this?

 

 

Source:

https://www.game-debate.com/news/26614/metro-exodus-geforce-rtx-2060-and-gtx-1070-benchmarks

[Fasauceome]

If you're not using the same system with the same exact settings, the results will be amiss. In a case like this, I would think anti aliasing of some kind was to blame, as that usually makes for a pretty large performance hit. What do you have to do to get their results? 

[porina]

Assuming you got the same settings applied, that still leaves possible differences in game patch level, and GPU driver. They aren't static and will be updated over time.

[Constantin]

Use Afterburner 

[DismissedFetus]

1 hour ago, Constantin said:

Use Afterburner 

I have as that was my FPS counter.

 

[DismissedFetus]

1 hour ago, porina said:

Assuming you got the same settings applied, that still leaves possible differences in game patch level, and GPU driver. They aren't static and will be updated over time.

I am definitely not the best at these kind of thing but I belive 50 FPS difference is too big of a gap to take this as the case.

[jones177]

A i7 5820k has 6 weak cores.

A i7 8700k has 6 strong cores.

 

For example a i7 5820k has a Cinebench single core score of 139 or lower than a R5 1400. An i7 8700k has a score of 208. Higher than a R7 3700X.

 

This is a 1080p CPU usage test I did with Metro E. This is at low settings. The game is using all the cores so the the old 6 core would be incredibly weak in this title compared to your CPU.

Test is with a i7 8086k at 5ghz and a 2080 ti.

 

 

 

[Sakkura]

You get 80FPS with everything ultra. Techspot got 80FPS at ultra.

 

https://www.techspot.com/review/1795-metro-exodus-benchmarks/

[Zando_]

Just now, jones177 said:

A i7 5820k has 6 weak cores.

A i7 8700k has 6 strong cores.

 

For example a i7 5820k has a Cinebench single core score of 139 or lower than a R5 1400. An i7 8700k has a score of 208. Higher than a R7 3700X.

*angery 5820K user noises*

You talking stock or OCed benches? 5820K is a solid meh at stock, OCed it's a great CPU. The test the OP mentioned annoys me, they didn't say what the CPU was clocked at. At stock clocks it's not gonna push much in modern titles, at 4.2-4.5GHz I push 100+ fps in most titles fine. Still need to test my really hungry bois like Assassin's Creed Odyssey though. 

OP, if they're running a stock 5820K for those tests then no wonder fps is so low, that CPU chokes on most newer stuff at stock clocks. Sadly I don't have a 2060 or Metro to test, but it would be interesting to see how my 4.5Ghz 5820K compares to the 8700 (IIRC those boost to up to 4.7 or so and have a better IPC). 

[DismissedFetus]

1 hour ago, jones177 said:

A i7 5820k has 6 weak cores.

A i7 8700k has 6 strong cores.

 

For example a i7 5820k has a Cinebench single core score of 139 or lower than a R5 1400. An i7 8700k has a score of 208. Higher than a R7 3700X.

 

This is a 1080p CPU usage test I did with Metro E. This is at low settings. The game is using all the cores so the the old 6 core would be incredibly weak in this title compared to your CPU.

Test is with a i7 8086k at 5ghz and a 2080 ti.

 

 

 

I belive I can take this as I a good answer but I searched a little more and found a benchmark with I9-9900K and the issue was the same as they put it on Ultra rather than Extreme and I belive the Ray-tracing was turned off as they have non-RTX GPU's and they get 80 FPS while I get 120-130 FPS.

[DismissedFetus]

1 hour ago, Sakkura said:

You get 80FPS with everything ultra. Techspot got 80FPS at ultra.

 

https://www.techspot.com/review/1795-metro-exodus-benchmarks/

I get that 80 with Extreme and Ray-tracing.

[DismissedFetus]

1 hour ago, Zando Bob said:

*angery 5820K user noises*

You talking stock or OCed benches? 5820K is a solid meh at stock, OCed it's a great CPU. The test the OP mentioned annoys me, they didn't say what the CPU was clocked at. At stock clocks it's not gonna push much in modern titles, at 4.2-4.5GHz I push 100+ fps in most titles fine. Still need to test my really hungry bois like Assassin's Creed Odyssey though. 

OP, if they're running a stock 5820K for those tests then no wonder fps is so low, that CPU chokes on most newer stuff at stock clocks. Sadly I don't have a 2060 or Metro to test, but it would be interesting to see how my 4.5Ghz 5820K compares to the 8700 (IIRC those boost to up to 4.7 or so and have a better IPC). 

Well as another user replied mentioning Techspot they used I9-9900K which is certainly much better than my I7-8700.

 

[DismissedFetus]

1 hour ago, Fasauceome said:

If you're not using the same system with the same exact settings, the results will be amiss. In a case like this, I would think anti aliasing of some kind was to blame, as that usually makes for a pretty large performance hit. What do you have to do to get their results? 

Techspot is again the example.While I used less powerfull CPU and put the max avaible on impactful  settings I belive I 80FPS that I got equaling theirs is problematic.

 

[Zando_]

1 minute ago, DismissedFetus said:

Well as another user replied mentioning Techspot they used I9-9900K which is certainly much better than my I7-8700.

 

If it's at stock then it has an extra 2c/4t, all core turbo is 4.7Ghz (your 8700's all core turbo is 4.3Ghz). So 400Mhz, a noticeable difference but not too massive, the 8700/8700K (same all core turbo at stock) is still no slouch in games. 

