Support NeoGAF

[EGM1966]

I was wondering if they would "go the other way" in this area. Seems they will. In their shoes I would have too. If Scorpio is perceived as just MS version of Pro it's not going to do much for them other than make solid revenue from existing install base upgrading.

They need some points of difference and trying to give view its more open than Pro is one.

I doubt it'll result in much difference in actual game development decisons but I get why they'd take this PR route.

 

[oneils]

I can't see destiny pvp being 30fps on one Xbox and 60fps on another. Doesn't make sense. Pvp, at least, will probably have parity between the two xboxes.

 

[Bits N Pieces]

It will only be first party and independent developers that will really push the new hardware. Expect performance parity from the likes of EA, Ubisoft, Activision etc.

 

[Shiggy]

Wouldn't some mandating be good? Otherwise why buy a Scorpio if developers aren't making use of the increased hardware specs?

 

[c0de]

The additional technological ressources of Scorpio may be used, without any restriction from MS's side, to up the resolution and add some effects, but as a consequence the game may run worse on Scorpio than Xbox One, which might be something some owners of Scorpio would not be too happy about. I don't get why, in a discussion about this, you outline exactly in what way Scorpio is more powerful than Xbox One.

How exactly would a game run worse on Scorpio than on Xbox One?

 

[bennibop]

You are not going to see 60fps and Scorpio and 30fps on x1 mp no developer in their right mind would do this.

Microsoft policy is no different to Sony in regards letting developer decide how to use enhancements. The developer are not going to let part of their audience have a large advantage over part of a user Base.

 

[Floody]

The whole point is that it should be up to the developers. If they want to do it then let them. They want to take the risk let them. Sony and Microsoft shouldn't ever stick its nose into their business. Mandating stuff like that is stupid. Let them mandate in forcing devs to do achievements or small stuff like that. And even that is dumb tbh.

Also the 970 released like 3 years ago and its still gonna be roughly in the same ballpark of power as the Pro/Scorpio depending on your CPU. That even more just shows how old and outdated the vanilla 2013 consoles are.

Ah, get you now, my bad. But if developer wouldn't do it anyway, don't see what either MS or Sony would gain by coming out and saying "developers can leave the majority of you behind if they want", just be bad press.

I do agree if in like 2 years there's no PS5/Xbox 1-2 that Sony and MS should allow exclusives on Pro & Scorpio though, just don't think it's needed yet.

 

[Yoshi]

How exactly would a game run worse on Scorpio than on Xbox One?

By using additional ressources on higher resolution textures, higher screen resolution, additional effects, to the detriment of performance because of pushing Scorpio a bit further than the base version is pushing the Xbox One. I really don't see what's so hard to understand here, as far as I know, worse performance has been achieved on PS4 Pro in the past as well, when compared to PS4.

 

[Yoshi]

Wouldn't some mandating be good? Otherwise why buy a Scorpio if developers aren't making use of the increased hardware specs?

Every developer would require a Scorpio dev kit as well in that case to properly test Scorpio versions, which could be detrimental to indies especially in the beginning.

 

[mrklaw]

I was wondering if they would "go the other way" in this area. Seems they will. In their shoes I would have too. If Scorpio is perceived as just MS version of Pro it's not going to do much for them other than make solid revenue from existing install base upgrading.

They need some points of difference and trying to give view its more open than Pro is one.

I doubt it'll result in much difference in actual game development decisons but I get why they'd take this PR route.

How are they going 'the othe way' - isn't this the same as Sony's policies? You can and do have higher framerates on pro games. Unless you're arguing that MS' comments mean devs could release games that perform *worse* on Scorpio which I really can't see happening.

 

[oneils]

How exactly would a game run worse on Scorpio than on Xbox One?

Developers might prioritize graphics over performance.

 

[c0de]

By using additional ressources on higher resolution textures, higher screen resolution, additional effects, to the detriment of performance because of pushing Scorpio a bit further than the base version is pushing the Xbox One. I really don't see what's so hard to understand here, as far as I know, worse performance has been achieved on PS4 Pro in the past as well, when compared to PS4.

You can push Scorpio above what the bone can do, easily.
But then it's of course the fault of the developer.
An Xbox One game that releases on Scorpio only with a higher resolution should not perform worse but instead better.

 

[MosquitoSmasher]

Why don't they do it like Gears 4? Where it's 30fps in Horde and 60fps in MP. So for Destiny 60fps in MP or crucible they call it? And 30 for story.

