Endwalker benchmark is significantly better than Shadowbringers

Shadowbringers
Average Frame Rate: 111.3384
Minimum Frame Rate: 45

Endwalker
Average Frame Rate: 146.6522
Minimum Frame Rate: 62

Finally, 60 fps minimum! On the same hardware.

After running the benchmark, there's a button on the launcher to save results. Clicking that opens the text file in Notepad (and saves it in the benchmark folder. don't have to save it again). That text file contains all the info.

Wonder what or what they are not processing in the background anymore that is taking up less CPU space. Maybe a gfx card driver update?

My scores were within a margin of error from each other. Might want to make sure your settings are identical.

Mine ran better this time as well, but I'm not sure if it's because everything is up to date or something. My hardware hasn't changed since 2017 when I bought my PC except for a new speaker. I don't use it hard though so that's probably why I haven't had many issues.

To be clear, I ran both benches back to back, with identical settings. If old scores for ShB are being used it stands to reason there will be differences.

They did say they were adjusting things with their lighting system didn't they?

no controller support and screenshot button only works sometimes really my only issue with it

Quote:

---

Originally Posted by Arrhin

To be clear, I ran both benches back to back, with identical settings. If old scores for ShB are being used it stands to reason there will be differences.

---

you're right. that was bad science of me to use an old shadowbringers score, however i'm happy to report the same result: massive increase in performance for me. dunno what makes the difference.

here are my current shadowbringers results:
Average Frame Rate: 85.51546
Minimum Frame Rate: 33

i made sure the settings were the same this time

full report:
FINAL FANTASY XIV: Shadowbringers Benchmark
Tested on: 7/13/2021 6:25:23
Score: 12860
Average Frame Rate: 85.51546
Minimum Frame Rate: 33
Performance: Extremely High
-Easily capable of running the game on the highest settings.
Loading Times by Scene
Scene #1 1.153 sec
Scene #2 2.354 sec
Scene #3 2.716 sec
Scene #4 1.656 sec
Scene #5 0.801 sec
Total Loading Time 8.68 sec

DAT:s20210713062523.dat

Screen Size: 3840x2160
Screen Mode: Borderless Windowed
DirectX Version: 11
Graphics Presets: Maximum
<i snipped the settings details here because the Maximum preset defines them>

System
Microsoft Windows 10 Pro (ver.10.0.19042 Build 19042)
AMD Ryzen 9 5950X 16-Core Processor
32710.625MB
NVIDIA GeForce RTX 3090(VRAM 24348 MB)

The latest big patch absolutely destroyed my frame rate, I hope EW will improve it again.

and here is endwalker again, so they are back to back. this is at maximum setting, so there is the most room for improvement

FINAL FANTASY XIV: Endwalker Benchmark
Tested on: 7/13/2021 6:42:15
Score: 20838
Average Frame Rate: 146.5652
Minimum Frame Rate: 62
Performance: Extremely High
-Easily capable of running the game on the highest settings.
Loading Times by Scene
Scene #1 1.093 sec
Scene #2 1.952 sec
Scene #3 2.95 sec
Scene #4 1.165 sec
Scene #5 0.567 sec
Total Loading Time 7.727 sec

DAT:s20210713064215.dat

Screen Size: 3840x2160
Screen Mode: Borderless Windowed
DirectX Version: 11
Graphics Presets: Maximum

System
Microsoft Windows 10 Pro (ver.10.0.19042 Build 19042)
AMD Ryzen 9 5950X 16-Core Processor
32710.625MB
NVIDIA GeForce RTX 3090(VRAM 24348 MB)

Interesting, so here's mine:

FINAL FANTASY XIV: Endwalker Benchmark
Tested on: 7/13/2021 8:28:59
Score: 13378
Average Frame Rate: 91.06842
Minimum Frame Rate: 59
Performance: Very High
-Easily capable of running the game. Should perform exceptionally well, even at higher resolutions.
Loading Times by Scene
Scene #1 1.63 sec
Scene #2 2.975 sec
Scene #3 4.049 sec
Scene #4 1.468 sec
Scene #5 0.862 sec
Total Loading Time 10.984 sec

DAT:s20210713082859.dat

Screen Size: 2560x1440
Screen Mode: Borderless Windowed
DirectX Version: 11
Graphics Presets: Maximum

FINAL FANTASY XIV: Shadowbringers Benchmark
Tested on: 7/13/2021 8:20:35
Score: 13764
Average Frame Rate: 91.79177
Minimum Frame Rate: 36
Performance: Extremely High
-Easily capable of running the game on the highest settings.
Loading Times by Scene
Scene #1 1.277 sec
Scene #2 3.073 sec
Scene #3 3.107 sec
Scene #4 1.931 sec
Scene #5 1.027 sec
Total Loading Time 10.415 sec

