How should devs handle ports between consoles? *spawn

onQ

Veteran
Mod: This thread assumes the consensus of hardware designs for XB1 and PS4 are correct. Discussion as to what's in the boxes belongs elsewhere - this discussion is just a hypothetical investigation based on what we currently know

Like in a scenario where the Xbox One game is 1080P 30FPS how should the 3rd party developer go about making the PS4 port if the same code is already running 1080P 45FPS un-unoptimized on the PS4 hardware?

Should they lower the frame rate to 30FPS while adding effects or should they lower the effects a little & push for 60FPS?


Scenario #2

PS4 game 1080P 60FPS being ported to the Xbox One with the unoptimized code running at 1080P 45FPS, how should the developer go about optimizing the Xbox One port?

Should they lower the effects or resolution & go for 60FPS or should they drop the frame rate to 30FPS & add more effects?


I'm guessing in a 3rd scenario where the PS4 game is 1080P 30FPS the developer isn't going to have much of a choice when porting to the Xbox One they will have to lower the effects/resolution to reach 30FPS.

How do you think they should handle the ports in these situations?
 
I expect dynamic image quality more often with the next generation consoles. A blurrier image in heavy scenes is better than slowdown.
 
Far more important question is which console is going to serve as the lead platform? Last time around it was the 360 because it came out a year earlier and was easier to develop for, which doesn't apply this time. Will it be the Xbox One and everything will "scale up" to the PS4 and PC or will it be the PS4 and everything "scales down" to the Xbox One? Or is it just dynamic resolution such that the versions are largely the same?

Or an even simpler question is whether it's actually even porting when the CPU and GPU architectures are identical?

If it ends up that the Xbox One runs the same game in 720p and the PS4 runs in 1080p, then I pretty much declare it a wash considering the average TV size is 42" and from 10 ft away, you can't really tell the difference between the two (besides native 1:1 pixel mapping actually gives you far greater sharpness).
 
so we don't know and you speculate about, ok
All speculation without hard facts has a degree of uncertainty. However, if you want to discuss hardware then at some point you have to come to an understanding, and technical know-how and prior leaks and previous experience show us that the expectations have an extremely high probability of being correct. Look at Wii and Wii U for how wild speculations as to what could be in there, drawing on vague PR remarks and uncertain tech documents, were completely unfounded and the hardware was exactly as those arguing on technical merit suggested. Look at the earlier rumours of Durango have multiple SOCs and how wrong they were.

At this point in XB1's life, unless we get something fairly concrete to contrary, it is safe and sensible to take the accepted hardware design as the reference for discussion, for both consoles (otherwise we can speculate Sony put a super-duper DSP in PS4, over have overclocked it since someone somewhere rumoured as much). The thread for discussing what is in XB1 is here. To stop endless, circular discussion across the board, I suggest we keep that speculation there and use the consensus for both hardwares as points of discussion for other topics until proven otherwise (whether hardware changes or cloud computing or whatever).

-----

As for the actual topic of this thread, I'd prefer better IQ and framerate over more eyecandy. XB1 eyecandy should be plenty, and it would be automatic to scale games up in framerate. So this solution means the least possible effort for developers while still delivering the favourable outcome (hypothetically, not that I've put my name to get any console next-gen).
 
All MS and Sony can demand is equality/parity withing the possibilities of the hardware.
PS4 strengths can't be overlooked simply because "MS wills it" and vice versa.
 
Last edited by a moderator:
Not really a technical expert, just a layman gamer giving my opinion.
They should make it as good as possible on all platforms. Just as they've allways done, for the most part.
The best thing is that now all have Bluray, and all consoles is having lots of RAM and Xbone and PS4 even have similar CPU's.
It shouldn't be as hard as last generation to port to between platforms, and eventual extra times gained by similar hardware could be spent on optimising for platforms lagging behind.

Yes, there will be framerate-differences, but it's not a issue in a realworld-scenario, for me.
yes, I can see the difference of 30 and 25 FPS sitting 10 inches from my PC-monitor watching Digital Foundry videos - but on the living-room TV, 4 meter from the screen, it's not noticeable for me if it's running 28 FPS sometimes, or 30 FPS.

So if your making a game like Yakuza, you suddenly have much less issues next-gen.
Plenty storage, plenty RAM, similar architecture..
Get on that localisation Sega, we westerns might not be able to buy enough copies on PS3 alone, but now you soon can sell it to PS3, PS4, Wii U and Xbone. :p
 
What if MS demands at least equality?
The only chance Xbox One to achieve equality could happen if developers treated the console as a TBDR machine and divided textures in grid patterns.

With that method and using the eSRAM cache there could be some parity.

But I can only see 1st party titles trying this.
 
I don't know how much this will save on the X1 (not a computer tech guy), but the least noticeable way to scale down the graphics would be to reduce the resolution IMO. As we've seen in current gen games, it doesn't have to be 720p or 1080p. I'm sure both systems will have good scalers.
 
I'm also for the dynamic IQ solution rather than drops in framerate.
 
Ok, but what if the dev is already using dynamic resolution on the PS4 version to get better results? Is the XBone going to scale down to SD levels or even sub SD levels to keep up?
 
Well I think the only time that might be an issue is when a dev targets 720p30 on ps4 as there isn't really anywhere to go in that instance. Luckily for ms, sony seem to be targetting 1080p60 so 1080p30 is probably going to be the lowest output which leaves lots of scope for reduction should devs need it.
 
No, I'm talking dynamic resolution. If the XB1 can get a 50% performance jump from it then why can't the PS4 get the same? If it's such an unnoticed win, why not? And if the game on the PS4 is already using a dynamic resolution then what happens with the XB1 version to keep up? SD? Sub-480 when the shit really starts flying?
 
No, I'm talking dynamic resolution. If the XB1 can get a 50% performance jump from it then why can't the PS4 get the same? If it's such an unnoticed win, why not? And if the game on the PS4 is already using a dynamic resolution then what happens with the XB1 version to keep up? SD? Sub-480 when the shit really starts flying?

Because differences in hardware and the memory subsystems will dictate the limits of a performance jump. XB1 without using the esram is at a clear bandwidth disadvantage. Using the esram allows devs to use other tricks to achieve similar effects without relying on the limitations of main memory bandwidth pipe for greater gain. To emulate that you would have to lock off part of the ps4 main ram and treat it like a cache effectively reducing the amount of available memory to the system.

The same gains just would not be there.
 
Because differences in hardware and the memory subsystems will dictate the limits of a performance jump. XB1 without using the esram is at a clear bandwidth disadvantage. Using the esram allows devs to use other tricks to achieve similar effects without relying on the limitations of main memory bandwidth pipe for greater gain. To emulate that you would have to lock off part of the ps4 main ram and treat it like a cache effectively reducing the amount of available memory to the system.

The same gains just would not be there.

You wouldn't lock of any of the GDDR5 as a cache, i wouldn't be surprised if devs had a hard time filling it at first.

The only major problem I can see with the porting is the fill rate and FLOPs difference if they can get around that they should be okay.
 
Back
Top