PS5 vs Series X Performance Comparison Explained
The Xbro vs BaeStation Showdown
The brand new consoles from Microsoft and Sony have officially launched. And gamers around the world are eagerly anticipating buying an Xbox Series X and a PlayStation 5, aka Xbro and BaeStation. And as such, with every new generation comes the inevitable comparison of which Console is the most powerful.
So with that being said the Xbox Series X completely dominates the PlayStation 5.
Wait, so you’re telling me that’s not right. And you’re saying that both consoles have practically identical graphics.
Well, the Xbox Series X is the only console to achieve native 4K in every video game, and as Xbox Executives have said it’s the most powerful console in the world.
Wait, so you’re saying that’s not right either. And that the Xbox Series X is actually running games and dynamic 4K just like the PlayStation 5 and some cases lower resolutions, and Xbox Executives changed their slogan to World’s most powerful Xbox.
Well, the resolution doesn’t matter because all that power has to be put into running games at frame rates no other console could ever achieve, because according to Xbox Executives it’s the BEAST that eats monsters for breakfast.
Wait so you’re saying that the PlayStation 5 is also running games at 120 frames per second and in some cases keeping a higher frame rate than the Xbox Series X version of games.
Hello and welcome to Briscoe games and our first segment known as WTFreak. This is a segment that will cover multiple topics and help explain the reasoning behind why things are the way they are, or why some endeavors didn’t turn out the way people theoretically expected them to. And our first segment is none other than the PlayStation 5 versus the Xbox Series X.
On paper, the Xbox Series X should be the clear winner when it comes to hardware specs, however, the PlayStation 5 has been outperforming the Xbox Series X in several games which is definitely not what was marketed or advertised to fans of the Xbox brand by Microsoft.
Now let me be perfectly clear and say that this is not a bash Xbox article/video, as I’ve personally been a fan since the original Xbox, and you can clearly see in our review of the Xbox Series X we think the console is an amazing piece of hardware even as we were working on this topic. However I do consider myself a journalist, and as a communication major, a computer science minor, and a developer myself I simply love talking and discussing the truth about tech, games, and hardware and so this is simply to explain what exactly is going on with the Xbox Series X that’s preventing it from achieving the results that on paper it should be capable of, and how the PlayStation 5 is punching above its weight. So without further ado let’s dive into the four reasons why the PlayStation 5 is punching above its weight and the Xbox Series X is not.
Terminology: Rumor-Fact, a statement that has not been confirmed, but is 99.9% true.
The Devkits, but more so Tools
Now, this is the official statement, released from Microsoft on why the performance of the Series X is falling slightly behind the PlayStation 5, and fans have run with it from both sides as a legitimate reason, or simply an excuse. Xbox has gone on record to state that they were behind Sony in manufacturing their console. According to Xbox president feel Spencer Xbox was months behind the competition with manufacturing their console.
“We started manufacturing late summer. We were a little bit later than the competition because we were waiting for some specific AMD technology in our chip. We were a little bit behind where they were, where Sony was, in terms of building units. We started in late summer. When you do that, then you have to ship them to all the right retailers and distributors. There’s a time lag, even when you start and even when they’re coming off the assembly line, [until they’re] sitting at retail shelves.”
Phil Spencer, The Verge
But that’s being taken out of context as it’s specifically talking about consumer console manufacturing. There’s no finite date as to when developers actually began receiving devkits or targeted hardware released by Microsoft. Based on leaks we know developer kits for the PlayStation 5 can be traced back as far as November 2019 therefore developers likely had them prior to that (rumor-fact: they actually several months before then almost the beginning of 2019), but we were still unsure of Series X devkits.
