Encyclopedia Brown & The Mysterious Case of the PS4 and the Missing Anisotropic Filtering

Status
Not open for further replies.
function said:
You're mistaken.

Exclusives are not relevant to any "versus" discussion, but are relevant to the "does AF cost" debate. Which is actually the debate that's interesting.

For example, Drive Club using reduced / low AF in-game, but high in photo mode, shows that high AF is not free. It is not even effectively free. And this is for a big budget, single platform, long-in-development, first party technical showcase.

There's is an insane amount of effort going into not acknowledging this.
Click to expand...

AF is not free I never said this. I am sure all PS4 and Xbox One game can have 16xAF but some sacrifice elsewhere need to be done.

After I said than for DMC, Strider or Unfinished Swan The PS4 has less AF than PS3 and it is strange.
 
Exclusives games are different beast. They do the compromise they want...

The only exclusive game to have 16xAF is TLOUR and it is a remaster of a PS3 title...

After the only example of explanation about porting a title on PS4 is The Crew. They improve GNMX on this title and the number of people working on the port was low...

It is easier to port a PC title to the Xbox One. I think the main machine is PC this time not like last gen...
 
chris1515 said:
AF is not free I never said this. I am sure all PS4 and Xbox One game can have 16xAF but some sacrifice elsewhere need to be done.

After I said than for DMC, Strider or Unfinished Swan The PS4 has less AF than PS3 and it is strange.
Click to expand...

Well yes it certainly is strange, I'll give you that! And for some games or engines it might be an API issue or bug, but for Spark and Unreal Engine and Cryengine all to have exactly the same bug or misunderstand of the API / documentation and for Sony (with 20+ million PS4s sold) to be unable to explain this and for no-one in the game and engine developers to have noticed that GCN isn't doing what GCN does ...?

... and for even big budget first party games to be making very similar compromises by dropping AF to low levels in demanding games (e.g. Driveclub and The Order - which on some surfaces seems to have low AF and in other parts seems to have none).

It'll be interesting to see both Unreal Engine 4 and Uncharted 4 roll round.
 
function said:
Well yes it certainly is strange, I'll give you that! And for some games or engines it might be an API issue or bug, but for Spark and Unreal Engine and Cryengine all to have exactly the same bug or misunderstand of the API / documentation and for Sony (with 20+ million PS4s sold) to be unable to explain this and for no-one in the game and engine developers to have noticed that GCN isn't doing what GCN does ...?

... and for even big budget first party games to be making very similar compromises by dropping AF to low levels in demanding games (e.g. Driveclub and The Order - which on some surfaces seems to have low AF and in other parts seems to have none).

It'll be interesting to see both Unreal Engine 4 and Uncharted 4 roll round.
Click to expand...

AF on exclusive is a choice. I can tell some non sense like Xbox One is not good on AF because Forza Horizon 2 has low AF... It is a compromise like everyone do on machine with fixed power...

On KZ SF AF an exclusive demanding game AF is pretty good 8x for some texture like ground... And Alien Isolation is 16xAF on PS4 with good framerate.

Only devs like Fafalada on GAF have an idea about the problem...
 
DrJay24 said:
Apparently Dying light patched in better AF. I noticed it was patched the other day. I'm guessing it didn't affect the performance, which was always good for an open world game.

http://67.227.255.239/forum/showpost.php?p=155567641&postcount=1022
Click to expand...
Are you sure the game doesn't run at 20fps now instead of 30fps? That should be roughly the bandwidth the GPU should sacrifice going from 2xAF to 8xAF based of what I am reading.

Some people can't distinguish (or with difficulty) between 20fps capped and 30fps capped on a 60hz monitor, for instance Zelda Ocarina of Time (N64) and Metal gear Solid 3 (3DS) are capped at 20fps. (60 being divisible by 20 so no additional judder because of consecutive identical frames).
 
Can we just focus on the issue here and discuss why some games like Unfinished Swan don't have AF on PS4? Did anyone notice? It's pretty awful.
 
Globalisateur said:
Are you sure the game doesn't run at 20fps now instead of 30fps? That should be roughly the bandwidth the GPU should sacrifice going from 2xAF to 8xAF based of what I am reading.

Some people can't distinguish (or with difficulty) between 20fps capped and 30fps capped on a 60hz monitor, for instance Zelda Ocarina of Time (N64) and Metal gear Solid 3 (3DS) are capped at 20fps. (60 being divisible by 20 so no additional judder because of consecutive identical frames).
Click to expand...

You can calculate the bandwidth sacrificed on a closed system by code you do not have access to (along with required profiling tools) and then translate that bandwidth into a 10fps delta? Call me skeptical...

And no the framerate did not change, I was playing for hours last night. Anyone can see the tearing or stuttering induced by drop to 20fps.
 
Globalisateur said:
Are you sure the game doesn't run at 20fps now instead of 30fps? That should be roughly the bandwidth the GPU should sacrifice going from 2xAF to 8xAF based of what I am reading.

Some people can't distinguish (or with difficulty) between 20fps capped and 30fps capped on a 60hz monitor, for instance Zelda Ocarina of Time (N64) and Metal gear Solid 3 (3DS) are capped at 20fps. (60 being divisible by 20 so no additional judder because of consecutive identical frames).
Click to expand...
Sorry. I love you, but WHAT?!
 
Globalisateur said:
Are you sure the game doesn't run at 20fps now instead of 30fps? That should be roughly the bandwidth the GPU should sacrifice going from 2xAF to 8xAF based of what I am reading.
Click to expand...
It would have to be a fairly nasty pathological case to cause such a large difference in overall performance.

At any rate, I'm confused as to how you'd calculate what the difference ought to be, considering there's an incredibly complex hardware system including a cache structure, with demands varying by scene depending on a huge number of factors.
 
I wouldn't be so quick to dismiss the quote from Fafalada. Though it would be nice for further clarification, if possible. The man was a wizard with the PS2 hardware.
 
Status
Not open for further replies.

Similar threads

Shortbread
The Great PS4 Missing AF Mystery *spawn
6 7 8
Replies
143
Views
18K
AlNom
A
A
Dynamic framebuffer resolution
2 3
Replies
42
Views
17K
chris1515
C
Back
Top