Hi there
I’m not sure but i think there are much differences in physics on different devices and/or corona simulator. Can anyone confirm that?
For example if i let fell down an object and collide it with another… if i get the forces from the collision they are much different from the simulator to my HTC desire… Why and what can i do about this? (One Idea maybe to calibrate it…) how you make this?
Matt [import]uid: 44329 topic_id: 16856 reply_id: 316856[/import]
I see you are a pro developer, which means you have ios and android licenses.
Have you run the same test vs an ios device (iphone 3gs, 4, 4s, Ipad1/2)? Just wondering as you are talking SIM vs device, but only mention android.
ng [import]uid: 61600 topic_id: 16856 reply_id: 63120[/import]
I tried it until now only SIM vs Android… is it there a known issue? Has someone tested it on several android devices? Are there also differences between different Android devices? Is is a solution to “calibrate” it and store as a factor? [import]uid: 44329 topic_id: 16856 reply_id: 63168[/import]
Different Android devices are configured in as many ways as there are grains of sand at the beach.
I tested the drag boxes example on Ipad2 / Iphone 4s. I then gave the code to a buddy who has a pro license who did it on both. Tested with a Droid X and a Samsung fascinate.
He said there wasn’t a difference, but did notice the Droid X2 took longer to load and wasn’t as responsive as the GalaxyS 2 when touching.He said the falling of objects was identical.
Not sure what to tell you, but Android unlike Apple has MANY configurations, some have 500mhz cpu’s while some have 1.2 ghz CPU’s some have 1 core, some have 2 cores. Some have 400mhz CPU…etc etc etc.
[import]uid: 61600 topic_id: 16856 reply_id: 63171[/import]
But what you think about the “calibration”… for the force, would that be a possibility? [import]uid: 44329 topic_id: 16856 reply_id: 63172[/import]
Any other solutions? [import]uid: 44329 topic_id: 16856 reply_id: 64018[/import]
How do the android indies solve this? [import]uid: 44329 topic_id: 16856 reply_id: 65260[/import]
We are just doing some early testing of Corona 2D physics and see some rather disturbing results.
Different devices run the physics at different speeds to each other.
There seems to be 2 speeds (roughly) in play here:
Windows simulator = Galaxy S2 (faster simulation)
Nexus One = HTC EVO 4G (slower simulation)
To put this into the context of the falling objects. They fell at very different rates on these 2 groups of devices. Sort of looks like double the physics speed.
How is it possible to run a physics game consistently with Corona’s frame based implementation?
Actually I can think of some ways to get there, but they are not ideal. [import]uid: 28534 topic_id: 16856 reply_id: 66847[/import]
Hey guys - are you actually seeing differences on different devices (phones and tablets), or only between the simulator and devices?
Please let me know - any details help.
Peach [import]uid: 52491 topic_id: 16856 reply_id: 76860[/import]
Hi Peach,
With our game and tests we see roughly two different speeds as detailed in my post above.
So…
Yes the simulator is different to some devices
and
Yes there are differences between devices too.
The tablets tested (Xoom and Galaxy 10.1) both ran the faster simulation. So essentially like the Galaxy S2 and Windows Simulator.
Our Game now in the market, while runs similar on all devices it is not accurate and is only doing so by us messing with things based on the device speed. Not ideal nor a very good approach.
You can see it here:
https://market.android.com/details?id=com.cmglabs.Flicky
[import]uid: 28534 topic_id: 16856 reply_id: 77239[/import]
Hey Ben, can you use this form please?
http://developer.anscamobile.com/content/bug-submission
If you could post the bug # here I can bring it up in our next meeting as well.
Peach [import]uid: 52491 topic_id: 16856 reply_id: 77261[/import]
No Problem Peach. The bug has been reported…
(Case 11306) Physics Simulation Speed Errors
Thanks,
Ben. [import]uid: 28534 topic_id: 16856 reply_id: 77418[/import]
Thank you [import]uid: 52491 topic_id: 16856 reply_id: 77490[/import]
Using the new physics.setTimeStep( 0 ) I was able to get this problem to be much less of an issue. Probably for our purposes it is good enough.
Ben [import]uid: 28534 topic_id: 16856 reply_id: 78004[/import]