How Microsoft's marketing is killing the PS5 hype – Reader’s Feature

A reader examines the marketing battle over the tech specs of the Xbox Series X and PS5 and how both Sony and Microsoft are bending the truth.

While it’s clear from marketing analysis that there is a greater public interest in the PlayStation 5 right now, it would certainly appear that Microsoft have made significant ground in their early marketing campaign. To be specific, Xbox Series X is dominating the conversation online.


‘12 teraflops of power, true 4K gaming and 8K ready, up to 120 fps, 1TB custom SSD.’ Are the key words promoting the Xbox Series X, clear and concise. It’s a much easier sell than the ‘imagine the possibilities’ that Sony are promoting their SSD with, without a hands-on demonstration. Microsoft are very keen to sell the power behind their new console; the word power itself appearing 24 times on the Xbox Series X homepage.

While teraflops is a metric pertaining to computing speed and therefore power, having more teraflops is purely a semantic advantage. TFLOPs are spurious indicators of performance but nevertheless have become an important number in advertising consoles and graphics cards. This is something that the marketing teams of AMD and Nvidia are well aware of, and it seems that Microsoft too have had this in mind for a long time.

The TFLOPs to performance outcomes vary drastically from one GPU to another, depending on the architecture. Even GPUs in the same series reveal that there is no 1:1 ratio for power and performance. To give context for this, we can observe the differences between two of the top cards on the market, both in the same series.

GeForce RTX 2080 Ti compared to the RTX 2080 Super has: 21% more teraflops, 25% more memory bandwidth, 38% more ROPs, 38% bigger L2 cache, 38% more RAM.
Benchmark performance difference? It’s between 10 and 15%, depending on the site.

To give another and more relevant example of AMD cards, specifically RDNA (the precursor to the Xbox Series X and PlayStation 5 GPU):

The RX 5700 XT compared to the RX 5700 has: 23% more teraflops, equal bandwidth, equal ROPs, equal L2 Cache, equal RAM.
Benchmark Performance: <10% on average

Xbox Series X compared to the PlayStation 5 has: 18% more teraflops, 25% more memory bandwidth (5/8), 25%(-) Less bandwidth (3/8), equal ROPs (TBC), equal L2 cache (TBC), and equal but less flexible RAM.

Comparing those three scenarios and a reasonable estimate would be the Xbox Series X will have better average frame rate performance at the same resolutions by 9% or less. This means the two will basically run the same settings albeit with different feature sets, and is the reason the most informed commentators suspect the two consoles are very similar. But the story doesn’t end there.

There’s also the APIs (application programming interface) to consider. The Xbox Series X uses DirectX12 Ultimate, which not only provides a unified codebase between Xbox Series X and PC, but has multiple features that cleverly boost performance. Meanwhile PlayStation 5 uses PSSL: their own programming language featuring the Geometry Engine. Not much is known about it other than it has access to similar features but with an emphasis on primitive shaders.

A possible advantage that PlayStation 5 has here is that whereas DirectX12 Ultimate is not optimised specifically for Xbox Series X, the opposite is true of PSSL for PlayStation 5. This advantage could perceivably make the performance difference completely negligible, but then there’s also the possibility that the updated PSSL doesn’t match up to DirectX12 Ultimate. Once again, this isn’t the full story of how each console might perform.

Sony miscommunication and misinformation

Sony have really let the narrative get away from them at present. According to Jason Schrier of Bloomberg, Sony’s initial developer conference went over really well with developers. However, there has been significant confusion in the general community over some of the concepts that Mark Cerny mentioned during the talk, especially over the relationship between clock speeds and variable frequency.

Cerny later attempted to clarify the issue in an interview with Digital Foundry, but even with that it has perhaps been better explained by independent commentators. Unfortunately, the vacuum left by the uncertainty of Cerny’s explanation has led to fairly baseless speculation that the sustained performance of the PlayStation 5 is actually 9 teraflops, fuelled by prior leaks and developer comments about the PlayStation 5 developer kits, which have locked profiles not reflected by the actual consumer hardware. These figures have now been hijacked by commenters doing a lot of Microsoft’s work for them.

