I can see that in the PS3 era. When companies tried new things. For what I’ve heard those dual issue computing operations are not suit to graphics and are better for simulation and IA. So you could have a innovative game exclusive for Pro and PS6/nextBox that looked like ass. Fat chance. We’re in the safe generation now.
They are PS4 games at heart.
I would say Dual-Issue is used for AI/ML upscaling and PSSR but who knows.
It’s the exact same engine as Part 2, which runs just fine on a PS4. Returnal is also a PS5 exclusive game. So is Ratchet for the matter, and while that is heavier on PC then your average game, it’s not to the extend of TLOU.
Nope it's a custom GPU.
This version is built from the ground up on PS5 unlike the various versions of TLOU2.
It’s built on the same game that launched on PS3 with slightly tweaked visuals from Part 2. If they wanted to do a PS4 version it could be done. But even if not, Returnal and Ratchet were PS5 exclusives.
It's a custom GPU, but the 45% figure given by Sony already gives us a clear idea of where rasterization performance will land.
And since its from Sony themselves I would bet money that's a best case scenario
If i remember correctly you said that you heard in term of real performance in games the Pro will be in the ballpark of a 4070 ?
Yes real world performance not brute strength
Nope built from the ground up on PS5.
Yet some early examples show more than a improvement of only 45%
It's closer to a mix of 7700XT/7800XT but again it's custom so we can not directly compare.
If you have to point out someone isn't a god theyre usually a bad motherfucker at what they do
Thanks for the answer, it makes me even more curious to see how the Pro will do in different games even if it will necessarily depend on the involvement that the developers want to invest in it.
No need to drink the marketing cool aid. We can all see it’s pretty much identical in feature set to Part 2.
With any new GPU architecture, it makes sense that there will be titles that show greater improvements than others. But if the average performance improvement was above 45%, then Sony would have provided a higher figure. (At least I can't think of any good reason for Sony misrepresenting the performance improvement to make the Pro seem less impressive than it really is).
As per the video that I posted above...He and a Dev stated Sony can afford to post a worse case scenario for its specs and still have it punch well above it's specs. Because PlayStation has no competition.
Koolaid? More like you being on the outside being in denial.
I'd say they were broadly honest in this instance and this should be the expected result. It checks out with the GPU specs.
Their "competition" is the PS5, since they need to convince people that the upgrade is worthwhile. It makes no sense in this context to hold back performance.
And we’re back to square one.
What better way to sell a $700 product than to undersell its capabilities? Companies are totally not known to EXAGGERATE in order to boost sales, but honest guy Sony is so concerned about its customers that it does the opposite.
Right you are in denial.
A worse case 45% advantage over PS5 is worthwhile.
There will be a section of people who find a 45% increase underwhelming and would be more impressed by say a 60% increase (or more). So as far as I can see, the logical consequence of what you're saying is that Sony is deliberately making the PS5 Pro seem less impressive, to get less sales.
I already explained Sony not having competition is giving a worse case scenario as stated by ML and the developer in the video.
PS5 Pro : 56.17fps (45% up) + PSSR + 3x RT + 28% more memory bandwidth
The PS5 Pro will deliver great performance with stricter power limits than PC.
S0ULZB0URNE
right now: 
Makes no sense whatsoever. Whether people buy a Pro or PS5, it's all the same to Sony. It's money in their pockets and more customers. On the contrary, they have a small window to really sell the Pro before the marketing dries out and people more or less forget about it. Underselling its capabilities is against common sense. It's a very pricey product without a disc drive and we're sitting here arguing that Sony is trying not to market it too much and keep it in the background? Come on man. The Pro's reveal video has a ratio of dislikes to likes of 3:1. If Sony could have helped it, they would have done so. You want to present such a product in the best possible light, not hold it back. Unless we believe the entire marketing department is made up of a bunch of morons, assuming they undersold it is nonsensical.
The reason it punches above its weight is because of PSSR don't be disillusioned about the 45% raster performance that's actually a really good number but it seems people are equating that number as overall performance and that's obviously not the case.Right you are in denial.
A worse case 45% advantage over PS5 is worthwhile.
They said it punches well above I agree by looking at the specs and some of the early improvements.
PSSR if it's like DLSS can boost frame rates by up to 30% against native if it reconstructs from 1440p. That's an 88% improvement over a base PS5 compounded with the 45% boost for a similar or better image quality. A game running at 40fps on the PS5 at 4K native will run at 75fps on a PS5 Pro using PSSR for a similar image quality. The big problem is that it's quite tough to sell that with shitty compressed video feeds to people watching on a smartphone.
So I went ahead and listened to the whole segment from MLiD and nowhere does he give a reason for Sony to try to sabotage their launch by misrepresenting the performance improvement. He's giving the lack of competition as the reason for the higher price, but Sony clearly still want to sell as many consoles as they can at that price. In order to do that they should give the highest performance figure that they can justify.
There's a big difference between them not marketing it and them deliberately sabotaging it.
In this particular case, perhaps not because it's aimed at the hardcore crowd anyway. However, the Pro had not been received well following the reveal and I think it'd be disingenuous to pretend otherwise. It only slowly recovered after we saw high-quality videos and comparisons.
But we've all seen them attempting to zoom in stills to sell the console. Yet, you're telling me that those same people would then turn around and undersell the 45% better performance when 1. it was told to the devs and 2. it would completely undermine what Cerny was trying to show off. You're postulating that they purposefully self-sabotaged, but we can see that's impossible because they tried their hardest to show the difference in those zoomed photos, something someone who's not trying to sell you a product wouldn't do.
He mentioned lack of competition for Sony's modest 45% increase.
S0ulzborne was saying they were being modest with 45%. I'm saying they weren't. It's 45% in rasterization...but there's another 20-30% added by PSSR, and the 2-3x faster RT. I don't think they undersold anything with the specs sheet. I do think they fumbled the reveal.
But I'm saying the initial reveal was junk and that following this, the coverage from tech outlets has been slowly turning the opinions around. We're saying the same thing here.
In that sense, I agree. I was merely laughing at the post saying that Sony undersold the 45% faster rendering. If the Pro could do, say, 60%, Sony would have run with that figure instead. I think 45% is accurate, but it's one piece of the puzzle and perhaps not even the most important one.
You fall in the same trap as many others comparing a console GPU to a PC GPU. You just can't isolate a consoles GPU and compare that part to a PC GPU. A console is greater that the sum of it's parts, while a PC is lower that the sum of it's parts in comparison. Is the PS5 Pro's GPU better than a RTX 4700? No, but can a PS5 Pro compete with a PC with a RTX 4700? Yes. It's obviously easier to design a console with less bottlenecks than a PC. A console is like a normal distributed muscular man, compared to a PC which would be a bodybuilder who skipped legday.
Nah, I just know that being objective and not trying to push an agenda would be using multiple datapoints as reference, instead of one specific one that supports my narrative.
But it is over 30TFLOPs dual issue compute but the fixed function pipeline is now 1:2 with the 32-bit float compute pipeline.
www.playstation.com
www.playstation.com
No, the closet PC GPU would be the RX 6700 or GeForce 2080.The Last of Us Part I for PC - PC Requirements & Features | PlayStation
Endure and survive. The Last of Us for PC – see PC requirements and features – now available to buy and download on Steam and Epic Games Store.
The closest PC GPU to a PS5 is the Radeon RX 5700
Yet the Avg. performance: 60fps @ 1080p. On PS5, this delivers 1440p60.
