OK OUYA... I'm hoping to get a lot of support behind me for this.
Don't do a yearly refresh!
Seriously, don't release an OUYA 2 next year. Wait. Give it time. Give developers more time to develop for the OUYA 1. Work out the flaws.
Wait just one extra year. Then release OUYA 2 with Tegra 5. Because at that point, you'll have learned from your mistakes, worked out a better OS... and then when Tegra 5 is available you release OUYA 2 which will objectively outperform the Xbox 360 and PS3 just in raw GFLOPS.
Don't believe me? Tegra 5 so far has been announced to peak at 400 GFLOPS when clocked at 1GHZ. The Xbox 360 tops out at just under 200, and PS3 at around 218. So objectively it will outperform them by about twice as much.
If you don't release the next one a year from now with Tegra 4, it will actually have a bigger impact. For one, a lot of developers and gamers simply won't be keeping up with a yearly refresh (I'm sorry to say I certainly won't). For another, if the power jumps from about Wii-level, to suddenly curbstomping the PS3, I think it would actually get the attention of a lot of gaming media.
Thoughts?
You didn't remember the plot of the Doctor Who movie because there was none; Just a bunch of plot holes strung together.
Killa_Maaki
Posts: 504Member
Comments
Additionally, yeah this is kind of about power but on the other hand I think a twice-yearly refresh makes more sense from a consumer standpoint. I don't want to buy a new OUYA every year, and most likely won't. Either I'll be left behind if the other developers all switch, or the OUYA 2 won't get any traction if all the developers stay on OUYA 1.
:)
Seriously though Ouya can't survive with only Ouya 1 for longer than a year (in my humble opinion).
OUYA have failed to communicate that, so most people have taken it at face value / first impressions. Even if there is a big update a month later, and the whole thing plays and feels completely different, that doesn't matter, because the consumer has already made a judgement
People aren't used to the idea that they might put something in a cupboard, then when they pull it out 6 months later and it's magically got a whole lot better. Trying to persuade them to take it out and give the OUYA a second chance will be incredibly difficult. "Well.. it was a bit meh last time, and I can't see why that will have changed"
OUYA 2 will be perceived as different enough from OUYA 1 that people will be prepared to take a second look. OUYA will have already (hopefully) made the fixes to OUYA 1 and more, but OUYA 2 is the opportunity to advertise that to a sceptical public who associate OUYA 1 with all the flaws
The hardware is actually irrelevant so long as people perceive that it has changed enough that OUYA 2 isn't just an OUYA 1 painted a different colour.
A good example of all this is with the OUYA controller, which I explain here how and why it needs to be seen to be radically overhauled in time for OUYA 2
As for timing, OUYA need to do this as quickly as they possibly can, but at the same time, need to have actually solved all the issues/flaws (real and perceived) with the current OUYA. Once PS4 and Xbox One hit in November, OUYA aren't going to get any sort of press or coverage, and they don't have enough time between now and then anyway to fix things.
OUYA 2 has to come long enough after those systems are released to not get drowned out by them. Early summer 2014 is probably the earliest they can do that.
Leave it any later, and OUYA will probably have lost too much momentum (and money) in the meantime to recover. There is also the thing with the yearly Tegra release cycle where in the second half of 2014, Tegra 5 will already be on the market in many devices, but not have come down in price enough that it can fit in the OUYA for $99.
Also the power of the hardware is not really important if OUYA can pitch the games right. Just like the Wii was underpowered, the OUYA can still appeal to a casual audience, kids and as a party games console. (So long as they only imply they are like the Wii, and don't actually explicitly compare themselves to it, as people will rip into how arrogant OUYA are to compare themselves to the biggest selling console of all time).
In terms of dev relations, rebooting the image of OUYA in the form of an OUYA 2 next summer, will give developers more confidence in the system (remember, many are themselves consumers and will be swayed by what the industry publications and their fans are saying)
Taking a clear direction about being the new party games console, will help developers understand who exactly the OUYA is for, because they sure as hell don't know at the moment. That's something OUYA can start doing now, and can build up to be a really strong image in the lead up to OUYA 2.
Both those above things I think will be far more important to attracting developers than some worries about Tegra 4 vs Tegra 5, especially considering Android ensures backward compatibility, and high graphic fidelity games that take advantage of that power will be in the minority anyway
Website
Website
I'll probably switch to my fully deferred renderer for the Tegra4, my current OUYA engine uses an FBO as a very small G buffer combined with a standard forward renderer for lighting/shadows.
My only real concern with the Tegra4 is that it still seems to be light on overall memory bandwidth and hence I am expecting similar fill rate issues as the Tegra3. Obviously instead of rendering to 1920*1080 we'll be able to reduce the resolution to overcome this potential issue as we do now, just not as drastically.
Come to think of it, at low resolution it should be fast enough. I am already running my game at Wii resolution to avoid fillrate issues (I plan to have plenty of particle effects, so fillrate is important). I suppose Tegra 4 should handle deferred rendering just fine at 480p, right?
When I first received my OUYA in early January, the first example I had running on it, (after San-Angeles), was a deferred renderer of an ocean scene that I developed at work. It ran at around 1.5 fps compared to 150+ on the desktop, but it ran fine, (after suitable modification to the shaders, ie removing tessellation, "OpenGL ES'ing" them etc).