MS can't be that dumb to not show REAL gameplay this time.
Not trying to sexualize her or anything but i suspect there is some serious Booty hidden under that outfit.
Not trying to pick you apart really, but the main issue is your usage of "power" and how X is 20% more powerful than Y. That conflates theoretical tflop count with how the actual software performs. That is not a correlation you can easily do and one thing does not imply the other. It would be true if all other moving parts were equal, which is why you do PC hardware benchmarks in machines designed to eliminate all other bottlenecks. You don't have such scenario here, you need to sum all the parts and then compare performance.There's nothing incorrect about it. It is what it is. It's a theoretical advantage, sure, but it's there and it's 20%.
And you're right about the I/O architecture, but as you said yourself earlier: "Please re-read my statement".
Exactly, I only have so much blood in my body and either it goes to the big head or the little one. If only we had the smartshift technology of the PS5 and we could shift blood flow to keep both organs operating at near maximum performance. Where a 10% lower blood flow to the brain would only translate to a couple percentage points of lower brain capacityI am always trying to learn technical stuff on these forums but every bloody page i turn, people are always distracting me with awesome screenshots. Regardless if it's Spider man, Aloy butt pics or Amanda Cerny, there is just to much eye candy on display.
When will this Madness end.
Exactly, I only have so much blood in my body and either it goes to the big head or the little one. If only we had the smartshift technology of the PS5 and we could shift blood flow to keep both organs operating at near maximum performance. Where a 10% lower blood flow to the brain would only translate to a couple percentage points of lower brain capacity![]()
Not trying to pick you apart really, but the main issue is your usage of "power" and how X is 20% more powerful than Y. That conflates theoretical tflop count with how the actual software performs. That is not a correlation you can easily do and one thing does not imply the other. It would be true if all other moving parts were equal, which is why you do PC hardware benchmarks in machines designed to eliminate all other bottlenecks. You don't have such scenario here, you need to sum all the parts and then compare performance.
Meaning, I think it's fair to wait for a few multiplatform games and put them side by side to count fps. You then can say X outperforms Y by this <----> much. In that other game maybe Y outperforms X by <---> that much.
But looking at a tflop count, in isolation, especially for 2 totally different APU architectures is a futile exercise. As an example, a pipeline that won't use all 52 (is it 52?) CUs on the XSX will mean the performance gets reduced significantly when comparing with an architecture running lower CU counts but much higher clock speeds. In the PC space you could compare to Intel and AMD a few years back when AMD would run laps around Intel in multi-threaded performance due to its higher core count, but the Intel would still edge out in gaming performance due to higher clock speeds on fewer cores.
Sony Patent Could Mean Big Things for PlayStation 5 UI
https://gamerant.com/sony-patent-playstation-5-ui/
![]()
Alright, I guess we'll agree to disagree on this oneYou're definitely not picking me apart - don't worry about that.
I'm not talking about how software performs, what the games will look like, how many FPS games will have etc. I'm simply saying that going with what we know, the XSX is 20% more powerful than the PS5. You can talk about AMD vs Intel, software or "totally different APU architectures" (hint: they aren't) all you want, that's not going to change much.
Yeah, I'm playing now Red Dead Redemption 2 and my friend we spend a great time of our time in the horse.This generation saw both the rise and the fall of open world games, at the beginning everyone was rejoicing about the sheer size of maps and number of missions, literally hundreds of hours of gameplay. But from middle to end, people began to suffer OW fatigue due to colectatons, dull side missions and mostly empty worlds. It's not about quantity, it's about quality of time again, not every game can be a Witcher 3.
This situation reminds me of TV series, what do you prefer: a 24-episode full of fillers series or a 10-12 episode one that goes straight to the point?
To illustrate how TF alone can't be used, here are some numbers (from https://www.resetera.com/threads/pl...al-discussion-ot.231757/page-13#post-37912777):There's nothing incorrect about it. It is what it is. It's a theoretical advantage, sure, but it's there and it's 20%.
