Digital Foundry Article Technical Discussion Archive [2014]

[Shifty Geezer]

Uhh? You are getting more frames pr second, the PS4 is still ahead even with Microsoft helping Blizzard to achieve the 1080p checkmark.

DrJay24's point is more that, assuming PS4 is capable of a lot more than XB1 is achieving due to having more GPU power, it would have been nice for the devs to crank up the Graphical Whizz a bit.

But as you say, devs aren't likely to scale up the visuals (depends whether the PC version has higher quality options beyond the current-gen consoles). Furthermore, if the target is an absolutely rock solid 60 fps, you need the average amount of effort to be well below that, and the hardware to not be stressed. Then, when things get busy, the hardware can manage it in time and the framerate doesn't take a hit.

That's where the MP stress test is going to be important here. It'll show if the PS4 is managing only a small increase in framerate, or substantially more, which'll be an interesting tech preformance comparison.

 

[specwarGP2]

We don't even know if the PS4 can maintain a solid 60 fps under all conditions, and the Xbone could drop further than 8 fps too.

Like many other folks, I'd like DF to push on and test the bits that really thrash the engine. As the first post-June and MS involved 1080p title (as far as we know) this could be an interesting title to dig into.

I can confirm PS4 isn't a solid 60 fps on single player.

 

[Scott_Arm]

Well, you should watch their video then. There are some "rare" drops yes but also whole rather long areas that are hovering at ~55fps on XB1. They even acknowledge it on the article:

Hmmm, I'd have to read again, but I thought later in the article they said they had to comb hours of footage to find the sections with frame drops.

 

[Shortbread]

You don't know that. You don't know about other system limitations. You're simply stating blind wishes.

"Absence of evidence is not evidence of absence!"

 

[function]

"That which can be asserted without evidence can be dismissed without evidence."

 

[DrJay24]

Evidence: The min frame rate is 60.
Assertion: The average and max are higher than 60.
Deduction: There is overhead going to wasted frames.

Of course all we have are the numbers from DF, which may not hold up for later in the game, but I was far from making things up.The only small leap is that the average and max frame rate is higher than the min, it is highly unlikely the game is hitting 60fps, no more or less.

 

[BRiT]

DrJay24, that's all good, but right now you have no idea how much spare capacity the system has, so your statement of 80fps is baseless.

 

[Dave Glue]

I can confirm PS4 isn't a solid 60 fps on single player.

IME, this is true. It's 60fps 95+% of the time through 15+ hours, but there are the few drops when things get nuts in the later stages.

Now, add in 3 more co-op players, all casting various complex graphical effects on top of that - it's questionable how much more "headroom" the PS4 would have when co-op is a big draw for this title.

 

[Jedi2016]

Even the PC version drops now and again. I had down to around 15fps a couple weeks ago when there were around 85 enemies on screen. That's pretty rare, though. If the PS4/XBO can keep it higher than that under that kind of stress, that's not bad, IMO.

 

[specwarGP2]

IME, this is true. It's 60fps 95+% of the time through 15+ hours, but there are the few drops when things get nuts in the later stages.

Now, add in 3 more co-op players, all casting various complex graphical effects on top of that - it's questionable how much more "headroom" the PS4 would have when co-op is a big draw for this title.

Its the areas that have a fog - lvl 2 sewers (high 50s), lvl 5 swamp area before first checkpoint (low 50s - high 40s), lvl 5 area where you free myriam (mid 50s). With one player using crusader darklight build.

Probably drops a lot more when you have multiple crusaders and WDs spamming theirs skills at the same time.

FYI playing on a Sony TV on impulse mode (BFI) so any framerate drop is very noticeable.

 

[djskribbles]

DF should test these areas.

 

[shredenvain]

Yes they should go back and test the Ps4 version but they most likely wont.
Its all good though both versions seem to provide a great experience with occasional drops to the mid 50s. The Ps4 version seems to be more consistent.

 

[TomRL]

"That which can be asserted without evidence can be dismissed without evidence."

Hitchens ftw

 

[djskribbles]

Yes they should go back and test the Ps4 version but they most likely wont.
Its all good though both versions seem to provide a great experience with occasional drops to the mid 50s. The Ps4 version seems to be more consistent.

Why just the PS4 version? That wouldn't be fair to only test the PS4 version during parts that really stress the engine.

From what I gather from people who have played the PS4 version, it doesn't have occasional drops like XB1; it runs at 60fps 99% of the time. But there are a few parts that seem to really stress the engine and cause frame drops. We can already see the XB1 struggling to hold 60fps for an extended period of time in less stressful parts. I want to see how both versions run in the above mentioned parts.

 

[shredenvain]

Why just the PS4 version? That wouldn't be fair to only test the PS4 version during parts that really stress the engine.

From what I gather from people who have played the PS4 version, it doesn't have occasional drops like XB1; it runs at 60fps 99% of the time. But there are a few parts that seem to really stress the engine and cause frame drops. We can already see the XB1 struggling to hold 60fps for an extended period of time in less stressful parts. I want to see how both versions run in the above mentioned parts.

iThere is no doubt the Ps4 version has the higher on average framerate.
The reason for the discussion was that Df admitted to spending several hours pouring over the Xbox One footage looking for frame drops. They should have done the same for the Ps4 version. Infact they spend very little time discussing the ps4 version in the final faceoff.
There are a few B3D members saying that the Ps4 version drops frames in certain places on the previous page. Yet df doesnt go into detail about it at all. There is no doubt from the faceoff that the Ps4 consistently hits 60 fps.

 

[djskribbles]

DF probably looked for spots that caused frame drops on XB1 because more than likely they would stress the PS4 version as well. However the PS4 version holds a steady 60fps in all of the tested footage.

The people in this thread are some of the people I was referring to. There appears to be a few particular moments later in the game that really stress the engine, enough to even cause a few drops in the PS4 version. More than likely these drops will be present on the XB1 version as well. These are the spots I would be interested to see tested, so that we can see how the two versions handle them.

 

[Deleted member 11852]

There are a few B3D members saying that the Ps4 version drops frames in certain places on the previous page. Yet df doesnt go into detail about it at all. There is no doubt from the faceoff that the Ps4 consistently hits 60 fps.

The reason seems to be that DF barely played the game at all, i.e. they didn't get to the areas that would have put the engine on both machines under stress.

Given the game starts slow (I've only played the PC/Mac version), testing those starting areas of the game is borderline pointless because the meat of the game is in the higher-populated areas of the game when things get crazy

 

[Malo]

I'd have to agree, based on some of my 4 player Rifts on PC, especially with a WD around. Even on my decent overclocked i5/7870 PC the framerate can tank. That would be a better test between the systems to really distinguish them.

 

[Arwin]

iThere is no doubt the Ps4 version has the higher on average framerate.
The reason for the discussion was that Df admitted to spending several hours pouring over the Xbox One footage looking for frame drops. They should have done the same for the Ps4 version. Infact they spend very little time discussing the ps4 version in the final faceoff.
There are a few B3D members saying that the Ps4 version drops frames in certain places on the previous page. Yet df doesnt go into detail about it at all. There is no doubt from the faceoff that the Ps4 consistently hits 60 fps.

The DF article mentions that the PS4 version can still drop some frames when new data is being streamed in or netcode is doing something.

 

[orangpelupa]

frame drop due to netcode, isnt that new and only happen in D3?