Tip:
Highlight text to annotate it
X
>>Ian: Hello, and welcome to the Friday review of games. As always I'm your host. Ian Ni-Lewis.
And with me today is our special guest Steven Spielberg.
>>Dan: [laughs] Thank you very much. Dan Galpin here as always to help out with playing the
games that you see here online. And giving you guys a lot of great feedback. Thanks a
lot as always for nominating the games to our Moderator. And today we are gonna be doing
driving games. Which areó
>>Ian: Absolutely, driving games which are actually some of my favorite games. But I
had never really played them on Android. I played Fortza, the entire Fortza series on
my Xbox. Actually started out with Project Gotham racing.
>>Dan: Absolutely.
>>Ian: Actually did some work on Project Gotham Racing III. Fixed their sound. Well fixed
the Xbox so it could play their sounds.
>>Dan: Exactly.
>>Ian: So love these games. Lots of great titles. But before we get started, wanted
to make some special mentions today. In the studio with us we have our new engineer. Engineer
Dan.
>>Dan: Mm-hm.
>>Ian: The other Dan. Dan Fam.
>>Dan: Yeah.
>>Ian: and producer Reto Meier. So can you introduce yourselves guys.
>>Dan F: Hey you guys. This is Daniel Fam. Just wanted to say hi.
>>Reto: And this is Reto. And you may know me better from the Android App clinic. Which
is on immediately before. And I'll be keeping an eye on all of the live streams and hopefully
passing any questions you have straight on to the guys behind the desk.
>>Ian: Absolutely. Now let's see the producer cam.
>>Dan G: That was the idea. It's to show the producer cam. Alright.
>>Ian: We got the producer cam guys?
>>Reto: hit Hangout. On the loweró
>>Dan G: Ah, there we are.
>>Ian: Yep.
>>Reto: Second from the top--
>>Ian: There you go.
>>Reto: No, not that one. No
>>Dan G: No, we're getting good though. Absolutely. Awesome.
>>Reto: Try it again.
[laughter]
>>Ian: That's not it either. But. Oh, oh oh, now, yah, yah. One more step. There we go.
>>Reto: Hi!
>>Dan G: Alright.
>>Ian: Alright.
>>Dan G: Producer cam. Excellent.
>>Ian: Whoo-hoo! Everybody was complaining. They said why don't you hire a professional.
And we did. So you can just stop complaining now. From now on the show's gonna be great.
>>Dan G: That's right. It's gonna suck less. I really like that like that.
>>Ian: Alright. So as far as sucking less goes, why don't we go ahead and open up some
liquid humor.
>>Dan G: [laughs] Oh wow, this is a big bottle.
>>Ian: Yeah. Well, you know it is a double chocolate stout. So today we're drinking Irondale
Double Chocolate Stout. From, uh, I have no idea actually.
>>Dan G: Yeah.
>>Ian: Swanton Rea's. Apparently.
>>Dan G: Yeah, brewed in bottle by Hermitage Brewing Company in San Jose. So there we are.
>>Ian: Beautiful. So this is local.
>>Dan G: Yeah, that's fantastic.
>>Ian: Even though it's got a label, it's all like, I don't know, English-y or something.
>>Dan G: Absolutely.
>>Ian: Whoa. Well, well.
>>Dan G: Exactly. There we are.
>>Ian: There you go.
>>Dan G: Excellent. Thank you sir.
[bottle cap clinks]
>>Ian: Alright, although apparently we are limited in the amount of time we can show
what we're drinking. Because after a few number of seconds. The Man considers it a beer commercial.
[laughter]
>>Dan G: I'm still wondering who The Man is.
>>Ian: Well in this case the man is our boss Billy. Hi Billy!
>>Dan G: Hi Billy.
>>Ian: So alright. Let's go ahead and talk about some games. Who do you wanna look at
first Dan?
>>Dan G: Alright well I've got two games cued up here. We're actually gonna try a little
bit of a different tack this time. We have two Transformer Prime's rather than one. And
the first game I wanted to look at was Real Racing.
>>Ian: Absolutely. Whoa. Whoa. You're, I mean you're really, you're gonna make it hard for
anybody to follow this, OK?
>>Dan G: I know. I know. This game is awesome. I mean.
>>Ian: OK, let's go ahead and take a look at it.
>>Dan G: Let's do one. Let's do Courier. I'm gonna do easy.
>>Ian: Alright.
>>Dan G: "Are you sure you wanna do easy?" Yes.
>>Ian: Very easy. OK.
>>Dan G: Let's bring up the camera on that. So here we are with Real Racing and we'll
start off with our club division. This is exactly how you begin the game. And Real Racing
is from Firemint Studios down in Australia. And they really have I think set the bar for
realistic graphics on a racing game.
>>Ian: Right.
>>Dan F: Camera two.
>>Dan G: And alright, we hit the home key.
>>Ian: Nice.
>>Dan G: Yes.
>>Ian: Well let's see how they recover from that.
>>Dan G: Exactly.
>>Ian: See, that was totally planned.
>>Dan G: That was our first test. Right. Exactly. Or it's possible that it died. Hopefully.
Oh. Come on. Come on. Fingers crossed. That's what's great about doing a live program. [sucks
in breath] Ooooohhh. Man.
Alright, so first thing.
>>Ian: OK, So.
>>Dan G: Checking the license every single time you run.
>>Ian: Yeah.
>>Dan G: Oh. Sad.
>>Ian: You might also do that. And the main reason is because you might be somewhere where
you can't get network connectivity. So have a license policy that lets you check the license
intermittently. And if you don't get it, if you don't get a contact with the license server
for a couple times, it's not a big deal. Most people aren't gonna be able to contact the
license server every single time they play.
>>Dan G: Yeah, the server manager that we have built into you know the license servers.
That's the way to go. It, you know we actually think that it makes a lot of sense. That people
can still get their license and yet play the game on a plane or something like that.
>>Ian: Right.
>>Dan G: Let's try a quick race instead. Let's see how we do.
>>Ian: Sure. Go easy on the camera two there, hot shot.
>>Dan G: [laughs] Exactly. So and now we're at the exciting loading screen agaaaaaain.
>>Ian: I love this loading screen man. But you know, but look at this. It's cool. I mean,
obviously you know, you're always gonna have constraints with a mobile device. Mobile devices
have power and heat and battery and size constraints that a full console or desktop computer don't
have. But with all that considered, its' a really nice looking game, I mean they've done
all of the interstitials and the hide and everything. Just like I remember on my Xbox.
It's great stuff.
>>Dan G: And you can actually, one thing that's not easy to figure out is you can actually
change the view here. And you'll notice, you know you actually would think that you would
get a higher frame rate when the full HUD was up. But actually you don't. Andó
>>Ian: Yeah, show that again.
>>Dan G: Yeah, here's our different views. Here's our full screen. There's our front.
And there's with the.
>>Ian: and what's the problem here, Dan? Whyó
>>Dan G: And the theory that I have, and I haven't thrown this one through all of my
tools. Is that in fact if you notice the out, the border of the HUD actually has some translucency
on it. You can see the outside of those windows. And so my theory is that they're actually
drawing this HUD last in order to get that a little bit of transparency. And what they
probably should be doing is drawing just the translucent parts of the HUD last. And actually
drawing the rest of the HUD first. 'cause you'll actually see the frame rate is just
killer when it's not there.
>>Ian: Right, they probably can't see it. Because I don't think we can capture video
for YouTube
>>Dan G: Yeah.
>>Ian: At the highest frame rate this device can put out. But take it, take Dan's word
for it. Because Dan has sort of a fifth sense when it comes to frame rate.
>>Dan G: [laughs] No, it looks amazing. And actually I've found that I really love playing
it this way. I actually get a sense of speed playing at full screen without the HUD.
>>Ian: Yeah, that's some nice texture down there.