But console development happens over years, and looking back at the timeline and Xbox statement of they were waiting for specific RDNA 2.0 tech, we can find an approximation of when devkits were likely in hands of developers as well as another rumor fact. Multiple events that happened during the end of 2019 and the beginning of 2020. For example, AMD hosted at CES in January 2020 and revealed the RX 5600XT as one of the final GPUs launching from RDNA 1, and just two months later reveal the foundation of RDNA 2.0 the same time that Xbox would allow select media to go hands-on with the Series X. But going back even further Xbox had a full render of what the Xbox Series X was going to look like in preparation for the Video Game Awards in December 2019 meaning the design was more than likely completed prior to the show, and being that far into a reveal the system was technically locked into specs at that point meaning developers likely had devkits or at least targeted hardware in November 2019. But going back even further the code name Anaconda can be dated back to late 2018, which turned out to be the exact code name Xbox were using for their console and they were spot on because targeted hardware was already discussed with and being sent to developers first and major 3rd party developers (rumor-fact).
So that brings the question of if Series X targeted hardware was available for almost 2 years, what was really the problem? And the answer is the Tools…somewhat.
The Xbox series console launched with a new toolset known as GDK. GDK will go on to replace the SDK that the Xbox One used, being “available” for PC February 2020 and not releasing four Xbox platforms until June 2020. This is where the game development tools topic became a talking point, and this I can confirm has been an issue for some developers, but not all of them.
The PlayStation 5 has been praised by developers in the industry, as well as the developers that I spoke to personally, primarily because they don’t have to relearn an entirely new toolset as the environment is basically a PS4+. This is due in part to a statement that Mark Cerny made during The Road to PS5 presentation where he stated that the focus of the PlayStation 5 was both evolution and revolution. And not only did they apply that to the hardware they also applied it to the software and tools. The PlayStation 5 development kit is an evolution of the development kit for the PlayStation 4 that revolutionizes it with new and added features to support, optimize, and maximize the benefits of the new PlayStation 5 hardware and architecture; hence evolution and revolution. Look no farther than Richard Leadbetter’s comments regarding how developers he spoke to praised the PlayStation 5’s tools.
“There are two ways to look at this. First of all, I can tell you every single developer I’ve spoken to developing for PS5 has been evangelizing how easy it is to work for. It’s essentially the same development environment as PS4 and you scale up from there, for the new power, the new features, and whatnot. I can’t stress enough how happy developers seem to be with this situation.”
Richard Leadbetter, EGX
However, on the Xbox side of things, GDK replacing SDK hasn’t been nearly as smooth of a transition. While many features of SDK have moved over to GDK offering some sense of familiarity and perform much better, there are still some features that perform worse or flat-out missing. This is due in part to SDK being designed specifically for Xbox One consoles, meanwhile GDK has been designed to take advantage of Xbox One, Xbox One S, Xbox One X, Xbox Series S, Xbox Series X, and PC. So PlayStation developers have had a significantly easier time developing for PlayStation 5 due to having years of experience already developing on familiar tools for the PlayStation 4, Xbox developers have had a mixed bag as far as performance goes with some developers having no real issues while others developers are experiencing somewhat of a learning curve. According to Dirt 5’s Technical Director David Springate, the tools for the Series X are different but familiar, but again Richard Leadbetter has heard mixed things from developers regarding the Series X tools.
“It's not barebones, they brought a lot, no, most of the tools over from of SDK in terms of familiarity and how it works, in fact, stability and speed improved a lot on GDK, but yeah, I am happy with it, if anything there is very little small things I would like, I have weekly meetings with XBOX and my account manager get a lot of feedback from me, I am actually very happy with it.”
David Springate, GameOnDaily
“Speaking to developers, the development environment that they are dealing with, some people seem to be extremely happy with it, other people are having problems with it, because they’ve moved away from what was previously the XDK, which was specific for Xbox, to the GDK, which is a more general development environment for PC and for Xbox, and even Xbox One is encompassed by the GDK. So the long-term win with the GDK is you’re going to be able to develop for it and deploy across all of these systems, but in the short-term, I have heard developers having some problems with it. And whether that is all going to sort of manifest in final games is the big question.”
Richard Leadbetter, EGX
So yes the tools are an issue for developers, but not specifically because they only got them in “June” of 2020, that’s when the tools became open to everyone after the engineering team was “done” with them. The issue was the tools simply weren’t the same as SDK and, therefore, some developers simply didn’t have the experience to get things as optimized as they would have liked. But tools aren’t an end-all explanation of why the Series X isn’t performing quite on par with what you’d expect and the PS5 is winning in performance.
