Xbox cares a lot. They heavily marketed the Series X around the narrative. This is common knowledge for anyone who was half awake during its launch.
Last edited:
DF Weekly: If Xbox Series X is more powerful, why do some PS5 games run better?
- Thread starter Fenton
- Start date
OP
OP
It is so weird this last pic...
The guy says that talked with Devs and they said the power differece is staggering.
It is so bullshit.
Jason (probably one of the few that really talk with devs) in a podcast way before the launch said:
And of course he was right.
Bonus... people at ERA making jokes lol.
Jason Schreier: Playstation 5 more superior piece of Hardware in a lot of different ways despite of what you see in Spec Sheets
https://podcasts.google.com/?feed=aHR0cHM6Ly9yc3MuYXJ0MTkuY29tL2tvdGFrdS1zcGxpdHNjcmVlbg&episode=Z2lkOi8vYXJ0MTktZXBpc29kZS1sb2NhdG9yL1YwL2kwUmJiWmUwMGdrSEJ0MkRnUXpfMjVFeWVhaFBEODB0b2ZKdjBoVDJOakU&hl=de&ved=2ahUKEwiY-bHO36boAhWMEMAKHWM4AqgQieUEegQIChAE&ep=6 A very interesting quote from Jason...
www.resetera.com
Microsoft created the illusion to Xbots that Series X could run Minecraft in full path-tracing (basically everything in RT)
Remember the tools excuse when the first games launched like Dirt and DmC ran better on PS5
It become a meme.
It a fucking lie to try to damage control at time.
Edit - Eurogamer did a entry interview to talk about the "tools".
I forget to say a bit more...
What Microsoft did with the memory setup of Series is something so atrocious that nVidia got (and lose) several lawsuit from gamers when they did something similar with GTX 970 and it 4GB of RAM that generate performance issues when you used more than 3.5GB due 512MB having half of the speed
nVidia had to pay $30 do each buyer of GTX 970 in US.
www.guru3d.com
But for Microsoft nothing happened
What Microsoft did with the memory setup of Series is something so atrocious that nVidia got (and lose) several lawsuit from gamers when they did something similar with GTX 970 and it 4GB of RAM that generate performance issues when you used more than 3.5GB due 512MB having half of the speed
nVidia had to pay $30 do each buyer of GTX 970 in US.
Nvidia settles GeForce GTX 970 memory scandal in the USA
Nvidia gives American owners of GTX 970 may receive $ 30 back. That is the outcome of a settlement that the company has made in a class action case from disgruntled buyers after the scandal about the ...
www.guru3d.com
But for Microsoft nothing happened
Always stating it in a way that affirms Xbox as the more powerful console. If it was a more powerful console it would've been consistently beating the PS5 for the past 4 years and that didn't happen at all,, it gets beat by the PS5 all the time including on huge titles.
Last edited:
why there are always that one braindead?
Thread from last year
icon-era.com
Digital Foundry finally admits that PlayStation 5 is better designed than Xbox Series X
Something that I would say most of us already knew but what an admission from DF All that bullshit about TFLOPS.
icon-era.com
Oh, they know perfectly well - and I'd argue they've known perfectly well since the first year when results started coming in and PS5 was even or better in many games. It's just a mixture of ego, being paid by MS PR, and hatred of Sony, that means that they will never openly admit that the PS5 has superior engineering.
Even after all these years the DF team is still unable to admit they were wrong to just pick on the compute throughput numbers and assume that meant cOnSoLe pOwEr, even after so many tech experts and software developers came out against that very obviously wrong idea.
Not only that, they also went with the "9 TFLOPs average" made-up FUD that kept being spread at the start, even after the PC RDNA2 GPUs came out with core clock numbers largely surpassing the PS5's 2.23GHz.
Ignorant to what the numbers meant, they simply went with parroting Microsoft's marketing material because Sony didn't give them any exclusive insight.
The truth always was that the PS5 has ~15% lower theoretical compute throughput but:
- It runs its effective throughput closer to its theoretical 10TFLOPs limit because it has the same amount of Shader Engines and Shader Arrays (meaning more caches, registers and work schedulers per Compute Unit).
- It has the same amount of ROPs but clocking 22% faster, and it uses the older RDNA1 ROPs that are less area-optimized but have higher throughput-per-clock under certain conditions (transparency? can't remember..). This means the PS5's fillrate is significantly higher.
- It has the same number of geometry processors but clocking 22%, higher meaning it does triangle setup 22% faster (and the hardware triangle setup is still used even in UE5's Nanite for larger triangles).
- It has 20% lower memory bandwidth, but the Series' uneven memory pools with different bandwidths created problems on their own.
Also, the higher memory bandwidth would have helped the most with 2 things: running at very high 4K resolutions (which neither console ended up doing because of temporal upscaling becoming widely adopted), and heavy RT (which neither console ended up doing because RDNA2 was never designed for heavy RT anyways).
This is not to say the Series X's SoC isn't good foor what it was meant to do. At the cost of increased die area, its wider GPU + lower clocks allow it to be more power efficient than the PS5. This makes sense if Microsoft was thinking of a future based off cloud gaming services where they're the ones paying the energy bill and not the consumers.
There's a reason why the Series X SoC was first presented by their "Azure Silicon Architecture Team" and not a 1st-party game developer like 343i or The Coalition.
Not only that, they also went with the "9 TFLOPs average" made-up FUD that kept being spread at the start, even after the PC RDNA2 GPUs came out with core clock numbers largely surpassing the PS5's 2.23GHz.
Ignorant to what the numbers meant, they simply went with parroting Microsoft's marketing material because Sony didn't give them any exclusive insight.
The truth always was that the PS5 has ~15% lower theoretical compute throughput but:
- It runs its effective throughput closer to its theoretical 10TFLOPs limit because it has the same amount of Shader Engines and Shader Arrays (meaning more caches, registers and work schedulers per Compute Unit).
- It has the same amount of ROPs but clocking 22% faster, and it uses the older RDNA1 ROPs that are less area-optimized but have higher throughput-per-clock under certain conditions (transparency? can't remember..). This means the PS5's fillrate is significantly higher.
- It has the same number of geometry processors but clocking 22%, higher meaning it does triangle setup 22% faster (and the hardware triangle setup is still used even in UE5's Nanite for larger triangles).
- It has 20% lower memory bandwidth, but the Series' uneven memory pools with different bandwidths created problems on their own.
Also, the higher memory bandwidth would have helped the most with 2 things: running at very high 4K resolutions (which neither console ended up doing because of temporal upscaling becoming widely adopted), and heavy RT (which neither console ended up doing because RDNA2 was never designed for heavy RT anyways).
This is not to say the Series X's SoC isn't good foor what it was meant to do. At the cost of increased die area, its wider GPU + lower clocks allow it to be more power efficient than the PS5. This makes sense if Microsoft was thinking of a future based off cloud gaming services where they're the ones paying the energy bill and not the consumers.
There's a reason why the Series X SoC was first presented by their "Azure Silicon Architecture Team" and not a 1st-party game developer like 343i or The Coalition.
Knowing how to make games doesnt mean you know everything about the industry or mean your not a dishonest Microsoft nuthugger