DAT:s20210713082035.dat

Screen Size: 2560x1440
Screen Mode: Borderless Windowed
DirectX Version: 11
Graphics Presets: Maximum

System
Microsoft Windows 10 Pro (ver.10.0.19043 Build 19043)
AMD Ryzen 5 5600X 6-Core Processor
32683.359MB
NVIDIA GeForce GTX 1080(VRAM 8079 MB)

If you look at the minimum framerates there is a huge difference. I expect that EW is better able to utilize the hardware, and yet, it just so happens that my GPU doesn't punch high enough for it to skew the score. CPU wise we should be very close, XIV really only kicks at a single core so...

Hey try not to run that benchmark in windowed mode.

Windowed score:
https://i.imgur.com/BfEQClZ.jpg

Full Screen Score:
https://i.imgur.com/IeGHHfP.jpg

Also, since your running 4K, try changed the resolution to 4096 x 2160 or True 4K instead of 4K UHD. The game may look better for you as I feel the game is stretched at 16:9

I just made a post about this too.

Example:
https://i.imgur.com/kH8rcBx.png

Quote:

---

Originally Posted by Ath192

Hey try not to run that benchmark in windowed mode.

Windowed score:
https://i.imgur.com/BfEQClZ.jpg

Full Screen Score:
https://i.imgur.com/IeGHHfP.jpg

---

As XIV has issues with fullscreen mode most everyone compares bench scores in borderless windowed, as that is the mode we play in. A benchmark isn't a benchmark if it does not apply to real world application.

Quote:

---

Originally Posted by Arrhin

As XIV has issues with fullscreen mode most everyone compares bench scores in borderless windowed, as that is the mode we play in. A benchmark isn't a benchmark if it does not apply to real world application.

---

You're empirically wrong about what is being tested here. They are testing the difference between the benchmarks of SHB and EW, so regardless in what mode gets tested, if there is a real difference between both benchmarks it should be apparent in any mode tested. Given I ran tests and came with results that are absolutely impossible I have good reason to believe there is something wrong with the windowed mode in the EW benchmark. My full screen testing results are closer to what I've seen other players report playing at 4K maximum. Therefore, his data may not be correct if it is getting tested in windowed mode.

I also didn't know you spoke for the entire FFXIV community about the mode we play in. I can certainly attest you don't speak for me.

Quote:

---

Originally Posted by Ath192

You're empirically wrong about what is being tested here. They are testing the difference between the benchmarks of SHB and EW, so regardless in what mode gets tested, if there is a real difference between both benchmarks it should be apparent in any mode tested. Given I ran tests and came with results that are absolutely impossible I have good reason to believe there is something wrong with the windowed mode in the EW benchmark. My full screen testing results are closer to what I've seen other players report playing at 4K maximum. Therefore, his data may not be correct if it is getting tested in windowed mode.

I also didn't know you spoke for the entire FFXIV community about the mode we play in. I can certainly attest you don't speak for me.

---

I know exactly what is being tested here, as you may have noticed I recommended recent runs with identical settings so we could in fact, compare the two without driver or game update changes skewing the results. And I speak based on the accepted norms in the community. The issue you speak of is outside this scope and is with full screen mode, not windowed, as it has been since the launch of the game, I'm sure this is a known quantity at this point. My testing showed a similar delta between windowed and full screen in both benchmarks. As for the different modes that is the very reason we attach the system info you see so that we know if it's apples to apples or not, or if people are cheesing the run with lowered settings.

Can you point to where is the certified document where the norms in the community are ratified and what body certified it? Or are you just making things sound more official to fit your narrative? Either way. My results speak for themselves and it would be ignorant of you to dismiss them as a possibility that is what is happening to OP here.

It is also counter productive to be dismissive when someone comes forth with information that may be of use as well.

It seems OP is experiencing what I experienced running the program windowed. For whatever reason. Meaning my claim is actually what could help solve the riddle as to why his scores differ so much.

So after reading comments on this topic, I decided to compare running the benchmark under borderless window and full screen. Interesting that fullscreen mode actually gave me a lower score than my borderless window mode. I was expecting full screen to have better result lol. Anyone why it is like this?

Borderless window: 11k
FullScreen: 8k

Thank you Garet. I do not know why, but since I game in 4K and have tried FS, borderless and windowed and experience minor hiccups in supremely demanding scenarios I am inclined to believe the lower score, as with 14K on my PC that seems about right, whereas the 22k score should not have me dipping below 60 FPS no matter if I’m in the middle of a hunt with 200 people or a 24 man raid with all effects popping off. But it is a mystery.