The GPU & Hardware
Now if you look at the hardware of both console on paper at least the Xbox Series X is clearly the more powerful console, but while both consoles use the same overall architecture such as Zen 2 processors, RDNA 2 GPUs, 16GB of GDDR6 RAM, and an NVMe SSD’s there are still some key and important differences between the two.
Now, we can get into the entire Xbox Series X using a server CPU thing and PlayStation 5 having unified L3 cache and therefore Zen 3 or Zen 2.5 rumor, but overall the CPU portion of the console isn’t really a huge talking point as far as the performance difference goes, even with the slightly different CPU clocks. Even though the Series X is using a Server-based CPU it still functions as a regular CPU just with additional server-based code and operations. Also, conducting my own test, something you can do as well, you will find out that the PlayStation 5’s CPU has a similar cut-off as Zen 2 CPUs even with the rumored unified L3 cache confirming that the PlayStation 5 is utilizing Zen 2 and not a Zen 3 architecture. The SSD and RAM are also areas where things differ but we will speak to that later on in the article.
What we’re going to be focused on in this section is the GPU of the consoles and find out what benefits and advantages each GPU has over the other. Many people think it’s a clear win for the Xbox Series X and its 12 TFLOP GPU, however, that’s not 100% the case.
The Xbox Series X is a 52CU GPU clocked at 1.825 GHz, and the PlayStation 5 is a 36CU GPU clocked at 2.233 GHz. You can clearly see the CU advantage that the Xbox Series X has over the PlayStation 5. The Xbox Series X has a 44.4% lead in overall CU count over the PlayStation 5, which is a mere 69.2% of the overall CU of the Series X, or you can say the PlayStation 5 has 30.8% less CUs than the Series X (see how easily you can manipulate numbers to fit whatever narrative you want as each of those percentages are true and represent the exact same thing but look drastically different).
But before we go too far let’s take a look at the TFLOPS of the consoles and introduce a constant. So for anyone who’s taken a science class in the last few years you probably know what a constant is. But for those of us who need a bit of a refresher, a constant is basically a value or a part of an experiment that doesn’t change, or in our case that remains consistent to the other components we will be observing.
So we will be using the RX 6800 as our constant in this experiment. The RX 6800 has 60CUs and a base clock of 1.815 GHz with a boost clock of we will say 2.25 GHz for convenience which it can easily achieve and can be found on partner cards.
Now looking at those two clocks speeds and comparing them to the Series X and PlayStation 5 they instantly look very familiar as the Series X once again has a 1.825 GHz clock speed and the PlayStation 5 has a 2.23 GHz clock speed putting the Series X at 0.01 GHz higher than the base clock of the RX 6800 and the PlayStation 5 at 0.02 GHz lower than the RX 6800’s boost clock which isn’t a coincidence on why the console GPUs ended up being what they are.
But let’s see what the difference in TFLOPS will be on our constant utilizing the Xbox Series X GPU clock speeds and the PlayStation 5 GPU clock speeds.
To get that we need to use the formula for finding TFLOPS which is the number of stream processors (in this case shading units), multiplied by 2, and multiplied by the clock speed of the GPU. You can replace the stream processors with the number of CUs multiplied by the number of stream processors per CU (usually 64) and that will also get you the TFLOPS which would look as follows:
RX 6800: 60CUs * 64sp * 2 * 1.825 GHz = 14.016 (14.0TFLOPS)
RX 6800: 60CUs * 64sp * 2 * 2.23 GHz = 17.14944 (17.1 TFLOPS)
Now, just on clock speeds alone, our constant has shown a +22.1% performance increase in boost clocks over base clocks again using the PlayStation 5s clock speed in comparison to Series X. So this means that the 17.1 TFLOP RX 6800 should have a 22.1% performance boost in games compared to the 14 TF RX 6800 right? Wrong. In fact the 17.1 TFLOP RX 6800 only has around an 8% – 12% advantage in gameplay with an average of around 10%.
This is why teraflops aren’t the end-all of specs when it comes to gaming because teraflop comparisons aren’t truly 1:1 comparisons. A Teraflops is simply a measure of computing specifically floating operations per second, so while TFLOPS can sometimes be used to find approximate performance differences between GPUs of the same architecture to indicate performance in games, it is not a 100% constant especially when dealing with different architectures or GPU customizations.