The clarifications infer that Sony’s machine will most likely be running 10.28 teraflops, mostly as advertised. But to take that with a pinch of salt, it is very possible that the PlayStation 5 doesn’t close the performance gap to the extent my initial analysis would suggest, and it could be that Xbox Series X does in fact perform more than 9% better. But oh no, the story doesn’t end here either.

The difference between graphics cards in the same series is usually achieved by steady increments to the structure of the GPU; clock speed, compute units, ROPs, etc. However, the difference between the Xbox Series X and the PlayStation 5 GPUs is one not well accounted for with graphics cards as a reference. The main reason for that is that the PlayStation 5 GPU is bonkers. The capped 2.23GHz clock is completely unparalleled and perhaps impossible in PC architecture without extremely diminishing returns. The effects of overclocking in PC architecture would not necessarily be the same as what happens in a console, but it is at this point a difficult estimation.

The PlayStation 5 and the Xbox Series X GPUs are very different in structure. Mark Cerny made the assertion with an example that a narrower GPU with higher clock speeds would outperform a much wider GPU with the same teraflops. Observing the differences in graphics cards suggests that teraflops do in fact improve performance less linearly with added compute units.

In reality there is little to verify this, but if true, It could potentially see the PlayStation 5 GPU on an absolute par with or even beyond the Xbox Series X. It’s all complete speculation and could just as easily go the other way, giving Xbox Series X even more of an advantage. Regardless, even in the event this was an oversell by Cerny, it hasn’t landed in the gaming conversation to anywhere near the extent of Microsoft’s boasts about teraflops. Not only has Microsoft been keen to promote this fairly arbitrary metric, but they have been very confident in demonstrating the system also.


Whereas Sony have clearly won the conversation over SSD speeds, the tangible effect of these still remains unpronounced despite the showing of multiple games using their unique features. Microsoft on the other hand have displayed their graphical fortitude by means of a benchmark test with a clear and comparable demonstration. I am of course referring to the Gears 5 test that led Digital Foundry to claim that the Series X was on par with Nvidia’s GTX 2080. This definitely caused excitement amongst Xbox fans and a stir amongst PC gamers.

Whilst it was another brilliant piece of marketing by Microsoft, it was ultimately a very misleading display. The test compared a RTX 2080 Ti down-clocked to the Xbox Series X running an un-optimised Gears 5. The biggest flaw with this test is that Gears 5 is a game that notoriously underperforms on Nvidia hardware, as a game that was originally programmed specifically for AMD (Xbox One) GPUs. There are other unverified caveats to this test but ultimately this has proven to churn the wheel of hype on a platform many gamers still defer to as an authority on gaming. A very smart move on Microsoft’s part.

What can Sony do?

Sony still remains the dominant force in consoles and arguably it is only their absolute dominance that is being threatened this generation. Ultimately, if what Cerny and some third party developers claim is true, the best thing Sony can do is release the PlayStation 5 as soon as possible before their competitors, in order to set a standard and demonstrate their machine.

Further than that they need to solidify the console as the lead platform so that the Xbox Series X can’t gain perceivable third party design advantages with its superior CPU. But given the strong link Microsoft has made with the PC market for the next generation this will be tougher than ever for Sony to achieve. The next generation will certainly be a battle to behold.

By reader Twiggy Smalls

The reader’s feature does not necessary represent the views of GameCentral or Metro.

You can submit your own 500 to 600-word reader feature at any time, which if used will be published in the next appropriate weekend slot. As always, email [email protected] and follow us on Twitter.

Follow Metro Gaming on Twitter and email us at [email protected]

For more stories like this, check our Gaming page.

Source: Read Full Article