 

[Hesemonni]

I'm really interested how they word their "true 4k" choice of words at E3.

 

[Tertullian]

Most people seem to be saying this is good, but I'm a little confused. Don't PS4 Pro owners actually wish there were more rules in certain areas? Like, all those games that arbitrarily disallow downsampling for 1080p TVs. Wouldn't it be better if Scorpio were required to allow downsampling on games that can run at higher resolutions, say?

 

[geordiemp]

This kool-aid is really syrupy

The Scorpio's CPU is 9% faster than Pro's

Thats not fair, its 9.52 %.

Posters thinking that Pro or Scorpio can do PC frame rates is hilarious.

An I5 has 1.7 x the IPC of Jaguar, so a 3.4 Ghz I5 is something like 3 x faster performance than Scopio or pro.

THREE TIMES FASTER. Let that sink in, and lets hope in 2019 we get proper CPU's of i5 class on next gen....ZEN save us !

I have a Ps4 pro, and at 2.1 Ghz and crap IPC., I do not expect much from the limitations of the tiny and old technology Jag chip.

Most people seem to be saying this is good, but I'm a little confused. Don't PS4 Pro owners actually wish there were more rules in certain areas? Like, all those games that arbitrarily disallow downsampling for 1080p TVs. Wouldn't it be better if Scorpio were required to allow downsampling on games that can run at higher resolutions, say?

Only rule I know of is 1080p and Pro must be better, that is the same as saying do what you like, all pro and scorpio can do is better IQ, its not often we get 60 FPS in 30 FPS games and only for GPU limited games.

Jaguar cannot do large map, lot of enemies and high AI games at 60 FPS anyway, even if it was 2.5 GHz. Jaguar is the real leveller here, pro and scoprio maybe have 20 or 30 % more than OG models, but they are within 10 % of each other.

 

[oneils]

You can push Scorpio above what the bone can do, easily.
But then it's of course the fault of the developer.
An Xbox One game that releases on Scorpio only with a higher resolution should not perform worse but instead better.

Sure, but that's not what you asked.

 

[c0de]

This kool-aid is really syrupy

The Scorpio's CPU is 9% faster than Pro's

While I highly doubt that the game will be 60 fps, the clock advantage is per core so it's definitely more than that. And yes, you also cannot just all the frequencies but it's also false to assume a single core processor.

 

[oneils]

I know.

I"m saying the Destiny Marketing could by why Bungie won't talk about the Scorpio version.

Yeah, there really isn't any incentive to discussing a game being released on a platform that itself has not been released. By the time destiny 2 is out, there will be over 100million xboxes and ps4s out there. They will focus their energy on that. I even wonder if they are making a version for Scorpio at the moment.

 

[Ushay]

FUCK.YES. This should be a given based on how they anticipate the "no generations" generation to go. Also, console parity in a world of 900p vs full 4k was a laughable thought.

This is what stands out to me the most, how can they (Bungie) be thinking about parity when there will be such a huge gulf in fidelity.

 

[napata]

While I highly doubt that the game will be 60 fps, the clock advantage is per core so it's definitely more than that. And yes, you also cannot just all the frequencies but it's also false to assume a single core processor.

What? 9% is the upper limit. Depending on the bottleneck a CPU that's clocked 9% higher will give you at minimum no performance gains and at maximum 9%. Usually it's somewhere in between.

Why would you think it would more than 9%? Where would this magical increase in performance come from?

 

[CerebralTiger]

Don't confuse bandwith with latency. Latency differs a lot between connection types. Player A can see event X a few frames before player B sees it due to connection latency on top of the latency introduced by the display.

Higher latency or ping is usually the result of geographical distances or just poor routing/a higher number of hops in a traceroute between connection A and B. There's no workaround for this, and it's the only uncontrollable variable here. Display lag can be reduced by playing on a low-latency monitor, but of course not everyone will ditch their big screen TVs for such an advantage.

Internet latency + display latency are already two variables in the mix. Adding variable frame rate (30/60) to it only widens the potential gap between player A and B. Definitely not a good idea imo.

 

[c0de]

What? 9% is the upper limit. Depending on the bottleneck a CPU that's clocked 9% higher will give you at minimum no performance gains and at maximum 9%. Usually it's somewhere in between.

Why would you think it would more than 9%? Where would this magical performance come from?

