The Caveat Wars: Gap Width Measurements *spin ad nauseam*

[function]

I think it's most interesting to compare the evolution of game engines, game series', and games from the same developer. You actually have something reasonably meaningful to compare over time this way, instead of a pick and mix crapshoot where you can pretend the facts say anything that you want. Much better to look at:

MGS GZ -> MGS 5
Watch Dogs 1 -> Watch Dogs 2
Tomb Raider DE -> ROTTR
COD whatever -> COD whatever
etc

The general case is clearly that X1 had a very rocky start and things improved massively in the first couple of years, although not all games show this depending on how .

One of the most interesting things to look at is DICE's output. One because they are one of the most technically accomplished studios in the world, two because because they aren't afraid to give kick ass presentations at GDC and three because they have three titles out on these platforms now spanning a period of time each with incremental improvements.

If you look at DICE's output you can see a clear move from much lower res and much lower framerate (with bonus missing effects on X1), through to simply lower resolutions but with the edge on frame rate.

 

[Nesh]

The general picture is quite clear for me.

We just have to read/see DF articles/videos : http://www.eurogamer.net/?topic=digital_foundry + https://www.youtube.com/user/DigitalFoundry/videos

Games where the resolution is not the same from the most recent comparisons until the last E3 (june14/16 2016) : Dishonored 2, WD2, BF1, Infinite Warfare, TF2, Mafia 3, Deux Ex, Batman RE, F1 2016, Carmagedon, Tellate Batman

Games where the resolution is the same from the most recent comparisons until the last E3 : ROTR, Asseto Corsa, Skyrim, Bioshok collection, Dead Rising RE, RE4 RE, Marvel ultimate alliance, Hawken, RE5 RE, Lego Star Wars, Mighty N°9

I didn't count the patched games. I put the results that should not exist in black (obvious poor optimization) otherwise the trend is quite consistent.

The difference between TR DE and ROTR is simple : TR DE is capped at 30fps on XB1 and capped at 60fps on PS4. In ROTR both games are capped at 30fps. So, you can't know what would be the difference if both games were capped at 60fps.

And you didn't understand me : i said that the gap was about the same as before. Obviously, i don't count the 1080p/720p scenario because it's not relevant to me. Also, for a game like BF1 where the difference is smaller than the usual 900p/1080p difference + close framerate, you have games like Doom, Black Ops 3, TF2, Project Cars, etc. There are more AAA games above the normal difference than the contrary (below the normal difference).

I could continue with other arguments, but people might think that i'm going in a war thing while i only try to have a rational approach based on facts.

Ok, this was my last answer even though i already said it before.

Do you agree that there are more games today that demostrate a smaller gap than they used to of which some are demanding? Yes or no?
Facts tell you that there are games that demonstrate the same gap and games that demostrate smaller gap. This is the fact. This is the rational approach. End of story.
As I said earlier I am not trying to convince that the XBOX will eliminate the gap completely. But it does show a significant improvement. Again I fail to understand what is your point. In the past the XBOX One was struggling even more.
Your own examples (plus others you didnt mention) show this.

 

[Tkumpathenurpahl]

I don't think the PS4 XBoxOne debate is destined to really go anywhere. Both have such similar architectures, but one with half the GPU, that the performance barometer will only ever wiggle.

Last generation, at least it was interesting to see how the Cell was used to prop up the PS3's shitty GPU. I suppose we'll see something of that nature if the Switch is successful: we get to see a new mobile chip take on... older laptop chips.

 

[Nesh]

From what I see, demanding games still show the same typical ~900p/1080p-type gap, and there are a few rare cases where the gap is either slightly smaller, or slightly bigger than usual. It depends on which console was the lead platform, and also whether or not the game is CPU limited. If the game is CPU limited and XB1 is the lead platform, there's a good chance that XB1 can perform closer to PS4. If not, then it's usually the same old situation.

If the game is CPU limited or not doesnt make the point any differfent. If the XBOX can benefit from some technical advantage whatever that reason is, what matters is the result.

 

[Globalisateur]

...
One of the most interesting things to look at is DICE's output. One because they are one of the most technically accomplished studios in the world, two because because they aren't afraid to give kick ass presentations at GDC and three because they have three titles out on these platforms now spanning a period of time each with incremental improvements.