2 minutes ago, DismissedFetus said:

Techspot is again the example.While I used less powerfull CPU and put the max avaible on impactful  settings I belive I 80FPS that I got equaling theirs is problematic.

How so? 

[jones177]

9 minutes ago, DismissedFetus said:

Well as another user replied mentioning Techspot they used I9-9900K which is certainly much better than my I7-8700.

 

I have a i9 9900k overclocked to 5ghz on all cores in the house and it does not beat my i7 8086k at 5ghz on all cores in games. If it did I would upgrade. 

 

Put an all core 5ghz overclock on your 8700k and you will have i9 9900k performance in games for now.

 

[Zando_]

2 minutes ago, jones177 said:

I have a i9 9900k overclocked to 5ghz on all cores in the house and it does not beat my i7 8086k at 5ghz on all cores in games. If it did I would upgrade. 

 

Put an all core 5ghz overclock on your 8700k and you will have i9 9900k performance in games for now.

OP said he has an 8700 non-K, so best it will do is stock 8700K perf (same all core turbo). But yeah, I don't think many games take proper advantage of the extra cores on the 9900K right now, so in a lot the 6c CPUs will still compete. 

[5x5]

1 hour ago, DismissedFetus said:

I have:

RTX 2060

i7-8700

16 GB RAM DDR4

 

The issue is most benchmarks done by third-parties specify that I'll get 30-40 FPS on the title Metro:Exodus at extreme graphics on 1080p resolution.

I downloaded Afterburner to check that for myself.I turned ALL the settings including ray-tracing to ultra and got stable 80 FPS.I was honestly was scared to crank to extreme before as thought it would crash.

The source I took the benchmark from used i7-5820K while I don't think this matters as most benchmarks display the same amount (being lower than what the RTX 2060 is capable of)

Could someone shed some light to this?

 

 

Source:

https://www.game-debate.com/news/26614/metro-exodus-geforce-rtx-2060-and-gtx-1070-benchmarks

Your issue is using gamedebate. It's known for imaginary false numbers. Same with userbenchmark, gpusboss, cpuboaa and the tons of random YouTube videos with imaginary performance. It's all faker than Kim Kardashian

[DismissedFetus]

 

1 hour ago, Zando Bob said:

OP said he has an 8700 non-K, so best it will do is stock 8700K perf (same all core turbo). But yeah, I don't think many games take proper advantage of the extra cores on the 9900K right now, so in a lot the 6c CPUs will still compete. 

You took the words out of my mouth in the first part but still about 6 cores:

6 cores is great for games and I belive it still beats the stock 8700 I have without overclock.

 

[DismissedFetus]

1 hour ago, 5x5 said:

Your issue is using gamedebate. It's known for imaginary false numbers. Same with userbenchmark, gpusboss, cpuboaa and the tons of random YouTube videos with imaginary performance. It's all faker than Kim Kardashian

We did move on from that and took the Techspot as the source but still it's false by a longshot.

[Zando_]

Just now, DismissedFetus said:

 

You took the words out of my mouth in the first part but still about 6 cores:

6 cores is great for games and I belive it still beats the stock 8700 I have without overclock.

The 8700K vs the 8700? They both have the same 6c turbo speed and are the same CPU (8700K just has an unlocked multi), there should be no performance difference. 
 

1 minute ago, DismissedFetus said:

We did move on from that and took the Techspot as the source but still it's false by a longshot.

How is it false? Because you got the same score? That just means Metro isn't as hard on the CPU, thus why you're pulling the same fps with the same GPU. 

[jones177]

1 minute ago, Zando Bob said:

OP said he has an 8700 non-K, so best it will do is stock 8700K perf (same all core turbo). But yeah, I don't think many games take proper advantage of the extra cores on the 9900K right now, so in a lot the 6c CPUs will still compete. 

Thanks. I missed the non k part.

 

I did my test just after the Metro E came out so I did not test anything newer.

This video got about the same results I did with ACO, Fortnite and Shadow of the Tomb Raider.

 

[Pauleft]

3 minutes ago, Zando Bob said:

The 8700K vs the 8700? They both have the same 6c turbo speed and are the same CPU (8700K just has an unlocked multi), there should be no performance difference

Perhaps. Sorry had to do it 

[DismissedFetus]

1 hour ago, Zando Bob said:

The 8700K vs the 8700? They both have the same 6c turbo speed and are the same CPU (8700K just has an unlocked multi), there should be no performance difference. 
 

How is it false? Because you got the same score? That just means Metro isn't as hard on the CPU, thus why you're pulling the same fps with the same GPU. 

A user mentioned that the i9-9900K doesn't beat the overclocked 8700K and I stated that I have 8700 stock.

 

Another user said that the FPS might be CPU related so we changed the source to a testbench with 9900K.

And I still got same FPS at higher settings than their tests. 

[DismissedFetus]

1 hour ago, jones177 said:

Thanks. I missed the non k part.

 

I did my test just after the Metro E came out so I did not test anything newer.

This video got about the same results I did with ACO, Fortnite and Shadow of the Tomb Raider.

 

So 9900K is less powerfull than 8700K when both overclocked fully while 9900K outperforms in other workloads such as rendering because of its cores?

 

[DismissedFetus]

1 hour ago, Pauleft said:

Perhaps. Sorry had to do it 

You did good.