>>Dan G: Yeah, it's beautiful. You can even see on the road, what I love is that they're
using a beautiful job with filtering there. And you can see MIT mapping. That texture
is just beautifully mit mapped so that when it's close it does not look blurry and when
it's far away it is not clogging down the GP with lots of extra drawing. And so you
know, I mean this really is just kind of a boiler plate of how to put the game. Even
when you see these buildings that we pass by. The textures might be a slightly bit blurrier
but we have such a great feeling of speed and the frame rate is high enough that we
donít' even notice it. And even little things like the effects of the sun flaring on the
windshield. I mean it's just beautifully, beautifully done. And the, the other optimizations
or changes you can make are pretty simple. You can turn on and off the HUD if you wanna
have just, be that much more immersive.
Now it's interesting, this game actually does not have a back button on the screen. Which
is awesome. You actually have to hit the back key to get here. And you can actually go and
change in the settings to go and say that I don't want to have the heads up display.
And now we can actually go back. And again it's all driven using the standard Android
navigation system which is cool. And so now there we have, we're really just full screen
which is kind of awesome. Like just being able to get to this view I think is supper
killer.
The other thing is that they don't have an accelerator. All we can do is brake. We brake
by tapping the screen. And it is actuallyó
>>Ian: You can actually put it into a mode where you need to both accelerate and brake.
>>Dan G: Mm-hmm.
>>Ian: Which can be nice actually. I prefer it that way because I like to slide around
corners
>>Dan G: Yeah.
>>Ian: By just decelerating a bit. But I think they made the right choice as far as making
it easy to jump in and play.
>>Dan G: Yes.
>>Ian: There's a lot of assists on. Which is why Dan is doing so well.
>>Dan G: [chuckles] Yes. I'm in easy mode.
>>Ian: Why don't you play this game with the assists off? Show 'em what it feels like.
>>Dan G: OK.
>>Ian: Uh, just go into your settings andó
>>Dan G: OK, let's get back into settings. Dah. Dah. And under, uhó
>>Ian: Settings. Yeah.
>>Dan G: So I can't change to "difficult" from here. I'd have to restart.
>>Ian: Oh, you'd have to restart the race.
>>Dan G: Yeah, exactly. So that was what I was afraid of. But yeah, the game really just
have a great sliding scale. But if you wanna just jump into it it's great too. There's
nothing that prevents you. And again as you're looking at this freeze frame you can just
see how high quality that road texture is.
>>Ian: yeah, let me take a look at that actually. So this is, um, like you really see what'só
>>Dan G: Yeah, the MIT mapping right there in the center of the screen.
>>Ian: Well, what's really amazing is that they've actually done a good job of simulating
an ISO.
>>Dan G: yes.
>>Ian: I don't think they're actually doing an ISO 'cause I don't think this hardware
can do it at that level.
>>Dan G: Yes.
>>Ian: For a lot of MIT map stuff you'll actually see sort of a line you know, about, let's
say an inch from the bottom of the screen you'd actually see it turn really blurry.
Or get some you know, just visual artifacts and general other [inaudible]
>>Dan G: Hm. The attention to detail is awesome on this game. So yeah, so this game also was
originally released as essentially a free demo that was unlockable. And this version
is now the premium version. And it's totally worth it. For 99 cents it's an absolute steal.
And if you like racing this game really does have the best visceral feel I think of any
racing game.
>>Ian: Yeah, so really I think the only thing we're saying about it is, if you're going
to paste a big *** HUD on your screen [Dan chuckles] do it first so that these parts
can z-reject pretty fast.
>>Dan G: Mm-hm.
>>Ian: And there are no mobile chips that are really good at overdraw. No Android GPU
is going to be happy with tons and tons of overdraw. So just don't do it. The one thing
that I'm missing in this game and, it's actually in every game that we've looked at.
>>Dan G: Mm-hmm.
>>Ian: I'm really missing good damage.
>>Dan G: Yeah.
>>Ian: And I don't see why not. I think the hardware is perfectly capable of running a
damage model. A simple one. I'm actually because nobody has it, especially the two Gameloft
games I'm looking at, I'm actually wondering if they couldn't get the license.
>>Dan G: Well that's what I was wondering too. You have to get a specialó
>>Ian: Yeah, damage license. So that is, yeah.
>>Dan G: But real cars in this game are beautifully rendered. It's again, it is the real car game,
the real car racing game on Android.
>>Ian: Yeah, and it is tough.
>>Dan G: It is tough. Especially if you're in easy mode. But it is nice the easy mode
for people who just wanna drive. So I'm going to switch over.
>>Ian: Yeah, as a contrast. This one, I would call this next one, "Forza III meets Mario
Kart".
>>Dan G: [laughs] So we're now looking at Asphalt 7. Let's, and I switched the tablets
so we can make a very very very fast switch. These are big big games.
>>Ian: Whoo!
>>Dan G: So I'm gonna let Ian play on this one.
>>Ian: Oh, ok you're gonna cue up the next one?
>>Dan G: Yeah, I'm gonna cue up the next one.
>>Ian: OK. So for this one, the game mechanics are similar in a lot of ways and since we're,
it's another one of these games where you start with a little bit of money. And that's
gonna let you buy a car. And then you're gonna compete in races for more cars and more upgrades
and things like that.
>>Dan G: There we are. We've switched. Now we're really on Asphalt 7: Heat. [laughs]
>>Ian: Nice. Alright.
>>Dan G: So, yeah. And so this is using, is what is not a free title. This is a paid title.
But it's a 99 cent title as well. And it's, it is very playable without additional purchases.
But it does strongly encourage you to spend additional money on the game.
>>Ian: Yeah. Well, you're gonna grind a lot to get more stars.
>>Dan G: Mm-hmm.
>>Ian: And in fact you're gonna pay a lot to get more stars too.
>>Dan G: Yeah.
>>Ian: So I noticed, for instance, so the first thing you get which is your little tiny
Fiat Abarth? I guess. I don't actually know how to say that. But this is like the entry
level car for every racing game ever. [Dan laughs] You wanna go up a level. To a range
rover.
>>Dan G: Mm-hmm.
>>Ian: Or even just a Mini Cooper S. That's gonna cost you 30 stars. Now 30 stars is like
ten or 15 bucks actually in there.
>>Dan G: is it really? Wow.
>>Ian: It's quite a lot. Let's see. Get more stars. Um, yeah, yep, there you go, ten bucks.
>>Dan G: Yeah.
>>Ian: Or you get one or two stars per race.
>>Dan G: Mm.
>>Ian: So you can see, you're gonna grind a lot in this game. I mean, these are not
cheap. So but you know, aside from that it's actually really fun. The races themselves
are a good time. I'm gonna go andó
>>Dan G: Mm. and you know to be honest, as long as you can grind in the game, I think
it's actually kinda fun and does give it sort of an RPG element. And I think that's actually
kind of a not entirely a bad thing. In terms of this.
>>Ian: yeah. And you've got this interesting thing where you like pay 5,000 for a sponsor.
Which is a little weird 'cause I thought the sponsor would pay you.
>>Dan G: I pay you, I know. [laughs]
>>Ian: OK, fine. They've got some goals here. They've got a little bit of a style system.
Like Project Gotham used to have.
>>Dan G: Mm-hmm.
>>Ian: Where you get some extra points for doing things like drifting and getting air.
[pause]
Mmm. And they also have some really beautiful interstitials.
>>Dan G: They absolutely do.
>>Ian: Yeah, it's very professional. It's exactly the kinda thing you'd expect. Now
once again, this is a very simple game. I think it is an auto accelerator. You have
to, if you wanna brake you have to hit the screen. But they have a nitro bonus system.
Because of course they do right?
>>Dan G: Yeah, which is that's the Mario Kart element. And actually just to point out there
are many many control schemes for Asphalt 7.
>>Ian: Yeah.
>>Dan G: So you can use tap on the screen. You can use Virtual Wheel. There's just tons
and tons of different ways you can set youró
>>Ian: Yeah. You control. But I tell you, especially with the newer Android tablets
having a significant improvement in--Whoa!
