Support NeoGAF

[CrustyBritches]

He also said it would be 8TF

Nope. Gonzalo was 1.8GHz, Oberon 2GHz, and the final version in Prospero is variable 2.2GHz. They were on the trail the whole time.

 

[Deto]

Nope. Gonzalo was 1.8GHz, Oberon 2GHz, and the final version in Prospero is variable 2.2GHz. They were on the trail the whole time.

Digital Foundry predicted that the new generation would be 5700XT, it made sense to say that it would not be 36CUs because that would be "8TF" and nobody imagined that with RDNA2 the clock could be increased to [email protected] 10.28TF

True. DF has gone to Sony numerous times to ask about VRS, and Sony refused to answer.
Sometimes you learn more from what people don't say, than they say.
If the PS5 has VRS, and they have been asked to clarify it by DF, and they don't, I can only guess that's because it doesnt have it.

Why should a Sony answer something to DF?


After Cerny made a presentation saying:

- SSD is not just for loading

- any downclock will be minimal.

- increasing the clock improves everything on the GPU.

DF shit out of his mouth:

- SSD is only for loading.

- PS5 never runs with CPU and GPU simultaneously at full clock.

- clock increase does not improve anything, because the 5700XT above 2ghz does not improve anything.

would you talk to any journalist who calls you a liar?

And guess what, DF isn't interested in listening to Sony, because it didn't hear anything from Cerny's lecture as I just demonstrated.

If I was interested in listening, we would not have the FUD festival encouraged by DF, and by that time DF would have already made a video about Geometry Engine

GE >> VRS.


===========

Linus apologized after talking nonsense about the PS5's SSD.

where is the "dictator" apologizing after saying that the SSD is only for loading? That the PS5 is never at full clock?

Will Leadbetter apologize after implying that Cerny lied, that raising the clock above 2ghz doesn't improve the GPU when AMD releases "6900XT" at 2.2~2.3 ghz?

 

[Deto]

You can't find any actual example of bias in that article can you? Claims without supporting evidence are meaningless. The article you posted as proof is anything but, in fact the article is perfectly reasonable and even-handed.

"In less than 5s I showed random proof" No. You showed a link to a Gaf thread where people, who think like you, were thinking like you. That's not proof of anything.

Show us the quotes, show us the examples, be specific. If DF is so biased you should have no problem.

while the 900p/1080p differential turned out to be much less pronounced than the raw maths suggested. Indeed, as the major titles rolled out in Q4, we saw resolution parity in key titles such as GTA 5, FIFA 15, Destiny and Assassin's Creed Unity. On other tentpole games like Far Cry 4 and Call of Duty: Advanced Warfare, PS4 maintained a resolution advantage, but raw pixel count wasn't the most crucial element of image quality (though it played much more of a role in the multiplayer portion of COD).

- 900p/1080p differential turned out to be much less pronounced than the raw maths suggested.

- major titles rolled out in Q4, we saw resolution parity in key titles such as GTA 5, FIFA 15, Destiny and Assassin's Creed Unity

- PS4 maintained a resolution advantage, but raw pixel count wasn't the most crucial element of image quality

will we have an article from DF saying that the difference between 1440p and 2160p is also "much less pronounced than the raw maths suggested"?

because yes, the difference between 900p and 1080p on a 1080p TV is MUCH BIGGER than between 1440p and 2160p on a 2160p TV

the best was the damage control to say that GTA 5 has resolution parity, and that resolution doesn't matter. GTA 5 is an example of PS4 running graphics better at the same resolution.


Ryse is still one of the most visually striking titles on Xbox One, and as this PC comparison demonstrates, its image quality holds up, despite the resolution drop compared to the computer version.

LOL

The Xbox one 900p doesn't even lose to the PC.

This behavior of the DF is disgusting.


Even I didn't know that DF was such an xbox fanboy, I was disgusted.

DF is a cancer of the gaming world.

 

[CrustyBritches]

Digital Foundry predicted that the new generation would be 5700XT, it made sense to say that it would not be 36CUs because that would be "8TF" and nobody imagined that with RDNA2 the clock could be increased to [email protected] 10.28TF

DF was following the trail of the actual PS5 hardware. Fanboys can not reconcile this in their minds, because they believe they are right even when they were way off they will just move the goalposts.

Oberon was listed at 2GHz and Dark10x was explicitly talking about RT being used for select effects. The difference between 36CU @ 2GHz and 36CU variable 2.23GHz with 1-2% drop in frequency is nothing. Only a fanboy thinks 100-200MHz means anything.

“Bu...bu...but Matt/Kree/BG/Osiris said dat...” hahaha.

 

[CrustyBritches]

Mobile double post

 

[Deto]

Only a fanboy thinks 100-200MHz means anything.