If you look at DICE's output you can see a clear move from much lower res and much lower framerate (with bonus missing effects on X1), through to simply lower resolutions but with the edge on frame rate.

It's just the case for one game though which is very CPU limited in some modes: BF1. And minimum resolutions still show same usual gap. (720p / 900p).

Also in the case of BF1 different modes than the much discussed CPU heavy 64 players (like Deathmatch) are actually running very similarly on PS4 and XB1.

 

[djskribbles]

If the game is CPU limited or not doesnt make the point any differfent. If the XBOX can benefit from some technical advantage whatever that reason is, what matters is the result.

Ok I'm jumping into this debate a little late. So what is the point you're trying to make? It has been known since day 1 that the XB1 has a slight CPU clock advantage. Either not many games are CPU limited, or developers are finding workarounds, because we aren't seeing that many games where XB1 has a performance advantage. And even if there is, it's just a few FPS at most.

 

[Nesh]

Ok I'm jumping into this debate a little late. So what is the point you're trying to make? It has been known since day 1 that the XB1 has a slight CPU clock advantage. Either not many games are CPU limited, or developers are finding workarounds, because we aren't seeing that many games where XB1 has a performance advantage. And even if there is, it's just a few FPS at most.

My point is, is the result that counts. Explaining that a game performs better on the XB1 because it is more CPU intensive as a counterargument is irrelevant. We dont care how. The point is what is displayed on screen. We arent discussing strictly about a specific hardware component where one console has a clear edge over the other.

Just to sum up, not all games demonstrate the same huge gap where the PS4 has a 50%ish lead in resolution or in performance or in a combination of the two.
There is a very significant list of games that show a much smaller gap or a negligible gap in addition to the games that show a gap consistent with the theroetical performance advantage.

And for the record I am a PS4 owner only, I dont regret my purchase, nor do I regret not purchasing an XB1, so I have no reason to be biased. This is clearly what I see.

edit: Now that I reread your post I think we are saying the same thing

 

[Nesh]

This is a list of some of the games that demonstrate a smaller gap. Some of these titles are demanding. If the extra performance partly doesn't show due to framerate caps is irrelevant. We know the PS4 is more powerful, we arent questioning that. What we get back as consumers is what eventually matters.
The question is: is the PS4 capped because the potential framerate advantage is not large or stable enough for a good gameplay experience?
Or is it a policy to avoid showing a bigger gap between the two versions and commercially harm one of the platform holders?

Battlefiled 1 Resolution improved to dynamic with a base res at 900p over the 720p found in Battlefront and Battlefield 4, plus a slighly better performance on XB1
Rise of the Tomb Raider 1080p and 30fps on both. Slight improvements in certain areas on PS4. No framerate improvement like in TRE which shows that framerate might have been more unstable. How well the PS4 might have performed uncapped is hypothetical
Dirt Rally Almost or perfectly Identical
Overwatch Almost or perfectly Identical with slight polish on PS4
Mad Max Almost or perfectly Identical
Division PS4 version permanently operating at 1080p, XBOX One version operating at 792x1008 and 1728x972 in outdoor areas,1080p in simple and indoor areas, an improvement over the 1600x900 found in other games.
Hitman 1080p 30fps on both. Slight performance edge on PS4
Need for Speed Almost or perfectly Identical
Transformers 1080p, 60fps on both with more occasional framerate drops on XB1
Dragon Age Inquisition higher resolution on PS4, better performance on XB1
Lego the Hobbit Almost or perfectly Identical
Fallout 4 Almost or perfectly Identical
The Crew Almost or perfectly Identical
Destiny Identical with a very slight performance edge on PS4
GTA V Almost or perfectly Identical
MK X 1080p on PS4 and dynamic resolution on XB1 ranging at around 1360x1080 to 1344x1080.
Assasins Creed Unity 900p on both, slight performance advantage on XB1
Assasins Creed Syndicate 900p on both, slight performance advantage on PS4
Alien Isolation 1080p on both, some small performance disadvantage on XB1
F1 2016 1080P on PS4 and 1440x1080p on XB1, some minor graphical refinements on PS4, some framerate drops on PS4 (low 50s on PS4) noticeable framerate drops on XB1 on dynamic weather conditions (45fps lowest)
Far Cry Pimal 1080p on PS4 and 1440x1080p on XB1, some minor visual improvements on PS4,
Far Cry 4 1080p on PS4 and 1440x1080p on XB1, some minor performance advantage on XBOX One
Titan Fall 2 PS4 resolves at around 1000p, and Xbox One around 810-828p. Clearly smaller gap than the typical 1080p vs 900p. Also lets not forget that Titanfall 1 was 1408x792 and suffering with significant frame drops on XBOne. This time performance is also improved
Hawken 1080p on both, slight differences in performance (different implementation of V-Sync)

