If you talked to me seven years ago, the concept of phone games beyond Snake and basic flash games would've been beyond me. That wasn’t really on my mind, especially because playing games like Battlefield 2 effectively required a desktop. I did try it on a Dell XPS M170, with an Nvidia 7800 GTX graphics card, 1 GB of RAM, and a Pentium M processor, but even then, it was no guarantee that it would load the game consistently. Crashes to desktop were common because of the RAM requirements, and hitting the page file would cause frustrating freezes. Of course, things have completely changed since then. Mobile became one of the fastest growing sectors in the tech industry, 40-60mm thick, 14 pound laptops were no longer necessary to play  games at acceptable IQ/FPS levels, and the slate-style smartphone has gone from distant curiosity to everyday necessity.

So the port of Half-Life 2 and Portal to the Shield was interesting, because I’ve seen the sheer breadth of experiences that Source can provide, from Resistance and Liberation to Nightmare House 2. Nvidia has stated that Half Life 2 on Android is a port of the Linux version with OpenGL support to OpenGL ES, and based upon a casual playthrough of both Half Life 2 and Portal, it’s not immediately obvious that there are any issues with the engine port itself. In fact, it runs quite well. When immersed in playing the game, the experience is incredible, especially compared to the experience that one usually gets on handheld consoles like the PS Vita and 3DS. While I was concerned that the control scheme would be difficult to adapt to as a PC gamer, it turns out that with auto-aim and some other compensation mechanisms that gameplay is perfectly workable. The one issue that I did have was targeting things like headcrabs in mid-air, as pixel-precise pointing still isn’t quite there for joystick control systems. Battery life was also great. I managed to go around four hours of nonstop gaming at a mid-level of brightness, although this is an extremely rough rule and can vary greatly. I hope to get more precise data on this soon.

Of course, the bigger question is how well it actually performs. While timedemo functionality can’t be accessed, video settings are locked out, and there’s no console access like in the PC versions, it’s still possible to gauge the approximate experience. Based upon a quick playthrough of Ravenholm, I managed to take a closer look at the FPS instead of relying purely upon subjective judgment. It turns out that there’s a pretty even distribution of FPS across the board from 20 FPS to 60 FPS. The median FPS was 43, and overall I’d say that it’s quite accurate, as in less intensive scenes it will stay capped at 60 but anything with effects such as fire and explosions will often cause the gameplay to stutter noticeably, as in the scene above. Portal was similar, but due to the relatively simple level design FPS remained higher in general. Overall, it appears that the CPU usually is only at around 50% utilization throughout the game, which is indicative that either the GPU is the limiting factor in performance of this game or that the engine is not well-threaded for the port. There was the possibility of thermal throttling, but unlikely because exhaust air temperatures weren’t hot enough for concern. Unfortunately, no data on battery voltage nor temperature is given by the Shield. Ultimately, while this is helpful information for someone that wants to buy the game, the real question is how well IQ and performance compare to PCs, and will require further investigation to get a rough idea of how well modern ARM SoCs compare with CPUs and GPUs from the 65-90nm era.

POST A COMMENT

40 Comments

View All Comments

  • Morawka - Thursday, May 22, 2014 - link

    and just imagine what Tegra K1 will be able to port. Nvidia's CEO recently admitted Shield 2 is in the works. One can only dream what games we'll see ported. It will definitely produce stiff competition in the hand held market. What other handheld can play N64, SNES, PSP, Gameboy (all of them), Android Games, and stream PC while bringing console ports.

    Its only a matter of time before we have a Assassins Creed on this puppy.
    Reply
  • Impulses - Friday, May 23, 2014 - link

    Well, you don't really need a Shield to pay N64, SNES, PSP, Gameboy, or Android games... Even an older phone can handle those emulators with ease and Bluetooth gamepads are now much more commonplace.