But getting back to the GPUs the instant standout on what is allowing the PlayStation 5 to remain competitive with the Series X is obviously the higher clock speed of the PS5s GPU. And while CUs and clock speeds are both measures to find out teraflops, which again are not 1:1 comparisons, showing the difference of how CUs and clock speeds can dramatically affect overall GPU performance is still something to explore. So let’s conduct another quick experiment to see just how important the PlayStation 5 clocks are for achieving parity with the Series X.
Let’s take our same constant the RX 6800, the Series X, and the PlayStation 5, and give all the GPUs the Series X clock speed of 1.825 GHz. That would be as follows:
RX 6800: 60CUs * 64sp * 2 * 1.825 GHz = 14.016 (14.0TFLOPS)
Series X: 52CUs * 64sp * 2 * 1.825 GHz = 12.1472 (12.2 TFLOPS)
PS5: 36CUs * 64sp * 2 * 1.825 GHz = 8.4096 (8.4 TFLOPS)
Now TFLOPS again aren’t 1:1 but as you can see that’s a pretty big drop for the PS5 if it used the same GPU clock as the Series X. In real-time performance, this would put the PS5s GPU on par with an RTX 2060 super, which is still a solid midrange performing GPU, but a noticeable 23% less powerful in actual gaming performance than its current RTX 2080 equivalent performance discovered by Digital Foundry.
Now giving the PlayStation 5 its original clock speed back at 2.23 GHz you end up with a GPU with the following specs below, and in practice performance closer to an RTX 2070 super which is around 5% less than the RTX 2080 in gaming performance, so much closer to it’s achieved performance.
RX 6800: 60CUs * 64sp * 2 * 1.825 GHz = 14.016 (14.0TFLOPS)
Series X: 52CUs * 64sp * 2 * 1.825 GHz = 12.1472 (12.2 TFLOPS)
PS5: 36CUs * 64sp * 2 * 2.23 GHz = 10.27585 (10.3TFLOPS)
So that explains the basics of the GPU but it still doesn’t answer the differences and areas where the PS5 actually pulls ahead because so far all we know is that the clock speed is allowing the PS5 to compete, but why is it winning. And the answer is still the clock speed.
As Mark Cerny also stated in The Road to PlayStation 5 the benefits of having a higher GPU clock makes almost every component in the GPU run faster.
“Teraflops is defined as the computational capability of the vector ALU that’s just one part of the GPU. There are a lot of other units, and those other units all run faster when the GPU frequency is higher. At 33% higher frequency rasterization goes 33% faster, processing the command buffer goes that much faster, the L2 and other caches have that much higher bandwidth, and so on. About the only downside is that system memory is 33% further away in terms of cycles but the large number of benefits more than counterbalance that. As a friend of mine says a rising tide lifts all boats.”
Mark Cerny, Road to PS5
And we’re clearly seeing that here as the PS5s rasterization has been the main area where the PS5 wins in comparison to the Series X, higher rasterization is tied to higher FPS and that is why PS5 is slightly edging out the Series X currently in higher frame rate modes. But it’s not only that, higher clocks also produce a higher Pixel Rate, which is the speed at which pixels are rendered on screen. The formula for pixel rate is Render Output Units also known as ROPs * the GPU clock.
Pixel Rate = ROP * GPU clock
Now both the PS5 and Series X have 64 ROPs, but as we discussed the Series X has a 1.825 GHz GPU clock and the PS5 has a 2.23 GHz GPU clock meaning we get the following Pixel Rate for both consoles.
Series X Pixel Rate: 116.8 GPixels/s
PS5 Pixel Rate: 142.72 GPixels/s
This basically means to an extent the PlayStation 5s GPU is better at rendering resolution than the Series X GPU thanks to having a higher clock GPU (hence some cases of dynamic resolutions favoring PS5). But on the flip side, we also have Texture Rate which is similar to Pixel Rate but is the number of textured pixels a GPU renders per second. And the formula for texture rate is almost identical to Pixel Rate except you measure it with the Texture Mapping Units also known as TMUs * the GPU clock.