Sony is her own fanboy

 

[Xplainin]

Digital Foundry predicted that the new generation would be 5700XT, it made sense to say that it would not be 36CUs because that would be "8TF" and nobody imagined that with RDNA2 the clock could be increased to [email protected] 10.28TF




Why should a Sony answer something to DF?


After Cerny made a presentation saying:

- SSD is not just for loading

- any downclock will be minimal.

- increasing the clock improves everything on the GPU.

DF shit out of his mouth:

- SSD is only for loading.

- PS5 never runs with CPU and GPU simultaneously at full clock.

- clock increase does not improve anything, because the 5700XT above 2ghz does not improve anything.

would you talk to any journalist who calls you a liar?

And guess what, DF isn't interested in listening to Sony, because it didn't hear anything from Cerny's lecture as I just demonstrated.

If I was interested in listening, we would not have the FUD festival encouraged by DF, and by that time DF would have already made a video about Geometry Engine

GE >> VRS.


===========

Linus apologized after talking nonsense about the PS5's SSD.

where is the "dictator" apologizing after saying that the SSD is only for loading? That the PS5 is never at full clock?

Will Leadbetter apologize after implying that Cerny lied, that raising the clock above 2ghz doesn't improve the GPU when AMD releases "6900XT" at 2.2~2.3 ghz?

Linus apologized for misrepresenting what Epic said about the SSD. They said "best in class", which Linus read as "fastest SSD".

Mark Cerny gave DF a one on one interview about the PS5. So while you might think DF is biased, Cerny obviously doesn't. They asked a follow up question for him to clarify if the PS5 has VRS, and they never got a reply.

It's not a long bow to draw to assume its because it doesnt have VRS.
And VRS and GE have no equivalency.
Its like saying GPU > RAM.

 

[Tripolygon]

Hang on. Is this a satire account? If so apologies, I've been taking it seriously all this time.

The whole article was a damage control to show the diminished importance of resolution. The article postulates that resolution is a mere means to compare the relative performance and not a factor that shows a perceptible difference in real life. Do people not know how to read anymore?

In some ways it's a grim state of affairs - especially for Microsoft - because as 2014 progressed, resolution as a meaningful, differentiating issue of the gameplay experience between multi-platform titles became less important. The amount of visually compromised 720p/792p titles appearing on Xbox One dwindled as the year progressed, while the 900p/1080p differential turned out to be much less pronounced than the raw maths suggested. Indeed, as the major titles rolled out in Q4, we saw resolution parity in key titles such as GTA 5, FIFA 15, Destiny and Assassin's Creed Unity. On other tentpole games like Far Cry 4 and Call of Duty: Advanced Warfare, PS4 maintained a resolution advantage, but raw pixel count wasn't the most crucial element of image quality (though it played much more of a role in the multiplayer portion of COD).

The question of whether resolution actually matters is explored nicely in GameSpot's recent Reality Check video, in which Cam Robinson oversees a 'blind taste test' of sorts across three platforms - PS4, Xbox One and PC. Far Cry 4 and COD are the main examples here, illustrating that despite a yawning chasm in resolution between PS4 and Xbox One, it's very hard for the majority of those participating to tell any difference in actual gameplay conditions. What's crucial in this case is that not only are both COD and Far Cry 4's res reductions well-handled on Xbox One, they also have performance profiles equivalent to or even better than their PS4 counterparts - and we're firmly of the belief that frame-rate difficulties have much more of an impact on the overall experience than resolution.

If you are wondering what reality check video they are talking about, it is a video where well you watch it for yourself.

Robinson could have chosen titles where the difference is more pronounced - we would have enjoyed seeing the same test undertaken on the 720p vs 1080p Metal Gear Solid 5: Ground Zeroes, for example - but as an indicator for how close multi-platform titles are becoming, the games are well chosen. Call of Duty on Xbox One operates mostly at 1360x1080, yet looks remarkably close to the full HD PS4 version in motion. Similarly, Far Cry 4 runs at 1440x1080 on Xbox One, while again running at an uncompromised 1080p on the Sony console. The PS4 version is cleaner and more pleasing to our eyes, but not to a revelatory degree.

And the conclusion?

In the meantime, perhaps the biggest takeaway from the survey data is that it's the Wii U owners that are having the most fun from their gaming hardware

Lets try another article from Richard

However, clearly it's still early days, and right now these machines remain very much uncharted territory - even for those who've been working with prototype hardware for a long time. Microsoft tells developers that the ESRAM is designed for high-bandwidth graphics elements like shadowmaps, lightmaps, depth targets and render targets. But in a world where Killzone: Shadow Fall is utilising 800MB for render targets alone, how difficult will it be for developers to work with just 32MB of fast memory for similar functions? On the flipside, Xbox One's powerful custom audio hardware - dubbed SHAPE (Scalable Hardware Audio Processing Engine) - should do a fantastic job for HD surround, a task that sucks up lots of CPU time on current-gen console. How does PS4 compare there? And just how much impact does the GDDR5 memory - great for graphics - have on CPU tasks compared to Xbox One's lower-latency DDR3?

