tempest in a teacup

Apparently it is upon us

I’ve been wondering whether something like this was feasible at all. The prospect of being able to play the best PC games without actually requiring the hardware to do so is quite a tantalizing one. On top of that, unlike the Phantom, it appears to be an actual product (it was test-driven by gaming reporters at GDC). There are a few things that might turn into problems, though.

Scalability

One of the major selling points of cloud computing is the ability to scale your applications as demand increases. This is all well and good when your app is a web application that’s designed to be scalable from the ground up, but I imagine making an app designed to run on a single machine into something that can run into a cloud must be pretty difficult. I’m wondering if they got the content providers to actually go back and make their apps scale across multiple machines (seems rather unlikely), or if they just have the apps running on virts with a graphics API implementation that just turns around and passes rendering/processing tasks off to a giant server farm (also seems unlikely).

On the other hand they could just be spinning up a virtual machine every time someone starts a session, in effect running multiple gaming sessions on a single box. This actually sounds like it might be easier to implement, although scaling horizontally would require purchasing more hardware as traffic ramps up. On top of that I don’t know how far support for dedicated graphics on virtual machines has progressed.

I’d love to know how they’re tackling this problem (seeing how I deal with a company that has to deal with massive scale all the time)…I wonder if they’ve published any technical documents about their solution.

Customizability

A major advantage PC gaming has over its console counterpart is the ability to tweak settings to improve performance. I can see a service like OnLive allowing you to fiddle around with the settings present from within the game executable itself, but what about stuff like custom scripts and binds for multiplayer FPSes, installing mods, and otherwise tweaking your settings? It doesn’t sound like any of that will be possible with a service like this.

Networking

OnLive will apparently require at least a 1.5 Mbps connection for standard def (480p) resolutions and a 5Mbps connection for hi-def (I’m assuming 720p) resolutions. Assuming they’re actually rendering the source game at 1280×720 in 32-bit colour, each frame will be about (1280 * 720 *32)/10^6 = 29.5 Megabits. To get a playable framerate you need at least 30 frames per second, which translates to about 885 Megabits of data transmitted every second. This clearly vastly exceeds the throughput of a 5Mbps connection, so I’m led to believe they must be either employing compression on the framebuffer output, rendering at a lower resolution and upscaling, or a combination of the two. Either way it sounds like at least graphics-wise, OnLive won’t be a substitute for the real thing at all. On top of that ISPs these days are incredibly finicky about bandwidth usage, and a service like this sounds like a bandwidth hog that’d be liable to get throttling slapped on your account.

And of course there’s the issue of input lag – your keystrokes and mouse inputs need to make a full round trip to the OnLive servers before they’ll be registered. I don’t have any data on typical mouse and keyboard response times, but I am a little skeptical about whether they’re significant compared to typical network round trip times.

In any case, there’s an interview with the CEO of OnLive at GDC here, so go have a watch if you’re interested:

http://www.gametrailers.com/player/47079.html

Most interestingly he mentions a solution that deals with the input lag problem I mentioned that took seven years to develop. I have no idea what this solution is – I suppose I’ll need to actually try it out to be convinced one way or the other.

1 comment

This post might be a little disjointed – I’ve been playing a bunch of games recently and thought I’d just pen down my thoughts in one single post since I don’t really want to write five separate posts in one night.

Team Fortress 2

Valve has patched the game twice since the Scout update, yet the Sandman remains unfixed (although they did manage to break rocket jumping while trying to fix an exploit that made it harder for Snipers to get headshots, and they did break Natascha’s slowdown effect yet again). I’ve played in a few big games since the update, and I’d say the stun is overpowered even if you discount the fact that it affects ubercharges. I have not yet seen a single Scout that doesn’t try to tag me with the ball the moment I see them, and in a few circumstances they’ve managed to stun me for extended periods at close range, allowing them a free kill. One particular incident that comes to mind occurred on the second point of the second stage of Dustbowl – I was playing Soldier and standing on the point when a Scout ran out from the central tunnel, ran up to the point and threw his ball at me. I was stunned for a good 3 seconds or so, and this was from being tagged at close range.

The response from the competitive community has been very clear. CEVO has banned the Sandman, and from what I’ve heard a bunch of the other leagues like ETF2L have followed suit. Keep in mind that this is the first unlockable weapon that they’ve actually banned. If I recall correctly they didn’t even ban the Pyro’s Backburner back when it granted a ridiculous 50-point health bonus.

