Seems multiplats will be the same

@Zero_epyon: However the CU count difference is 10% on the 5700 to the 5700XT the difference in CU count in the PS5 to the XSX is just over 30%. If the core count was closer on the PS5 vs the XSX then this might be a useful video.

This is true. And I'm not saying that the difference isn't there. What I'm saying is that the difference won't as significant as some are expecting. Some are expecting 2080 Super vs 2060 Super. They're setting themselves up for disappointment.

true

I will wait until these systems are out in the wild and being tested by people like DF before I pass judgement. Even then I think the differences will be minor and not cause for concern

PS5 may load some games a second or two faster and the XSX may have slightly better graphics in some multiplats but I think most non anal people will be able to enjoy games on whatever system they choose

@regnaston:

Bullshit and even more of it turn out rumors about RDNA2 are true, and it has 2.4GHz speed, that would prove without doubt that part of RDNA2 performance against, is not based on CU but on much higher clock speed.

From 2.4GHz to 1.8GHz there are some 600mhz of speed difference.

Performance is not just CU, is also speed of those CU , again if you downclock the series X to 1GHz is not a 12TF machine any more even with 52 CU.

So yeah the 5700 can beat the 5700xt, nut just so you know the gap between the 5700 stock and 5700xt is like 22% and the gap in performance between both is minimal.

Brah, SW is all about the anal.

Wait, shit!

Wait...

@Chutebox: the post above yours proves that both in terms of the poster's mental makeup and sexual desires

Dirt 5, 4K 120 modes. Not seing that power difference Lems ... https://mobile.twitter.com/dirtgame/status/1318582859918180352?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1318582859918180352%7Ctwgr%5Eshare_3%2Ccontainerclick_1&ref_url=https%3A%2F%2Fs9e.github.io%2Fiframe%2F2%2Ftwitter.min.html1318582859918180352

@Sagemode87: Haha lol you think Dirt is going to be native 4K on PS5

@the6millionlies: lmao you guys are funny. It's not even an impressive looking game. Lems narrative is just one big troll job at this point. I'm sorry you guys bought Microsofts hype.

It reminds me of how your faction bought into the hype of the SSD. 😎

According to DF, Dirt 5 has a dynamic scaler in play for both 4K/60 modes. The 120hz mode rendered at 1440p with drops into the 90's for them. 1440p is due to their display limitations.

So it doesn't look like either version is going to guarantee a locked 4K/60 and that seems like it's the devs fault.

Lems said the same thing about the PS4 and Xbox1 (launch sku).

Oh, how history repeats itself...

The DF comparisons are definitely going to be interesting on the upcoming multiplats.

Cherno is rendered nearly speechless

But not in a good way...

@Zero_epyon: Why should anyone care about this guy?

He's been running an educational youtube channel for software engineering, specifically game development, for the last 8 years. He's also a former EA software engineer of 4 years and he's currently building a game engine on his own while continuing his educational videos. He provides some good insight as an actual game developer.

@dabear: how? Games ran at clearly different resolutions. That's not the case here. What difference will you be happy with from 16 to 18 percent. The desperation is insane.

@Zero_epyon: https://youtu.be/4cmclCjS5jg

https://youtu.be/4cmclCjS5jg

You don’t have to take this conversation seriously. But it’s true. The GPU inside the Xbox Series X is about a 2080 Super.

The power differences will be noticeable and XSX will have a massive performance advantage.

no do the math its not 30%. the math does not add up to 30% so how is it we need to accepted it.

fact is fans boys made up the 30% because of the 9.2 teraflops. its not its 10.3.

fact is the ps5 loads faster who cares. 20% more powerful who cares. sony first party is going to create some great looking games next gen are they not?

I already read that quote and it quoted out of context. No where did AMD said that both component will run attheir max clocks at the same time. They only said both can be boosted, meaning they can be boosted as long as they are below their threshold and max power. As long as one hits the ceiling they can no longer be boosted simultaneously and only one will run at the max clocks.