Quote:

---

Originally Posted by Garet

So after reading comments on this topic, I decided to compare running the benchmark under borderless window and full screen. Interesting that fullscreen mode actually gave me a lower score than my borderless window mode. I was expecting full screen to have better result lol. Anyone why it is like this?

Borderless window: 11k
FullScreen: 8l

---

Full screen was the opposite for me. 11K Borderless, 13K fullscreen

Quote:

---

Originally Posted by Ath192

Can you point to where is the certified document where the norms in the community are ratified and what body certified it? Or are you just making things sound more official to fit your narrative? Either way. My results speak for themselves and it would be ignorant of you to dismiss them as a possibility that is what is happening to OP here.

It is also counter productive to be dismissive when someone comes forth with information that may be of use as well.

It seems OP is experiencing what I experienced running the program windowed. For whatever reason. Meaning my claim is actually what could help solve the riddle as to why his scores differ so much.

---

*Rubs temples* Ok, so yes, we know there is an issue here, and your information IS of value. What I am trying to tell you is that there is a wealth of knowledge on the topic of fullscreen vs. windowed going back to ARR release on this and many other forums, and yes, there are accepted community norms in terms of this benchmark as well as others, this is what I was referring to. You do you, and I'll do me, that's the best part about PC gaming. Now, that aside I do suggest you monitor your CPU and GPU utilization while running in fullscreen mode. If it's anything like what we've seen before, and what I see now, neither will be at 100%. Typically in a non v-synced setting either CPU or GPU will be at 100% utilization depending on where your system bottleneck is. Once we kick over to either of the windowed modes you'll see the GPU kick up substantially, in my case, to a steady 100% excepting for load screens. Now this is what we expect to see. As for what is going on, IF my memory serves me it has to do with Windows and how it does vsync (which seems to be constantly changing). Now in regards to the difference between the two benchmarks I do recall that ShB had issues with crashing in high framerate situations, or with high refresh rate displays. Now it's possible that this has been remedied and implemented in the EW bench but I do not know. All I do know is there is a massive difference in lowest framerate, where that comes from I have no idea, I'd have to try and do more testing but I'd rather just enjoy the performance boost. Anyways cheers, and if you find anything to add from other posts I think we'd be interested to see it.

i ran the benchmarks in borderless window because, as arrhin pointed out, i don't use fullscreen when i play. the game sometimes freezes if i tab out for discord or to look something up on the web, both of which happen often while playing ffxiv, because it's an mmo (i could handle fullscreen with something like tomb raider where i'm playing by myself). i have run the shb benchmark in fullscreen before and gotten slightly better performance--but not enough that i care

the game doesn't appear stretched to me. my monitor is actually 5120x2160 (4k ultrawide, otherwise known as 5k2k). the game looks great at this resolution. but the benchmarks don't do ultrawide so i'm stuck with running the benchmarks at 3840x2160. i don't have cinema 4k in my settings for fullscreen. even if i did, it wouldn't be pixel perfect, and if i manually set that resolution in windowed mode, it'll just change the window and not the aspect ratio of my monitor

My performance has actually gone down and I'm not sure why its doing that for the benchmark.

FINAL FANTASY XIV: Shadowbringers Benchmark
Tested on: 7/13/2021 14:06:15
Score: 14192
Average Frame Rate: 101.7494
Minimum Frame Rate: 44
Performance: Extremely High

FINAL FANTASY XIV: Endwalker Benchmark
Tested on: 7/13/2021 14:14:09
Score: 14015
Average Frame Rate: 98.86352
Minimum Frame Rate: 53
Performance: Very High

EDIT: To remove the terrible screen tearing that occurs in the benchmarks I changed to borderless windowed and lose FPS. Results are shown below.

FINAL FANTASY XIV: Shadowbringers Benchmark
Tested on: 7/13/2021 14:25:40
Score: 13765
Average Frame Rate: 96.99744
Minimum Frame Rate: 46
Performance: Extremely High

FINAL FANTASY XIV: Endwalker Benchmark
Tested on: 7/13/2021 14:33:17
Score: 13594
Average Frame Rate: 94.76842
Minimum Frame Rate: 53

Quote:

---

Originally Posted by Inosaska

My performance has actually gone down and I'm not sure why its doing that for the benchmark.