>>Dan G: Nice. Air.
>>Ian: Significantly improved their gyroscope and accelerometers, so you can get the attitude
sensor to work really really well.
>>Dan G: Now I'll tell you a couple things about this games. One of them is, this game
actually has shortcuts in the races. Which are totally fun, there's a lot of stuff to
find. There's far more than just one route through any of the individual races. My only
complaints about this game are really that, you'll see that if you look at the quality
of it, if you hit the pause key actually so we can take a look at one of these frames.
>>Ian: Oh.
>>Dan G: you can't actually look at the frame I guess.
>>Ian: Can you pause?
>>Dan G: I guess you can't. I thought you could. Anyways.
>>Ian: Oh, here we go.
>>Dan G: No. No.
>>Ian: No, you can't just pause.
>>Dan G: Yeah. Anyway, you can see if you look down, the quality of the road just is
not anywhere near what Real Racing is putting out. And if you look at the vegetation going
by. You can see that the stuff that's near the camera is really really pixilated.
>>Ian: Right. But you never notice that because you're going too damn fast.
>>Dan G: Right. Exactly. Also from a frame rate perspective.
>>Ian: [shouting at game]
>>Dan G: it is not, we're not getting consistently high frame rates. It's pretty good. I mean,
this is definitely, but I think that there's aó
>>Ian: It doesn't bother me. And actually if you play it on, like I played it on the
Nexus 7, I never noticed a frame rate problem at all.
>>Dan G: Yeah.
>>Ian: Of course, I don't have your supernatural skill in judging frame rates.
>>Dan G: So it is interesting. So this one I did actually throw into perf HUD to take
a look at it. And I was actually going to do sort of an interactive perf HUD view on
the show today but I think tható
>>Ian: No, but Dan chickened out. Trying toó
>>Dan G: I did. I chickened out. But partially because it wasn't interesting enough with
this sort of game play. But I, one thing that is interesting is you'll, on this game you'll
notice the cars are drawn last. And so we areó
>>Ian: Yeah. Um. [inaudible]
>>Dan G: Which is a little unfortunate cause it is taking up a large chunk of the screen.
>>Ian: Well, for the player's car.
>>Dan G: Yes.
>>Ian: You'd probably wanna draw the other things last, well, of course. This is something
that is, oh yeah, check it out. I'm in adrenaline mode!
>>Dan G: Aaaaaaa!
>>Ian: Who! Knock down!
>>Dan G: So.
>>Ian: So adrenaline mode is like you're in Tron mode. You can do no wrong. It's awesome.
I have no idea how I even got there.
>>Dan G: [chuckles]
>>Ian: Ooooh. Wrecked. Now see that's my problem. I just wrecked. My car is completely smashed
up. But it looks perfect. Why? Why? Did Fiat make you do that?
>>Dan G: [chuckles] It's very possible. Um.
>>Ian: So, the yeah. In generally what you wanna do is draw your foreground elements
first. So that you can take advantage of hierarchical z. And in this case what you probably wanna
do, you know for a racing game, is you can assume that the players car is always gonna
be in the foreground. Anything else is probably not worth it. The other cars are almost never
in the foreground.
>>Dan G: Yeah. You really see when you see the vegetation on the left hand side of the
screen where we're going now. That's where you really see the quality.
>>Ian: Yeah, I think that's a textural density issue.
>>Dan G: Yeah.
>>Ian: They just aren't loading up big enough textures. Um, although, you know, they have
a lot of varied textures.
>>Dan G: They do.
>>Ian: and for the areas of the track that are father away it's actually very well done.
I think.
>>Dan G: Yes. Absolutely. I just think there's probably one level of MIT mapping that's missing.
>>Ian: Right.
>>Dan G: And tható
>>Ian: Well and knowing the game development process I wouldn't be surprised if that level
was created and then thrown away at the last minute in order toó
>>Dan G: Yeah. Very possible and you can also see there is a bit of popup going on. If you
look at that in the distance. And--
>>Ian: yeah, I did notice some popup. And that's an issue. And it's actually a little
bit surprising considering for instance, this particular GPU is really excellent at geometry.
>>Dan G: yes.
>>Ian: compared to its pixel performance. You actually probably get more out of pushing
your far z plane out a little bit but you know.
>>Dan G: Yeah. When youó
>>Ian: That's not gonna be the right choice for everyone.
>>Dan G: And also that is only good if you're drawing in the reverse draw. If you're drawing
in order you're going to, by pushing your z plane out you're still gonna have some pretty
big performance problems.
>>Ian: Well, if you're going back to front. Right.
>>Dan G: Yeah. If you're going, if you're goingó
>>Ian: Which you should never do. Right? I mean.
>>Dan G: Exactly.
>>Ian: I mean drawing your stuff back to front is always the wrong choice unless you have
to do Alpha. What you should be doing is just like these guys here is you should be drawing
that alpha area last.
>>Dan G: Mm-hmm. Absolutely. So that's just basically something that everyone needs to
look at. That being said, this game is totally fun.
>>Ian: Absolutely.
>>Dan G: And that's, and it is beautiful looking.
>>Ian: It's a kick. Andó
>>Dan G: I just wanna see it that much better. Like that's--
>>Ian: Yeah. I really do too. And I would love to see more you know, some more collisions.
Some more particles. My guess is that the very top end of Android devices, especially
tablets could handle more than what we're seeing here.
>>Dan G: Mm-hmm.
>>Ian: I would love, I would pay extra to get let's say a Tegra optimized version or
this. Or a tablet optimized version of this.
>>Dan G: Mm-hmm. Yeah.
>>Ian: To play around with 'cause I think this game could be even cooler than it is.
>>Dan G: Totally. Totally. And you know. Just interestingly enough, hitting the back key
multiple times just gets us out to the menu. Kind of the way we'd like. They've done a
good job of making the navigationó
>>Ian: Mm-hmm. Yeah, the navigation is great.
>>Dan G: And boom, we're back out.
>>Ian: Perfect.
>>Dan G: So that's that. Let me see, I was gonna try to cue up the next game on this
tablet. And we'll see if I actually managed to do that.
>>Ian: OK. Before we start the next game I think Producer Reto has something to say.
>>Dan G: Oh, exactly. He is on the screen. So.
>>Reto: Exactly. Thanks guys. So just a couple comments on YouTube and on the G plus stream.
So Leif points out that Asphalt 7 does notification spam, and after not playing it for a few days.
>>Dan G: Ooooh.
>>Ian: Oh. How many people don't do notification spam?
>>Dan G: Now the real question is actually whether or not you can actually turn off notifications?
>>Ian: Right.
>>Dan G: And right away Game Loft assures me that it's actually, the game is intended
to be playable offline. If for some reason, you saw that network errors occur, we tried
cancel. They say that's a bug.
>>Ian: So what we're asking people to do now is, OK, we understand that some people like
notifications. We don't know who those people are, but apparently some people do like notifications.
What we want to do is make sure that users who don't want notifications or even users
who do want notifications but they don't want all of them, have the ability to control that.
Now of course we added new features to Jellybean to make it easier to control the kinds of
notifications that you get from different apps. But that's sort of an all or nothing
thing.
>>Dan G: Mm-hmm. Yeah.
>>Ian: You can say "This app is never allowed to make, to send me notifications" or "This
app is allowed." Now me personally, I don't mind being told for instance that there is
a massive sale on some in-app item of a game I like. But I do hate personally, I do hate
those notifications that are like "You haven't played me in a while. Why don't you come back?"
I mean.
>>Dan G: Yeah.
>>Ian: If I wanted somebody whining at me about coming to play, I'd play GTA4.
