Stop blaming long console cycle for lack of "next-gen" PC games

"Consoles are crappy. Current-gen consoles have crappy hardware and are holding back PC game development"

How many times have we read variations on this this old tune?

Over the years and across countless forums and blogs, I have heard again and again, that console gaming has held back PC gaming. Well, I'm mad as hell, and I'm going to spend a few thousand words refuting the claim. It's my life to waste.

Platform of Innovation

In the heyday of PC gaming (1990s), all you needed to play the latest PC game was a relatively recent PC which was already shipped with a decent VGA 2D graphics card. The delta between last years PC and this years PC was 2x or less. PC sales were increasing exponentially so games makers could focus on the most recent PCs and a good percentage of the PCs could run even the most modern games as PC sales just kept escalating, so the older PCs that got left behind were not a significant share of the market.







In those days, new PCs were significantly better than old PCs to the point that old PCs simply stopped being used. This had the effect of ensuring that on average, most PCs were relatively new, and the performance delta between fastest and slowest PCs was perhaps 4x at the top 70% of the market. You could easily build games to scale games to perform well on a 486 50Mhz and a Pentium 90, typically via scaling the gameplay window or adjusting certain quality controls in the software renderer.

Enter 3D



As the PC market stabilized, games started to require 3D cards, and some PCs had 3D cards and some did not. That was the first major feature fragmentation of the market since the introduction of VGA graphics. Now there were PCs with and PCs without 3D graphics.

For many many years, PCs, especially notebook PCs shipped without any kind of 3D acceleration whilst FPS games increasingly required the latest and greatest 3D cards. A greater and greater percentage of PCs in current use were not modern game capable, shrinking the market.

PC games makers failed to notice that the money was not at the cutting edge, where it had once been, but at the centre of the hardware configuration normal distribution (they should be targeting the average spec PC). New games now catered to the hardcore crowd that had to upgrade their own graphics cards, but still, PC was receiving the most cutting edge games.

Console Replaces PC

It was around the point of the original XBox that the average spec gaming console exceeded the average spec PC in current use (as notebooks were on the rise, and they typically had no form of 3D acceleration at all until around 2003 - I know from bitter shopping experience).



By now, the market for cutting edge PC games was deeply in decline. Most PCs were just not game capable, meaning that they shipped without 3D hardware that was fit for purpose. The delta between the average business PC and the average gaming PC in terms of floating point (required for 3d gaming) performance was several orders of magnitude (~100x). It became impossible to target the average, so now games were targeting a smaller percentage of the PC market all the time, both in percentage, but also in absolute terms, as many enthusiasts wanted out of the PC hardware rat-race and moved into consoles. This was not the fault of the consoles, but a natural end to the era when most PCs were new PCs and new PCs had similar gaming performance to most other new PCs.

This, alongside the increased level of piracy brought about through torrents and newsgroups, which were disproportionately used and understood by the very enthusiasts that were the target market of the struggling PC gaming market ensured the slow decline continued. Taking into consideration the arduous task of testing PC games and its almost infinite configurations of hardware and software, the PC became a difficult platform on which to make a profit.

Still Alive

But the PC was still a massive platform, and the base of the platform just kept getting better and better. That is, total amount of available memory, total amount of memory available to a competent 2D graphics card, and very very slow adoption of competent 3D graphics cards.

Casual/Indie games flourished in such a market. They didn't require cutting edge graphics hardware, fast cpus, and were compatible across the vast majority of in-use PCs. These games mostly made use of 2D graphics acceleration because then everyone was a potential customer. 2D gaming never really died and the PC gave it a place to flourish after failing to make an impact in the console space.







Upon this backdrop then, of a hardcore gaming legacy in ruin, and the phoenix of casual/online gaming rising from the ashes, the 360 arrives.

The 360 was massively more powerful in terms of GPU compute power than even the average PC GPU of today. It has a powerful dedicated GPU, 6 hardware threads, a decent amount of gaming memory and drivers tuned to make the most of the hardware. Most PCs sold today are notebooks and less than $500. Try to find a notebook PC with a graphics card faster than the GPU inside the 360. Most PCs sold today do not even match up to the 360, and once you factor in older PCs, you appreciate the hardware fragmentation situation that killed PC gaming.

The vast majority of cheap PCs have low-end discrete GPUs or integrated GPUs (such as Intel HD Graphics 3000/4000). Netbooks have even less power. So there are huge quantities of PCs with graphics that are not even up to the level of a console introduced in 2005, but credit to Intel, the HD3000 was the first integrated graphics that actually allowed for playable frame-rates (at low resolutions). Consoles are not the lowest common denominator my friends, PCs are.