 

[djskribbles]

Lego, Destiny, Transformers and Overwatch aren't what I would consider demanding games. My PS4 is as quiet as it is idling while running Destiny. (edit: my bad, missed the part where you said that only some of these games are demanding.)
Division has a slight performance advantage on PS4.
MK X is basically 900p on XB1 in terms of the number of pixels.
For Titanfall 2, the pixel gap is actually larger than 1080p vs 900p.
Far Cry and the F1 games are just slightly higher than 900p on XB1 (+8%), but as you said, they all have more refined visuals on PS4. Plus as far as F1 goes, it has at least a 15% framerate advantage.
Hawken seems to have a considerably higher FPS on PS4 (looks like ~+20%), more refined AO and is completely V-sync'd on PS4, whereas XB1 constantly tears. XB1 does feel more consistent due to the V-sync on PS4 and uncapped framerate. If they disabled v-sync or went with an adaptive V-sync on PS4, the performance gap would probably be larger. Either way, this game doesn't look all that demanding either and I wouldn't use this as an example for anything.

MS had marketing rights to many of those games (BF1, Division, Dragon Age, Fallout 4, RoTR, AC Unity, TF2), and at least a few of those games are probably CPU limited (BF1, Hitman and AC Unity to name a few).

 

[Nesh]

Lego, Destiny, Transformers and Overwatch aren't what I would consider demanding games. My PS4 is as quiet as it is idling while running Destiny.

.....Some of these titles are demanding.

MS had marketing rights to many of those games (BF1, Division, Dragon Age, Fallout 4, RoTR, AC Unity, TF2), and at least a few of those games are probably CPU limited (BF1, Hitman and AC Unity to name a few).

It is irrelevant if they are CPU limited or not. If the XBOX One can output improved results for whatever reasons in some occasions so be it.
CPU or GPU or memory or whatever is the case they are part of the whole that contribute to performance.

 

[Recop]

No discussion. Just to correct some facts :

Rise of the Tomb Raider : 1440x1080 during cutscenes + motion blur on/off + missing shadows + worse AO + worse specular reflection + adaptive Vsync + worse input lag + worse framerate on XB1

http://www.eurogamer.net/articles/digitalfoundry-2015-rise-of-the-tomb-raider-tech-analysis

http://www.eurogamer.net/articles/digitalfoundry-2016-rise-of-the-tomb-raider-ps4-face-off

Division : Dynamic resolution from 1728x972 to 1920x1080 + adaptive Vsync + worse performances on XB1

http://www.eurogamer.net/articles/digitalfoundry-2016-the-division-face-off

Lego the Hobbit : Downsampled from 1920x1280 on PS4

http://www.eurogamer.net/articles/digitalfoundry-2014-lego-the-hobbit-face-off

GTA V : Massive foliage difference in some areas + worse performances on XB1

http://i.imgur.com/Csa01f2.gif

http://i.imgur.com/RWA2KMR.gif

http://www.eurogamer.net/articles/digitalfoundry-2015-has-rockstar-really-downgraded-gta-5

MK X : Resolution at around 1344/1360x1080 + worse performances on XB1

http://www.eurogamer.net/articles/didigtalfoundry-2015-mortal-kombat-x-face-off

Alien Isolation : adaptive Vsync + much worse performances on XB1

http://www.eurogamer.net/articles/digitalfoundry-2014-alien-isolation-face-off

F1 2016 : 1440x1080p + worse mortion blur + no AO + much worse performances on XB1

http://www.eurogamer.net/articles/digitalfoundry-2016-f1-2016-face-off

Far Cry 4 : better performances on PS4 according to NXGamer (he made a much longer test)

