User banned (3 days): Trolling. Platform wars.
It seems disingenuous to take sony peak performance numbers vs ms sustained numbers on paper, when we know sony will operate closer to 9 teraflops in reality
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
-
We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
Side-by-side official next gen flagship specs: Xbox Series X and PS5 spec sheets
- Thread starter BAD
- Start date
Where do we "know" that from? Do you have hard evidence?
SSD (defualt on both systems) and Direct Access of entire 100GB assets of a game on SSD as if it's all stored in Ram (feature confirmed on both systems) will totally alter how game devs design their games from the beginning.
Having no more loading screens, or fake elevators, or corridors, or checkpoint stutters, having actual geometry and hyper detailed (and reachable) draw distances instead of painted backgrounds, having thousands of NPC on screen at the same time (e.g. in big cities) and smoothly traverse all of this with a car or a plane or just flying ala Iron Man without a single hiccup is game changing, as it would have been impossible in the last 20 years of consoles.
But this will be possible both on PS5 and XSX.
It's the huge additional speed jump of PS5 SSD compared to the XSX one that probably won't make any relevant difference in most games.
Wow I've seen so many posts saying this and it is false. Its misinformation to make the gap look larger between the consoles. I will believe Cerny until its proven otherwise. Seriously how can anyone think it will throttle and decrese performance by 1tf, I'd like to know the reasoning for this other than github.
I have been saying this for months in the next gen speculation thread.
Putting the obvious load time benefit aside, if the XSX has a slower (but still fast) SSD, what would the real world benefit be for players in-game for games on the PS5? Or is the benefit largely in the development side?
In order for me to understand this a little bit better, could someone come up with a comparison %-wise between the PS4 Pro and the Xbox One X, and how things are now between the PS5 and The Series X?
It seems to me that's the most logical comparison, since that's the beginning of the trend, right? Just to know how power choices are evolving.
Thank you :)
It seems to me that's the most logical comparison, since that's the beginning of the trend, right? Just to know how power choices are evolving.
Thank you :)
And where is your evidence that the Ps5 will work at 9 Tflops most of the time? Saying something is sustained also doesn´t mean shit without evidence that it does.
As a person who has a gaming PC I was always going to get PS5 and I like what I'm seeing. Obviously I want to see how things run.
People should really watch the DF video and listen carefully to that part where they talk about the "boost" clocks. Stop spreading false information please!
Xbox One X to Ps4 Pro about 40% difference
XSX to Ps5 about 16% difference
In the df video leadbutter explains how the clocks are locked and that ms stressed there is no boost clock or variable component speed
You don't seem to understand the boost technology at all, do you? If a dev wants 10.28 TF, they will get exactly that at 100% of the time.
A potential variables is how much RT hardware is present per CU. If PS5 has 2x the amount of RT hardware per CU, for example, then it would be better at RT despite fewer CUs. Now, I don't think that's a likely scenario, and I agree that the CU count is a strong suggestion that the XSX has stronger RT capabilities.
Yea I'm not sure what people are expecting there isnt going to be a major resolution difference like there was with some 1x and pro games this gen as both will do 4k easily enough. 16% or add another 4% if ps5 throttles for some reason and it's still not going to be a major difference.
I'm really excited for both and its gives me an excuse to buy the series X due to its slight power advantage and pretty much cheap first party due to gamepass. PS5 will be a must for the exclusives.
Right, i think anyone here should watch that video as it really seems apparent that ms knew the specs sony was using.
It is variable... that usually means real world performance is probably an average of the min and max values.
Cerny is a liar. Again.
Just like how he lied about SSD performance and RT hardware.
Honestly is this stuff not ban worthy? It's literally FUD.
This gap is also bigger than between PS4 and SNES by your definition because one is 1.8TF and other is 0.001TF. Just subtract one from other and apply it infinitely at any scale.
What are the min values? Can we all read it some where?
So assuming DF and MS´s claims are right, how does that fit into your claim:
I really wonder.
Also you have yet to show recepits for your claim that the Ps5 will operate at 9 TFlops. When DF and Sony made it a clear point that the Ps5 will work at 10,28 TFlops.