[laughter]
>>Dan G: Yeah, no exactly. And actually one of the things I was looking for, and I didn't
actually check this game before we started. And I was looking to see if there was a way
to disable that within the game. Doesn't look like there is. We strongly encourage developers
to make that something they can turn off. And 'cause a lot of users prefer it. Now it's
interesting, on one of the other Game Loft games if you turn off notifications, every
single time you start up the game it pops up a dialog and says "Hey, you realize you've
turned off notifications? Would you like to turn them on?"
>>Ian: Wow. That's like iTunes bad. "Hey I noticed I'm not your default game player for
this machine? Would you like to change that?"
>>Dan G: Yeah. [chuckles]
>>Ian: Come on.
>>Dan G: So.
>>Ian: There will come a day when a, those people will be the first up against the wall.
But even before that there will come a day when that is no longer considered feature-worthy.
>>Dan G: Yeah, we really wanna make sure the user experience is, and addressing the user
concerns are the number one priority for developers. And you know, we understand that notifications
are useful. There's a lot of game players who actually really like them. And certainly
on a lot of the games or resource management is an issue. Or their daily bonuses. That's
really awesome. You know like, playing Dungeon Hunter and going "Hey man, there's like if
I play today I have an opportunity to get 5,000 gold for doing this." That's kinda cool
as long as I can turn it off.
>>Ian: Right.
>>Dan G: Well we do have the other game, do you wanna go on? Or are there other questions.
>>Ian: Reto maybe had one more question.
>>Reto: There is, there's one more. So this is an actual question, a programming question
from Alex.
>>Dan G: Right. OK.
>>Reto: On G plus stream. And he says "You guys always say to pay attention to the resolution
of the art aspects within games so that they don't look blurry or overstretched."
>>Dan G: Mm-hmm
>>Reto: "I'd like to know what you think of pixel art style of graphics. Do you think
it's ok to do all the art on a canvas of say 400 by 240 height with pixel art style?" He's
asking because pixel art looks nice when stretched but it's all messed up when shrunken. And
that goes against standard best practices on game art where you obviously have higher
res assets and then shrink them down. Although making two separate groups of the artwork
both big and small seems to be a better option it's not practical 'cause it's going to take
too many resources.
>>Ian: Well, remember that the reason it looks bad when it' shrunk is because we're doing
the scaling very quickly. And it's usually, for instance OpenGL just implements a bi-linear
filtering which isn't very good. Now, if you run your art through something like Photoshop
and it does a Lanczos filter on it, you're gonna get art that looks pretty much great.
>>Dan G: yeah, exactly. I wouldn't worry too much about that. You also actually can actually
go into bitmap factory you can set the options for higher quality scaling as well. Within
the actually scaling of the bitmaps and that. That may help you out as well. But the truth
is, yeah, my feeling is high resolution pixel art looks amazing.
>>Ian: Mm-hmm.
>>Dan G: And the games we saw, if you looked at the last weeks show we were doing tower
defense games. We saw some amazing high res pixel art. That just, that looked mind blowingly
good on these devices and actually scale down reasonably well.
>>Ian: Yeah.
>>Dan G: So I think it comes down to art design and style. There definitely are outlying pixel
art really doesn't scale very well. If you're trying to have lots and lots of really delicate
line details, those are gonna get lost alot of the time.
>>Ian: Yeah. You need to do that.
>>Dan G: But it reallyó
>>Ian: Well, and you know who does actually very careful line drawing? Is Rovio.
>>Dan G: Mm-hmm. Absolutely.
>>Ian: and they actually put a lot of effort into making sure that at every single resolution
the blackout line around the birds is exactly one pixel wide.
>>Dan G: Yeah. And they're very good about that. And so I think it's certainly possible.
It's more of an art design question I think than a technical question when it comes to
that.
>>Ian: I think a lot of people will do something in pixel art thinking that's actually gonna
be easier. And it's not. Doing pixel art well and making it look great on a modern device
and also appealing to modern users is actually really tough. Because you want it to be old
school but you don't actually want it.
>>Dan G: Actually want itó
>>Ian: To be old school. You want it to be evocative rather than representative of your
limitations.
>>Dan G: Absolutely. But I mean we certainly love pixel art. We've given major thumbs up
to a lot of pixel art titles in terms of you know what we, the presentation. And I think
the real key thing is just again, like anything else, it all depends on the quality of your
artist as much as anything else. And your design. Andó
>>Ian: Absolutely. So. Producer Reto, I have to ask, was, so is that HPC2000 guy on the
live chat?
>>Reto: I haven't seen any questions from him yet.
>>Ian: OK. Yeah, he was gonna ask you a question about open gl. I told him I'd answer it. But
I'm not gonna answer it if he's not watching. 'Cause come on.
>>Reto: Exactly. I will keep an eye out. I will let you know if he is actually here.
>>Ian: Sounds good. Let's go ahead and move to our next game.
>>Dan G: Alright, so our next game isn't really a driving game. It would be aó
>>Ian: Is there driving in this game?
>>Dan G: There is driving.
>>Ian: There is a lot of driving in this game.
>>Dan G: It isn't really a racing game. I should say.
>>Ian: That's true. We didn't say racing.
>>Dan G: It is a driving game.
>>Ian: we said "driving."
>>Dan G: Did we?
>>Ian: Yes.
>>Dan G: OK.
>>Ian: And the only thing that we put off limits was golf, as I recall.
>>Dan G: That's true. We said that it was not that kind of driving. [laughs] So here
we are. It is GTA. They've done a great job of bringing it to Android. It controls very
very well. I actually really like the fact that the default set of controls is actually
tapping the screen. Rather than accelerometer. I think it works really well for the kind
of arcade-y style of driving that is in the game. And so you know, since it is a driving
game we're doing to hit the car. Now there are a lot of on screen buttons in this game.
And I would say a lot of that is because it wasn't really designed for this kind of system.
I think if you were designing it today, you would not have nearly as many button actions
as this game has.
>>Ian: Yeah, this is definitely more of a port than a remix. I think it's not reallyó
>>Dan: Wow, am I driving badly.
>>Ian: It's not really re-imagined for tablet, it's just totally ported. So you know what
this plays awesomely with is like a Nyko controller or an Xbox 360 controller. Plug into the USB
port.
>>Dan: And it's great, this port that if you're on an Android 3.x device or you're on something
like the Xperia Play 3.x or higher device they really do a great job of making sure
those are there for you guys. But other than that they did a very good job of porting the
game and the graphics over.
>>Ian: You know where I played this is, I met with the Wikipad guys.
>>Dan G: Oh yeah.
>>Ian: You might have seen Wikipad online. They were at GDC. And they handed me a prototype
and it was playing GTA three and I'm like, wow. It's a completely new experience to be
able to play that on tablet with a controller that's built in. Do I'm hoping that we see
more of those. And there's several that, Wikipad, Nyko's, there's one other I can't remember.
>>Dan G: Yeah, Moga is also doing it.
>>Ian: Moga.
>>Dan G: Yeah, it's got a pretty cool controller andó
>>Ian: And of course we support USB controllers. Basically any USB controller that Linux supports.
>>Dan G: Yeah and Bluetooth as well. So USB and Bluetooth controllers. You know, there's
really gonna be a lot of choice out there. Let's run into a few more trees here in Grand
Theft Auto. So what we're gonna say about this other than it's a port done well. I would
love to see a more native version designed for touch controls. I think that you could
take it really really far. But they've done about as good a job as you'd expect to do
given where they started with a fairly complicated joystick interface. Nice, my car is on fire.
And Iím in the first, I'm in training mission. Um.
>>Ian: This is just, it's a fun game. What more to say?
>>Dan G: Yeah, exactly.
>>Ian: Let's not waste any more time. They did a good job. Oh, but there is actually
one thing that's unique about this game.
>>Dan G: Um-hm.
>>Ian: If they still have it. They still have the persistent notification?
>>Dan G: Yes. They do, they just kind of whack you with that. So if we go to the home screen
you'll see that it actually lets you know whether it's still running and allows you
to get back to the game right away. I don't know how exactly I feel about this. 'Cause
things do run in the background and that's the expectation.
