Support NeoGAF

[DarkLordMalik]

Explosions and physics bring the FPS down. But from the single example in the vid it seems the players only uses 2 C4 charges. You can later upgrade to set 4 charges at the same time - i assume framerates take a further hit when you place all 4 around a structure and detonate them.

The example seen in the video is from a mission in the very beginning of the game. It is also one of the extreme example as it literally destroys a bridge, that hardly happens elsewhere in the game (atleast from what I played). The bridge can be down for a while and magically gets rebuilt after a while. I guess the developers intended it to be this way because the bridge is essential if we are driving a car. I accidentally managed to drown my car thinking the bridge was rebuilt but it was still destroyed.

 

[Alienous]

Heh, PS4 version turned OK,X1 in the other hand.

X1 with lower Res and Fps.

Nah. Dipping to 26 because you blew up a couple of cars in a game about blowing shit up is far less than 'OK'.

 

[Nzyme32]

Regarding the PC version:

"What we can say right now with a fair amount of confidence is that the GTX 950 paired with a Haswell Core i3 processor can easily run Just Cause 3 at 1080p maxed with SMAA anti-aliasing and 8x anisotropic filtering and stay well north of 30fps. Our 'go to' card for budget PC gaming - the GTX 750 Ti - does struggle though, dipping to a minimum 23fps even with shadow quality pared back to the high setting (though overclocking can help). These are value-orientated parts - a more powerful mid-range gaming PC should have few problems running this game well. We'll report back on that as soon as we can."

 

[etta]

Of course it does.

Over and over again, for reassurance.

What? Can't that be said about the consoles, too?
And there are some shitty port jobs like Batman, even though they are rare.

 

[Gipsy Danger]

Does this even need to be said?

It needs to be said as much as the PS4 versions being superior to the X1 version. Both are foregone conclusions 99% of the time.

 

[Loris146]

Regarding the PC version:

"What we can say right now with a fair amount of confidence is that the GTX 950 paired with a Haswell Core i3 processor can easily run Just Cause 3 at 1080p maxed with SMAA anti-aliasing and 8x anisotropic filtering and stay well north of 30fps. Our 'go to' card for budget PC gaming - the GTX 750 Ti - does struggle though, dipping to a minimum 23fps even with shadow quality pared back to the high setting (though overclocking can help). These are value-orientated parts - a more powerful mid-range gaming PC should have few problems running this game well. We'll report back on that as soon as we can."

CPU or Gpu limited? Or both ?

 

[george_us]

Ugh wish I had more money. This game looks incredible but Xenoblade takes precedent.

 

[KainXVIII]

To summarize.

Lowest drops to 24 fps on PS4, 20 fps on Xbox One.
PS4 is closer to the target 30 fps than Xbox One and maintains it longer. Xbox One can drop randomly when nothing explosive is going on.
Graphical quality is same for both versions of the game.
Resolution is 1080p on PS4, 900p on Xbox One.
Both version of the game run with adaptive Vsync.
IQ is equal to the high settings on PC.

The game also requires a decent GPU as Digital Foundry weren't able to get the game working at closer to 30 fps with GTX 750ti. Some PC folks were sure that this would be the case, lol.

PS4 has minor framepacing issue.

 

[Aceofspades]

So when people thought the XB1 > PS4, the higher clocked CPU was the reason.


Now what?

This has been proven wrong since the beginning, I think it happened with Witcher and Unity then both games got patched and run better on PS4.

 

[BennyBlanco]

Regarding the PC version:

"What we can say right now with a fair amount of confidence is that the GTX 950 paired with a Haswell Core i3 processor can easily run Just Cause 3 at 1080p maxed with SMAA anti-aliasing and 8x anisotropic filtering and stay well north of 30fps. Our 'go to' card for budget PC gaming - the GTX 750 Ti - does struggle though, dipping to a minimum 23fps even with shadow quality pared back to the high setting (though overclocking can help). These are value-orientated parts - a more powerful mid-range gaming PC should have few problems running this game well. We'll report back on that as soon as we can."

https://www.youtube.com/watch?v=fQrudYDyLds

Looks to be running well over 30 fps on an i3 + 660 at console settings.

 

[LordOcidax]

Ugh wish I had more money. This game looks incredible but Xenoblade takes precedent.

Agree, going to grab this game later, for sure.

 

[Kralamoonard]

Just a warning guys, please keep off derailing the thread by mentioning NX Gamer and turning it into a pissing contest/shitposting against him or Digital Foundry as they both do great work in what they provide.

(Also, bear in mind he didn't actually test the PS4 version, so people claiming now that JC3 "isn't as bad as NXG made it out to be", have to understand that he was testing the XB1 version).

Thanks.