The joy of the coming of the 360/PS3, was that now games could be targeted to extremely capable game hardware, and those games could actually stand a chance of making a profit. Games were targeting hardware WAY beyond average PC hardware, and console games pushed forward gaming technology in a way that the damaged PC gaming market (fragmentation+piracy) could not possibly have sustained.

Now there was an opportunity for games developers/publishers, not to develop for the tiny high-end PC market, but to invest a little bit extra money and port the console games to the PC so that they could generate extra sales, likely not anywhere near enough to have supported the entire development, but more than the cost of the port.

Enter Steam

PC gaming was also facing a distribution problem. The market was very small, so PC games shelf space disappeared, almost overnight.



If not for Steam, then I'm sure that some other distribution distribution platform would have appeared, but Steam was bolstered by being the distribution platform of choice for Half Life 2 and Counter-Strike. Through early availability and canny business savvy it ultimately evolved into the de-facto digital marketplace for tent-pole PC game releases as well as smaller indie and casual titles. Now games makers could eliminate the cost of manufacturing and distribution and the risk of unsold stock from PC game publishing. Now the cost of producing PC games was reduced even further making the value proposition of porting console games even better. PC gaming was back.

The "Consoles have held back PC gaming" argument



Its about here that the hardcore PC gaming crowd goes crazy. "Lazy ports". "Consolified controls". "Last-gen graphics".

"Lazy Ports"

The first complaint is perhaps the most valid. A lot of ports have been relatively lazy. It might be that the UI has not been updated for the PC and not changed some of the messages. Any reference to "console" within a game seems to annoy some PC gamers that somehow feel that they are better than owning a console and that console is a dirty word. They don't realize that collectively, PC owners have pretty awful gaming machines compared to the gaming hardware Zeitgeist.

"Consolified controls"



This is perhaps a little true too, although I would dispute that this a bad thing in all cases. Console games are designed to use game pads so therefore there can be some design choices that are optimized for that form of input. Nearly all console ports support mouse/keyboard though, and without these console ports, there might be nothing but casual/indie games on the platform.

"Last-gen graphics"

Last-gen graphics seems to be a common criticism on why console game ports are holding back PC gaming. On the surface, there is some merit in the argument. Most games target consoles first, consoles are weaker and have less hardware features than up-to-date PC gaming hardware. But the very notion that if games were not targeting consoles, that they would somehow be targeting PCs is absolute nonsense.

As stated earlier, it is not economically viable to target the few PC gamers that have a cutting edge next-gen graphics card. It costs tens of millions of dollars to produce a AAA game, right now, targeting 360/PS3 level graphics is the only way to make money. PCs with much better specs than the 360 benefit from higher resolution graphics, better textures (in some cases) and anti-aliasing, all making the PC versions of most console ports, the best version available. But PC gamers should not forget that the hardcore PC gaming machine install base is tiny compared to the collective install base of PS3s and 360s which ALL have the capability of playing modern games - well.

Consoles as Saviour



From one perspective, both the Xbox 360 and PS3 can be viewed as the saviours of traditional gaming on the PC. Without the large install base of consoles, then the install base of modern-game-capable PCs could not have supported the development of such games as Oblivion, Rage, Skyrim, Bioshock, Crysis 2+3, Arkham Asylum/City, Darksiders, Saints Row, Portal, Portal 2, Far Cry 2/3, and on and on and on.

Of course, there were a few AAA games that managed to release on PC first, Crysis, Far Cry, Half Life 2 being good examples. RTS and MMORPG games of course are also exceptions to the rule, although certainly in the case of WoW, the low hardware requirements and low poly design certainly contributed to its success, which is exactly the counter-argument of the assertion that console ports are holding back PC gaming. The most successful game in the history of gaming held itself back, to be able to reach a larger market.

Half Life 2 was released with relatively low hardware requirements and certainly could not be said to have advanced the state of the art at the time, but it was an outstanding game, and proof that PC games do not need to advance the state of the art to be exceptional games. Crysis apparently made a profit, but it was a huge risk for its developer Crytek, and there was huge piracy of the title. Crytek were unwilling to repeat the experience and hedged with simultaneous PC/Console releases for the sequels.

Dedicated gaming PCs outclass the current consoles by an order of magnitude. The PC ecosystem is modular so better graphics at the same cost arrive every year, if you care about such things. If you are a enthusiast, you already have a rig that is at least as fast as the successor to the 360, but obviously, without software, much of the potential of the hardware is untapped.

"This Console Generation is too long"



Since around 2009, there have been calls for there to be new hardware soon. "We need next generation games. PCs are so much more powerful than the current consoles." "The current consoles GPUs were old in 2005, now they are archaic (back in 2009)." "Now is the time for new hardware, etc, etc, etc."

Well, the current console generation is exactly the right length, its all about the heat.

You can't just introduce a new console and expect it to be 10x more powerful because 4 years have passed. Consoles of a certain size can only dissipate a certain amount of heat, and in general, that determines the clock speed and performance of the console.