Whether it's moot or not, if they base clocks were always at the peak then Cerny would have said so not avoid answering it. Machine Learning or manual control is again irrelevant because no how you choose to control it you will still be running just one of them at max clock at the same time and it has been proven in practice.

So, you along with your fellow cows who keep babbling and misquoting it, can tell me if both can be run at their max simultaneously then where will the power be routed to and from what? If both can run simultaneously at max then why Smartshift is needed? All other mental gymnastics is irrelevant, answer the simple question.

so you got the ps5 numbers to right? we know for a fact what the ps5 will be doing?

its all guess work with out them testing each one. it might be a quiverlant to this or that is all made up without testing them. because we don't know the features they have added to the custom gpu's. same with series x. there is no way to compare it to pc gpu's without the machine to test it.

the series x and ps5 will do more raytracing of a super because of the developers will code it in there games. just like how gears 5 has more raytracing in it over the pc version until the update comes.

@Zero_epyon: Sorry if I don't take his input too seriously. 😎

Again missing the point. Let me break it down:

1. 5700 vs XT isn't a good comparison for PS5 vs XSX. One has just a difference of 4 CUs while the other has 16.

2. In the video you posted the games are running at 2K or 1080p, can't really tell but definitely not at 4K and definitely not at max performance. Once you have enough head room due to low resolution it won't show you much difference. This can be confirmed from the lowest FPS differential.

3. The better comparison would be 7970XT with 7870 OC or something like that. I forgot the actual two models but the OC version has higher TFLOPs but lower CUs and it didn't perform well in games.

4. Add RT and the difference will be higher. Also, pretty sure Richard didn't test the RT's effect on memory bandwidth. Regardless, at higher clocks with less cores memory becomes a bottleneck.

5. Lastly, if Cerny is such a genius on why higher clocks are better then do tell why Nvidia and AMD for their flagship GPUs have more cores and faster memory over clocks. Suddenly Cerny knows more about the actual creators of GPUs.

The card inside the XSX is not 2080 Super. It won't even be 2080. Add in RT and ML and it will be way less, actually for ML and RT, XSX was performing in the ballpark of 2060 if I remember correctly. You consolites needs to stop overhyping your hardware and use Gears 5 as a benchmark of anything.

@Pedro:

What a shock pedro de fake developer doesnt take seriously a real one.😂😂😂

Epic.

@regnaston:

Thats funny for some one so invested in downplaying the PS5 to the point of spreading false information.

The two links are to the same video and they only discuss the hardware in it. You might be trying to refer to this?

https://www.eurogamer.net/articles/digitalfoundry-2020-inside-xbox-series-x-full-specs

However, even basic ports which barely use any of the Series X's new features are delivering impressive results. The Coalition's Mike Rayner and Colin Penty showed us a Series X conversion of Gears 5, produced in just two weeks. The developers worked with Epic Games in getting UE4 operating on Series X, then simply upped all of the internal quality presets to the equivalent of PC's ultra, adding improved contact shadows and UE4's brand-new (software-based) ray traced screen-space global illumination. On top of that, Gears 5's cutscenes - running at 30fps on Xbox One X - were upped to a flawless 60fps. We'll be covering more on this soon, but there was one startling takeaway - we were shown benchmark results that, on this two-week-old, unoptimised port, already deliver very, very similar performance to an RTX 2080.

And if that's what you're thinking if then I say be careful with how you interpret this. For one he says 2080 and not 2080 Super. Also when the One X was revealed, Richard made a similar claim that it performed like GTX 1070 based on a demo of a Forza title. DF later walked some of that back in later comparisons where they claimed that the One X was more like an RX 580.

Then consider this:

This is Gears 5's Benchmark feature running on an GTX 1080ti at 4K Ultra. The benchmark runner stresses that he's using Nvidia's Shadow play which takes performance down between 5-10%. So the game should hit 60 fps with so lows in the low-mid 50's. Also, consider that the Benchmarking tool really taxes the hardware and doesn't represent typical gameplay performance. So it's absolutely possible to get 4K Ultra 60 fps on Gears 5 with a 1080ti. That's why I'm always a bit hesitant to believe Richard when he makes points like this.