 

[Shenmue]

What? When did anyone claim that the XBox version was running better than PS4? I don't remember any of that from the other thread, maybe I missed something? Or are people creating perceived slights?

 

[Bunta]

The PS4 version looks pretty solid in the video. The XO version seems to perform better than it did in the NXGamer video. I wonder why.

Tested different areas in the game, maybe? I only watched 1-2 mins of NX and haven't watched DF's yet, so just throwing out a possibility.

 

[DarkLordMalik]

PS4 has minor framepacing issue.

Where. It wasn't mentioned in the article?

As for the tearing, I have yet to actually see it or get bothered by it. It is not really noticeable at all for the majority of the game TBH. PS4 version here.

 

[ISee]

Regarding the PC version:

"What we can say right now with a fair amount of confidence is that the GTX 950 paired with a Haswell Core i3 processor can easily run Just Cause 3 at 1080p maxed with SMAA anti-aliasing and 8x anisotropic filtering and stay well north of 30fps. Our 'go to' card for budget PC gaming - the GTX 750 Ti - does struggle though, dipping to a minimum 23fps even with shadow quality pared back to the high setting (though overclocking can help). These are value-orientated parts - a more powerful mid-range gaming PC should have few problems running this game well. We'll report back on that as soon as we can."

...with CPU power the likely culprit for bottlenecked performance there

.

So an i3 can outperform the AMD 6-7 jaguar cores in the consoles? Seriously?

 

[geomon]

I can live with the occasional dip to 24fps. Preorder retained. ☺

 

[Kralamoonard]

On topic though, it is incredibly disappointing (and odd!) considering how well optimised Just Cause 2 was for consoles regarding how amazing it looked (and still does, particularly the sea + lighting) and how stable the performance was (from what I can remember so rose tinted glasses etc).

I really hope they patch it up; it's even more puzzling considering it went gold like a month and a bit ago so there shouldn't have been any reason for this in that they had time to patch it, or it shouldn't have gone gold until they fixed it (for which again they had time to). I know it's first time on new hardware and all, but damn has the engine not held up well. Perhaps it was too well optimised for last gen (or it may be an entirely new engine for all I know)!

Or just tone down the 'splosions a little lol.

 

[Nzyme32]

CPU or Gpu limited? Or both ?

No idea, but I'd assume for the 750ti going to high settings across the board would have helped rather than only parring back the shadows

https://www.youtube.com/watch?v=fQrudYDyLds

Looks to be running well over 30 fps on an i3 + 660 at console settings.

Sounds about what you'd expect. Typical avalanche affair where the PC version is pretty good at scaling. I think the bigger concern is if they sort out the weird Sqenix issues with the leaderboards when servers are not available. That may or may not be an issue when it launches

 

[DarkLordMalik]

https://www.youtube.com/watch?v=fQrudYDyLds

Looks to be running well over 30 fps on an i3 + 660 at console settings.

This isn't shocking, and DF tested it with GTX 750ti and faced drops to 23 fps at the lowest. So it certainly isn't better on the GTX 750ti + i3 combo.

Sounds about what you'd expect. Typical avalanche affair where the PC version is pretty good at scaling. I think the bigger concern is if they sort out the weird Sqenix issues with the leaderboards when servers are not available. That may or may not be an issue when it launches

I don't know what were they thinking when they implemented it. Did they learn nothing from the DriveClub fiasco? =/

 

[Saty]

The example seen in the video is from a mission in the very beginning of the game. It is also one of the extreme example as it literally destroys a bridge, that hardly happens elsewhere in the game (atleast from what I played). The bridge can be down for a while and magically gets rebuilt after a while. I guess the developers intended it to be this way because the bridge is essential if we are driving a car. I accidentally managed to drown my car thinking the bridge was rebuilt but it was still destroyed.

Yes, but if you can upgrade to 4 charges at the same time them maybe setting all 4 means double the calculation and worse framerate.

 

[Portugeezer]

How does this compare to JC2? I remember it not running too well on X360 in the more urban areas.

Was hoping they'd be able to maintain 30fps minimum, I can just get it on PC anyway.

 

[Inuhanyou]

.

So an i3 can outperform the AMD 6-7 jaguar cores in the consoles? Seriously?

Even the weakest i3 core is stronger than the Jaguar cores in the consoles. That is a fact

 

[Kezen]

A very modest 950 apparently has no problem with high settings at 30fps, bodes very well for 60fps on higher-end systems (R9 285/280x/GTX 680/770).

 

[DarkLordMalik]

Yes, but if you can upgrade to 4 charges at the same time them maybe setting all 4 means double the calculation and worse framerate.

I am trying to figure out what you mean here? Because the charges that we set on bridge are rather minor. It the start, we begin with 4 charges atleast so what are the extra charges? They only set two on the bridge because that is what the missions states. We can set more as well.