While next generation of consoles finally arrive in a matter of months, the launch games will have mostly been developed on incomplete hardware - a state of affairs that was blatantly obvious from titles seen so far. On paper, Sony retains a clear specs advantage, but it was difficult to see that reflected in the quality of the games at E3. Based on what we're hearing about the approach to next-gen development, it could be quite some time before any on-paper advantage translates into an appreciably better experience on-screen.

Another article from Richard

The Xbox One vs. PlayStation 4 graphics spec comparison is stark to say the least. Both systems utilise AMD's GCN (Graphics Core Next) architecture, but Sony's rendering tech has 50 per cent more raw computational power than the Xbox One equivalent - and that's factoring out other differences between the systems. The question is, what is the impact in actual gameplay conditions?

Behind the scenes, developers have suggested to us that we shouldn't jump to conclusions about the extent of the PlayStation 4's superiority, and that the 50 per cent boost in GPU power emphatically won't result in a likewise boost to in-game performance.

"The point is the hardware is intentionally not 100 per cent round," Cerny revealed. "It has a little bit more ALU in it than it would if you were thinking strictly about graphics. As a result of that you have an opportunity, you could say an incentivisation, to use that ALU for GPGPU."
An interpretation of Cerny's comment - and one that has been presented to us by Microsoft insiders - is that based on the way that AMD graphics tech is being utilised right now in gaming, a law of diminishing returns kicks in.

Wanna know whats funny, remember when Mark Cerny says raising clocks is better than more CU?

The results pretty much confirm the theory that more compute cores in the GCN architecture doesn't result in a linear scaling of performance. That's why AMD tends to increase core clock and memory speed on its higher-end cards, because it's clear that available core count own won't do the job alone.

All of a sudden more CU is better than higher clocks now. I wonder why?

According to inside sources at Microsoft, the focus with Xbox One was to extract as much performance as possible from the graphics chip's ALUs. It may well be the case that 12 compute units was chosen as the most balanced set-up to match the Jaguar CPU architecture. Our source says that the make-up of the Xbox One's bespoke audio and "data move engine" tech is derived from profiling the most advanced Xbox 360 games, with their designs implemented in order to address the most common bottlenecks. In contrast, despite its undoubted advantages - especially in terms of raw power, PlayStation 4 looks a little unbalanced by comparison.

I don't mind when someone has a bias but we can't deny that it is not there.

On a different note. Y'all need to give it a rest already.

First PS5 was RDNA 1 or 1.9, to it does not have hardware accelerated ray tracing, then it is really a 9TF console and now it does not have VRS or Mesh Shaders. Yes it does, this is a standard RDNA 2 feature set and there is no conceivable reason why PS5 will not have it. It is spelled out in the Road to PS5 video as part of the new feature set of the updated geometry engine.

 

[rnlval]

Yup you have boost and OC, the poster I was countering was suggesting the extra GHz of the 2080 does nothing vs the extra TF of the Ti. I dont look too much at PC performance as I game on console, so thanks for that.

Your data shows clearly both cards effectively reach the same GPU boost clocks anyway, so proves my point that Ps5 will have some benefits of faster GPU clocks to make up ground on the TF differential.

NVIDIA's paper spec boost clocks for RTX GPUs are conservative while it can't be said for AMD RX 5700 XT.

 

[rnlval]

Where do I start ? 14 gbs is the speed of the memory chips, they are the same, XSX has wider bus for the 10 GB, Narrower for the 6 GB, go work it out for yourself there are 2 bandwidth numbers.

Go read up on absraction and apis if you dont understand, google works.

Frame buffer's render targets has largest benefit for higher memory bandwidth.

Using PS4's example

Render targets' memory storage is small while it has the highest memory bandwidth consumer.

XSX was designed for the highest render resolution target with the highest performance for $$

 

[Panajev2001a]

If it’s all about who hits the report button more, whoever has the biggest fan base will get banned less, thus gaining an even bigger amount of support on the site and dominate the report button even more. It’s a terrible stance for the site to hold

SonyGAF conspiracy? Again?!?

 

[Panajev2001a]

The whole article was a damage control to show the diminished importance of resolution. The article postulates that resolution is a mere means to compare the relative performance and not a factor that shows a perceptible difference in real life. Do people not know how to read anymore?




If you are wondering what reality check video they are talking about, it is a video where well you watch it for yourself.




And the conclusion?