That's why it's impossible to call it x% power advantage without looking at each game. I'm sure the XSX will be better at some things (probably number of rays generated) if all CUs are used, but the PS5 will be better at other things (more calculations and higher refresh rates) due to higher clock speeds. The biggest unknown being how the storage solution for the first time in... well, forever, might actually impact the graphics quality by actually allowing the gameworld to be rendered "on the fly" without needing all the textures stored in VRAM.To illustrate how TF alone can't be used, here are some numbers (from https://www.resetera.com/threads/pl...al-discussion-ot.231757/page-13#post-37912777):
Triangle rasterisation is 4 triangles per cycle.
PS5:
4 x 2.23 GHz ~ 8.92 Billion triangles per second
XSX:
4 x 1.825 GHz - 7.3 Billion triangles per second
Triangle culling rate is twice number triangles rasterised per cycle.
PS5:
8 x 2.23 GHz - 17.84 Billion triangles per second
XSX:
8 x 1.825 GHz - 14.6 Billion triangles per second
Pixel fillrate is with 4 shader arrays with 4 RBs (render backends) each, and each RB outputtting 4 pixels each. So 64 pixels per cycle.
PS5:
64 x 2.23 GHz - 142.72 Billion pixels per second
XSX:
64 x 1.825 GHz - 116.8 Billion pixels per second
etc.
So those 20% also goes the other way, right?
To illustrate how TF alone can't be used, here are some numbers (from https://www.resetera.com/threads/pl...al-discussion-ot.231757/page-13#post-37912777):
So those 20% also goes the other way, right?
To illustrate how TF alone can't be used, here are some numbers (from https://www.resetera.com/threads/pl...al-discussion-ot.231757/page-13#post-37912777):
Triangle rasterisation is 4 triangles per cycle.
PS5:
4 x 2.23 GHz ~ 8.92 Billion triangles per second
XSX:
4 x 1.825 GHz - 7.3 Billion triangles per second
Triangle culling rate is twice number triangles rasterised per cycle.
PS5:
8 x 2.23 GHz - 17.84 Billion triangles per second
XSX:
8 x 1.825 GHz - 14.6 Billion triangles per second
Pixel fillrate is with 4 shader arrays with 4 RBs (render backends) each, and each RB outputtting 4 pixels each. So 64 pixels per cycle.
PS5:
64 x 2.23 GHz - 142.72 Billion pixels per second
XSX:
64 x 1.825 GHz - 116.8 Billion pixels per second
etc.
So those 20% also goes the other way, right?
I was only giving a summary of the video because it's 15min long.
My apologies to all Sony fans that I didn't mention that he discussed the SSD in the previous video, next time I'll make sure to include the summary of the 5 previous videos too.
You're definitely not picking me apart - don't worry about that.
I'm not talking about how software performs, what the games will look like, how many FPS games will have etc. I'm simply saying that going with what we know, the XSX is 20% more powerful than the PS5. You can talk about AMD vs Intel, software or "totally different APU architectures" (hint: they aren't) all you want, that's not going to change much.
This world is more bigger than I think it was. When Cerny said "its more easy to use 36 CUs in parallel than 48" i thought they said "hey guys we are using all 36CUs in PS5" and mainly when he spoke about triangles that's is difficult to fill all those CU when you have more CUs. But to make a decision to go to 36 against 48, is because you can use more of the 36 than the 48 or the 36 higher clocks makes it better. I don't know everything but I think the API has something so special to make that parallel thing he said, even still suffering of CU occupancy.Are we discussing bandwidth vs. Tflops by CU occupancy?
CU occupancy above 60% in a real program is considered good programming. 100% occupancy is only possible in benchmark. No real non-benchmark program will ever achieve 100% at least not for now and for the next decades.