>>Ian: I know, I know, I am so torn about this. Because on the one hand, it violates
a lot of the philosophy that we have around Android. On the other hand, it's done for
such a good reason. Because basically everything that we've done around Android is for a completely
different use case.
>>Dan G: Yes. Andó
>>Ian: and the use case of having something that is really really big, and really really
expensive to initialize, but also something that you might wanna interrupt every now and
then to let's say, get a phone call. It's tough. So they've done something interesting.
They basically say, OK we're gonna run all the time. But instead of just doing that,
which is horrible,
>>Dan G: Yeah.
>>Ian: They put up a persistent notification that says essentiallyó
>>Dan G: Our process is still in memory.
>>Ian: Yeah. GTA III is sitting here taking up resources and draining resources so you
can go back to it fast. Do you want it to stop that?
>>Dan G: It is, yeah. It is actually not draining your battery. But it is still resident in
memory.
>>Ian: Mm. Well.
>>Dan G: So it doesn't actually keep the game running. The threads areó
>>Ian: The threads are stopped. But it will end up making things harder for you. You know,
there will be more swapping out.
>>Dan G: Yeah, I mean ultimately Android will kill the process if you try to run too much
stuff. And we should actually just try that. Let's try to launch Real Racing and see what
happens.
Now I'm, testing fade here. We're obviously running two enormous titles. What should happen
is we should see GTA get kicked out of memory. Always gonna do these things live. [chuckles]
>>Ian: Let's see what happens.
>>Dan G: Exactly.
>>Ian: Well my point is that it's not consuming CPU, but it is gonna consume the battery because
what's gonna happen is that small wraps are gonna cycle in and out faster.
>>Dan G: Yes. And the thing is, once it gets kicked out it should just be kicked out. It
doesn't get kicked back in unless they're actually running a service. And hopefully
they're not doing that.
>>Ian: Well they are running a service. They've got a persistent notification, it's really
hard to tell.
>>Dan g: yeah, I actuallyó
>>Ian: You're gonna have to go out of here and then go back into GTA and see what the
[inaudible]
>>Dan g: Yeah, well you see the notification is still there. So it'só
>>Ian: well that's true but that doesn't really mean anything 'cause all that means is their
service is still alive. It doesn't necessarily mean their main processor is.
>>Dan g: Yeah, let's take a look at what happens. Boom. There we are.
>>Ian: Wow. Well you know.
>>Dan G: So we had, this takes a lot of ram. [laughs] So.
>>Ian: well that's true.
>>Dan G: And anyways, again it is possible to swap between two rather large games. I
donít think I wanna push it and actually swap between more than that. But there we
are with Real Racing as well. So. Awesome. Nice multitasking going on there.
>>Reto: Couple of quick points from YouTube. Crispy Core wants to know if you can try and
see if the tilt controls work while driving. He asked that just a couple of minutes ago
so I'm guessing that's for GTA.
>>Dan g: The tilt controls for GTA?
>>Reto: There seems to be a lot of problems with that.
>>Dan g: I think there may be an issue and as I recall this is one of those games that
actually has I think accelerometer issues. So let's turn accelerometer on. And let's
use take analog.
>>Ian: Use analog steering yeah.
>>Dan g: Analog steering.
>>Ian: And turn that on. OK.
>>Dan g: OK. And lets' take a look. As I recall, this may have been one of those games that
actually does not correctly, we look at the default orientation when interpreting the
accelerometer. But I think as I recall, and it's been a while, I think they do it correctly
for tablets. And they may not do it correctly for--.
>>Ian: so that may actually be something the type of device instead ofóso what'll happen
is the orientation is always done with respect to a default orientation of the device. You
know, in other words, which orientation the home screen expects to display in. and that's
not always what the game expects. Ironically, this isn't very difficult to deal with. Where
people get into trouble is when they try to figure out what the default orientation is.
And then figure out what the accelerometer is telling them rather than just believing
the quaternion they can get from the accelerometer.
>>Dan g: Yeah, the quaternion is actually always going to be correct. And that's really
the key here. But anyway, the tilt is working fine here. We should try it on a phone device
and see whether it's also handling that correctly. I shouldn't say phone. What I should say is
a portrait default device. The real question is what happens on Nexus 7 here. So we can,
I think I have it on this guy.
>>Ian: We'll cue that up and see if there's another question while we're doing that.
>>Reto: Exactly, so while Dan's getting that going, HPC2000 is in fact here, so we would
love to hear the answer to his question. So if you could repeat the question.
>>Ian: Right. I believe the question was about how, oh, boo.
>>Dan g: Whoa.
>>Ian: Look at that. I downloaded all these files last night. At least I thought I did.
Maybe I didn't.
HPC2000 asked, how easy is it to take an open GLES game that's made for some other system.
He named a system, I hadn't heard of it.
>>Dan g: yeah.
>>Ian: But whatever, use what you want. If you wanna port from another system, that's
an interesting question. I'll tell the truth. It's something that I think we have mixed
feelings about. And when I say we, I mean the Android organization in general. We want
people to bring all of the best content on Android. At the same time, we don't want to
just see random crummy ports. We wanna see things that are built for Android. And for
its unique abilities. So that said, it's not very difficult to port and open GLES game
that's written in C to Android. There are some issues. A lot of people think that since
Android is Linux, it should be very easy to port any piece of Linux software. And that's
not entirely true. Because Android like all other Linux systems is build out of a lot
of different components. And there are some things you can pick and choose in terms of
the libraries, internal chains. We have a specific C runtime that we use called bionic.
That a lot of other people don't use. And all of this adds up to more hassle. Now, normally
for a Linux system this isn't a hassle because you use something like apt or you know, macports
or one of these things that compiles everything and runs through a bunch of compatibility
scripts. And if you've ever used a port system, you know that there's like a million different
tasks to figure out where your C runtime lives. And where your pilot lives and all these other
things. Android of course doesn't do that. We distribute binaries to the phone. Or to
the device. So you are not able to count on all of the things that your desktop Linux
system has. Or even all of the things that are in the Android open source project right
now. What we do instead is we release something called the NDK. The native development kit.
There's a lot of things in the NDK. But the real core of it, the irreplaceable piece of
it is a set of headers that form a contract that Google has decided we will always fulfill.
>>Dan g: Yeah, it has an NADI, an application binary interface. And the key thing about
that is that for your open GL calls, those open GL calls will be essentially identical
because open GL is a standard. It actually provides a contracted ABI itself. And we're
adhering to that contract. So from an open GL perspective all the open GL code should
port over incredibly easily. Obviously, if you're taking advantage of any architecture
specific implementation details, especially extensions. Those'll have to be looked at.
But the vast majority of the open GL code will run absolutely identically on both platforms.
>>Ian: You will need to be aware of one thing though.
>>Dan G: Yes.
>>Ian: Which is because Android is an open system, there's a huge number of OEMs. You
can't make as many assumptions about the GPU hardware as you can with some other systems.
>>Dan g: Absolutely.
>>Ian: So even though the open GLES ABI is standard, some of the extensions that you
may have come to rely on or possibly some of the behaviors or even bugs that you may
rely on one system are not necessarily going to be the same on other systems. A good example
of that is that PVR GPUs do geometry sorting. So they're almost completely insensitive to
the way you submit geometry. But they are very sensitive to how many state transitions
you do.
>>Dan G: yes.
>>Ian: Where as Tegra GTUs are almost the opposite. They change state very fast but
you want to always sort front to back.
>>Dan G: exactly and so the real key, and that's certainly, it's only one issue. There's
also things like if you wanna use texture compression. You know, there is not a standard.
>>Ian: Everybody's got their own thing.