---

Don't go by the ratings "Extremely High" and "Very High". Yoshi-p said during the recent live letter that they upped the score requirement for Extremely High, but that doesn't have any bearing on your performance (i.e., they won't call it Extremely High unless it meets a higher bar). If you cap your frames at 60, then your performance has actually gone up, because your minimum framerate in Endwalker is higher and you'll never see the difference in Average Frame rate.

and... even if you don't cap your frames, i doubt you'd notice the difference between 98 and 101. i still say you've got better performance in endwalker

Quote:

---

Originally Posted by cheolsu

Don't go by the ratings "Extremely High" and "Very High". Yoshi-p said during the recent live letter that they upped the score requirement for Extremely High, but that doesn't have any bearing on your performance (i.e., they won't call it Extremely High unless it meets a higher bar). If you cap your frames at 60, then your performance has actually gone up, because your minimum framerate in Endwalker is higher and you'll never see the difference in Average Frame rate.

and... even if you don't cap your frames, i doubt you'd notice the difference between 98 and 101. i still say you've got better performance in endwalker

---

I just turned off 1 setting that is Screen Space Ambient Occlusion and my FPS increased by 14! The game also looks way better with it off and less dark. That setting is terrible and I wish I had turned it off sooner.

Quote:

---

Originally Posted by Nothv13

Full screen was the opposite for me. 11K Borderless, 13K fullscreen

---

Yes. I do not get it. I was expecting full screen to really give better scores as well. Wonder why it is opposite for me.

BORDERLESS WINDOW MODE:
FINAL FANTASY XIV: Endwalker Benchmark
Tested on: 7/14/2021 3:08:24
Score: 11091
Average Frame Rate: 79.1565
Minimum Frame Rate: 42
Performance: Very High
-Easily capable of running the game. Should perform exceptionally well, even at higher resolutions.
Loading Times by Scene
Scene #1 3.372 sec
Scene #2 6.377 sec
Scene #3 9.889 sec
Scene #4 4.291 sec
Scene #5 2.408 sec
Total Loading Time 26.337 sec

Screen Size: 1920x1080
Screen Mode: Borderless Windowed
DirectX Version: 11
Graphics Presets: High (Desktop)

FULLSCREEN:
FINAL FANTASY XIV: Endwalker Benchmark
Tested on: 7/14/2021 3:16:53
Score: 8519
Average Frame Rate: 56.47453
Minimum Frame Rate: 42
Performance: High
-Easily capable of running the game. Should perform well, even at higher resolutions.
Loading Times by Scene
Scene #1 4.339 sec
Scene #2 6.072 sec
Scene #3 9.415 sec
Scene #4 4.232 sec
Scene #5 2.285 sec
Total Loading Time 26.343 sec

Screen Size: 1920x1080
Screen Mode: Full Screen
DirectX Version: 11
Graphics Presets: High (Desktop)

Edit: It seems my Full screen mode is being hampered by my Vsync setting. I turned it off and it produced a higher score now

FULLSCREEN WITH V-SYNC OFF
FINAL FANTASY XIV: Endwalker Benchmark
Tested on: 7/14/2021 3:35:35
Score: 11393
Average Frame Rate: 81.81217
Minimum Frame Rate: 42
Performance: Very High
-Easily capable of running the game. Should perform exceptionally well, even at higher resolutions.
Loading Times by Scene
Scene #1 3.419 sec
Scene #2 6.24 sec
Scene #3 9.805 sec
Scene #4 4.557 sec
Scene #5 2.291 sec
Total Loading Time 26.312 sec

Quote:

---

Originally Posted by Inosaska

I just turned off 1 setting that is Screen Space Ambient Occlusion and my FPS increased by 14! The game also looks way better with it off and less dark. That setting is terrible and I wish I had turned it off sooner.

---

glad to hear that worked! and, i know what you mean. it's terribly expensive and just makes things darker. but when i need to reset my graphics i just choose Maximum from the presets and forget that it's on. with it off, the game is brighter and looks more anime ^^

To give some heads up, sometimes the graphics driver may only kick in Vsync or frame limiter in full screen mode as games are suppose to run in true fullscreen mode as standard, sorry to the above but true fullscreen is the default, borderless window is there to make tabbing out easier but FF14 is the only game that has issues with tabbing out in fullscreen mode.

Also you need to be careful when going borderless window mode to check the resolution its set at, it might be rendering a lower res sometimes.

It ran marginally better for me, but more importantly: That was by far the coolest of the benchmarks so far.

Quote:

---

Originally Posted by Limecat

It ran marginally better for me, but more importantly: That was by far the coolest of the benchmarks so far.

---

Agree. The benchmark is like a real cinematic trailer of the battle to come ala-avengers endgame :)