Lets try another article from Richard





Another article from Richard





Wanna know whats funny, remember when Mark Cerny says raising clocks is better than more CU?

All of a sudden more CU is better than higher clocks now. I wonder why?

I don't mind when someone has a bias but we can't deny that it is not there.

On a different note. Y'all need to give it a rest already.

First PS5 was RDNA 1 or 1.9, to it does not have hardware accelerated ray tracing, then it is really a 9TF console and now it does not have VRS or Mesh Shaders. Yes it does, this is a standard RDNA 2 feature set and there is no conceivable reason why PS5 will not have it. It is spelled out in the Road to PS5 video as part of the new feature set of the updated geometry engine.

Leadbetter has a long pattern of hyping or reasoning issues away for one side and raising doubts and concern for the other. We can add to that how they were fanning the fire “software only RT“ concern only for one console, small things such as referring to the X1X CPU as custom x86 CPU from AMD while PS4 Pro has just a higher clocked Jaguar (B.S.: both being customised Jaguar cores, but Jaguar has a bad reputation, so... off the specsheet vs chart), look at his PS3 reveal conference reaction video, the PS4 reveal conference reaction one, etc... Leadbetter and Battaglia constantly try to snatch critique from the jaws of praise only when talking about one console manufacturer. Which is what biased tech bloggers do: one side is where hope and positivity almost springs eternal, the other one always constantly needs to prove itself and even when things are good there is some “concern” to be raised.

His stance in his articles or even the videos he partakes in is not unbiased. DF gives a lot of console fodder out.

 

[Redlight]

The whole article was a damage control to show the diminished importance of resolution. The article postulates that resolution is a mere means to compare the relative performance and not a factor that shows a perceptible difference in real life. Do people not know how to read anymore?

This is just another example of the bias being in your interpretation, even your description of what the article is about is, in my opinion, wrong. There is nothing in the articles, or the quotes that you pulled from them, that offers proof of a bias by the author.

What is untrue about what he said or what you chose to highlight? He just didn't pander to your particular set of beliefs.

Here's some other quotes from that original article...

"...but resolution is a parameter by which the differing power levels of the machines can be addressed, and was certainly the way the specialist press - ourselves included - went about it."

"The PS4 version is cleaner and more pleasing to our eyes..."

"Xbox One not only runs at a slower frame-rate, but its imagery is delivered in an inconsistent manner, resulting in further judder."

"Battlefield Hardline launches later this month with a 720p Xbox One resolution and unimpressive image quality as a result, while we suspect we'll see the same situation with Metal Gear Solid 5: The Phantom Pain later on in the year..."

Doesn't sound like an unrepentant Xbox-loving rescue mission to me. The bias you're seeing is in your reading of the material, not the material itself.

 

[ethomaz]

Guys.

The guys asking for receipts, and receiving an example article just didn’t read it because it is against their narrative.

And others followers wants to ask ban for those that showed examples of how bias Richard and Alex are

There is no way anybody can deny DF hype one side with speculations while create concern of the other side with the same speculations.

They before did damage control to Xbox but after 2016 if I’m not wrong they started to do the PR marketing for Xbox and that is the situation today... it is a very old issue with them since DF was created.

 

[rnlval]

The whole article was a damage control to show the diminished importance of resolution. The article postulates that resolution is a mere means to compare the relative performance and not a factor that shows a perceptible difference in real life. Do people not know how to read anymore?

If you are wondering what reality check video they are talking about, it is a video where well you watch it for yourself.


And the conclusion?

Lets try another article from Richard


Another article from Richard

Wanna know whats funny, remember when Mark Cerny says raising clocks is better than more CU?

All of a sudden more CU is better than higher clocks now. I wonder why?

I don't mind when someone has a bias but we can't deny that it is not there.

On a different note. Y'all need to give it a rest already.

First PS5 was RDNA 1 or 1.9, to it does not have hardware accelerated ray tracing, then it is really a 9TF console and now it does not have VRS or Mesh Shaders. Yes it does, this is a standard RDNA 2 feature set and there is no conceivable reason why PS5 will not have it. It is spelled out in the Road to PS5 video as part of the new feature set of the updated geometry engine.

XSX is reaching about RTX 2080 level performance with Gears 5's built-in benchmark. Like other Unreal Engine 4 games, Gears 5 is not Forza/Battlefield V level AMD GPU friendly game.

ASUS ROG RX 5700 XT Strix already has about 10.27 TFLOPS with 448GB/s memory bandwidth

ASUS Radeon RX 5700 XT STRIX OC Review

The ASUS Radeon RX 5700 XT STRIX OC is a huge improvement over the AMD reference design. It comes with an excellent cooler that reduces temperatures and noise levels at the same time, matching NVIDIA's offerings. Idle-fan-stop is included, too, and the factory overclock nets additional performance.

www.techpowerup.com

ASUS ROG RX 5700 XT Strix at 2007 Mhz average clock speed yields about 10.276 TFLOPS.