?
The CPU contains 2 packages with 4 cores each.
Assuming proper multi tasking and multi threading you have more than the 9% increase.

 

[Kayant]

Why don't they do it like Gears 4? Where it's 30fps in Horde and 60fps in MP. So for Destiny 60fps in MP or crucible they call it? And 30 for story.

Could be they prefer consistency across modes and/or as some people have suggest it being P2P there be requirement of a host at anytime to run some calculations etc could be a limiting factor.

 

[jroc74]

Thats not fair, its 9.52 %.

Posters thinking that Pro or Scorpio can do PC frame rates is hilarious.

An I5 has 1.7 x the IPC of Jaguar, so a 3.4 Ghz I5 is something like 3 x faster performance than Scopio or pro.

THREE TIMES FASTER. Let that sink in, and lets hope in 2019 we get proper CPU's of i5 class on next gen....ZEN save us !

I have a Ps4 pro, and at 2.1 Ghz and crap IPC., I do not expect much from the limitations of the tiny and old technology Jag chip.



Only rule I know of is 1080p and Pro must be better, that is the same as saying do what you like, all pro and scorpio can do is better IQ, its not often we get 60 FPS in 30 FPS games and only for GPU limited games.

Jaguar cannot do large map, lot of enemies and high AI games at 60 FPS anyway, even if it was 2.5 GHz. Jaguar is the real leveller here, pro and scoprio maybe have 20 or 30 % more than OG models, but they are within 10 % of each other.

1. Its amazing the misinformation some ppl have about the Pro, especially compared to the Scorpio.

2. Ppl need to keep their expectation in check about the Scorpio, lol. Focus on the ram difference between the 2, but dont expect it to be some next gen beast.

3. This post just helped me decide to get a base PS4 now and wait for the PS5.

Kids been begging me to get another PS4, I've been holding out for a Pro.

 

[geordiemp]

?
The CPU contains 2 packages with 4 cores each.
Assuming proper multi tasking and multi threading you have more than the 9% increase.

If the logic was not reliant on the previous task it could be done by GPGPU.

If 7 parallel tasks are done 9.5 % quicker, its still 9,5 %. If a serial task is done 9.5 % quicker, guess what ?

1. Its amazing the misinformation some ppl have about the Pro, especially compared to the Scorpio.

2. Ppl need to keep their expectation in check about the Scorpio, lol. Focus on the ram difference between the 2, but dont expect it to be some next gen beast.

3. This post just helped me decide to get a base PS4 now and wait for the PS5.

Kids been begging me to get another PS4, I've been holding out for a Pro.

I got 2 Pros, but god I was annoyed at the Jaguar but it was the only choice. I had the spare cash but was under no illusions.

The IQ is nice above 1080p if you have a big TV, but 1440 / 1800c / 2160p is hard to differentiate IMO on 55 inch at 8 ft. We need to shift from GPU terraflop talk to CPU IPS and speed talk as its the true bottleneck now.

BUt you get a few more games more stable at 30 as opposed to dipping under occasionally, but expectations are something else.

I had a dream of playing Bloodbourne or Witcher TYPE games at 60 FPS, but lets face it, jaguar. Maybe in 2019 / 2020

 

[oneils]

This is what stands out to me the most, how can they (Bungie) be thinking about parity when there will be such a huge gulf in fidelity.

Because gamers are babies, and bungee won't want to deal with 100million crybabies complaining that the kid down the street has a better toy than them

I'm only partly kidding, why would they want to deal with that headache? There seems to be no incentive to making multiple versions.

 

[napata]

?
The CPU contains 2 packages with 4 cores each.
Assuming proper multi tasking and multi threading you have more than the 9% increase.

So? If you add a fixed % to a group of numbers, add them all up and compare them to the sum of the previous numbers you still get that fixed %.

Imagine a 10 core CPU at 1. If you'd clock it 10% higher you'd get 10 cores at 1.1. Now if you add them all up you have 10 vs 11. What's the difference between 10 and 11? How much % is that?

I know math isn't everyone's strong point but man...

 

[IHaveIce]

Saving this thread for when they announced D2 is 30fps too.

Should be hilarious.

Well that is on bungie not MS

 

[c0de]

So? If you add a fixed % to a group of numbers, add them all up and compare them to the sum of the previous numbers you still get that fixed %.

