Digital Foundry and console warriors alternate realities *spawn*

However you seem to be making arguments as to why the number would actually be lower on the assumption that a significant proportion of those users who own an expensive gaming GPU and have the worlds biggest gaming client installed on that PC are not using the PC for gaming. But to be honest, I'm not even sure that matters since the circle-jerk arguments usually revolve around how many units there are out there anyway, not how many of those units are being used to game on or not (which applies to both PC's and consoles). Also bare in mind Steam doesn't represent the entire PC market so while we can reduce the Steam number a little on the basis of some of those PC's are not used for gaming, we should also uplift it a little to account for the segment of the market that doesn't use Steam.

Honestly, gpu sales numbers are no indication anymore of who's a gamer and who isn't. There are plenty of people gaming on laptops now since that's what they own, and there's also plenty of people that simply do not want to spend money on new hardware and will game on whatever they have. Simply drop game settings to low and off they go, not everyone cares about getting all the visuals maxed out at 60fps, some people are perfectly content just playing the game in whatever settings work. I don't think making assumptions on who is a gamer based on gpu class works anymore. Especially with Intel gpus becoming faster and such a large part of the market now, most companies will test against those and have a settings profile that will work on them for most games. And also as you said, as much as Steam is often treated as the default pc gaming mechanism, it's really just one of many so it's data is not pure gospel anymore as there are plenty who pc game without it.

TLDR: Gpu is no longer an indicator of someones gamer status.
 
This current discussion (DSoup vs PJBLiverpool, 2015) wasn't about who's a gamer. It's about how many PCs > PS4 in specs are being used for gaming.
 
Colleague of mine sold his mining machine with four GPUs to a guy who wanted the best system his money could buy to play Battlefield 4 with ... Don't think many people still mine considering electricity costs more than mining yields ,..
 
However you seem to be making arguments as to why the number would actually be lower on the assumption that a significant proportion of those users who own an expensive gaming GPU and have the worlds biggest gaming client installed on that PC are not using the PC for gaming.
Not quite, let me clarify.

Firstly I think it depends on the question. If you're asking which platform has more gamers now, PC or PS4, then I'd say PC by sheer numbers but also because PC have a far larger indie scene. The question is, are you counting any type of game or are people more interest in choice of PC and PS4 (or any console) for the type of games common to both, like AA and AAA and at console equivalent quality - which is itself subjective.

If you're looking at the former, it's somewhat easier in terms of numbers because there are the Steam install base numbers but in terms of the choice of PC/console, it's more complicated because many games aren't available for consoles. Plus you can't reliably measure intent of the user, i.e. was gaming a significant consideration to the purchase or was gaming something do because they can. I didn't buy my MacBook Air 11" to game, but was surprised that I could run StarCraft 2. But I do play some Minecraft on it and more typical indie titles like Pixel Piracy, Project Zomboid, FTL, The Long Dark, Sunless Sea, Winward and so on. This is also an apples or oranges comparison.

If the latter then you have the complex technical assessment as what combination of PC hardware constitutes console quality and, for those PCs with modern graphics, you still don't know if those PCs are actually being used for games that are selling on consoles. Many of the low-resource indie titles I play on my MacBook Air I also play on my MacBook Pro (Quad i7, 16Gb RAM, 2Gb GF650M) and iMac (Quad i7, 16Gb RAM, 4Gb GF780M). And while I consciously bought machines with GPUs that "didn't suck" so I could game if I wanted too, a bigger part of the decision for those GPUs was OpenCL performance.

I'd say there's at least as much likelyhood that someone would buy an XBO for it's TV/communication capabilities without caring about games as someone buying a high end gaming (as opposed to pro) GPU, then installing Steam on that PC and not using that PC for games.

I agree. As above, it depends on whether the question is focussed on the common games or any games.

That sounds like an assumption to me ;)

It does but it's based on the fact that in most markets, the low-end of the market (i.e. the number of people who buy the cheapest solution to a given problem) outnumbers the high-end. The PS3 was, for some time, both the best and cheapest Blu-ray player on the market and sales the PS3 dovetailed with increasing sales of Blu-ray discs, ergo there is considerable evidence that a number of people bought the console for it's movie playback capabilities. Whether that was their primary reason, or just a significant reason, is another question!
 
Not quite, let me clarify.

Firstly I think it depends on the question. If you're asking which platform has more gamers now, PC or PS4, then I'd say PC by sheer numbers but also because PC have a far larger indie scene. The question is, are you counting any type of game or are people more interest in choice of PC and PS4 (or any console) for the type of games common to both, like AA and AAA and at console equivalent quality - which is itself subjective.