MS did NOT reveal XSX GPU's non-CU areas such as ROPS unit count to enable XSX GPU to jump above RX 5700 XT level GPUs into RTX 2080 level.

A reminder from AMD's road map

Notice NAVI's scalability improvements with AMD's road map.

NAVI's scalability will be important to compete against NVIDIA's Turing RTX 2080/2080 Ti and Ampere RTX 3070/3080 level GPUs. It's unlikely "BiG NAVI" will rival RTX 3080 Ti.

 

[Panajev2001a]

ASUS ROG RX 5700 XT Strix already has about 10.27 TFLOPS with 448GB/s memory bandwidth

ASUS ROG RX 5700 XT Strix at 2007 Mhz average clock speed yields about 10.276 TFLOPS.

Subtle RDNA1.2 dig at PS5 again ?

MS did NOT reveal XSX GPU's non-CU areas such as ROPS unit count

They are likely the same as PS5’s GPU and thus not useful, or wrong narrative setting (lower clock, same number of units, you do the maths... even if they were more they would still need to be enough to cover the clockspeed gap) in the super long deep dives they did where they enunciated lots of other details (they went into details more than Cerny did on many things on the Road to PS5 video).

 

[Redlight]

This behavior of the DF is disgusting.

Even I didn't know that DF was such an xbox fanboy, I was disgusted.

DF is a cancer of the gaming world.

DF Stating the truth in an even-handed manner is not proof of bias. I'd be happy to have that discussion with you if I thought you were genuine or reachable, however I'd have more chance of convincing an anti-vaxxer that Bill Gates is okey-dokey.

 

[Tripolygon]

This is just another example of the bias being in your interpretation, even your description of what the article is about is, in my opinion, wrong. There is nothing in the articles, or the quotes that you pulled from them, that offers proof of a bias by the author.

What is untrue about what he said or what you chose to highlight? He just didn't pander to your particular set of beliefs.

Here's some other quotes from that original article...

"...but resolution is a parameter by which the differing power levels of the machines can be addressed, and was certainly the way the specialist press - ourselves included - went about it."

"The PS4 version is cleaner and more pleasing to our eyes..."

"Xbox One not only runs at a slower frame-rate, but its imagery is delivered in an inconsistent manner, resulting in further judder."

"Battlefield Hardline launches later this month with a 720p Xbox One resolution and unimpressive image quality as a result, while we suspect we'll see the same situation with Metal Gear Solid 5: The Phantom Pain later on in the year..."

Doesn't sound like an unrepentant Xbox-loving rescue mission to me. The bias you're seeing is in your reading of the material, not the material itself.

Don't end it there, continue to the next paragraph.

Those titles should be outliers though,

and fingers crossed that across 2015 we'll see enough progress that next year's re-run of the Nielsen survey sees the quality of the gaming experience as the motivating factor behind investing in console hardware.

Its the quality of the gaming experience that matters, (or like some fanboys love to proclaim, better controller, better online, friends) not the resolution. Lmao

The article was made to downplay resolution. I don't care if Richard has a bias but we cannot deny that he doesn't. The dude started the PS4 is unbalanced rumor, started the PS4 only has 4.5GB RAM for games, he also started the PS5 can't sustain high clocks on both the CPU and GPU.

 

[Panajev2001a]

DF Stating the truth in an even-handed manner is not proof of bias. I'd be happy to have that discussion with you if I thought you were genuine or reachable, however I'd have more chance of convincing an anti-vaxxer that Bill Gates is okey-dokey.

Oh the irony here...

 

[Redlight]

Don't end it there, continue to the next paragraph.



Its the quality of the gaming experience that matters, (or like some fanboys love to proclaim, better controller, better online, friends) not the resolution. Lmao

The article was made to downplay resolution.

Nonsense. You'r having to torture the article to eek out the bias you want to see and then, in your brackets, invent an implication that simply doesn't exist in the article, just in your head.

 

[ethomaz]

Nonsense. You'r having to torture the article to eek out the bias you want to see and then, in your brackets, invent an implication that simply doesn't exist in the article, just in your head.

The article probably doesn’t exists... it just in our read.
You can’t even read it because you know it doesn’t exists.

Maybe DF doesn’t exists at all? All these biased articles are just a massive alternate reality created by gamers

 

[rnlval]

Subtle RDNA1.2 dig at PS5 again ?


They are likely the same as PS5’s GPU and thus not useful, or wrong narrative setting (lower clock, same number of units, you do the maths... even if they were more they would still need to be enough to cover the clockspeed gap) in the super long deep dives they did where they enunciated lots of other details (they went into details more than Cerny did on many things on the Road to PS5 video).