As chip fab technology gets better, transistors shrink in size (roughly half in size every 18-24 months), and roughly speaking, you can double performance at the same level of heat dissipation, or reduce heat generated and maintain the same level of performance. That is why first-gen XBox 360s are large and had a lot of heat related issues (as they had 90nm transistors), and XBox 360 slims are smaller (and have 45nm transistors - roughly 1/4 of the size).



The XBox 360 slim used the improved chip fabrication technology to reduce power consumption and to reduce heat dissipation meaning a smaller unit, smaller heat sink, quieter operation, but identical performance. It was not possible to have an XBox 360 as small, as quiet, and that consumed as little power as the slim model back in 2005. Technology moves forward fast, but you can't import future tech into the present.

In 2005, XBox 360 moved forward from the original XBox by packing more transistors into the box than the original XBox, using smaller more efficient transistors, clocking it higher, and as a result, it was more than 10X faster than the original XBox but the transistors were not 10x as heat efficient. The remaining performance was essentially bought at the price of higher power consumption and higher heat dissipation, effectively an escalation of the power envelope that it was deemed for a console to run within. By allowing the 360 to consume more power and generate more heat, it would run faster than if it was designed to emit the same amount of heat as the XBox (1). By it running 10x faster as opposed to say 5X faster, it would look next-gen as opposed to an incremental upgrade. And it worked.

Now the problem with the "Next-gen-now" crowd is that they didn't seem to understand that next-gen-level-performance PC graphics cards up to 40nm consume huge amounts of power. It just was not feasible to get a decent performance boost from 40nm unless the next XBox was going to consume 400 watts of power and have an expensive heat-sink and fan to boot. Realistically, a console the size of an living room console has to consume no more than 250 watts of power for the whole system, not just the GPU. If Microsoft launched in 2010 or 2011, only 40nm would have been available to them, in 2012, the more efficient 28/32nm fab processes were extremely low yield and therefore expensive. 2013 was and is the first year where its feasible to launch on the lower sized and more efficient transistors. For PCs, a system can consume 400 watts and its fine, they are noisy and expensive. A console MUST consume no more than 250 watts if they want to be welcome in the living room.



All other things being equal, a next-next-gen Xbox based upon 28nm transistors will be able to fit in 8x as many transistors at the same power envelope as the original XBox 360 (larger sized). If Microsoft finds a way to improve the fans/heat sinks, they could fit in as many as 10x the number of transistors whilst fitting within a 250 watts power envelope. 10x the number of transistors does not necessarily mean 10x the performance but it is in that ballpark, alongside improvements in other system components and architecture.

Had Microsoft tried to launch in 2012, there would have been massive supply issues, and the price might have had to have been double. Had Microsoft tried to launch in 2010/2011, then 28nm would not have been available and the perforamance differential would have been approximately half the 2013 performance leading to accusations of "not-next-gen-enough".

Had Microsoft not already raised their power consumption on the 360 (versus the original xbox), then the successor to the Xbox 360 could have improved performance by raising the threshold versus the original, unfortunately, because they already raised the threshold back in 2005, the only way to improve of the successor to the 360 without increasing the power consumed was to rely on transistor fabrication process enhancement and architecture enhancements. That alone determined that late 2013 had to be the release date of the next-next-gen XBox.

2013 and Beyond



So, it is 2013, and with the 28nm fab process reaching workable yields, the new console generation will most assuredly arrive. This marks a first increment in the minimum target specs that games developers will be using in 8 years.

Finally owners of up-to-date gaming PC hardware will see some true next-gen games, and it will happen very quickly. By December 2014 I'd be surprised not to see at least half a dozen titles that have tricked down to the much smaller PC gaming market (likely via digital distribution). Middleware supporting these new features will appear quickly.

Finally hardcore PC gamers will have some cutting edge games using their fancy DX11 tessellation API features, and using hardware compute APIs for AI/Pathfinding, etc. The next gen of gaming is here, and the PC will benefit from it just as much as the consoles that facilitated a business model where its possible to target these features and have a reasonable amount of sales.

So - PC gamers, it is my assertion that consoles have not been holding you back, they have been dragging you forward. Without consoles, then Epic may have already went out of business. Without consoles, there would be far less gaming choices on the PC than there is today,and as such, the state of the art would be less-than. Without consoles, there would be far less innovation in digital distribution on the PC platform, necessity is the mother of invention. Consoles have dragged the market forward, as they provide the certainty that allowed developers to take that quantum leap forward, safe in the knowledge that there is a market for the games.

Having the best hardware means absolutely nothing without a market large enough to support games that support its features.

I thank the current generation of consoles and look forward to the next generation of consoles, and the next generation of PC gaming.