You can squeeze their brains with the following heavy reading:
https://developer.nvidia.com/blog/t...lysis-method-for-optimizing-any-gpu-workload/
The coherence engines and the cache correctors will improve the use of the CU, as well as having a faster clock in the entire I / O complex (if the rumors are true and in PS5 the I / O controller is actually based on GPU CU tall clocks will improve the use of CU in several ways). I have read from leaks in ERA that inactive or standby CUs on PS5 are able to assist with I / O management to dedicated CUs, this will further enhance 2-way occupancy.
A stopped CU will not be because it will be managing I / O and on the other hand the I / O management aid will improve the times to occupy more CU.
It is a very brilliant idea and Cerny is a genius. It won't be long before we see CU GPU-based I / O management from here on out. Maybe improve utilization with good programming to 70% 75%? I would love to have a study in the future that talks about it. But I remember that a 10% -15% improvement on PS5 is being able to enjoy 1-1.5 more Tflops of power. Hopefully we could know the real impact, in August we will be able to know more about XVA and we will be able to know what SX has and see if MS made the right decisions. For now XVA is unknown. (I am speculating on the improvements, but I want to think that the improvement must be greater than the cost of the CU (2 CU on PS5) used for I / O at least)
The maximum theoretical speed of the PS5 SSD is 5.5 GB / s and I do not think it can be reached, however the actual use of the SSD will be through the I / O controller, Cerny spoke of a maximum I / O of 22 GB / s and that think 8-9 GB / s on average. That is the average data possibly achieved without Oddle Textures, possibly quite conservative and closer to 10 GB / s real according to all patents.
You have to think about the entire I / O group because the I / O was designed as a whole group. If you are worried about BC, it will probably load everything 10 times faster. 30 sec -> 3 sec 60 sec -> 6 sec. It is a very important gain, but a BC game will only serve to wait less in the loads it will not have any change beyond, so I think it is not important.
I hope the translator does a good job.
To illustrate how TF alone can't be used, here are some numbers (from https://www.resetera.com/threads/pl...al-discussion-ot.231757/page-13#post-37912777):
Triangle rasterisation is 4 triangles per cycle.
PS5:
4 x 2.23 GHz ~ 8.92 Billion triangles per second
XSX:
4 x 1.825 GHz - 7.3 Billion triangles per second
Triangle culling rate is twice number triangles rasterised per cycle.
PS5:
8 x 2.23 GHz - 17.84 Billion triangles per second
XSX:
8 x 1.825 GHz - 14.6 Billion triangles per second
Pixel fillrate is with 4 shader arrays with 4 RBs (render backends) each, and each RB outputtting 4 pixels each. So 64 pixels per cycle.
PS5:
64 x 2.23 GHz - 142.72 Billion pixels per second
XSX:
64 x 1.825 GHz - 116.8 Billion pixels per second
etc.
So those 20% also goes the other way, right?
So those 20% also goes the other way, right?
Bo_Hazem : Isn't that a Kena screenshot?
That game is coming to PC and has minimum requirements like "AMD Radeon HD 7750/NVIDIA GeForce GTX 650 or Equivalent", I doubt that game couldn't come to this gen?
All those games look great to me though.. how they look/feel in action maters more though.
Bo_Hazem : Isn't that a Kena screenshot?
That game is coming to PC and has minimum requirements like "AMD Radeon HD 7750/NVIDIA GeForce GTX 650 or Equivalent", I doubt that game couldn't come to this gen?
All those games look great to me though.. how they look/feel in action maters more though.
Yes, we need to see screenshots of it running on PC first, those are potato mode requirements.Andodalf and it's not coming to PS4, official.
Bo_Hazem : Isn't that a Kena screenshot?
That game is coming to PC and has minimum requirements like "AMD Radeon HD 7750/NVIDIA GeForce GTX 650 or Equivalent", I doubt that game couldn't come to this gen?
All those games look great to me though.. how they look/feel in action maters more though.
Yes, we need to see screenshots of it running on PC first, those are potato mode requirements.Andodalf and it's not coming to PS4, official.