I don't expect miracles from "RDNA 2" when XSX GPU only rivals RTX 2080 with 448GB/s memory bandwidth while MS/AMD throws higher memory bandwidth against it. I expected more from AMD.

I'm still waiting for RTX 2080 level efficiency from AMD e.g. NVIDIA's real-time memory compression is still superior when compared to AMD's.

You are forgetting memory bandwidth scaling with any TFLOPS increase.

XSX has ten 32 bit memory controllers and RDNA basic design has L2 cache in front of memory controllers e.g. 5MB L2 cache instead of RX 5700 XT's 4MB L2 cache.

Your "They are likely the same as PS5’s GPU " speculation is not superior to my own speculation.

Both X1X and PS4 Pro has forked from Polaris 10 baseline design

PS4 Pro includes RPM for CU and 40 CU design. Sony focused on compute bias improvement.

X1X includes 2MB render cache for ROPS and 44 CU design. MS focused on raster and some compute improvements.

Both improvement areas have appeared in Vega.

 

[thismeinteil]

Is VRS the technology that MS is going to use to help the Series X sell more than half as much as the PS5? Or make them actually pump out quality titles for more than half a gen? No? Oh...

 

[Panajev2001a]

I don't expect miracles from "RDNA 2" when XSX GPU only rivals RTX 2080 with 448GB/s memory bandwidth while MS/AMD throws higher memory bandwidth against it. I expected more from AMD.

I'm still waiting for RTX 2080 level efficiency from AMD e.g. NVIDIA's real-time memory compression is still superior when compared to AMD's.

You are forgetting memory bandwidth scaling with any TFLOPS increase.

XSX has ten 32 bit memory controllers and RDNA basic design has L2 cache in front of memory controllers e.g. 5MB L2 cache instead of RX 5700 XT's 4MB L2 cache.

Your "They are likely the same as PS5’s GPU " speculation is not superior to my own speculation.

Both X1X and PS4 Pro has forked from Polaris 10 baseline design

PS4 Pro includes RPM for CU and 40 CU design. Sony focused on compute bias improvement.

X1X includes 2MB render cache for ROPS and 44 CU design. MS focused on raster and some compute improvements.

Both improvement areas have appeared in Vega.

PS4 Pro actually has more ROPS than X1X but that actually exceed bandwidth.

Do you have details about the extra 2 MB render cache on X1X?

The difference between that comparison is the next generation systems is that in the previous case the other design had an extra 12 months of R&D put into it and thus is like another mini GCN revision.

We will see, quite confident that you will see the number of ROPS and ACE’s being the same in both consoles. Not an area MS needed to waste money on.

 

[The Hamburglar]

The whole article was a damage control to show the diminished importance of resolution. The article postulates that resolution is a mere means to compare the relative performance and not a factor that shows a perceptible difference in real life. Do people not know how to read anymore?




If you are wondering what reality check video they are talking about, it is a video where well you watch it for yourself.




And the conclusion?


Lets try another article from Richard





Another article from Richard





Wanna know whats funny, remember when Mark Cerny says raising clocks is better than more CU?

All of a sudden more CU is better than higher clocks now. I wonder why?

I don't mind when someone has a bias but we can't deny that it is not there.

On a different note. Y'all need to give it a rest already.

First PS5 was RDNA 1 or 1.9, to it does not have hardware accelerated ray tracing, then it is really a 9TF console and now it does not have VRS or Mesh Shaders. Yes it does, this is a standard RDNA 2 feature set and there is no conceivable reason why PS5 will not have it. It is spelled out in the Road to PS5 video as part of the new feature set of the updated geometry engine.

Thank you for going to that trouble. It’s quite obvious Richard at DF and EG in general have an Xbox bias. All those articles you posted I remember reading them years ago and thinking they’re not even trying to hide it.

it’s a shame all the posters in this thread can’t seem to read or remember.

 

[SatansReverence]

Guys.

Guy.

Your "evidence" does not, in any way, support your claim.

 

[Clintizzle]

Lmao that DF article was basically what every other games media were saying that time.

I guess everyone was biased towards M$ but just never bought an Xbox hahha

 

[Dodkrake]

The FUD continues

I'm going to repeat myself:

The PS5 does not have VRS as this is a MS Tech. Link for those that want to be bothered to read a bit. It's possible the PS5 has a similar tech implemented, but it would still not be VRS.

18 pages of non-discussion because fanboys can't do a simple google search

 

[Thirty7ven]

Lmao that DF article was basically what every other games media were saying that time.