That's fine. I'll remember that the next time you give your opinion lol

1. As I said to @regnaston earlier, I'm not saying that the clock nullifies the difference, but that the difference won't be as larger as people expect.

2. You are correct that it's 1440p. I thought it was 4K. However, the first video I posted shows little differences between the 4K modes in those games, including games like Metro Exodus. If that doesn't convince you then you're free to do so.

3. That comparison isn't better though. The 7970 had an extra GB of Ram than the 7870 XT. The 5700 and 5700 XT are identical enough that the VBios for the 5700 XT can be installed on the 5700 and work well. I'm not sure the same could be done with the 7000 cards. The 5700 XT and 5700 also share the same memory size, bandwidth, and bus sizes. I couldn't find any benchmarks but I'd imagine the 1 GB memory difference would play a big factor in performance.

4. He couldn't. The 5700 doesn't have dedicated RT hardware like RDNA 2 is supposed to have. So the Jury is still out on that one, but if it's supposed to be more beneficial to have for CUs then it is what it is. Again, not arguing that there will be no difference, just that it won't be as pronounced as some expect.

5. If these leaks are real, Navi 21 XT will have a boost clock of 2.4Ghz, just a bit higher than the PS5's boost clock along with what you said. I'm sure cost was a factor in going with the memory speeds but I can't speak for Cerny. https://technosports.co.in/2020/10/19/the-leak-of-amd-navi-21-gpu-is-one-the-biggest-flex-we-have-seen-in-2020/#:~:text=The%20veteran%20tech%20leaker%20Rogaine,of%20up%20to%202.4GHz.

Basically. Gears 5 can run near a Series X on a 1080Ti. Obviously, it won't be able to do any ray tracing stuff but it doesn't need to be a 2080 super to do so.

This is a fun thread. Really enjoying the conversation here lol.

@Zero_epyon: https://www.techpowerup.com/gpu-specs/xbox-series-x-gpu.c3482

https://www.techpowerup.com/gpu-specs/geforce-rtx-2080-super.c3439

Even Alex says “It’s much faster than a 2080.”

As far the video you and Torm love and hype. It perfectly demonstrates that the more you push clocks the less performance you get back.

In what world do you think the PS5 will win head to head in performance? Remember when RE3 remake came out and DF ran there analyst and found the PS4 Pro was out performing the Xbox One X in that game? Sony ponies made thread after thread about how power doesn’t matter. It’s not about TF. They only made a big deal about this because they knew PS5 was a lot weaker than XSX and they used that as an example that power doesn’t matter. That was until a week later when they patched RE3 and XBOX One X ran at higher resolution and more stable frames......Then you guys went silent.

Only time PS5 will beat XSX is if developers decide they don’t want to optimize the game for XSX fully. Just like they did for RE3 Remake until fans called them on it.

I don't know what those two links are supposed to demonstrate. You can't compare spec/metrics of different architectures since they mean different things on each architecture.

As for your Alex quote, no you misquoted it again.

The video starts at where they start talking about the 2080 comparisons. They said it's on par with the 2080 because the heavy benchmark ran close to the 2080. But as you saw, and conveniently failed to address, the 1080Ti can run that benchmark in ultra at 4K close to 60 FPS.

I remember that thread and funny enough, the reason why the One X version of RE3 ran poorly against the Pro was because the One X ran the game at Native 4K without a scaler while the Pro had a lower resolution and with a scaler I believe. I think the argument wasn't that power wasn't everything, but the push for 4K wasn't worth it if framerates tanked. After the backlash, Capcom patched it so that it ran with a scaler like the RE2 remake did, which fixed performance. So you have it totally backward (starting to see a trend lol).

The PS5 will perform on par or better than the Series X when the game is running workloads that benefit from higher frequencies than CU counts. But when it doesn't, those higher clocks will mitigate some of the performance gap between the CU differences, but obviously not completely. That's my take.

@Zero_epyon: But it won’t always perform on par right? I doubt it ever out performs the XSX unless the developers don’t really optimize the game. For reasons explained. The XSX has just too many advantages. Some developers have said PS5 is difficult to get to match XSX in performance. I know RE8 developers said the PS5 is struggling to match XSX as well. It’s struggling to hit 1080p 60fps while XSX is pushing 4k 60fps. So I don’t know if I believe your claim.