Ember Lab revealed Kena: Bridge of Spirits during the Sony PS5 "the future of gaming" reveal today. Kena: Bridge of Spirits is coming to PS5 and PS4 in the holiday season. It's a timed console exclusive and will come later on to PC. Kena is a story driven, action adventure game, mixing a 3D animation style with a rich and fantasy atmosphere.
I just don't know where you are getting "not possible on current gen" for all of these games; "not coming to the current gen" is pretty different than "not possible."
Games said to be not possible on current gen, next gen only.
So you meant only that one game?That's what I said.
The developers of The Medium have revealed that the game's core gameplay feature was not possible on current-gen. The Medium is a forthcoming horror game for the Xbox Series X and PC and is being developed by Team Bloober, the team behind Blair Witch and Layers of Fear.
.![]()
The Medium's Core Gameplay Wasn't Possible On Current-Gen
The Medium's gameplay is made for next-gen.screenrant.com
Picked games that aren't coming to current gen consoles first, cross gen after. Any question of "why" should be sent directly to the devs themselves, I guess.
"I can't tell you right now about our most important feature, because we will have to save it for later. But the gameplay mechanic built around those two worlds couldn't be fully realized on current-gen hardware. This is something that I think will distinguish our game from any other. It's also even patented. So yeah, the game simply cannot work the same way on current-gen platforms. And that's why we are developing the game for Xbox Series X, plus PC."
It's been said Kena is coming to PS4 also
![]()
Kena: Bridge of Spirits is a Story Driven Adventure Coming Exclusively to PS5, PS4, and Epic Games Store
Screenshots, reveal trailer, first details for Kena: Bridge of Spirits, a beautifully animated action adventure from indie studio Ember Lab.www.dualshockers.com
Are they wrong?
Yes, we need to see screenshots of it running on PC first, those are potato mode requirements.Andodalf and it's not coming to PS4, official.
I'll ask anyone here that has a PC and MSI Afterburner to run an intensive game and check GPU utilization. It's rarely, so rarely touching the 90+% utilization. I mean maybe at 4k with RT on, but i'm usually running 1080 / 1440 at high refresh rates anyway. I'm super curious as to how Navi will performWhat you have actually shown is that the difference is really 15.41% and not 20% ( when you calculate in the # of CU's per system )
Assuming all other things are equal ( which they are not) and 100% CU utilization all the time the XSX would be at best 4.5 FPS better at 30FPS or 9 FPS better at 60FPS
So you meant only that one game?
That's why I'm confused here... I don't think anyone claimed Kena wasn't possible on the current gen.
But The Medium dev claims something about the gameplay isn't possible on last-gen... so not even sure why that's being used for some graphics wars post lol
The EA game engine developer talks about this while reacting to "The Road to PS5" video. He talks about how adding more and more parallel processing (adding more CU's) doesn't mean you can use all the CU's. He also mentions that it's better to improve the per-core clock speed of the GPU rather than adding more CU's.This world is more bigger than I think it was. When Cerny said "its more easy to use 36 CUs in parallel than 48" i thought they said "hey guys we are using all 36CUs in PS5" and mainly when he spoke about triangles that's is difficult to fill all those CU when you have more CUs. But to make a decision to go to 36 against 48, is because you can use more of the 36 than the 48 or the 36 higher clocks makes it better. I don't know everything but I think the API has something so special to make that parallel thing he said, even still suffering of CU occupancy.
LMAO "Offical"
Ember Lab
Ember Lab is driven by a passion for telling great stories. We’re currently working on our first game, Kena: Bridge of Spirits.www.emberlab.com
You might want to let the Kena devs know
screenshot of the press kit for fun
![]()
Official PS channel says otherwise. Well, another exclusive for PS4 won't hurt.
Not mentioned by them as well on the official PS blog:
![]()
Kena: Bridge of Spirits from indie studio Ember Lab announced for PS5
This beautiful, story-driven action-adventure game is a passion project from the small team at Ember Lab.blog.playstation.com
It might come at the same date to PS4, after sometime, or never.