The frame rate drops because the physics of the bridge come into play here, and it gets destroyed along with the cars on it. It won't get destroyed in a different way if we set 4 different charges on it.

 

[BennyBlanco]

.

So an i3 can outperform the AMD 6-7 jaguar cores in the consoles? Seriously?

Yea, this has been happening routinely since the consoles launched.

 

[KainXVIII]

Where. It wasn't mentioned in the article?

In youtube video.

 

[hawk2025]

What? When did anyone claim that the XBox version was running better than PS4? I don't remember any of that from the other thread, maybe I missed something? Or are people creating perceived slights?

Originally Posted by JigglesBunny

Played both the PS4 and Xbox One versions, XBone has a much more solid framerate but looks slightly muddy and aliased, but PS4 is a lot clearer IQ wise, but drops frames like it's going out of style.

On both consoles, loading times are vile. Like, 5 minutes before respawning. I got up, made a sandwich, ate it, and it finished 30 seconds or so afterwards, and no, I'm not joking.

Actual game is just JC2 with some ramped up visual flair and a loads more destruction, which for me is worthy of GOTY haha, but YMMV.

Looks like it is running fine on Xbox One to me.

Aw man.. Xbox version looks to run a lot better. Whats up with that :/. Id take 900p over 1080 of it means the game runs good

Yep, Xbox One version seems to be running great with all of that destruction and rain.

Fair enough, it's mostly ironcreed, and I should know better at this point. But the "it's fine" brigade based on little to no information (not only exclusive to Xbox or PS4, that's not the point) is always strong until actual, factual information comes out.

 

[MDSLKTR]

The game also requires a decent GPU as Digital Foundry weren't able to get the game working at closer to 30 fps with GTX 750ti. Some PC folks were sure that this would be the case, lol.

Ouch and I was debating going pc with this one. Avalanche better have some performance patch in progress for every version

 

[psychotron]

Where. It wasn't mentioned in the article?

As for the tearing, I have yet to actually see it or get bothered by it. It is not really noticeable at all for the majority of the game TBH. PS4 version here.

It was mentioned in the video. I'm avoiding this game until it's patches further.

 

[Shenmue]

Fair enough, it's mostly ironcreed, and I should know better at this point. But the "it's fine" brigade based on little to no information (not only exclusive to Xbox or PS4, that's not the point) is always strong until actual, factual information comes out.

Ahhh ok makes sense why I missed some of those then.

 

[DarkLordMalik]

Ouch and I was debating going pc with this one. Avalanche better have some performance patch in progress for every version

Not really, it is a good port on PC. Go for it

I was just saying that GTX 750ti, which is usually considered for a comparison with PS4/XBO, can't really put better performance here despite a better CPU. The PC port sounds good so far, unless you have GTX 750ti?

It was mentioned in the video. I'm avoiding this game until it's patches further.

Just saw the video. TBH, the frame rate is perfectly fine in the video which is weird. I will wait for further stress test as the game plays fine and maintains its target 30 fps as seen in the video. It is when we go full "Just Cause" the frame rate goes bad. Hopefully it can be patched to some extent.

 

[Backfoggen]

Bummer for Xbox One only users and about the physics-based fps drops.

 

[Nzyme32]

This isn't shocking, and DF tested it with GTX 750ti and faced drops to 23 fps at the lowest. So it certainly isn't better on the GTX 750ti + i3 combo.


I don't know what were they thinking when they implemented it. Did they learn nothing from the DriveClub fiasco? =/

Their setup isn't the same as the consoles though for the 750ti, running it maxed with SMAA and 8xAF with only the shadows parred down to high.

As for the Sqeunix "online" issue, in theory it shouldn't be a hard thing to fix. There is also an offline mode available to avoid the issue if their servers are not working for the leaderboard stuff.

 

[BennyBlanco]

This isn't shocking, and DF tested it with GTX 750ti and faced drops to 23 fps at the lowest. So it certainly isn't better on the GTX 750ti + i3 combo.


I don't know what were they thinking when they implemented it. Did they learn nothing from the DriveClub fiasco? =/

i3 + 960 at higher than console looks to be running over 30 fps. Budget cards too stronk.

https://www.youtube.com/watch?v=GuEge1kf9CQ

 

[Justinh]

We managed to grab the game ahead of release from a local retailer

Hmm, I always thought that they got pre-release stuff for games.

This game's gonna have to be a PC only thing for me, it appears. Maybe I'll pick it up again later if issues get ironed out, but I'm not holding my breath.

 

[JigglesBunny]

I heard that some people even exchange their PS4 copies for Xbox One after threads with 3 minutes loading etc. I wonder what they do now =)

Having played both versions, they're both equally atrocious in regards to performance and load times.