I guess everyone was biased towards M$ but just never bought an Xbox hahha

Here’s a pattern of the western media. Xbox comes out with always online DRM, they write it’s how things are going to be for everyone. Xbox offers a lower resolution standard, they write the human eye won’t notice. Xbox sells less than the competition, they write don’t look too much into it, the numbers don’t tell the story. Xbox 1X releases, they write resolution is back on the menu, stunning 4K has to be seen to be believed. Xbox ends the generation dead last, they write you the top reasons why it won the generation. Etc etc

Maybe it’s because Xbox is the “underdog” story. Which is weird because Sony is definitely the “little guy” when compared to MS.

 

[semicool]

The FUD continues

I'm going to repeat myself:

The PS5 does not have VRS as this is a MS Tech. Link for those that want to be bothered to read a bit. It's possible the PS5 has a similar tech implemented, but it would still not be VRS.

18 pages of non-discussion because fanboys can't do a simple google search

Another VRS type solution, called by another name, most likely would still give "evidence" (which was not seen by DF) that's it's being used if in fact it's there and it's being used should it not?

Note, Geometry Engine on ps5 is culling geometry...ie..mesh shaders on X, not per pixel shading variation(VRS). Non equivalent but a different optimization technique with different results.

 

[sinnergy]

Here’s a pattern of the western media. Xbox comes out with always online DRM, they write it’s how things are going to be for everyone. Xbox offers a lower resolution standard, they write the human eye won’t notice. Xbox sells less than the competition, they write don’t look too much into it, the numbers don’t tell the story. Xbox 1X releases, they write resolution is back on the menu, stunning 4K has to be seen to be believed. Xbox ends the generation dead last, they write you the top reasons why it won the generation. Etc etc

Maybe it’s because Xbox is the “underdog” story. Which is weird because Sony is definitely the “little guy” when compared to MS.

Everything goes online ? It was the always online vision ... but they where a console generation to early ... see how many reacted positively for the all digital PS5. Humans can only change so fast ...

 

[killatopak]

Everything goes online ? It was the always online vision ... but they where a console generation to early ... see how many reacted positively for the all digital PS5. Humans can only change so fast ...

All digital is different from always online.

Every digital game I played can be played offline.

 

[Panajev2001a]

Here’s a pattern of the western media. Xbox comes out with always online DRM, they write it’s how things are going to be for everyone. Xbox offers a lower resolution standard, they write the human eye won’t notice. Xbox sells less than the competition, they write don’t look too much into it, the numbers don’t tell the story. Xbox 1X releases, they write resolution is back on the menu, stunning 4K has to be seen to be believed. Xbox ends the generation dead last, they write you the top reasons why it won the generation. Etc etc

Maybe it’s because Xbox is the “underdog” story. Which is weird because Sony is definitely the “little guy” when compared to MS.

Spot on...

 

[FranXico]

The FUD continues

I'm going to repeat myself:

The PS5 does not have VRS as this is a MS Tech. Link for those that want to be bothered to read a bit. It's possible the PS5 has a similar tech implemented, but it would still not be VRS.

18 pages of non-discussion because fanboys can't do a simple google search

Of course it's not the same exact thing. There are different kinds of VRS. Read Microsoft's patent.

 

[The Hamburglar]

I’m beginning to think that VRS is new secret sauce/power of the cloud.

 

[quest]

I’m beginning to think that VRS is new secret sauce/power of the cloud.

Im sure nvidia totally wasted transistors to help Microsoft market secret sauce lol.

Im sure the PS5 has vrs Sony could clear this up like many things with a quick phone interview with digital foundry. Don't blame digital foundry because Sony is being secret squirrels about anything that is not the SSD or tempest engine.

 

[Deleted member 775630]

Here’s a pattern of the western media. Xbox comes out with always online DRM, they write it’s how things are going to be for everyone. Xbox offers a lower resolution standard, they write the human eye won’t notice. Xbox sells less than the competition, they write don’t look too much into it, the numbers don’t tell the story. Xbox 1X releases, they write resolution is back on the menu, stunning 4K has to be seen to be believed. Xbox ends the generation dead last, they write you the top reasons why it won the generation. Etc etc

Maybe it’s because Xbox is the “underdog” story. Which is weird because Sony is definitely the “little guy” when compared to MS.

How do you explain all those analysis in the beginning of past gen where they clearly said you could see the difference between PS4 and XBox One in 3rd party games? CoD for example... It first ran at 720p and 1080p for Xbox and PS4 respectively, afterwards it got an update for Xbox to 900p. This really reads like they said there's no difference...

Each next-gen console version has its strengths and weaknesses, but clearly it is the PS4 game that offers a superior experience overall. The advantage here comes in the form of cleaner and sharper visuals that help to better realise the extra next-gen spit and polish on display over the 360, PS3, and Wii U versions of the game - something that Xbox One's upscaled 720p presentation fails to do in quite the same way. Source]

 

[Deto]

Linus apologized for misrepresenting what Epic said about the SSD. They said "best in class", which Linus read as "fastest SSD".