Now unlike ROPs, the TMUs are different between the two consoles with 1 TMU for every 16 shading units, and as we know the Series X has 44.4% more CUs and Shading Units already, so the Series X ends up with 208 TMUs while the PlayStation 5 only has 144 TMUs. So this is going to be a huge win for the Series X right? Let’s find out.
Series X Pixel Rate: 379.6 GTexel/s
PS5 Pixel Rate: 321.6 GPixels/s
So yes the Series X has a higher GTexel and to an extent is better at rending textures than the PS5, but looking at those two numbers it’s definitely not a 44.4% increase in favor of the Series X. It’s actually only an 18% increase in favor of the Series X over the PS5 once again thanks to the PS5s GPU clock speed. Once again this backs up Mark Cerny’s statement about how pushing faster clock speeds can dramatically increase performance because this is an area that the PlayStation 5 should have lost by a good margin excluding the clock speed difference.
So that’s what’s going on with the GPU side of things. While the Series X is the more powerful of the two, the PlayStation 5 is the faster of the two. A perfect example of this would be if two teams were competing in a race. Team A (PlayStation) has the 8 fastest people on the planet running for them, and Team B has the next 10 fastest people on the planet running for them (9 - 18). Both teams have to run a relay but only 8 people are allowed to run, in a like for like race/workload Team A/PS5 comes out ahead because they have the faster team/console. But if we change it to allow all of Team B to compete and change the competition to see which team can run a total of 10 laps around the track with each individual runner counting as 1 lap, then obviously Team B/Xbox would win because they have more runners/power and can have all 10 laps done by simply having each runner finish one lap, whereas Team A/PS5 will finish 8 laps faster than what Team B/Xbox was capable of, but they still will still need 2 of their runners to complete another full lap before they can reach the goal of 10 but Team B/Xbox will already be done. And that’s the difference between the two. Many GPU tasks require fast movement of data to render games and this is where the PS5 shines, while the more demanding task is where the Series X will flex its muscle such as hardware-accelerated raytracing.
And this brings us to the last component of the primary hardware comparison, the RAM. Again both consoles use 16GB of GDDR6 RAM, however, the bandwidth and how the pools are distributed are once again different. The PlayStation 5 is seamless and to the point, it has a 16GB pool of GDDR6 With a bandwidth of 448 GB/s for a consistent and ideal memory configuration.
The Xbox Series X on the other hand has a 16GB split memory pool with 10GB running at 560 GB/s and a slower 6GB pool running at 336 GB/s. Now, historically speaking having split pools of memory has consistently ended up running into problems. For example, the PlayStation 3 struggled in some areas and was unable to receive some OS features due to having two split 256MB pools of memory in comparison to the Xbox 360s shared 512 MB pool of memory. The GTX 970 was a GPU that eventually had some performance issues as it was a 4GB GPU, but had a 3.5 GB pool of memory and a 512 MB pool of memory. In games that started using four gigs of memory or more the performance of the GTX 970 would start to slightly decline and in some cases the 970 began to lose in comparisons to GPUs that offered a full 4GB of memory that the 970 had consistently beaten in prior comparisons. So for games that use 10GB or less memory, The Series X RAM will be the better solution; however, for games that use more than 10 GB of memory the PlayStation 5s memory solution will be the more ideal configuration.
Now, to the Series X credit the performance degradation from the pools a memory when it comes to games has thus far only been reported as single-digit percentage hits to performance generally 3% - 6% sometimes less, but rarely more. The issue however is games are starting to get more and more demanding when it comes to memory utilization as games like Godfall are already utilizing 12GB of memory on console and PC. So going back to that theoretical 18% advantage that the Series X GPU has over the PlayStation 5, in games with high memory utilization that 18% advantage now drops down to around a 10% – 15% advantage.
So looking at the hardware we can clearly see that the Series X while having a beastly GPU in RAW compute, the PS5 makes up for its disadvantage with a much faster GPU clock speed which helps the PS5s overall performance come much closer to parity to the Series X than many initially believed, and factoring in the differences in memory configurations with games requesting as much RAM as possible the Series X is going to consistently be seeing RAM utilizing be slightly lower performing than the PS5s in games that are not highly optimized for Series X (which is unlikely in 3rd party games hence the performance differences).