Imagine a 10 core CPU at 1. If you'd clock it 10% higher you'd get 10 cores at 1.1. Now if you add them all up you have 10 vs 11. What's the difference between 10 and 11? How much % is that?

I know math isn't everyone's strong point but man...

You are completely right and I don't know what rode me here. Forget my brain fart.

 

[geordiemp]

You are completely right and I don't know what rode me here. Forget my brain fart.

To be fair there is out of order execution, and maybe a serial 2 step task on 2 cores could go a little bit more than 9.5 %...but its clutching at straws and roll on the death of jaguar IMO.

 

[Dee Dah Dave]

60fps is going to be out of the question most of the time anyway. But the scorpio does have variable refresh, so games can have an unlocked framerate option for that instead of being locked 30fps like the standard XB1 version.

 

[FreedomFighter]

If you can have resolution differences, why must have for parity? Makes no sense to me.

 

[dh4niel]

Enjoy your 900p future Scorpio owners!

Spoiler

/s

 

[Doffen]

Seems like a lot of people expect 4k60fps for every single game.

Only for the games that matters, Forza Horizon 3 and Smite.

 

[napata]

You are completely right and I don't know what rode me here. Forget my brain fart.

What exactly was your train of thought? Did you multiply the 9% with 8 so you got 72%? To me that seems the most common mistake someone would make. I'm always interested in what kind of mistakes people make in math.

To be fair there is out of order execution, and maybe a serial 2 step task on 2 cores could go a little bit more than 9.5 %...but its clutching at straws and roll on the death of jaguar IMO.

Could you explain this a bit more? I get that out of order execution would achieve better results than serial workloads and that you'd get closer to a 9.5% performance increase but why would it surpass it? Aren't you still limited by the speed per core in parallel tasks?

 

[dose]

You mean, like on PC?

Yep, which is like i said later is one reason why I don't do MP games.

 

[oneils]

Yep, which is like i said later is one reason why I don't do MP games.

Well, to be fair, on pc you can scale back settings by disabling aa/af/shadows/effects and dropping resolution to eliminate the advantage. Can't do that on PS4 or Xbox One.

 

[EGM1966]

How are they going 'the othe way' - isn't this the same as Sony's policies? You can and do have higher framerates on pro games. Unless you're arguing that MS' comments mean devs could release games that perform *worse* on Scorpio which I really can't see happening.

Unless I've misunderstood isn't the point that MS has declared it won't enforce MP parity whereas Sony has mandated MP parity? Thus in theory Scorpio would allow owners a frame rate advantage vs XB1 owners whereas Pro and PS4 fps would be close to identical?

Perhaps I misread.

 

[Trup1aya]

Good news.

Parity in online MP is a myth- everyone has different tvs, controllers, connections, headsets etc.

A good dev, would sample inputs at 60hz across the board and run the framerate at whatever the consoles can handle.

 

[oneils]

Good news.

Parity in online MP is a myth- everyone has different tvs, controllers, connections, headsets etc.

A good dev, would sample inputs at 60hz across the board and run the framerate at whatever the consoles can handle.

Sure it's a myth, but myths are powerful and does bungie want to bother upsetting the apple cart? Like what is the incentive for them to make a 60fps version of the game for one console (a console that will have a very small install base comparatively speaking).

 

[Yautja_Warrior]

Unless I've misunderstood isn't the point that MS has declared it won't enforce MP parity whereas Sony has mandated MP parity? Thus in theory Scorpio would allow owners a frame rate advantage vs XB1 owners whereas Pro and PS4 fps would be close to identical?

Perhaps I misread.

Yes basically Sony originally said if a MP game is 30fps on base PS4, it can't run at 60fps on Pro, in case it gives some players an advantage. MS are just saying they won't enforce any rules like that and it's down to developers.

 

[Trup1aya]

By using additional ressources on higher resolution textures, higher screen resolution, additional effects, to the detriment of performance because of pushing Scorpio a bit further than the base version is pushing the Xbox One. I really don't see what's so hard to understand here, as far as I know, worse performance has been achieved on PS4 Pro in the past as well, when compared to PS4.

Why would a developer throw more things at the Scorpio than it can handle?

The problem you speak of isn't a result of the scorpio's additional hardware, it's the result of a developer not doing a good job.