Official PS channel says otherwise. Well, another exclusive for PS4 won't hurt.
Not mentioned by them as well on the official PS blog:
![]()
Kena: Bridge of Spirits from indie studio Ember Lab announced for PS5
This beautiful, story-driven action-adventure game is a passion project from the small team at Ember Lab.blog.playstation.com
It might come at the same date to PS4, after sometime, or never.
edit: In fact here's what he actually said, said, "couldn't be fully realized", but he's talking about a gameplay mechanic not graphics
I'm not dragging you through mud bro..Did I specify though? What's all this mud you're dragging me into? I'm done here.
![]()
Their website and press kit say it. Every media outlet reported it. That article never mentions launch platforms at all, just a bit about the game and how it uses the PS5. Its a cross gen game. Sorry if that fact hurts you in some way.
Too much talk, but let's compare. Games said to be not possible on current gen, next gen only:
XSX + PC (could be running on PC as well, it's vague)
![]()
![]()
![]()
![]()
![]()
Will not bring PS5 first party:
![]()
![]()
![]()
![]()
![]()
Now cross-gen on XSX (could be running on PC)
![]()
![]()
![]()
![]()
![]()
![]()
Then cross-gen running on PS5
![]()
![]()
![]()
![]()
![]()
![]()
Sources:
![]()
Xbox Series X | Xbox
Discover the fastest, most powerful Xbox ever with the Xbox Series X.www.xbox.com
![]()
PS5
Official PlayStation Blog for news and video updates on PlayStation, PS5, PS4, PS VR, PlayStation Plus and more.blog.playstation.com
The bandwidth might limit PS5 advantage there.
There's going to be so much hair-splitting in the console wars this gen. People currently downplaying PS5 games as being indiscernible from PS4 games are going to have a hard time saying there's a significant improvement on XSX. The opposite will be true, once Xbox has their presentation in July. The current gen is already very samey, so I don't see why anyone should expect any different with the next-gen. Each system will have its strengths and weaknesses, as you note, but it's not going to be anything noticeable in realtime. I think that's great. It should just be about the games now. No need to nitpick graphical differences anymore, just play what you like. If anything, the controllers might become the real difference-maker this coming generation.PS5 in a hardware so different from what is known, contains a high customization in many processes and it is possible that SX also has its customization, although according to what we can know in another sense.
The focus is so different between the two companies, one focused on power for lighting and effect enhancements and the other focused on data flow, textures, and assets.
Although the two are based on RDNA2, the differences that separate them are even in the concept of how to improve graphic quality. This tells me that we really cannot compare the theoretical power of one with the other. we won't possibly compare the games when we will see them either.
It is possible that the entire generation passes them with these doubts. If the game is faithfully optimized for each console and we see the results will not be comparable.
In SX the lighting, the effects and the fps will be enhanced.
In PS5 we will see active models, better distance and better textures. When you go to compare it is possible that you say ok SX has more fps better RT but it does not have those assets. SX in the distance is more blurred but has more reflections, PS5 has less reflections but the distance is very clear and the near and far textures have better shapes.
I sincerely believe that comparing graphics in a game that is fully optimized for each version would be unmatched and totally subject to personal preference.
Of course if we see a game optimized for the other console and doing a bad port, the optimized one will win. But here the commercial viability and the strategies that have nothing to do with hardware are being evaluated.
I'm not dragging you through mud bro..
But I'm about to... are you really gonna pretend your point wasn't about graphics? lol.. why show screenshots then?
MUDDD
(I think all the games look great... but most of that isn't gameplay anyways)
That profile picMS has no choice other than to improve the show. Expect for see content 3+ years away with CGI/In-engine trailers and some vertical slices.
Quality content is king and in that regard sony is going to deliver. MS should talk less and show more.
Yeah seems silly... unless they are going to strongly suggest 3rd parties offer a performance mode.I mean i hope for them they realize if most games won't run at 4K, 60fps/120fps like they're saying and promising they will got backlash, right?