>>Dan G: High performance ones used on Android, there's also you know, there are also issues
in terms of how you should write your shaders. And you know, making sure that you're really
querying very carefully. Also making sure that you do test those shaders on a variety
of GPUs. 'cause you will see different performance cliffs from device to device. That being said,
from a coding perspective, your code is going to, what you really have to look at is audio
'cause the audio standards are very very different. You're gonna look at touch and events which
are very very different. Networking actually if you're not using SSL, you can just use
sockets. And so your socket code will come over pretty much straight, to be honest. So
there's a lot you do get a lot. You can actually use even threading you can. If you're using
standard p-threads. If you, as long as you're not relying on making calls back into the
Dalvik virtual machine you can actually use those very easily. Or you can actually tie
those back into Dalvik threads. And there's actually a really nice library that Invidia
has provided for doing that. So there really are a lot of things you can do and really
one of the main reasons we do the NDK is to allow people to take advantage of tested code
sections that you, code libraries. That you have a written for another platform. So from
that perspective, that plus the standard APIs do provide a really nice basis. Especially
for games. Where it gets more challenging is when you want to take advantage of the
Android standard UI patterns. And those are, there's a lot of useful stuff in there. Even
if it comes to things like font rendering. Database storage isn't there. So if you wanna
use databases you're gonna have to make a call through JNI to getó
>>Ian: Oh, no, no. You're gonna get to make a call through JNI.
>>Dan G: [chuckles] That's right.
>>Ian: Because remember the java interfaces in Android are very very easy to use compared
to C and C++. There's' a lot fewer ways to shoot yourself in the foot. You'll end up
writing a lot less code. A good example of this is when you're doing your game play,
you know, every large scale C++ project that I've worked on has its own resource management
and reflection system. Which of course is built into Dalvik and into the java framework.
So you can actually save yourself a lot of time. Especially when you're dealing with
platform specific things. But even when you're not, by going into Dalvik or another virtual
machine.
>>Dan G: Absolutely. And there's plenty of places even in a game where it makes a lot
of sense to take advantage of Android framework resources. Things like notifications for example.
You know, networking. It's just nice to take advantage of the built in HTP stack. Built
in SSL implementation. All that kinda stuff. So there is a bunch of stuff that it does
make sense to actually move code over to the Dalvik virtual machine. But that being said,
for your core code, if there's things like physics engines that may be written entirely
in native code. Things like box td. There's really no reason not to just use that code
as it is. If it's a well-tested bit of code.
>>Ian: Absolutely. OK. I think that's enough on that. Producer Reto, are there any more
questions?
>>Reto: No.
>>Ian: OK.
>>Dan G: Excellent. [laughs] [inaudible]
>>Ian: Well done on the producer cam. That isó
>>Dan G: We're getting the hang of that today.
>>Ian: Yes. Good job.
>>Dan G: That's awesome.
>>Ian: Now, we're running a little low on time but something occurs to me which is that
we did not actually go in order of what people voted one today.
>>Dan G: That's true.
>>Ian: We sort of went in order of what we personally thought was cool. Especially in
order of these big profitable games.
>>Dan G: Yes.
>>Ian: Which I think makes us look a lot like tools of the game oligopoly establishment.
>>Dan G: Well, and I will say that all of these games really are fun, like these are
great games.
>>Ian: Yeah, absolutely. There's no question about that.
>>Dan G: And they're beautiful. They make, they makeó
>>Ian: But again, we don't wanna be tools.
>>Dan G: No that's true. And actually I didn't, I would have added Need For Speed: Hot Pursuit
by the way, to this list, had I been a tool. Because I actually love Need for Speed: Hot
Pursuit. I have ever since it came out.
>>Ian: I don't think it's a tool thing to beó
>>Dan G: And but I didn't nominate it to the list because there were so many other big
games on the list already. So let's--
>>Ian: Well, my main point is, we need to look at Tractor Digger.
>>Dan G: So Tractor Digger is not actually a racing game.
>>Ian: It's not.
>>Dan G: So I'm just going to bring it up very very quickly.
>>Ian: But it was voted number one. [inaudible]
>>Dan G: It was voted number one. And it is built using AndEngine.
>>Ian: and by the way you know who else is a tool? Anybody who watches this show and
doesn't vote. Vote for stuff.
>>Dan G: So this is actually a puzzle game and your goal is to move things into the tractor.
You've got a dual joystick control. And then we can actually take this stuff and move it
into this dumpster here. And it's actually pretty cool. It is not [chuckles] a racing
game.
>>Ian: We didn't say racing game. We said driving game. Are you driving?
>>Dan G: Sort of.
>>Ian: Are you driving a tractor?
>>Dan G: Yes.
>>Ian: I think you're driving a tractor. Um, now for some reason Dan is snapping his fingers
at us and I don't know why.
>>Dan G: Ah, the reason why is that I don't have this tablet plugged in. so let's actually
show you what I was doing rather than have me just looking down.
>>Ian: Yeah, see that's why we look at the monitor.
>>Dan G: Yeah, exactly.
>>Ian: And not the, yeah.
>>Dan G: That's why we have that. That's a good point. So anyways, oh it's backwards
though. That's really hard. [laughs] Never mind.
>>Ian: Maybe just glance at it. But--[inaudible]
>>Dan g: Yeah, exactly. So, um, so we can actually take advantage of physics here to
scoop up the rock into our little tractor and then we can actually flip it all the way
round to the other side in order to dump it back into the dumpster. But it is a puzzler
game. And it's cool. You're driving a tractor. And it's something I haven't seen before.
It's an AndEngine game. Which is always nice to see. If you haven't actually used AndEngine
before it is a really nice 2D game development platform. Written for Android. We actually
have them showing off the tools chain at Google IO this year. And it really does look, it
looks nice. And they're doing a lot of investment in it. Which is exciting. Oh by the way, you're
not trying to get the boulders into the truck. Just. Uh.
>>Ian: I did a good job of it though.
>>Dan G: Yes you did, you did. Now, at the controls. The controls work well, there's
definitely some parts of this game where I want to see higher resolution artwork. Some
more work done on some of the textures. So that it looks less this sort of repeating
pattern that you're seeing at the bottom. It just screams "I'm tiling something!"
>>Ian: Right. The right way to do that by the way is to create three or four random
textures and just randomize them.
>>Dan G: Yes.
>>Ian: For a texture like this, you can use, there's actually some photoshop plugins that
do this.
>>Dan G: Yeah.
>>Ian: You get something that has four or five different textures but they all tile
seamlessly along the edges.
>>Dan G: Mm-hm.
>>Ian: And then you just pick 'em at random and it looks really good.
>>Dan G: It really does. And that same thing I would say with the grass texture. It's just
like, tile, tile, tile, tile, tile.
>>Ian: Exactly.
>>Dan G: And I would love to see a little bit of work. 'cause there's so much work that
actually went into the vehicles in this game.
>>Ian: Yeah, they're cool. And you know what, I dig the background in this game even though
it's just a photo. Its' nice that it's out of focus.
>>Dan G: Yes.
>>Ian: It's got that depth of field effect to it. Good job there.
>>Dan G: Yeah, absolutely.
>>Ian: OK. Let's move on.
>>Dan G: Let's move on. Again, not a, not a racing game. Another one that's in the category
of somewhat--
>>Ian: [whispering off mic]
>>Dan G: Somewhat smaller games is Drag Racing from Nitro Nation.
>>Ian: I really like this game.
>>Dan G: And um, this game was, is kind of fun. but it really just seemed to be an exercise
in trying to end up getting me to buy stuff. Which was a little unfortunate. You know,
its' like, what I found was that pretty much no matter how well I did in the game, if I
did not have a car that was good enough and this was not an example of that. I was just
going to lose. Mostly because it's just kind of minimal controls here. And I actually played
this game not particularly well. I should actually be shifting.
>>Ian: I was gonna say, you know.
>>Dan G: See, you get good shifts.
>>Ian: If you shift like a moron you're not gonna get anything out of the race.
