The PS4 Pro: Just who is it for anyways?

Right; so the new PS4 Pro:


PS4 Pro


Like the original PS4, and the PS4 Slim, it’s a good looking console. From an aesthetic viewpoint there’s not really that much to say. Which is fine as people rarely buy consoles for the visual appeal of the box. It’s fair to say that from the very first E3 where the veil was lifted from the original PS4 that Sony’s console division was firing on all cylinders. The PS4 was not only cheaper than it’s direct competitor; it was more powerful; had a better line-up of exclusive games; and better console services. Sony had nailed everything that mattered; so of course all eyes were on the long rumored PS4 Neo.

Which, well, hasn’t been as well received as the original Playstation 4. Rather confusingly; the console now known as PS4 Pro doesn’t really seem to have an idea of what kind of “Pro” the console is aimed at.  Sony’s own PS4 Pro Ultimate Frequently Asked Questions blog post was inundated with hundreds of Playstation Plus subscribers all leveling complaints at the new system; a far cry from the wave of praise that greeted the original system. By far the most common complaint from a grep of the comments is the lack of a specification bump to the internal Blu-Ray drive to support the Ultra High Definition Blu-Ray format. To be fair the general response from Sony is that they don’t see UHD Blu-Ray as a huge selling point in the next couple of years; rather instead expecting streaming services like Amazon’s Prime Video, Netflix, or Sony’s own Playstation Oriented video services; to attract the audience that would have likely invested in UHD Blu-Ray systems. I am not so sure that is entirely true in the US, British, and EU markets where internet service providers are barely equipped to deal with rising numbers of Amazon and Netflix users requesting videos in 1080p/10bit@60fps; much less a bumps to anything in the 1440p/8bit@24fps  ~ 2160p/10bit@60fps ranges. So, fair shot, the lack of spec bump to the Blu-Ray drive is likely a fair complaint that indicates the PS4 Pro is only half-baked.

That, however, wasn’t among the issues I broached on Sony’s own blog. Rather; I started with the claims that the PS4 Pro is targeted towards rendering in QHD/UHD class resolutions; which would be 1440p/8bit@30fps up to 2160p/10bit@60fps. I don’t think the system is actually capable of rendering OpenGL 4.x / Vulkan class rendering in those resolutions.


Is this really 4k; or is just an upscale?

From what is reported the processor is a clock bumped oct-core built on the Jaguar architecture. While Jaguar systems with 6 or 8 cores never made it to the consumer market, 4 core designs did under AMD’s Kabini processors on Socket AM1 motherboards. Back in 2014 a number of benchmarks surfaced comparing Kabini processors to AMD’s Kaveri. As expected just based on AMD’s own promotional materials and pricing the two processors weren’t even in the same classification. What was interesting back then is that in terms of benchmark scores weighted against core count and frequency the Jaguar architecture was ostensibly superior to AMD’s Richland Architecture. However, it was inferior to Kaveri architecture. Now I, personally, have a Kaveri. The A-10 7850k. I routinely ding AMD on various social networks because the Kaveri processor just is not capable of pushing OpenGL 4.x / Vulkan class rendering in QHD resolutions; even when coupled with something like the Hawaii XT. Otherwise known as the Radeon R9 290x. Which was basically just die-shrunk and rebadged as the Radeon R9 390x because it’s just that good.

In fairness the Jaguar architecture that was used in Kabini has gone through a number of changes. The last available commercial version was a System-On-A-Chip design, or SOC design, called Puma+. Benchmarks of Puma+ systems were not exactly glowing; owing large part to almost all available systems shipping with a single-channel memory controller. So there’s no real world comparison that I’m aware of to directly reference the potential efficiency of a Puma+ derived system. I do know the relative tracking of performance in terms of IPC on the desktop APU side across the Steamroller and Excavator architectures; and those have only be changes measured in terms of low single digit percentages; not the massive double digit percentages that AMD’s Zen architecture is bringing.

Processor efficiency isn’t the only reason I’m not entirely convinced of the PS4 Pro’s capability to render into higher than 1080p screenspaces. It’s also the use of the Polaris Architecture. Now, AMD generally ignores journalists who have a Linux focus, so I wasn’t exactly on the list of people who got the top-end Polaris based RX 480 to play around with; nor am I one of those with access to AMD’s upcoming performance-targeted graphics chip known as Vega.  What I do know is that the top-end  RX 480 is a wonderful graphics card in OpenGL 4.x / Vulkan class rendering in the 1080p screenspace. I also know that the RX 480 isn’t that great above 1080p largely due to it’s use of only 32 Raster Operations Pipelines, or ROPs. One of the reasons why the legendary R9 290x was such a beast is that it toddled up to the plate with a whopping 64 ROP count.

In other words; I don’t buy the “4k” resolution promises. From known information the Puma+ based processor simply is not up to that task on it’s own merits; much less the limitations of what is currently the top-end Polaris chip. If the PS4 Pro processor was Zen based and there was a Polaris chip with 48 count or 64 count ROP units; I’d easily believe OpenGL 4.x / Vulkan class rendering in QHD+ Resolutions. Right off the top of my head; even 30fps in a 1440p screenspace is an iffy proposition; which might be why Sony’s own documentation shies away from suggesting that the console renders into that screenspace. My gut reaction right now is that Sony’s opening themselves up to world of hurt and perceived broken promises if the “4k” game support turns out to just be checkerboard sampled and upscaled from a 1080p screenspace.

