FINAL FANTASY XIV: Dawntrail Official Benchmark

[Det0x]

The 7800x3d is impressive. What memory settings were you running here?
My 14900kf is pretty much tapped out on this board (Apex running 8800 is gone).
View attachment 344119

Nice run

I were running my daily #1 profile (6600/2200) settings together with daily CO tune on the 7800X3D

(my daily #2 profile is 8000/2200) (haven't tested whats faster in this benchmark)


Can maybe match the 51k if pushed harder since i dont need full AVX stability to play games
These are my daily CO settings: (left side of screen)


*edit*
Sorry for the offtopic guys

 

[SuperMumrik]

I were running my daily #1 profile (6600/2200) settings

That's a mighty fine IMC
Is 6600/2200 usually faster in games?

 

[Det0x]

That's a mighty fine IMC
Is 6600/2200 usually faster in games?

Hard to say, i have pretty much only been playing hwbot the last year

But as Clam cache/memory benchmark show above both profiles are pretty even, so i guess it comes down to what the game prefer, latency or bandwidth

 

[kryptonfly]

Yeah, this makes NO SENSE at all vs a 5800x3d


The 7800x3d is impressive. What memory settings were you running here?
My 14900kf is pretty much tapped out on this board (Apex running 8800 is gone).
View attachment 344119

This is at 540p (1080p at 50% scale) and lowest settings ?

 

[SuperMumrik]

This is at 540p (1080p at 50% scale) and lowest settings ?

Yes

 

[sLowEnd]

And here's what it's like on something really slow, max settings and minimum settings
Athlon 3050u, 12GB (4+8) DDR4 2133, eMMC storage

Spoiler: Max

Spoiler: Min (100% scaling)

Definitely not combat-worthy, but probably tolerable for gathering/crafting in a pinch at minimum settings. It'll probably explode in Limsa Lominsa.

Edit: Oh wait, I didn't turn the resolution scaling as low as it could go....brb

Okay, 50 percent scaling is not a better experience. Even though the average is higher, it introduced massive stutters and frametime inconsistencies, while making the game look like a PS1 game.

Spoiler: Min (50% scaling)

Spoiler: 100% scaling

Spoiler: 50% scaling

 

[AVATARAT]

7900x + 7900 XT
A little bump over 1080p

 

[A Computer Guy]

7900x + 7900 XT
A little bump over 1080p

View attachment 344247

Which Taichi board is that?

 

[AVATARAT]

Which Taichi board is that?

The mobo is Asrock B650E PG Riptide WiFi.
I just use this CPU-Z skin because I like it

 

[AVATARAT]

7900x + 7900 XT
At 1080p the CPU hit is huge

 

[K2K]

i5 12500 | MSI B660M-A PRO | BLS2K8G4D30AESBK 3700 | RTX4070 Palit

1080p - 24873
1440p - 18586
2160p - 9469

 

[Arctucas]

 

[jiajo]

1080 maximum option DLSS

1080 maximum option FSR

 

[zack39]

Need an gpu update in the future:

 

[Kaira]

This is clearly still dx11, but why then there's a nvidia gameworks logo and the nvidia rtx logo if the game does not use RTX or anything of it?
they've updated some shaders (the skin for example), the textures still lack any depth like normal maps / height maps / POM, they aim to add something just before the release? possible that they're gonna release it with also rtx? dlss and fsr are nice, but i would prefer DLAA as the game AA sucks, i've seen some people saying "wow look at the detail of the hairs now" but this is ridicolous in the 2024

 

[glnn_23]

7900 xtx
7800x3d

 

[Space Lynx]

7900 xtx
7800x3d

View attachment 344891

Are you running 8000mhz ram on a AMD rig? I thought 6000mhz was max?

 

[Super Firm Tofu]

Are you running 8000mhz ram on a AMD rig? I thought 6000mhz was max?

They are (8000 MT/s, technically). It's doable on later AGESA versions, and mainly on two-DIMM boards.

There was certainly some skill in getting that dialed in, and also the clocks on their XTX.

If I'm not mistaken, that's the highest 1080p Maximum score in the thread so far, and beats my 14900KS/4090 combo.

Looks like X3D is the one to try to beat.

 

[Space Lynx]

that's insane, wow, well done!!! @glnn_23

 

[SuperMumrik]

If I'm not mistaken, that's the highest 1080p Maximum score in the thread so far, and beats my 14900KS/4090 combo

You are mistaken. I posted a 45k+ maximum result on page one (and have since improved it by 2500 points), but we test low settings since we occasionally hit gpu limits with maximum settings even on 4090

 

[Makaveli]

You are mistaken. I posted a 45k+ maximum result on page one (and have since improved it by 2500 points), but we test low settings since we occasionally hit gpu limits with maximum settings even on 4090

I think people also need to post their GPU Z screens also. My benchmarks are reference clocks and we have a mix of results now that are overclocked.

 

[Super Firm Tofu]

You are mistaken. I posted a 45k+ maximum result on page one (and have since improved it by 2500 points), but we test low settings since we occasionally hit gpu limits with maximum settings even on 4090

Can you clarify what Maximum and low settings mean? I was assuming Maximum is just choosing that default and by doing so disabling any type of up-scaling.

It'll be almost impossible to compare results if everyone is just picking settings at random.

 

[SuperMumrik]

Can you clarify what Maximum and low settings mean?

Sure, it's maximum preset when the end result says maximum alongside what upscaler is selected. If you make changes to settings outside the pre-defined presets it will say custom.

My benchmarks are reference clocks and we have a mix of results now that are overclocked.

I would assume most people benchmark overclocked systems? Mine was doing a "conservative" 6100Mhz on the low custom run and close to stock on the maximum preset with dayli 8266c34 memory for clarification.
Cpu can easily be pushed to 6300-6400 for benchmarks, so I might try that when I get on-shore

 

[RootinTootinPootin]

FSR seems to wildly have better FPS (not the best run)

 

[Det0x]

That's a mighty fine IMC
Is 6600/2200 usually faster in games?

So i have finished my dualrank 64 gigs performance exploration and went back to this benchmark, i found SR 2x16gigs adie @ 6600MT/s CL26-37-32-30-62 + 2200mhz FCLK 1:1 mode to about ~400points faster than SR 2x16gigs adie @ 8000MT/s CL32-45-40-44-84 + 2200mhz FCLK 2:1 mode in this benchmark.

So the verdict is that this benchmark prefers latency, atleast for my current X3D setup

Can maybe match the 51k if pushed harder since i dont need full AVX stability to play games

With some further tweaking i also managed to break 51k, but to my surprise, i still seem to be GPU limited even at 1080p lowest settings, on a tuned watercooled 3090
(saw nice scaling with +30mhz GFX core)