>>Dan g: Yeah, but even if you shift really really well, you know, there's a point at
which your car just isn't good enough. And you'll have to buy a new car. And the only
way to buy a new car is to get respect. And it's very challenging to get respect within
the game without actually paying money. And I kinda wish the game had sort of a deeper
grind factor. Maybe it's there and I just didn't see it.
>>Ian: Mm-hm.
>>Dan G: In my playing of the game. That being said, the game plays really well. It looks
pretty good. A little tiny bit blurry here on the size of a Transformer Prime.
>>Ian: It's definitely a button masher.
>>Dan G: It's, I wouldn't sayó
>>Ian: Yeah, it'só
>>Dan G: I wouldn't even call it a masher.
>>Ian: It's Pavlovian. It shows you the little upshift thing and then you press the button
and it says "Good job!"
>>Dan G: Exactly.
>>Ian: Good Job!
>>Dan G: So yeah, it's pretty limited. But I wanted to at least bring it up. You know,
and again, if you like this kinda game, it seems to be a pretty good version of that,
of an entry into that genre.
>>Ian: I mean, I've seen better. I really have. But not on tablet.
>>Dan G: exactly. Now another completely different kind of racing game is Reckless Get Away.
>>Ian: Oh, this is a great game. I love this.
>>Dan G: This is a super fun. it's, and it actually has a cute little story behind it
that you are trying to get away. Let's actually doó
>>Ian: Recklessly?
>>Dan G: Recklessly. Yeah. And even little details like.
>>Ian: I love that. "Please wait, robbery in progress." That's a great one.
>>Dan G: Robbery in progress. I love the money flying out of your car. There's all sorts
of cool jumps you can take. There is just tons and tons and tons of stuff to do in this
game.
>>Ian: Now I would have loved to have a little bit more control over it. but maybe that's
just me. Now I'm old and I don't have the greatest reflexes, but I felt like this control
scheme was actually really difficult to manage. I would have actually liked some really sensitive
tilt controls. Which is odd because as you know, I hate tilt controls. But--
>>Dan G: Yeah, driving games are one of the few places
>>Ian: Exactly.
>>Dan G: In which I find that tilt controls actually are not so bad. And I often prefer
them.
>>Ian: Well especially if they're reasonably sensitive.
>>Dan G: But this game is really well done. And it's, I love the perspective. You know,
we took a jump there.
>>Ian: and all the background work they've done. I mean this is, it's a professional
title. There's no question about that.
>>Dan G: Yeah, absolutely. So I mean as I said, I was glad to see this up there. This
game hasn't gotten a lot of love recently and I think it's just, I think it's a really
well done title that does something substantially different in this genre. You know. There's
really what I, three different kinds of games that we see really commonly in the genre.
We see sort of slot racing games. Which are over the head where you see a large percentage
of the track. We see the more traditional driving games. And this is a perspective we
don't see nearly as much. Which is kind of reminiscent of the perspective from like Spy
Hunter. In the original overheadó
>>Ian: that is what it reminds me of. Yeah.
>>Dan G: Overhead racing game. And it's awesome.
>>Ian: [inaudible]
>>Dan G: So you know, I definitely, here's by the way an example of the developer who
just did not know what to do with the back key. Here we are we've got a screen that says
"Play again" and "Continue" and there, so what do we do here with the back key. Well
actually in this particular case, Play again or Continue, there is actually a third option
which would be Exit to the main menu. And hitting the back key actually should exit
you back here. There's no reason why it needs to be Play and Continue. This is sort of a
design, so they made the back key do nothing. That's bad. It should always do something.
Especially at the end of the game when you might just wanna exit.
>>Ian: Yeah.
>>Dan G: And switch game modes and instead you have to target the continue button which
is really unfortunate.
>>Ian: Remember, the back button does two things. It moves you back to the previous
activity. Or it dismisses dialogs. So if you're saying to yourself, "Wow, what I'm looking
at here has no meaningful back thing."
>>Dan G: Yeah.
>>Ian: you know maybe it's because you need to think of it as a dismissing rather than
a moving on.
>>Dan G: Yeah. Continue actually does not. When I saw "continue", I think what does continue
mean? I had no idea. And then I realized what they, what "continue" actually does is take
you back to the selection screen. So "continue" is really not, doesn't make much sense there.
>>Ian: It's tough. Certainly there are some places where we stretch a little bit. But
you know what? Suck it up. Android has a back button and people expect to be able to use
it.
>>Dan G: So here's I wanted to bring up Tiny Little Racing.
>>Ian: Oh yeah, you know what, we were gonna look at this a while ago. And then we didn't.
>>Dan G: So first of alló
>>Ian: And I even said "Wow, we really loved this." And it turned out I was wrong.
>>Dan G: [laughs]
>>Ian: It turned out it was a different game with a really really similar name that we
really really loved. But this is still a good game.
>>Dan G: So let's talk about it. First of all, we see this.
>>Ian: Oooohh. Don't.
>>Dan G: [shouting] Don't ever do this!
>>Ian: Yeah, yeah.
>>Dan G: Don't ever do this because weó
>>Ian: At least autofill, come on.
>>Dan G: Well it's not just that. One is that I'm not using the standard Android keyboard.
>>Ian: Oh yeah.
>>Dan G: So the question is, I'm gonna stick this in the dock. And the keyboard doesn't
work.
>>Ian: [shouting] Number two! Camera two!
>>Dan G: Go to switch to camera two!
>>Ian: Number two! Camera two?
>>Dan: Wreee! Wreee! No, that's a no!
>>Ian: [inaudible] Alright. OK. Now look at what poor Dan is trying to do here.
>>Dan G: Yeah, I'm, I just quit the game. But I was trying to type my name. and I can't
type on the keyboard.
>>Ian: The point is, he's just amazingly frustrated with this whole thing.
[loud tapping sound]
>>Dan G: Because we're not using one of the default input methods of Android so I can't
actually type my name using the keyboard that's in front of me! So this is, so don't use your
own input method as a general rule, even though it looks really cool. And you can put Comic
Sans on there. Which also I say never use Comic Sans. But that's a different question.
>>Ian: Yeah, did you not get the meme?
[laughter]
>>Dan G: So. So that's one thing. This game is fun. I definitely wanna see some higher
resolution graphics. Everything here is really really blurry.
>>Ian: Yeah, this game wasn't really built for a tablet.
>>Dan G: No.
>>Ian: Also the models are really low res. So you get the feeling that maybe this was
tuned for a much less capable device. And that's fine. But what you might wanna do and
this is just a general game design idea, in general, it's not very difficult to make a
model that's very high polygon count. I mean, the hard part is reducing it to a low polygon
count.
>>Dan G: Yeah.
>>Ian: So if you keep your sources and just every now and then when the state of technology
gets there, put some better models in your game.
>>Dan G: Yeah, actually to be honest, the models I think the car models are really good.
It's the rest of everything else that I want to see.
>>Ian: Well, the car models are attractive. My point is that they're low poly.
>>Dan G: Yeah, and then they don't really look like anything else in the game. And I
think it's cute some of the design elements. But I would love to see some, even if the
backgrounds are gonna look like that it'd be kind of cool if they were a little bit
soft focused that kind of thing.
>>Ian: Now let's just move to camera two really quick. You'll notice that my excessively large
hands still have to kind of reach to get theó
>>Dan G: Yes. Sorry. We're losing HD Mic here.
>>Ian: Yeah, left and right button there. I mean, there, it really is a stretch. Notó
>>Dan G: This is an example of buttons that are not scaled based upon the pixel density
of the screen. They're stretched based upon the screen size. So we have two options here.
One is that we can look at pixel density. That helps a little bit. The other thing is
we actually can have a separate tablet layout that would help even more.
>>Ian: Yeah, remember that Android has some really nice features for putting together
layouts of different screens. And a lot of game developers say "Well that doesn't affect
me at all. Because I'm not making views and I'm not making horizontal layout."
>>Dan G: Yeah, exactly.