I had initially thought the Xbox One version to run at a more stable framerate until a certain instance involving a train caused the game to drop to what felt like single digits.

Both console versions are busted up.

 

[GifGafIsTheBestGaf]

one thing is for sure though, the game looks great despite its performance issues. i almost pre-order when i watched the 6 things you need to know from PSA this morning lol

i hope DF takes the time to test different hardrives in both machines, how a SSD performs etc et

 

[psychotron]

Not really, it is a good port on PC. Go for it

I was just saying that GTX 750ti, which is usually considered for a comparison with PS4/XBO, can't really put better performance here despite a better CPU. The PC port sounds good so far, unless you have GTX 750ti?


Just saw the video. TBH, the frame rate is perfectly fine in the video which is weird. I will wait for further stress test as the game plays fine and maintains its target 30 fps as seen in the video. It is when we go full "Just Cause" the frame rate goes bad. Hopefully it can be patched to some extent.

Frame pacing can be tough to spot unless playing. Bloodborne holds 30 very often but frame pacing gives it a stutter that really sucks.

 

[Sesuadra]

Does this game already use the 7.CPU core from the PS4?

 

[ISee]

Even the weakest i3 core is stronger than the Jaguar cores in the consoles. That is a fact

Yea, this has been happening routinely since the consoles launched.

I seriously didn't know that... I thought the game was just bad optimized and I auto assumed that it would run like ass even on descent i5s. Maybe that's why I am so 'shocked' that a two core i3 (+HTT) is already enough (at least for 30fps). Aren't console APIs supposed to take better advantage of multi cores?

 

[cleansock]

Little 950 doing work! I never seen one in the wild, but I bet they are adorable. Paired with an i3?! Does df have a photo of this beast of a rig?

 

[Y2Kev]

Am I the only one that really wants to know what happened to assassins creed unity last year?

 

[Cataferal]

Hmm, I always thought that they got pre-release stuff for games.

Not this time. We were given PC, but had to fetch the console copies ourselves.

i hope DF takes the time to test different hardrives in both machines, how a SSD performs etc et

We will =) Definitely high on the priority list for this one.

 

[Kralamoonard]

Am I the only one that really wants to know what happened to assassins creed unity last year?

Regarding the frame rate? I just finished the game this weekend (since I was LTTP), and it's really stable, except when when you enter really crowded areas which granted, doesn't happen as often. This was after a few patches released earlier in the year however.

Bear in mind, this is my just my own anecdote, I don't have any official frame rate readings, and I can't accurately say what it was at most times, but to my eye, it was a fairly stable 30fps as of this weekend.

 

[Inuhanyou]

Does this game already use the 7.CPU core from the PS4?

Who knows, but the game itself must be patched. This is way worse than launch Bloodborne load times even at their absolute worst, and the FPS issues aren't acceptable to this degree.

I seriously didn't know that... I thought the game was just bad optimized and I auto assumed that it would run like ass even on descent i5s. Maybe that's why I am so 'shocked' that a two core i3 (+HTT) is already enough (at least for 30fps). Aren't console APIs supposed to take better advantage of multi cores?

Well i can say it is badly optimized with those loading times, frame pacing issues and the like.

But its also a fact that the CPU's in the consoles are lower than i3's which are lower end intel CPU's.

Its up to the dev to work around that and provide acceptable performance before the game launches for every SKU. A majority of games that have had FPS or load time issues this gen have been largely fixed, but only after weeks or months of updates. I'm tired of this lazy attitude of launch now, patch later.

 

[patapuf]

I seriously didn't know that... I thought the game was just bad optimized and I auto assumed that it would run like ass even on descent i5s. Maybe that's why I am so 'shocked' that a two core i3 (+HTT) is already enough (at least for 30fps). Aren't console APIs supposed to take better advantage of multi cores?

at the end of the day clock speed still matters, a lot.

and most of the multi core stuff devs do on consoles they can do on PC.

 

[ReturnOfTheRAT]

Once again gaffers anecdotal evidence is crushed. So much for the PS4 version runs like shit as well narrative.

People were saying that?

 

[Nere]

Holy shit something kind off-topic but did anyone's else eye's hurt from the red coloured letters OP uses or it is only me with that problem?

 

[Kralamoonard]

Holy shit something kind off-topic but did anyone's else eye's hurt from the red coloured letters OP uses or it is only me with that problem?

Isn't it a consistent colour all around though? All I do is use the highlight thing I've seen some posters use to make points stand out by using the standard text thing on GAF: [hightlight][/hightlight]

Sorry if it bothers you, I mean I could remove it if you want, but we'll see if others are seeing the same problem. First time I hear of it tbh; maybe it's the monitor?