I think the answer to the first question is beyond dispute, 125 million active Steam accounts see to that. It's the second question I'm concerned with answering.

Plus you can't reliably measure intent of the user, i.e. was gaming a significant consideration to the purchase or was gaming something do because they can.

Absolutely right which is why I think it's kinda pointless to try. You can either take the number of physical PC's out there that are capable of playing those cross platform console games as well as the consoles (which is an easy enough number to derive) or you can take that number in combination with some general assumptions such as "Mac/Linux owners are less likely to be gamers and thus should be excluded" and "If someone owns a PC with both a reasonably high end GPU and Steam installed they are likely to be using it for gaming".

Since you have to make similar assumptions about the console market as well (i.e. the consoles are being used to actively game on and not either collecting dust or being used to view other media) you may as well just treat both platforms evenly and look at the pure numbers on the market.

Obviously the best way to measure would be by games sales but that's very difficult to come by.

If the latter then you have the complex technical assessment as what combination of PC hardware constitutes console quality

I don't think it's complex at all tbh. It's simply a matter of adding up the percentage ownership of all GPU's on Steam that are above console performance - a pretty simply task that comes in at around 14.5%. In fact that percentage alone is extremely interesting IMO since those circle-jerk discussions you reference regularly cite completely made up versions of exactly this number, generally hovering around the 1% and less mark. So now we know it's closer to the 14.5% mark. That's a big step forward in my view.

You can make the argument that we don't know all of those GPU's are paired with CPU's and system memory sufficient to match the consoles but given that the percentage of PC's out there with 4 cores or more is much, much higher (around 46%) and with 8GB of RAM is similar (around 44%) it's safe to assume that the vast majority of those 14.5% console+ level GPU's are going to be supported by sufficient CPU power and system memory to match or exceed the consoles - especially in light of the fact that 4 cores and 8GB of system memory aren't necessarily required to match or exceed console performance.

and, for those PCs with modern graphics, you still don't know if those PCs are actually being used for games that are selling on consoles. Many of the low-resource indie titles I play on my MacBook Air I also play on my MacBook Pro (Quad i7, 16Gb RAM, 2Gb GF650M) and iMac (Quad i7, 16Gb RAM, 4Gb GF780M). And while I consciously bought machines with GPUs that "didn't suck" so I could game if I wanted too, a bigger part of the decision for those GPUs was OpenCL performance.

None of your 3 Macs would have even made it into my count anyway. Two of them (the Macbook Pro and iMac) feature too weak GPU's and the Macbook Air GPU doesn't have it's own listing on Steam falling instead into the 17% 'other' category which I discounted. So in fact you highlight here that my 14.5% number (18% of all DX11 GPU's) could in fact be much larger if we had a proper breakdown of the 'other' category and those qualifying GPU's were included in the count.
 
Theres a world of difference between gaming PC's and general use/low end or business desktops. Is there any evidence the PC gaming market is shrinking?
According to JPR the PC gaming market in 2016 has grown and already hit the 2018 targets.

PC gaming hardware market now worth over $30 billion for first time
We have heard time and time again that the PC market is somehow ‘in trouble’, with declining sales and fewer customers to sell to. While that might be true for OEM systems from the likes of Dell and HP, it certainly isn’t the case for PC gaming hardware as the market has boomed in 2016, generating over $30 billion for the first time in history.

According to researchers working at analyst firm Jon Peddie Research, PC gaming hardware sales shot past the $30 billion mark in 2016 for the first time ever. Previously, the PC gaming hardware market wasn’t expected to hit this point until 2018, which shows how the market is growing faster than anticipated.

As the report notes, consumers “continue to embrace the PC platform for video games due to multiple factors”. The report goes on to praise higher graphical fidelity and greater control options as key reasons gamers choose to switch to desktop PCs: “The desktop is popular because the display distance offers increased detail when using HD and UHD monitors. Additionally, there is superior control with mouse and keyboard interfaces".
xPCGaming_PC_image_580.png.pagespeed.ic.YIPKzwcEAu.png

http://www.kitguru.net/desktop-pc/m...ket-now-worth-over-30-billion-for-first-time/
 
Last edited:
The flaw with the analysis is that JPR count a purchase of certain graphics cards as contributing to the gaming market. As somebody who manages a server farm I can tell you we spent more than £70m on 'gaming' cards for our site last year. For certain types of workload compute is a better solution. Plenty of farms are building compute clusters from 'gaming' cards.

I did buy a 1080 for home though!
 
