Square Enix shares GeForce RTX 20 Final Fantasy XV benchmarks

[HEXUS]

Compares the GeForce RTX 2080 (Ti) with a host of other cards at 1440p and 2160p.

Read more.

[CAT-THE-FIFTH]

An RTX2080TI is just under 14% faster than a Titan Xp and 21% faster than a GTX1080TI?? This does not sound right at all??

[Sixiepoo]

An RTX2080TI is just under 14% faster than a Titan Xp and 21% faster than a GTX1080TI?? This does not sound right at all??

imo, still to early to even conclude that as the drivers arent still mature, but at the same time, i dont see it improving by more than 30%, perhaps in higher resolution it might

[edmundhonda]

nVidia's own benchmarking suite had the 2080ti at ~35-50% over the 1080ti at 4k, so 21% for a real world sample isn't hugely surprising.

DLSS instead of TXAA might help if you're looking for effective 4k performance but that's it. These aren't superpowered cards.

[flearider]

if I'd of pre ordered I'd be saying ouch ..

[=assassin=]

Looking at the performance figures, and how they seem to relate between current gen and previous gen, my powers of deduction reckons that the 2070 will be a tiny amount ahead of a 1080, and the 2060 will be a tiny amount behind a 1070. That's just taking roughly a third of the previous generations card score and adding it on to that score. Be curious to see if that works out right...

[Safetytrousers]

The full results of the above tests:
http://benchmark.finalfantasyxv.com/result/

This is one game (which incidentally I'm not interested in playing).

[philehidiot]

The full results of the above tests:
http://benchmark.finalfantasyxv.com/result/

This is one game (which incidentally I'm not interested in playing).

I notice a few things about those figures. Firstly the new cards are present in some and absent in others. It's almost as if the results they published were only included if they show what the Nvidia marketing department want them to show.

Secondly I notice the disparity between the Vega 64 and its direct competitor cards (1070Ti) where the AMD cards underperform relative to the competition. We can therefore deduce that this game is being optimised for Nvidia cards and therefore will probably show the 20XX series at its best.

Thirdly, I expect this is on underdeveloped drivers Vs mature platforms so how representative this will be of final performance, we don't know. Performance could be artificially increased or suffering due to lack of optimisation.

[Safetytrousers]

I notice a few things about those figures. Firstly the new cards are present in some and absent in others. It's almost as if the results they published were only included if they show what the Nvidia marketing department want them to show.

The 2080ti is shown for every result from 1440p standard quality onwards.
It's equally plausible Sqaure Enix didn't test the card at lower settings knowing nobody will have bought a 2080ti to play below 1440p medium settings (in the absence of ray tracing).
What useful thing would you learn about the 2080ti at 1280 x 720, Lite Quality? Nobody would be doing that except for curiosity, and these results are about a game's performance, not a deep investigation of the 20 series.

[philehidiot]

The 2080ti is shown for every result from 1440p standard quality onwards.
It's equally plausible Sqaure Enix didn't test the card at lower settings knowing nobody will have bought a 2080ti to play below 1440p medium settings (in the absence of ray tracing).
What useful thing would you learn about the 2080ti at 1280 x 720, Lite Quality? Nobody would be doing that except for curiosity, and these results are about a game's performance, not a deep investigation of the 20 series.

The Ti is shown but I wouldn't ever consider one, I'm far more interested in the card which is mysteriously missing from some of the graphs. I don't think the argument about them not testing due to use case is valid as they've tested everything else on every conceivable setting regardless of whether anyone would use it. I doubt it's an oversight so either these are incomplete results which seems unlikely given the completeness of everything else. I suspect there's a marketing reason.

[Safetytrousers]

I doubt it's an oversight so either these are incomplete results which seems unlikely given the completeness of everything else. I suspect there's a marketing reason.

I haven't monitored the Square Enix page for FF XV benchmarks, but I would think all the results except for the 20 series cards were already there. Personally if I was the person doing the tests I would feel very foolish testing the 2080ti at 720p low settings.
What would be the point of surreptitious marketing within 48 hours of the reviews going live?