 

[Filthy Casual]

30 vs 60 fps in multiplayer would be bullshit. I don't care about the PC space because that's your choice to play in that environment and I doubt many of you who are supporting this choose to play at 30fps when everyone else is playing at 60 or above. If you do I'm sure you get spanked. Yes there are other potential advantages but 30 vs 60 is taking it too far. Not that it will happen anyway.

 

[Trup1aya]

Sure it's a myth, but myths are powerful and does bungie want to bother upsetting the apple cart? Like what is the incentive for them to make a 60fps version of the game for one console (a console that will have a very small install base comparatively speaking).

I don't understand the incentive NOT to upset the apple cart...

It's not as if people who have OG PS4s are going to skip Destiny because a tiny portion of the playerbase will be playing at a higher framerate.

When you are on a OG PS4 playing the game, you aren't going to notice, what system everyone else is on.

For a game that is mostly PvE it won't even matter.

That said this isn't just about Destiny, but gaming in general. It's crazy that console gamers still believe in parity, when differences in peripheral hardware and networking make it impossible and PC gaming proves that players with different hardware can coexist.

 

[Trup1aya]

Higher latency or ping is usually the result of geographical distances or just poor routing/a higher number of hops in a traceroute between connection A and B. There's no workaround for this, and it's the only uncontrollable variable here. Display lag can be reduced by playing on a low-latency monitor, but of course not everyone will ditch their big screen TVs for such an advantage.

Internet latency + display latency are already two variables in the mix. Adding variable frame rate (30/60) to it only widens the potential gap between player A and B. Definitely not a good idea imo.

Or It closes the gap if the 60fps player is dealing with latency that the 30fps player isn't.

Moral of the story is, parity is a myth in online games. It's time to bust the myth.

At the end of the day, having a high input sample rate (60hz or better) should make the game feel responsive to all players regardless of varied framerates between users.

 

[oneils]

I don't understand the incentive NOT to upset the apple cart...

It's not as if people who have OG PS4s are going to skip Destiny because a tiny portion of the playerbase will be playing at a higher framerate.

When you are on a OG PS4 playing the game, you aren't going to notice, what system everyone else is on.

For a game that is mostly PvE it won't even matter.

That said this isn't just about Destiny, but gaming in general. It's crazy that console gamers still believe in parity, when differences in peripheral hardware and networking make it impossible and PC gaming proves that players with different hardware can coexist.

Time and money could be a reason. Maybe its just time and money they believe are better spent on the current consoles. I mean, Scorpio isn't even out so why worry about creating the best possible console version of a game for a console that's not out yet. Guess we'll see at some point.

 

[geordiemp]

What exactly was your train of thought? Did you multiply the 9% with 8 so you got 72%? To me that seems the most common mistake someone would make. I'm always interested in what kind of mistakes people make in math.

Could you explain this a bit more? I get that out of order execution would achieve better results than serial workloads and that you'd get closer to a 9.5% performance increase but why would it surpass it? Aren't you still limited by the speed per core in parallel tasks?

https://en.wikipedia.org/wiki/Out-of-order_execution

So imagine 3 serial tasks, CPU might do a bit of prep work on say 2 and 3 while waiting for 1 to be done is how I understand it.

So can sometimes go quicker, but lets get real here, devs are just going to cap at 30 for 30 FPS games MOST of the time.

 

[Toni]

They might do 60 fps in single player, PVE, and do a rock-solid 30 fps for PVP.

They have to come through with premium features for the Scorpio users.

I can see this being the case on the Pro as well.

 

[EvB]

By using additional ressources on higher resolution textures, higher screen resolution, additional effects, to the detriment of performance because of pushing Scorpio a bit further than the base version is pushing the Xbox One. I really don't see what's so hard to understand here, as far as I know, worse performance has been achieved on PS4 Pro in the past as well, when compared to PS4.

That's more is a PS4 Pro issue than a Scorpio issue.

A 4K checkboard title requires double the pixels of standard PS4 + additional processing overhead to deal with the actual checkerboard stuff.

The PS4 Pro has twice the GPU and a faster CPU and 1GB of extra ram to just about cover this. That's assuming everything is as simple and linear as that.

Scorpio has 4.5X the GPU capability of Xbox One , Double It's ram, a huge increase in memory bandwidth AND a faster CPU.
There is significantly more headroom to reach native 4K without any performance drops from the base unit. That's before they think about increasing texture quality etc.

So as others have said, if a game performs worse, then that is entirely the fault of the developer pushing visuals too far.