>>Ian: But you know what? You don't have to use that for just views or dialog designs
or UI designs. What you can do is put constant values in there. And then pull those out based
on the resource id. And you'll get the different values based on which screen size you're on.
Or you know, whatever you filter it on.
>>Dan G: Absolutely. And again you could do a couple things. One is you could actually
get the pixel density. And you can just apply that to what you're doing in open GL. So you're
getting some thing that's the correct physical size. Which is really important. 'Cause what
you wanna make sure is also if this screen were very very small you'd still want those
things to be very very large. And this things looks like it's just stretching based on the
size of the screen. So that's definitely one thing. Making it automatic is really important
too. So--
>>Ian: Yeah, one thing you can do actually in open GL that's really super cool is define
your UI in meters. Which sounds stupid. But trust me. Don't ever do-- Like I've been on
projects where people like, "Oh, maybe we want a little bit more precision or whatever.
Let's use decimeters or something."
>>Dan G: Exactly.
>>Ian: And it just screws up your physics system. Don't ever use that. But define your
UI in meters and then just have a matrix that scales from meters to screen units.
>>Dan G: and you'll definitely have to test this on a wide variety of physical screen
sizes. But it's much much easier than actually trying to come up with lots and lots of individual
layouts to handle every single setup. Every single combination of pixel and density.
>>Ian: and I'm not gonna lie to you. This isn't easy. There is no simple thing that
you can to.
>>Dan G: Yeah.
>>Ian: Especially if you're dealing with different aspect ratios. But it will be worth it.
>>Dan G: Yeah, I mean, you know, the key thing is, if you're trying to make the HUD look
super attractive and perfect. It's definitely a bunch of work.
>>Ian: Oh yeah. Your artists who, I'll tell you exactly what they're gonna say. They're
gonna be like "We have to be pixel perfect."
>>Dan G: Yeah.
>>Ian: Otherwise it's no good. And that actually is, I said I would never swear on this show.
It is not true.
>>Dan G: [laughs] Very good. That was well, I knew, I heard it coming. And I heard like,
there's, the yeah, exactly.
>>Ian: But lemme tell you why it's not true. So that was extremely true back when we were
working with NTSC TVs, or 480P. or even the, you know, a really high resolution on a big
screen.
>>Dan g: but today everything is bi-linear filtered. At least get tható
>>Ian: Well not only that. But also we're talking about these really really dense displays.
OK, fine. If you're supporting some random old thing that has a very low density display.
You might want to think about this really carefully. Butó
>>Dan G: Here's an example. Let's take a look at Nexus 7.
>>Ian: You just can't tell.
>>Dan G: Nexus 7. All of the icons you see here for every single app are actually being
scaled down.
>>Ian: Mm-hm.
>>Dan G: We did not actually come up with new icons for every single app at this pixel
density. Because, you knowó
>>Ian: Right, this is a 213 dpi. And we scaled from what, 250, or something? 220?
>>Dan G: Yeah. And, 240.
>>Ian: 240, right.
>>Dan G: and so we're scaling down a small fraction. Honestly on this high of a pixel
density display you don't even notice it most of the time. You can actually go and create
icons at 213 dpi if you want to. But we looked at it and we said, it's just not worth it.
And so it think that says a lot about scaling down. If you're even, and again, icons are
the worst thing you can scale typically. Because they're really really information dense.
>>Ian: Except for text.
>>Dan G: Except for text. But even icons, a lot of icons contain text. And a lot of
icons contain line art.
>>Ian: Right.
>>Dan G: And details that you would normally think, oh this shouldn't scale. And the realityó
>>Ian: Don't put that in your iPods.
>>Dan G: Yeah. The reality is, that, and actually let's take one last look at the tablet. I
wanted to look at the icons today.
>>Ian: We are five minutes over time and Producer Reto is like packing up his stuff.
>>Dan G: [laughs]
>>Ian: And ready to go.
>>Dan G: So just one thing out of all these icons. There are two icons that actually I
think are reasonable Android icons. That is the tractor digger icon believe it or not.
And the GTA icon.
>>Ian: Yeah.
>>Dan G: Every single one of those.
>>Ian: Everyone looks like it's been maybe lifted from somebody else's design.
>>Dan G: Yeah, yeah, come on guys. Let's, let's try to make things. Every one of those
looks the same, it's hard to actually differentiate the content of all those other ones. But Tractor
Digger and GTA are really easy to differentiate.
>>Ian: Also I'm just gonna say, Asphalt 7, the game left icon looks like a six. It looks
a lot like a six. It looks like Asphalt 7 6. I'm just saying.
Alright, so let's go ahead andó
Dan G: Next, so yeah. Everyone make sure to vote next week. That's what I wantedó
>>Ian: Yeah, so next week's apps areó
>>Dan G: Sports! Because it's the Olympics. Exactly.
>>Ian: Will have been the Olympics.
>>Dan G: Yes.
>>Ian: So really it's sports games next week. Now if you don't know how this works, you
go to our page at Developers.Google.com slash live. Which is not down in the corner because
this is the wrong title. But it will be. Or you know, the link will always be in the event
that we put on our Google plus page. At that link, developer.Android.com slash plus. Go
there and there is at the very bottom, scroll down to the bottom. Ignore the big black video
thing that doesnít have anything in it because there is no video yet. Ignore that. Scroll
down. Scroll down further.
[Dan laughing]
Then at the bottom of the page there is a Moderator link. It's actually not a link.
It's actually embedded because the people that did it are really cool web people. It's
embedded in the page and you can suggest new apps. Or new games.
>>Dan G: Yes. You can flip to the tablet if you want to see what this looks like.
>>Ian: And you can also vote for the ones that you really like it. Yeah yeah. Exactly.
>>Dan G: So I put this up on our tablet. You can see everything we've looked at. You can
see we now are looking at Sports games. For next time. And right now, we already have
our first suggestions. Some of which don't look like sports games.
>>Ian: Yeah, exactly. So "Hi, when do you plan to review puzzle games." Not next week.
So let's do that somewhere else.
>>Dan G: [laughs] Yes.
>>Ian: This will be marked as "Inappropriate." Let's flag that as "Inappropriate." But thanks.
Because we do love puzzle games. At some point it'll happen. Just stay tuned. Just keep watching
the show.
>>Dan G: Exactly. So stay tuned.
>>Ian: If you donít watch the show every week, how are you gonna know?
>>Dan G: Right.
>>Ian: Exactly.
>>Dan G: So definitely continue to nominate, continue to vote. We're really excited about
a lot of what we saw this week. And I'm looking forward to seeing some great sports games.
>>Ian: Absolutely.
>>Dan G: And golf is a sport. So if you want toó
>>Ian: Mmmmm-kay.
>>Dan G: As an example. So we're thinking about things, I want mostly things that are
a lot of Olympic sports too. If there is something good that we've missed and it doesn't have
to be summer Olympics.
>>Ian: Oh yeah. If there's a great dressage app out there, I wanna see it.
[Dan laughs]
But the other thing that I wanna remind you is our goal here is to help developers make
better games. So we may not always review the most popular games. We may not always
be kind to some games. But our goal is to at least help you understand how a game could
be better. Now sometimes we don't accomplish that because the game is so great to begin
with Radiant Defense. But.
[Dan laughs]
IN general if we're critical of a game it's 'cause we think it's decent but could be better.
And for that reason actually our colleague Reto has renamed his show "The Friday App
Clinic."
>>Dan G: Mmm.
>>Ian: We may do something similar. I don't know. Leave a comment on our G plus page.
>>Dan G: Exactly. Let us know exactly.
>>Ian: OK, we gotta get out of here.
>>Dan G: We're way over time.
>>Ian: We're way over time.
>>Dan G: We started late though. So it's, we can be a little over.
>>Ian: So thanks once again to our producer Reto Meier. Our engineer Daniel Fam. And we
will see you next week.
>>Dan G: Cheers.
[bottles clink]