Mark Cerny gave DF a one on one interview about the PS5. So while you might think DF is biased, Cerny obviously doesn't. They asked a follow up question for him to clarify if the PS5 has VRS, and they never got a reply.

It's not a long bow to draw to assume its because it doesnt have VRS.
And VRS and GE have no equivalency.
Its like saying GPU > RAM.

 

[sinnergy]

Yeah, it must be totally worthless... waste of time engineering it.

PR that’s what it all is ...

 

[Dodkrake]

Another VRS type solution, called by another name, most likely would still give "evidence" (which was not seen by DF) that's it's being used if in fact it's there and it's being used should it not?

Note, Geometry Engine on ps5 is culling geometry...ie..mesh shaders on X, not per pixel shading variation(VRS). Non equivalent but a different optimization technique with different results.

Absence of evidence is not evidence of absence.

Just because you can't see something like this being deployed, it doesn't mean it's not occurring (nor it means it is).

 

[Deto]

Yeah, it must be totally worthless... waste of time engineering it.

PR that’s what it all is ...

expected, everything that speaks well on PS5 is PR.

lol

 

[martino]

this is called a nice dodge.
impossible this guy doesn't know mesh shader is a thing.

 

[Xplainin]

A CPU doesnt hold a candle to the processing power of a GPU.

Sigh.

 

[Deto]

this is called a nice dodge.
impossible this guy doesn't know mesh shader is a thing.

While VRS lowers the resolution to process fewer pixels, GE does not even process those pixels.

PlayStation 5: Does the new Sony console deliver the next-gen dream?

Yes, we finally saw the console, we know what it looks like and how gigantic it is - but crucially, yesterday's PlaySta…

www.eurogamer.net

another aspect of the Cerny presentation paid off spectacularly. One of the key features of the new GPU is the Geometry Engine, giving developers unprecedented control over triangles and other primitives and easy control over geometry culling. There's nothing new in terms of principles here - it boils down to removing the need to render triangles that end up being invisible in the final frame. The less geometry you process, the less work there is for the GPU, meaning that resources can be used elsewhere




Well, I went to research and everything I posted here has already been said.

This is just a topic from the "RDAN1" era trying to resurface with FUD to talk shit about PS5.

 

[FranXico]

A CPU doesnt hold a candle to the processing power of a GPU.

Sigh.

He is not talking about the CPU at all. You do know that the GE is also a GPU feature, right?

 

[Xplainin]

He is not talking about the CPU at all. You do know that the GE is also a GPU feature, right?

Umm, yeah. Looks like my post was lost on you.
Him saying that VRS cant hold a candle to GE, is like saying that a CPU cant hold a candle to a GPU.
They are all seperate things, and do different jobs, and can't be compared to each other.
Its a moot point.

 

[JimboJones]

Absence of evidence is not evidence of absence.

Just because you can't see something like this being deployed, it doesn't mean it's not occurring (nor it means it is).

I'd like to point out that the digital foundery article did not conclude that it's absence was indicative of the PS5 not having it.
They simply mentioned in one sentence that they didn't see evidence of it.
A fairly innocuous statement and well within line of their reporting on technical aspects.

Some people here might have twisted it to mean other things (rdna1 based there for massively weaker etc) when there is no definite evidence of that and most likely not true.
But at the same time others are taking to ridiculous defence tactics and trying to smear DF as some biased news source.

 

[semicool]

Absence of evidence is not evidence of absence.

Just because you can't see something like this being deployed, it doesn't mean it's not occurring (nor it means it is).

True, although it is by far the most likely explanation. Occam's Razor.

 

[ethomaz]

The FUD continues

I'm going to repeat myself:

The PS5 does not have VRS as this is a MS Tech. Link for those that want to be bothered to read a bit. It's possible the PS5 has a similar tech implemented, but it would still not be VRS.

18 pages of non-discussion because fanboys can't do a simple google search

Each party has it own software implementation of VRS.
Vulkan and OpenGL has VRS implemented via extension and it is not exactly like the software solution on DX12U.

That is the point anybody can implement VRS in their API since the hardware support the feature... that includes Sony.

 

[FranXico]

Umm, yeah. Looks like my post was lost on you.
Him saying that VRS cant hold a candle to GE, is like saying that a CPU cant hold a candle to a GPU.
They are all seperate things, and do different jobs, and can't be compared to each other.
Its a moot point.

Thanks for clearing it out.
He's arguing that since GE prunes geometry beforehand, it has just as much of an impact on performance as VRS.
Ideally, you should do both, of course.