Not sure if this was posted but it puts things into better perspective
Newzoo_2016_Global_Games_Market_PerSegment_Screen-1.png


Mobile gaming+tablet games far surpass PC or Console but we do know that what are being played on this devices is much different than whats being played on PCs and Consoles.
In addition MMOs like Warcraft and the likes should take a significant portion of PC revenue which again is not what defines the Console market. In addition PC games are more diverse. Revenue would include a low cost tiny casual games too.
Also notice that 5.2 Bn of PC revenue comes from casual webgames. Similar type of games also exist on Steam I guess.
It is impossible to distinguish the overlapping between PCs and Consoles or even distinguish how much revenue those special Console exclusives bring into the scene.

Last time I checked big budget games sold on both console and PC by big companies like EA, Capcom etc were doing better on consoles. But I dont have time to check if this is still true now
 
The flaw with the analysis is that JPR count a purchase of certain graphics cards as contributing to the gaming market. As somebody who manages a server farm I can tell you we spent more than £70m on 'gaming' cards for our site last year. For certain types of workload compute is a better solution. Plenty of farms are building compute clusters from 'gaming' cards.

I did buy a 1080 for home though!
The greater than good health of the PC is unquestionable. We must add to that that the data that the sum of those profits more than double that of all the consoles together.

And then there is the games. Using Metacritic: 179 games are considered very good (75 or more). And there are 46 exclusive games -no console comes close to that, this is being the worst consoles generation I know of-

2016122112031_2.jpg
 
Yeah the wider desktop market is cratering and while we may be a few years from that impacting PC gaming directly it will happen. PC gaming is a low volume, high margin fraction of the wider PC market. For years we've been reaping the benefits of Intel and AMD fighting over the general office PC market which paid for most of the features we wanted. The wider PC market wants are starting to diverge from what gamers need though such as highly integrated m/b with long battery life and 'good enough' performance, none of that aligns with what gamers want.


As gamers become an ever greater proportion of a shrinking market the entry level PC gamer can expect costs to start climbing in the worst way. If nothing else Intel is mad keen to make the whole market BGA and that really will kill the enthusiast PC (or make it significantly more expensive).
 
According to JPR the PC gaming market in 2016 has grown and already hit the 2018 targets.
Saw this on Arstechnica, I think we can safely conclude JPR are just pulling numbers out of their asses (*), we know from NPD etc data theres not massive changes each year for console hardware revenue, and PC sales data is only declining by ~10% each year, not suddenly doing worse than half, but heres JPR data for the previous 2 years

JPR data 2014: PC hardware revenue is 100% larger than console hardware revenue
JPR data 2015: PC hardware revenue is 15% smaller than console hardware revenue

(*) alternative facts
 
Saw this on Arstechnica, I think we can safely conclude JPR are just pulling numbers out of their asses (*), we know from NPD etc data theres not massive changes each year for console hardware revenue, and PC sales data is only declining by ~10% each year, not suddenly doing worse than half, but heres JPR data for the previous 2 years

JPR data 2014: PC hardware revenue is 100% larger than console hardware revenue
JPR data 2015: PC hardware revenue is 15% smaller than console hardware revenue

(*) alternative facts

When disputing arguments with alternative arguments it's generally good practice to post links to your version.
 
we know from NPD etc data theres not massive changes each year for console hardware revenue, and PC sales data is only declining by ~10% each year, not suddenly doing worse than half, but heres JPR data for the previous 2 years
The market for PC gaming hardware is rapidly expanding, more people are willing to spend top dollars to build powerful machines to satisfy their needs.

More from the report courtesy of ExtremeTech:

Comparing 2015’s market splits against 2016’s, we see the following changes: Low-end spending increased 5.7% (from $6.31B to $6.68B), mid-range gaming increased by 39% (from 7.64B to $10.61B) and high-end spending jumped 22% (from $10.7B to $13.05B). Most of the high-end gain was picked up by Nvidia, but AMD’s targeted strategy of refreshing its midrange products first should have given the company its own pick-up.

We agree with JPR’s hypothesis that the PC market is basically splitting between basic office and minimal Internet machines for most customers, and desktops for advanced power users. Since workstation and power users are the profit centers of the desktop space, you can bet plenty of companies will want to bring upmarket capabilities to these products — even if they don’t normally care much about gaming.
https://www.extremetech.com/gaming/243272-pc-gaming-surges-rest-pc-market-declines

I can tell you we spent more than £70m on 'gaming' cards for our site last year
Could I ask what types of GPUs were those? I am assuming they are a mix of med and high end GPUs.
 
Back
Top