Customizations and Lack Thereof
Moving onto reason number three why the PlayStation 5 is slightly outperforming the Series X in early comparisons is because of customizations or a lack thereof on the Series X side.
Now this one is a bit harder to go into as much detail as I did with the GPU side of things and even the tools simply because Sony hasn’t been as forthcoming as Microsoft has when it comes to what exactly they are using internally for the PlayStation 5, however, we do have various tidbits from developers who work on the console as well as just a general information Sony has made available to give us a picture of where things stand.
Starting with the NVMe SSD, and as everyone should know by now this is a key talking point for the PlayStation 5. Now generally from a PC side of things there are I’d say 4 main types of storage, a 5400 rpm HDD, a 7200 rpm or Hybrid HDD, an SSD, and an NVMe SSD. Now when it comes to running games they all do the job, except at different speeds. The faster the drive the faster the loading, and in this case either an SSD or NVMe SSD will allow drastically better loading in games and significantly faster boot times in comparison to the two HDD options. So when Sony discussed this as a key talking point gamers only saw this as being just a fast drive to try to reduce loading, as on PC moving from an SSD to an NVMe SSD which can be up to 10x faster in read and write speeds barely provides more than a few seconds of faster loading.
But what Sony has done is design a console where the bottlenecks for the true speed of the NVMe SSD is no longer an issue, and the reason developers are seeing this as potentially a huge deal is because the speed of the SSD along with the decompression engine has the potential to allow developers to use a portion of the SSD basically as a large pool of DDR3 level memory. This is effectively giving developers 16GB of GDDR6 on the PS5 and a pool of let’s say 20GB or more of DDR3 which is literally a game-changer for any developer in the industry.
Now, this isn’t to say the Series X isn't capable of doing this as well as theoretically it’s also possible however the Series X has a slower NVMe SSD than the PlayStation 5 reaching about 4.8 GB/s compared to around 9 GB/s for the PlayStation 5 and a rumored 12 GB/s theoretical peak performance in comparison to the PS5’s 22 GB/s theoretical peak. Basically with the Series X, we’re looking at potential DDR2 transfer rates on the low-end, or low speed DDR3 transfer rates at peak theoretical performance. While on the PS5 we’re looking at low-level DDR3 transfer rates on the low-end and transfer rates up to DDR4 2666 at peak theoretical speeds. I say theoretical speeds because, in theory, they’re possible, but in practice, they’re unlikely and were more likely to get something in between but closer to the low-end which still works for PS5, but is a big maybe on Series X.
But that’s theory on what may come, let’s talk about what is and what’s now, and that’s various other customizations for the PlayStation 5 in comparison to standardized parts in the Series X. Xbox Series X has a standard SSD controller where the PlayStation 5 has a custom SSD controller. PS5 also has a dedicated DMA controller with two I/O co-processors, has a 512 MB DDR4 cache of RAM for the SSD, and just has a bunch of other customized parts and feature that from what we know of the Xbox Series X doesn’t as they went for more of a standardized set of features compared to the PS5. The benefit of these custom parts is that they’re able to look at industry standardized parts and take everything that works well in those components and on top of that add their own customizations to it and enhance the original part to work even better for their specific needs. These features don’t drastically change performance individually, but collectively they contribute to the PS5 once again pushing above its weight.
But there is one customized feature that could potentially allow this disparity and that’s the highly customized Geometry Engine (GE). Now, a geometry engine is nothing new and is established in plenty of GPUs now, even the Xbox Series X has a Geometry Engine as it’s part of AMD's GPUs. However, Sony has highly customized theirs to take maximum advantage of their console and it’s just one of those things where it’s flat-out better. Their Geometry Engine just works for you so you can spend time coding and optimizing elsewhere. That’s not to say you shouldn’t spend the time optimizing the GE because you’d be leaving a significant amount of performance on the table, but the benefit of the PS5's Geometry Engine is that it helps with development and if you work with it, it basically frees up memory and compute for the console which in turn provides more overall performance for games. That’s how it’s been explained to me by a couple of developers working on PS5, and that’s what I’ve been able to see for myself over the holiday.
