So I have a question for folks who have put out Arm6 only binaries to stay in under the 20mg mobile download. Did it really help with your sales?
In general app sizes are getting larger with variables such as multi-version support, higher res map limits for iPad and iPhone4. With interactive storybooks and the like where there’s lots of images, and no realistic way to squeeze in under 20meg. But do you think ‘size’ really matters when it comes sales? I guess for the commuter impulse sales for sure, but has anyone noticed a tangible difference in unit sales between fat and skinny binaries?
If you ask the majority of people will tell you it makes no difference.
We however have a different experience…
Our first unity game was around 14 meg on the app store so was unable to be downloaded over the air (at that time the limit was 10 meg). It had a tiny bit of success with being featured by apple but after the dust settled it dropped out of mainstream contention on the USA store and had found a comfy little niche on the Australian store at around 35th in Sports.
Then one day I went to check the Australian charts as I did most days and to my surprise it had shot up to 10th spot overnight. I couldn’t believe my eyes, first thing I did was scour the internet for some kind of massive news article, nothing, then I searched for some major surfing event (it was after all a surfing game) nothing… Completely dumbfounded I came to these forums and someone had announced that Apple had lifted the OTA download limit to 20 meg overnight which in theory would put our game within OTA dl limits for the first time. I confirmed by re-downloading over 3G, yep all good.
The app then sat around 10th → 15th in sports for months afterwards.
Now there are several factors that need to be considered here.
Australia has pretty limited / terrible WIFI support in public areas.
The Australian iTunes store is a lot smaller than USA so a lift of say 20 units a day could have a massive impact on chart ranking.
(wildest claim of all) surfers may be the kinds of people who don’t bother hooking up to itunes later on once they realise they cannot download their app over wifi.
Either way, unless your app has no hope at all of being within 20 meg, I would always suggest you take the 20 meg limit seriously. People are lazy and spontaneous, and that even tho it may only be a small % of users who lose the impulsive urge after being denied OTA download it’s still a person who you lost a sale to and as such should be treated with respect.
I’ve personally been embarrassed when at a function and someone asks what the name of the game we made was, I told them and they instantly went to download it, was denied because of the OTA limit, and you could tell instantly there was no way in hell they would remember to get it when they got home at 10pm that night after several glasses of wine.
Thanks for the detailed info, Traction… very interesting. And Good suggestion Johnny, even if the full app is over, trying to squeeze out a <20mg lite version would appease the impulse buyer. Least the ones with their iPhone in one hand and a glass of wine in the other… 8)
The problem is you may never really get conclusive evidence… but that doesn’t mean it doesn’t have an effect.
Moving millions of units isn’t really evidence to the contrary, no one is saying you can’t make a lot of sales of an app over 20MB.
If we are going down the path of anecdotal evidence 9 of the top 10 selling apps are under the 20MB limit at the moment, the only exception being Madden which obviously has a lot of marketing behind it. 7 of the top 10 grossing apps are under the limit.
I would say be guided by practicality… if your app is close to making the limit when built as armv6 only then you might as well have a go at cutting it down.
If you do this but also want ipad native resolution then release a separate armv7 HD version.
Also, Fat binaries chart separately… If your fat is #8 in iPhone list, it can still be #150 in iPad. Thus, it makes no sense to have fat binaries (given the additional size of iPad textures) at all since it doesn’t help your chart position at all (and making money is about being in the charts).
Very interesting. We actually had our first universal binary approved today. I was on the fence about doing it from the begining. Having separate binaries allows for separate price points and as JTB points out, a fatty buys you nothing in collective ranking.
I guess the impetus was first, to see how a universal at a lower tier price than a stand-alone iPad would have been priced, might affect sales volume. And secondly, we’ve read quite a lot of negative feedback from customers who feel ripped off having to purchase a game twice for their iPad and then the iPhone. We’ll see how it goes, but may be our last foray into universals.
Damn, Disney releases each language version of their Toy Story book as a separate app…talk about milking it.
well be happy they don’t sell it through the ibookstore, otherwise you would pay additionally for each device
as for ripped off: I think that also depends a lot of the game
cheap ports that are basically just the same game in higher res to me also feel like ripoff and I just don’t buy them, as I have not the least bit of understanding to pay twice for the same thing.
now if it was opted and truely ported and brought to the ipad, its a whole different story and we can discuss about it (tower madness as an example)
iPad’s seem to be a very different demographic though.
If you have put genuine effort into the HD port, it should cost more. Also, the market is tiny in comparison, so its a hard sell unless you have something that’s truly going to be remarkable to bother specialising your app to it.
Agreed.
But just not scaling down the graphics you had in much higher resolution likely anyway does not go through as genuine effort.
An example of such a thing where not even better assets are present is Command Conquer thats just the same thing + the additional 5 mission mini campaign but sold at a premium price.
Thats something I would agree if the few games that tend to show up wouldn’t be blatant scaleups
But on the other hand, games have a hard standing on the ipad anyway, its a productivity platform primarily, not a gaming one.
The featured, staff picks and top lists show that very well.
but to come back to the topic: for a casual targeted game, the 20mb are very important.
the less “lightweight” it is, the less it is important as the content is rated higher in rpg and alike than “going below 20mb”
but if you are really worried, put the game with the first chapter online and offer all others for download through asset bundles for example
It has become more universal formats with variables such as support for multi-version, plus maps res limits IPAD and iPhone4. With interactive stories and others, where there are lots of pictures, and no realistic way to squeeze under 20meg.