I think XSX is just too powerful for the PS5 to keep up. Like I said and like DF said, it’s a 2060 Super vs 2080 Super. It’s not gonna happen buddy. Sorry.

@pc_rocks: He says he expects it to remain at or close to the peak. Take it for what it’s worth.

The fact that you believe RE8 can't do 1080p/60 on PS5 but somehow can do 4k/60 on series X because of a few more compute units tells me you don't really understand the hardware.

Also, DF never said 2080 super either so..

Like someone else said , if clock speeds are all and CUs aren't important , why the AMD and NVIDIA are putting more CUs to their GPUs? The 3090 is a beast with a lot of CUs/Cuda cores and clocked at 1700 Ghz

@Zero_epyon: I never said the PS5 could not do 1080p. I was quoting a development update from RE8 that said the PS5 is struggling to hit 1080p 60fps while XSX is running the game at 4k. So you need to relax.

You refuse to accept anything that doesn’t suit your video. The problem is, it’s just a video. Doesn’t mean anything.

https://youtu.be/oOt1lOMK5qY

Same thing. The more you push the card, the less you get back. Sometimes it can match performance and once in awhile yes the card can perform better by a couple percentage points. But most the time it does not. I don’t know how you don’t understanding that?

More core usually wins. Secondly, we are not talking about a 4 CU core difference. We are talking a massive CU disadvantage for PS5. Not only that but we are talking about a decent chunk in terms of raw power as well. Then add to the fact the XSX has bandwidth advantages, CPU advantages. The fact you blindly look away think to yourself..... yes, PS5 can overcome this because of a video I saw on YouTube is kind of troubling.

The other shocking disadvantage for Sony is it’s gonna struggle to manage its power consumption. The PS5 is energy hungry. The XSX uses less energy than X1X. The engineers for Xbox created a masterpiece of technology.

But yes. RE8 is struggling with PS5 development 🤓. At least according to them.

@BlackShirt20: You got owned up and down this thread but still won't admit fault. Pretty sad.

@Sagemode87: LOL we will see very soon.

Sony can’t hide forever.

Seriously dude

I love how you think that AF makes a huge performance difference.

And it's not 30%. The way you people take the rumored minimum clock as being the clock it runs at 100% of the time and as fact when it's unconfirmed in the first place is ridiculous.

You're arguing in circles, dismissing video evidence that counters your claims, and ignoring the fact that you've misrepresented and misquoted DF multiple times. Now you're resorting to rumors from a guy who said he spread the RE8 rumor because he thought press at the time was being one sided against Series X and wanted to even it out.

You also seem to be ignoring the countless times I've said to you and others here that the CU counts are an advantage that will provide some differences in performance between the Series X and PS5. But with the clocks being higher on the PS5, that gap will narrow, and the difference will end up not being so pronounced. You just want there to be a massive difference so bad that you've ignored the central point in all of this.

Interesting that you're making claims about the PS5 that haven't been tested yet. So tell me, how much power does the PS5 actually draw? In which games? What about any games that fully use the RT hardware in the CUs?

Don't even bother answering. I think we're done here because at this point I don't think you're interested in having a real conversation about this and you just want to cheerlead your preferred platform with buzzwords you heard on DF.

Good day...

Can you show me where the guy retracted what he said? You and others seem to be saying/implying that he said he lied. In fact he stands by what he said.

@tormentos: Still suffering from that Pedro envy. I am sorry I make you feel like a dunce.😂🤣

Now you know how much I view your opinion. 😉

@Pedro: enough to engage in a conversation apparently lol

@regnaston: didn’t say he retracted it. Him standing by it doesn’t make it true. He also said the PS5 would cost $600 to make around the same time a Bloomberg report said it costs $450. Rumors are just that. Rumors. But they’re being used here as fact.

@Pedro:

Na its just funny showing how much of a corporate b!tch you are while you desperate try to hide it.😂