    The differentiator for Shields is clearly PC streaming and NV's support... I just don't see the appeal of playing games made for larger screens and or mouse/kb on such a device, but that's just me.
    Reply
  • TheJian - Friday, May 23, 2014 - link

    But they won't play halflife2, portal, trine2, SSam2, nor any THD game and android gaming keeps getting better and better with companies like rockstar, gameloft etc aiming there now more than anything but PC and pc only tops it ~1.5%. On the other hand console dev is way behind at under 20% making something for any of them according to GDC 2014 survey of devs (PC/Mobile at 50%). If you're looking at devs on vita/3ds GDC survey pretty much shows they should just pack it in vs. mobile. The power on K1 etc just keeps going up vs these handhelds.

    OpenGL as others have noted will kill all of these as games will be made elsewhere first then ported to consoles. So you'll have no need to buy it later when they get around to porting to DirectX (xbox1/xbox360) or Sony's OpenGL tweaked api. Devs show nobody but the biggest dogs are aiming at console first today. Smaller firms simply can't wait around for 4-7yrs of console hardware sales to make money. Devs are not stupid and the audience on mobile is 1.2B today and growing to 2B+ in 2yrs. You have to aim here first or PC's for major AAA games today. Either way you have an easy port especially with kepler and a new desktop version yearly to throw at them. Just tone it down some with a few tweaks and boom, it's on mobile (or the other way around). You can become a millionaire overnight on mobile, where you have to have a HUGE AAA hit on PC's to do this in most cases.

    We are way past Angry birds these days. His stutters in shield 1 will be gone next month with r2/K1. We'll be talking M1 shield R3 next june. They will rev it yearly as it only costs them 10mil to dev it the first time and just a drop in of a new chip each rev. They are all made to fit inside the same envelope so no major design changes to change chips. Maybe a few mil to adopt new memory or screen. If they can do the entire design for 10mil (grid only cost 10mil also) then it's less to swap like chips into the same shell/fans etc.

    Rev yearly and port the most popular games as fast as possible while at some point start making their own games with new IP. I wish AMD/NV would both get into MAKING games as they know the hardware/drivers best. NV could spend 100mil on gaming and put out 25 AAA games (4mil each?, tons of great games on PC are made under 5mil, check kickstarter's most funded) that showcased their hardware and sells more cards. If you don't get your money back port to PC for a quick buck (amped up a bit on textures, object on screen etc) to recover or make money on them. If they can port in 2 weeks or less (as shown by ssam2, trine2, portal, halflife2) pump these out like crazy! They only have to break even to push the platform, so as long as you're not losing money pump out the games to draw the user base.

    I wouldn't even ponder a handheld today vs. shield and r2 is a month away which will absolutely destroy the perf of vita/3ds. R1 already does this + output to tv over hdmi, stream pc, play any movie codec android apps can etc etc. Tomshardware showed it streams fine from PC even on different ISP's in different countries. Handhelds are dead and sales numbers prove it. Consoles are already tanking too. MS Q1 off 50% just like I said previously ages ago. Wiiu had the same 50% drop off after 1Q. It is likely worse for MS since they don't tell SELL-THROUGH numbers. MS is at 400K/mo now (1.2m Q1 sales) and getting worse. Sony will be lucky to crack 11mil probably and MS won't hit 10 this year. PS4 is doing better but not a lot. It will get worse since no devs are paying attention to them first like last gen now. We'll have 20nm M1 etc next year in these android consoles. Scale 720p to 1080p using HDMI out to TV and consoles have problems (they already scale many games up) since you'll be close enough for many to not care.

    Porting in OpenGL is so fast they will have a pretty large library in another year. Shield already has 100+ games and can stream 300+ PC games in or out of the house and that's not counting the googleplay store's games which play on everything. How do you compete against $100-200 cheaper (assuming shield1 stays $200 while r2 goes for $300 again at debut), larger library, all android apps, pc streaming etc and they rev yearly. Casual users will not be buying standard consoles in years 3-7 like they normally do. You will be looking at a P1/V1 14nm SOC (pascal & volta names I guess?) just 3yrs into xbox1/ps4's lives and they may come with a huge cache then like Iris. I personally can't wait for consoles to die so that the time wasted on porting to these systems is spent on making the GAME PLAY fun.