There is possibly a caveat here though. QHD+ class displays still tend to be relatively expensive from reputable vendors. Sony’s long term plans might include another Playstation platform update in two years or so that is better equipped to deal with rendering into a 4k screenspace.


Let’s talk about that Pro Designation and the Storage Drive

The next issue I raised in Sony’s own blog was one that I know I brought to Sony’s attention when rumors of the PS4 Neo first started circulating. So I’m more than a little surprised that it was not addressed in the announcement or the FAQ posted to the blog. While Sony has bumped the PS4 Pro to a default 1tb hard-drive there is no slot for an m.2 solid state drive pinned for PCIE, or both SATA and PCIE.  Which is rather odd since Sony is ostensibly aiming the PS4 Pro at a, well, Pro Player.

This ties into a general focus from the console on improving the streaming gameplay capabilities.  The current PS4 offers a surprisingly wide array of tools and utilities to broadcast games live; and even edit movies taken while in game. For many of these streaming oriented users a solid state drive is a sensible upgrade; although many seem to favor an external capture device piped to what’s normally a Linux PC with a NVME capable PCIE drive.

While Solid State SATA drives are far cheaper; and thus more likely to be used by a casual consumer; PCIE Solid State drives won’t always be positioned as market rip-offs. Still, a “Pro” is likely to be willing to invest in an NVME Solid-State drive; and the PS4 Pro just does not address that type of player.


Pro Console … with No DisplayPort or USB Type-C?

I kept poking at Sony’s chosen nomenclature of PS4 Pro by noting the lack of DisplayPort support as well the lack of a USB Type-C port. While the official specification mentions the inclusion of a new USB 3.1 Gen.1 port, that could potentially be misleading. Generally speaking USB 3.1 Gen.1 was a rebadge/rebrand to standard USB Type A rectangle ports with USB 3.0 support. In other words; the text indicates it’s just a bog standard USB 3.0 port. Not really that exciting.

Now; I honestly did not understand why Sony skipped a DisplayPort output on the first revision of the Playstation 4. In terms of display output support the industry standard was already offering higher resolutions, refresh rates, color channels, and everything else that matters to displays including support for passing an HDMI+HDCP signal. Not to mention a little thing called Adaptive-Sync that I’ll come back to.

So, for the PS4 Pro, here comes a new console aimed at “Pro” users with 10bit color channel displays. So I kind of expected that Sony would dump the HDMI port and use the space to add a few USB Type-C slots. This would increase the port potential of the system; simplify the motherboard design; and still provide HDMI+HDCP support. Since USB Type-C also very recently added direct HDMI+HDCP support through Alt-Mode; a long expected upgrade; it’d be a nice win all around.

… But no Type-C ports at all? No DisplayPort output?  For a “Pro” console? Really?


About that Adaptive-Sync

The other reason that I was a little stunned the launch PS4 forgot to bring along a DisplayPort connection is that AMD had been working on a fix for screen tears and stutters at basically the level of the display industry. The branded AMD FreeSync was adopted into the DisplayPort specification as Adaptive-Sync. Which was a pretty big deal back in 2014 when the first PS4 launched. While it might be fair to say that the PS4 launched before Adaptive-Sync was ratified into the DisplayPort specification; there is no reason why AMD couldn’t have implemented the support into the PS4 under the already launched FreeSync branding.

So, PS4 Pro, the console aimed at the “Pro” who likely invested in a DisplayPort capable display for tear-free and stutter-free gameplay; doesn’t have a DisplayPort connection or Adaptive-Sync support. Nor does the console support AMD’s recently announced FreeSync Over HDMI. While HDMI equipped displays are still a bit off from entering the market; it’s a good question as to why Sony launched a console version that isn’t already bragging about support for that feature.

Now; there is some evidence here to suggest that Sony is capable of thinking about forward looking market potential. The launch PS4 hardware shipped with an HDMI port likely capable of draft HDMI 2.0 output. An upcoming firmware update to all existing PS4 consoles will theoretically add output in 10bit color channels; otherwise known as High Dynamic Range. Which basically means a deeper color gamut for more accurate color reproduction. To be really picky here; Sony hasn’t actually qualified whether or not HDR on the launch PS4 hardware is the HDMI 1.4a supported x.v.Color; or the HDMI 2.0a implied BT.2020.



Just who is the PS4 Pro aimed at anyways?

As I read through the various posts on Sony’s PS4 Pro I kept asking myself a question over and over and over. Just what Pro did Sony have in mind for those console anyways?  The console lacks key hardware features a Gaming Pro would be interested in; and isn’t really equipped to live up to the chosen marketing scheme.  I found myself questioning whether or not Sony should just put a lid on the PS4 Pro for a launch in 2016; go back to the drawing board and fix the missing features list; and shoot for a launch in 2017.

Not likely since the factory production runs have likely already been ordered; and Sony’s likely sitting on a few hundred thousand consoles ready to go in warehouses right now.

I will admit that the PS4’s bumped specs do suggest the capability to be a devastating contender in the 1080p screenspace. This could legitimately be the first console to reliably drive OpenGL 4.x / Vulkan class rendering @ 60fps; the same screenspace where the existing Polaris RX 480 is an impressive contender.

For what it’s worth then; I think the PS4 Pro is likely astounding value for money. If Sony reigns in the marketing rhetoric and takes a moment to reconsider how the marketing points refer to “4k” support; the new console could probably recover from it’s rather rocky launch reception.



Leave a Reply