Titanfall 2 : Dynamic resolution from 480p to 828p on XB1 against 720p to 1000p on PS4

http://www.eurogamer.net/articles/digitalfoundry-2016-titanfall-2-face-off

Hawken : adaptive Vsync + worse AO + worse performances but better frame pacing on XB1

http://www.eurogamer.net/articles/digitalfoundry-2016-hawken-performance-analysis

 

[Nesh]

No discussion. Just to correct some facts :

Rise of the Tomb Raider : 1440x1080 during cutscenes + motion blur on/off + missing shadows + worse AO + worse specular reflection + worse input lag + worse framerate on XB1

http://www.eurogamer.net/articles/digitalfoundry-2016-rise-of-the-tomb-raider-ps4-face-off

Division : Dynamic resolution from 1728x972 to 1920x1080 + no Vsync + worse performances on XB1

http://www.eurogamer.net/articles/digitalfoundry-2016-the-division-face-off

Lego the Hobbit : Downsampled from 1920x1280 on PS4

http://www.eurogamer.net/articles/digitalfoundry-2014-lego-the-hobbit-face-off

GTA V : Massive foliage difference in some areas + worse performances on XB1

http://i.imgur.com/Csa01f2.gif

http://i.imgur.com/RWA2KMR.gif

http://www.eurogamer.net/articles/digitalfoundry-2015-has-rockstar-really-downgraded-gta-5

MK X : Resolution at around 1344/1360x1080 + worse performances on XB1

http://www.eurogamer.net/articles/didigtalfoundry-2015-mortal-kombat-x-face-off

Alien Isolation : no Vsync + much worse performances on XB1

http://www.eurogamer.net/articles/digitalfoundry-2014-alien-isolation-face-off

F1 2016 : 1440x1080p + worse mortion blur + no AO + much worse performances on XB1

http://www.eurogamer.net/articles/digitalfoundry-2016-f1-2016-face-off

Far Cry 4 : better performances on PS4 according to NXGamer (he made a much longer test)

Titanfall 2 : Dynamic resolution from 480p to 828p on XB1 against 720p to 1000p on PS4

http://www.eurogamer.net/articles/digitalfoundry-2016-titanfall-2-face-off

Hawken : no Vsync + worse AO + worse performances but better frame pacing on XB1

http://www.eurogamer.net/articles/digitalfoundry-2016-hawken-performance-analysis

You are using the word "worse" in a way (perhaps deliberately) that you imply that the differences are bigger than they are and without providing proper context ulike the Eurogamer links you provide.

 

[Recop]

You are using the word "worse" in a way (perhaps deliberately) that you imply that the differences are bigger than they are and without providing proper context ulike the Eurogamer links you provide.

I'm not a native english speaker. To me, "worse" is the contrary of "better"... i don't try to imply anything, this is why i put the corresponding links that anyone can check...

And i said "much worse" when the differences are substantial.

 

[Globalisateur]

This is a list of some of the games that demonstrate a smaller gap. Some of these titles are demanding. If the extra performance partly doesn't show due to framerate caps is irrelevant. We know the PS4 is more powerful, we arent questioning that. What we get back as consumers is what eventually matters.
The question is: is the PS4 capped because the potential framerate advantage is not large or stable enough for a good gameplay experience?
Or is it a policy to avoid showing a bigger gap between the two versions and commercially harm one of the platform holders?