The response from the rest of the TF2 community has been less distinct. While there are a few players who recognise that the weapon is clearly overpowered, the vast majority of the community’s response has been ‘LOL LERN2TEAMWORK.’ By this they’re implying that somehow Pyros need to have their entire team with them when they try to circle behind enemy lines to attempt an ambush, so that they can successfully fend off a class that they were previously on a reasonably even footing against. And of course, if your Medic successfully builds up an ubercharge by being a good healer and deploys it on a friendly Demoman in order to take out a Sentry farm, only to have his uber rendered useless by a flying baseball, that’s his fault for not being a team player.

Surely the absurdity is obvious.

I’m hoping that, like they did with the Backburner, Valve will realise what a terrible beast they’ve unleashed and make some sort of fix. A popular suggestion seems to be to change the stun effect to something akin to the effect of Team Fortress Classic’s concussion grenades, but honestly I think they need to go back to the drawing board with this unlock. Never mind that there are a few achievements that depend on it – go back to the design phase and get it right this time. And for God’s sake stop getting your unlock ideas from the Steam forums.

I’m starting to wonder if I should be playing Fortress Forever instead…

Lost Planet

This was recently on sale on Steam for the irresistable price of $5. I’d only ever played the demo previous to this, and I thought it was alright, so I decided to see what the full game was like (even though the superior Colonies Edition is out now). I’ve played the first few missions, and it’s not too bad. I can’t quite get 60fps out of it, but shooting up giant bugs in snowy wastelands is pretty fun. My main complaints so far are that the game is kind of easy (I haven’t died once yet) and that Wayne’s default movement speed is a little on the slow side, even when he’s piloting one of the giant VS mechs. I guess I’ll see if these continue to be problems as I progress through the game.

Still, this has me interested in the recently-announced sequel – I wonder what improvements Capcom will bring to the table.

King of Fighters ‘98 Ultimate Match

I started playing KOF (and indeed, fighting games) with King of Fighters ‘99, but I respect KOF’98’s place in the order of things. Given that disclaimer, I rather like ‘98UM. It doesn’t have most of my favourite characters, but the (remarkably solid) system changes they’ve made to the original game make this a must-buy for any KOF fan, I’d say. Most of the changes they’ve made revolve around making Extra mode more interesting, and it seems to have worked. Extra mode users now benefit from the ability to cancel normals into dodges, and to cancel certain special attacks directly into MAX mode. On top of that they can choose when they want to break stock unlike vanilla ‘98 where the bar started draining as soon as it filled up.

In addition, the new Ultimate mode, which allows you to mix-and-match subsystems from both modes, poses some interesting possibilities. Do you want the mobility afforded by the roll, or will you trade that for the ability to dodge and quickly counterattack (and extend your combos using the quick dodge)? Do you want the ability to do SDMs at any life level, or would you rather have the ability to max out in mid-combo for the possibility of turning your otherwise staid B&B combo into a more damaging variant?

As for the quality of the port itself, fortunately the US version of the game seems to have turned out pretty well – progressive scan support is intact, and as far as I can tell the game has been brought over more or less unmolested, which should be a relief for anyone who was horrified by Ignition’s handling of the PAL versions of King of Fighters XI and NeoGeo Battle Coliseum.

So yeah, if you don’t have the import version already, go out and get this one. It’s $20 – you really have no excuse if you claim to be a KOF fan.

Street Fighter IV

I’m still getting used to the physics and timings, but at the very least my win ratio seems to have improved a little (in that it is no longer zero). My MadCatz Tournament Edition FightStick arrived two weeks late, but I used the Amazon gift certificate I was given as compensation to buy a PS2->PS3 controller adapter so I could use my old Tekken 5 Hori stick with the game until it arrived (makes me wonder why I didn’t do it earlier, actually). So far I’ve mainly been sticking to Ryu and Abel, with some failed attempts at using Fei Long and Dhalsim (both of whom apparently have a pretty steep learning curve). I’ve played a few games online, mainly against Orochinagi members; I’ve played a few random strangers, although I haven’t run into any of the Ken players of legend. I suppose I should be thankful.

On a side note, the MadCatz stick doesn’t work for PS2 games. I don’t know why I didn’t see this coming, as this was also the case for the Sega Virtua Stick and the Hori PS3 sticks when they were released. The best I can hope for is for support to be added in the next PS3 firmware update, I guess.