Console API vs Xbox aka PC API
And finally, the last reason why the Series X is only achieving parity or in some cases receiving slightly lesser versions of multi-platform games is due to the Series X's new API which is fundamentally the PC version of DX12 with slight Xbox enhancements. Similar to the tools for the Xbox Series X the API has also been changed for the console as the engineers at Xbox have decided to go with a variation of DX12 with Xbox customizations here and there rather than a fully customized Xbox API and this is possibly one of the biggest problems. Once again it makes compatibility across multiple Xbox consoles ranging from the Xbox One family, the Xbox series family, and PC significantly easier to develop cross-platform games on; however, it comes at the cost of developing highly optimize gains for the Xbox series console family. This is one area where the PlayStation 5 just flat out has an advantage because it doesn’t have to deal with having tools and an API that tries to be a jack of all trades and master of none. Xbox is giving up some of its maximum optimized performance for ease of compatibility across multiple platforms. There is still a level of optimization for the tools and API that Xbox has to eventually grow, but it’s unlikely that level will be to the same extent that the PlayStation 5 will be able to achieve once PlayStation moves on to only PS5 development, unless Xbox drops support for the Xbox One family of consoles, the Xbox Series S, and do so heavy customizations for Series X at the cost of some PC compatibility.
And it’s because of these 4 things combined that the performance disparity gamers expected has been flipped on its head.
Xbox was allegedly behind with their developer kits, but more importantly their tools and it’s giving PlayStation an advantage. Xbox knew they had the more powerful RAW hardware, but the PS5 engineers created the PlayStation 5 to be optimized for speed in every task while trying to remove bottlenecks developers had with the PS4 and multi-platform/PC development, and those benefits have allowed the PS5 to stay highly competitive with technically lesser hardware. Xbox used a combination of off the shelf products with a few customizations here and there, meanwhile the PlayStation 5 pushed for a higher level of customization to provide beneficial improvements for their hardware and software needs. And finally, Xbox has an API that’s very PC-like and designed to work across Microsoft targeted platforms (5 consoles and at least 2 PC specifications) in comparison to PlayStation who has only 3 consoles specifications to work across, and in some cases only the PS5.
Now I wanted to save my feelings on these topics until the very end to keep the journalistic integrity of the information, before giving an opinion on the content. The devkits, but specifically the tools and API, are legitimate issues they’re facing in comparison to the PlayStation 5, but they have no one to blame but themselves. They could have also gone with the same concept of evolution and revolution as Mark Cerny and the PS5 engineers did by simply listening to what developers wanted, but instead, they had to do what Microsoft wanted, which reading between the lines is focusing on content and services, not hardware, so they're building a console based on the budget allocated to them to help migrate people to their subscription service which is a great service for gamers currently. Now what they ended up with was still an amazing piece of hardware, but they weren’t able to build the system they wanted specifically when it comes to customizations. The Series consoles are basically underclocked Gaming PCs running a Windows Kernel with some Xbox specific tweaks here and there. It’s the most PC-like console we’ve ever had barring Steambox. And there’s nothing wrong with that, but what it does is it doesn’t offer the benefits of low-level access that a traditional console has because the API and Tools aren’t optimized for that, their primary feature is to have compatibility across multiple devices and I expect mobile to be added to the toolset in a year or two.
Once again this is not to say the Xbox Series X is a bad console, because it’s not. It’s still an insane value for the hardware packed into the box, and still a very powerful console. I just feel this was the Xbox team basically saying this is the console we had to make, and 90% of what we wanted is there, rather than the exact console we wanted because of Microsoft. Series X is equivalent to a high-end gaming PC at just $499 which again is an amazing value ($400 range GPU performance isn’t mid-range in our book). And in the end, it has and will continue to produce some key future wins for console gaming in Raytracing, amazing backward compatibility, effortless upgradability in games through firmware (game upgrades without fully recoding), and hopefully in the future offering top tier tools and exclusives that will allow developers to fully utilize the power of the Series X.