    You launch a game today and you port to xbox360, ps3, ps4, xbox1, mac, pc, maybe linux & android. Good grief. We need android, linux, pc, mac. That's enough and all opengl/es straight up. Desktop gpus in socs along with all that comes with that ends this whole story it's just a matter of how quickly. Unfortunately for consoles, the perfect storm for android was already underway before next gen consoles even hit (and it killed vita/wiiu sales obviously - or where did all those buyers go?). Next stop: PS4/Xbo1, it's just a question of which gen on mobile makes them pointless.

    Qcom/Samsung/Apple really haven't even joined the race yet either. If they all join to port to android/opengl/ios, consoles are screwed in a year or two instead of 3+. Those 3 have the money to match MS's 1Billion for xbox1 gaming. I'm shocked Apple hasn't made some massive announcement with 140B in the bank. I would have said "$5B for mobile gaming on apple with $500million each year for 10yrs - Buy our crapple stuff!" and that would have been ages ago. At worst I'd announce it with Iphone6 debut to slow android's march and help kill dx/wintel etc (which helps macs too right?). There are enough crapple (heh) devices to support this kind of investment and I doubt they lose any money overall while gaining gamers and probably end up the #1 mobile platform just because they can bleed you to death with 10-20B in gaming and still laugh it off.

    That said, I'll probably cry if apple takes my advice :) That would suck 10yrs from now. But business is ruthless and that is exactly what I'd do to you and worse with $140B+40B yearly in profits. My guess is Jobs would have done this by now, while Cook just seems to not want to screw anything up. I'd put out a 150w box for xmas and undercut MS/Sony by $100-150 and laugh as they stopped selling their boxes. If they lowered I'd drop the same until they died or realized the battle was pointless and just gave up. Buy NV and drop support for PC's when you gain more share on macs. We are only on PC's for apps and games. Macs have the major apps in a lot of cases, so get the games and we have a whole new ballgame all around. Again I hope Apple isn't listening...ROFL.
    Reply
  • djgandy - Friday, May 23, 2014 - link

    "But they won't play halflife2, portal, trine2, SSam2"

    No because no one plays those games any more. What is this obsession with straight porting? It's quite the waste of time, because people have done those games.

    As for the rest. TLDR. Half of what you wrote makes little sense, demonstrates a lack of knowledge and is formed with very poor grammar.
    Reply
  • kron123456789 - Saturday, May 24, 2014 - link

    But, it's not SSam2, it's SSam3, and SS HD: TFE/TSE(in future). Reply
  • Impulses - Friday, May 23, 2014 - link

    Not gonna argue with the while rhetoric, but Shield isn't really competing with handhelds... It's competing with the phones and tablets people already have or want. I don't see it ever being the kinda massive success you're dreaming up, not that I care much for consoles either. Reply
  • Flunk - Friday, May 23, 2014 - link

    I'm confused as to why anyone is drawing a line between shield and other Android devices at all. Shield isn't a competitor to Android phones, it is an Android phone without the phone. Reply
  • anandreader106 - Friday, May 23, 2014 - link

    It competes with whatever else I want to play mobile games on, including phones.

    I have a Moga Pro Controller and pair it with my Nexus 5. The combination essentially gives me a better version of Shield (Read: IMO) that is just as mobile with more functionality.
    Reply
  • kron123456789 - Saturday, May 24, 2014 - link

    But Nexus 5 can't run HL2 and Portal)) Reply
  • Ny_Wi - Saturday, May 24, 2014 - link

    No. Moga Pro + Phone cant have better feel then Nvidia Shield. I have Shield and MogA Pro + G2. A very noticable lack in combination between MogA Pro + G2 is the hinge of MogA pro has a limited range so when you play the game you cant have a better angle compared to shield. More over the weight balance on MogA Pro + G2 really not as good as on Shield. More over stream pc game on Shield really enjoyable. Even I play local on my home. I can go here and there inside my house to play walking dead season 2 and Child of light. Reply

Log in

Don't have an account? Sign up now