BlazBlue

Now, I haven’t actually had a chance to play this game yet, but PS3 and 360 ports were announced recently. A lot of fighter fans are understandably excited about this, but a potential issue with the port has already been raised. The arcade version of BlazBlue runs at 1280×768/768p, while both current-generation consoles are locked to outputting at 1280×720/720p (and indeed, won’t let you play the game at its native resolution). The full extent of the problem is laid out pretty clearly over at the Insomnia forums.

In short, it looks like the great sprites and backgrounds are in danger of being butchered by scaling. If the blurry upscaled sprites in all of SNK Playmore’s Atomiswave releases bugged you, well, this is just as bad. Possibly worse, since it’s one of the first high-definition 2D fighters, and really deserves more respect.

At this point, given that ArcSys can’t go back in time and re-program the game to output 720p in the first place, the only real thing they can do is to crop 48 lines from the top and bottom of the display to avoid affecting the sprites. This is what they seem to have done, but the screenshots still lack the definition of the arcade version. A rep from Aksys (the company handling the US release) has gone on the record as saying that apparently the screenshots they released of the 360 and PS3 versions were smaller in size to make them ‘download friendly’. This just seems ridiculous to me, given that stuff like this is typically distributed through special press FTP servers where presumably bandwidth wouldn’t be a concern.

We’ll have to see how this shakes out, but I am rather worried that we’ll end up getting a butchered port of one of the first high-def 2D fighters.

4 comments

A lot of gaming news sites and blogs trumpet Valve as one of the finest developers in the industry, regularly producing excellent games and otherwise shitting gold. I’ve certainly suggested that I think of them highly, both here and elsewhere.

Recent events have shifted this view of mine somewhat – particularly the Team Fortress 2 Scout update.

I don’t think I would draw much ire if I were to state that TF2 isn’t a great competitive game (regardless of what my review says; it was written when I wasn’t really knowledgeable about the game, and I should probably delete it). The game, in the form that it’s played on public servers, tends to revolve around explosive/flamethrower spam, choke points and the use of ubercharges to get past these choke points. The presence of the ubercharge is pretty much the only reason why the game even holds together at this level, if you ask me – it’s the only tool available to get through chokepoints and break stalemates, outside of perhaps an unusually coordinated Spy rush. And let’s not forget about critical hits and random damage.

(The competitive format of the game has none of these problems. obviously – the 6 vs 6 format and class limits on Demomen and Medics mean that spam never becomes an issue. And of course, crits and random damage are turned off)

So given that the ubercharge is pretty much a tool designed to break stalemates (a design goal specifically called out by Valve in their developer commentary for TF2), I have no idea why they thought that giving the Scout a new weapon that would basically nullify ubers would be a good idea. The motivation behind this weapon (and at least one of the other unlocks) seems to have been ‘the Scout doesn’t survive too well in the spam-dominated public game environment, so let’s give him something to cope with them.’

As if that wasn’t enough, the same weapon basically makes the Heavy useless. He already has trouble dealing with Scouts when he doesn’t have his gun spun up, and now he’s basically a walking bullseye target when he does have his gun spun up.

If anything, that betrays a key flaw in the way Valve is approaching TF2 balance – competitive games should be balanced with high-level play in mind. More often than not games that are balanced for high level play (Starcraft, Quake III Arena, any number of great fighting games) end up being good games at lower levels too. Instead of trying to balance the game for pubbers they really should be balancing it according to the needs of competitive players – for instance, making the Heavy, Sniper, Pyro, Spy and Engineer more viable in high-level play than they are now. Granted, if they had done this, the game would probably not be anywhere near as popular as it is now, but shouldn’t good design trump marketing considerations?

On another Valve note – Left 4 Dead.

I’ve been playing Versus mode a lot recently, and I have to say that playing as the Infected is great fun. There’s nothing quite like when a good plan comes together. Unfortunately, the other half of the experience – the Survivors – is considerably less interesting. The best strategy is to camp in a corner or a chokepoint while spamming your melee attack. Melee attacks don’t do that much damage, but they do keep you from taking damage rather effectively, and you can basically spam it infinitely. In theory the Smoker is supposed to counter this tactic, but smoked Survivors can easily be freed by hitting them with a melee attack, and you only ever have at most one Smoker on the Infected team. I really think that the melee attack needs to be given a longer cooldown, or they need to have some sort of timer that prevents you from melee-ing non-stop.

On a side note, I’ve recently been playing Quake Live, and if anything it’s revealed to me how terrible I really am at multiplayer FPSes. Of course, I’ll be playing it a lot more because it’s basically Quake III, and Quake III is awesome.

4 comments