The DF article say's the 2.23 GHZ is the typical speed of the GPU.
That is a contradiction. Then everybody would just run at 10.28 all the time. What is the the use of boost?
Also sony didnt mention the cpu having up to 16 threads. Using 16 threads vs 8 cores/8 threads is a 30% difference in performance by itself, not to mention the cpu is clocked higher on xsx in both modes
This isn't true. If you'd watched the presentation yesterday properly you would know this.
I know, but that's what Cerny told to them.
Tom Warren is also a super reliable source directly in contact with devs and insiders.
Variable is variable, as it is a "boost" compared to a baseline clock.
Sustained or Locked will always stays at that frequencies
We'll see.
Good question, I don't think we know them (yet). People assume it is the 9.2 number from GH.
Lol I'm starting to think alot of posters are just trolling now.
Probably when not under heavy load. its not locked so there isnt anyway to say its the typical speed. Peak speed is how to describe it.
As DF stated you can´t compare it to your typical boost you see on Pc´s etc. Really read the article, it is fairly clearly explained.
Did sony say the cpu supports 16 threads? I might have missed it
The blog post and DF video clear this. The cpu is 8C/16T, and the PS5 have a full implementation of HDMI2.1, the VRR is supported
Yeah it is clearly explained that the max setting cannot really be achieved. If you boost one, you lower the other one. This is what people are saying, that the actual performance is an average, since peak is impossible.
Agree, read the article. it's not a boost mode like PC, the frequency are stated by the DEV and fixed. All PS5 will run the same scene at the exact frequency. If the dev want 2.23GHZ all the time, the CPU will give some spare power. The thermal envelop is always at 100% and the CPU / GPU share these thermal envelop.
Finally the dev have a fine control off the CPU and GPU demand.
Posted on PS blog. Cerny didn't say a lot of things (like VRR, 120fps etc)
It's been answered above.
So what's the average?
Cable and connection support it, but it has to be supported also on a software level, which is not yet officially confirmed.
I mean, he did knock Sony for "we're nearly 4 months into 2020 and all we have is a logo" literally a day before Sony announced their ex-GDC stream, so I'm not sure I'd say he's the most well-connected insider around.
Why would you put your Gpu at a higher performance level when it´s not needed? What you say makes no sense at all.
There is no contradiction. If devs want 10.28, they get that. But there is a power budget, which means if the GPU runs at 10.28, they can't just put some heavy AVX-256 stuff on the CPU and expect it to run perfectly, they have to manage the workload. But that's the thing - they can do this. They have complete control of frequencies. And managing workloads to get stable framerates is something that's already being done in current-gen games.
I want real benchmarks, fps and frame times. Right now it doesn't really tell too except for getting only a vague idea of how powerful they're gonna be.
Guys and gals I understand that we're all a little riled up about things not going as expected for whatever your personal preferred systems are but please stop spreading misinformation to fit your narrative.
10.28 is the max for the gpu when it is at full boost, we don't know how long it can stay there or the percentage of performance lost if it drops since we don't know the base tf or frequency. Might be 90% of the time might be 60% might be variable depending on the air you feed your system.
Next 10.28 RDNA 2 tf does not equal 12.15 RDNA 2 tf in any world please stop saying this, the Xbox gpu is better, is it going to be taken advantage of who knows, but it is what it is, let's see what shakes out just like the ps5 ssd is superior to the Xbox series x, will it be taken advantage of we'll see
lastly, we don't know what memory configuration is better and it will likely come down to the size of a game, type of game, and dev preferences. If a dev only needs 10gb of ram the Xbox offers the faster solution if a dev needs more maybe they prefer the ps5 solution of it all being the same speed rather than 2 speeds. Wait till we know more as games release and hear from devs.
Accept the facts, argue as much as you'd like about what's better but please stop making things up to make each side look better or worse it doesn't help anyone
No idea.
Like I said, people claim minimum is the GH leak 9.2, so the average would be ~9.8.
Sure, but then 10.28 cannot be sustained for long, where XSX can have 12.15 sustained all the time, even enabling all the other components at max usage.