Battlefiled 1 Resolution improved to dynamic with a base res at 900p over the 720p found in Battlefront and Battlefield 4, plus a slighly better performance on XB1
Rise of the Tomb Raider 1080p and 30fps on both. Slight improvements in certain areas on PS4. No framerate improvement like in TRE which shows that framerate might have been more unstable. How well the PS4 might have performed uncapped is hypothetical
Dirt Rally Almost or perfectly Identical
Overwatch Almost or perfectly Identical with slight polish on PS4
Mad Max Almost or perfectly Identical
Division PS4 version permanently operating at 1080p, XBOX One version operating at 792x1008 and 1728x972 in outdoor areas,1080p in simple and indoor areas, an improvement over the 1600x900 found in other games.
Hitman 1080p 30fps on both. Slight performance edge on PS4
Need for Speed Almost or perfectly Identical
Transformers 1080p, 60fps on both with more occasional framerate drops on XB1
Dragon Age Inquisition higher resolution on PS4, better performance on XB1
Lego the Hobbit Almost or perfectly Identical
Fallout 4 Almost or perfectly Identical
The Crew Almost or perfectly Identical
Destiny Identical with a very slight performance edge on PS4
GTA V Almost or perfectly Identical
MK X 1080p on PS4 and dynamic resolution on XB1 ranging at around 1360x1080 to 1344x1080.
Assasins Creed Unity 900p on both, slight performance advantage on XB1
Assasins Creed Syndicate 900p on both, slight performance advantage on PS4
Alien Isolation 1080p on both, some small performance disadvantage on XB1
F1 2016 1080P on PS4 and 1440x1080p on XB1, some minor graphical refinements on PS4, some framerate drops on PS4 (low 50s on PS4) noticeable framerate drops on XB1 on dynamic weather conditions (45fps lowest)
Far Cry Pimal 1080p on PS4 and 1440x1080p on XB1, some minor visual improvements on PS4,
Far Cry 4 1080p on PS4 and 1440x1080p on XB1, some minor performance advantage on XBOX One
Titan Fall 2 PS4 resolves at around 1000p, and Xbox One around 810-828p. Clearly smaller gap than the typical 1080p vs 900p. Also lets not forget that Titanfall 1 was 1408x792 and suffering with significant frame drops on XBOne. This time performance is also improved
Hawken 1080p on both, slight differences in performance (different implementation of V-Sync)

As @Recop just wrote, you missed a few things, from the top of my head.

Rise of the Tomb Raider 1080p and 30fps on both. Slight improvements in certain areas on PS4. No framerate improvement like in TRE which shows that framerate might have been more unstable.

No. The framerate is basically locked on OG PS4 while the game sometimes drops at ~25fps on XB1 and often tears.

Overwatch Almost or perfectly Identical with slight polish on PS4

The game is basically locked at 1080p on PS4 while the XB1 game is dropping much more often its resolution and does have more framerate drops than PS4.

Dragon Age Inquisition higher resolution on PS4, better performance on XB1

Better effects on PS4 version too

Fallout 4 Almost or perfectly Identical

The game does in fact have a dynamical resolution on XB1 not seen on PS4.

GTA V Almost or perfectly Identical

The game has sustained drops on some GPU limited areas on XB1s not seen on PS4.

Far Cry 4 1080p on PS4 and 1440x1080p on XB1, some minor performance advantage on XBOX One

No. That was just using one scene. Others framerate videos showed a very different picture: The game performs as a whole better on PS4.

Titan Fall 2 PS4 resolves at around 1000p, and Xbox One around 810-828p. Clearly smaller gap than the typical 1080p vs 900p.

It's actually a very slightly bigger gap than 1080p / 900p even taking 828p for XB1 (~46% vs 44%). But because of the dynamic res it's in fact very hard to compare both games.

 

[iroboto]

I'm not a native english speaker. To me, "worse" is the contrary of "better"... i don't try to imply anything, this is why i put the corresponding links that anyone can check...

And i said "much worse" when the differences are substantial.

That's alright. You aren't the only non native speaker here.

It seems like everyone wants for this to keep going, so definitely going to request from mods a spin-off into "is the gap closing" or something. This certainly is not DF talk. It's certainly about using DF for something , but it's not DF game talk.

 

[milk]

if I were a mod I would have named the tread "the tight gap"

 

[ultragpu]

if I were a mod I would have named the tread "the tight gap"

Almost read that as the thigh gap

 

[BRiT]

That's reserved for the next DOA pixel counting thread Zilla creates.

 

[milk]

Almost read that as the thigh gap

that was the point.

 

[Commenter]

Microsoft are a software company so it's no surprise they created a system more biased towards the cpu, with DDR3 lower bandwidth, lower latency ram vs DDR5 high bandwidth, higher latency ram, and Sony being a primarily hardware company chose to focus on the technical specs. In many ways the One is a more balanced system, and let's not forget that the edram is actually an advantage in games like BF1 which are heavy on alpha effects.