Initial image position bogus for a few frames

Solar2D General Questions/Discussion
#1

I have a full screen background and a smaller image that I move from left to right on the screen. I start the small image at -200 and for the a brief period of time I see the image drawn on the RIGHT side of the screen. It should be halfway off the screen to the left (since the image is 400 pixels in width).

This is easy to replicate by just setting the x value for the image to -200 and leaving it (rather than moving it to the right). For a few frames you see it drawn on the right side of the screen, then it starts getting drawn on the left side where it belongs. NOTHING in the code changes the position from -200 to any other value.

If you need a small test case I can generate one, but the above description is accurate and easy to reproduce. [import]uid: 16734 topic_id: 17492 reply_id: 317492[/import]

#2

Mac or Windows? What version of Corona are you currently using?

Peach :slight_smile: [import]uid: 52491 topic_id: 17492 reply_id: 66417[/import]

#3

@peach, should that really matter? does the core perform differently between the two platforms?

cheers,

?:slight_smile: [import]uid: 3826 topic_id: 17492 reply_id: 66431[/import]

#4

Jayant,

There are always platform specific bugs when developing cross-platform apps or tools. I’m sure Peach is just trying to find the most expeditious way of reproducing the bug, and she is probably going to check the build notes since 591 to see if this is a known issue that has already been fixed. At least I hope it’s been fixed. [import]uid: 16734 topic_id: 17492 reply_id: 66509[/import]

#5

Peach,

Stable Build 591. This happens under Windows in the Simulator AND on the iPhone. Probably happens on the Android too, but I need to retest with my smaller test case to be sure.

I do not use the daily builds since we are close to release and the daily builds are not stable.

Also, I am using Director 1.4 [import]uid: 16734 topic_id: 17492 reply_id: 66478[/import]

#6

@Jayant - When someone has an issue these are the questions that we ask first. When we’re trying to look into a bug we want as much information as possible for testing purposes and to try and isolate, then remedy, the problem.

In any case, please don’t post if you aren’t contributing anything of value to the thread.

@Ken - Thank you for the info. I’m going to take a look at replicating this shortly as well as check whether or not it has been remedied since the last stable release.

In the meantime could you please file a bug report and post the case number here?

Thanks,

Peach :slight_smile: [import]uid: 52491 topic_id: 17492 reply_id: 66611[/import]

#7

@Peach, Do I see this as “It’s my playground, keep off?” It’s the first time you are talking about *not contributing*, interesting considering the amount of posts that we have on the forums and the content.

Secondly, I have asked a very valid question, I do understand that you want to know from the user to be able to reproduce the issue, but my question is also valid, Does the Core simulator differ from platform to platform. So if I test an app on a Mac and then move to a Windows machine, will I get different results? I thought that it worked almost the same. I am concerned about this situation, so instead of an answer I find you trying to *Police* the forum…

Ansca Playground… Ansca Staff… keep your playground, The reason a lot of people help here is this is open, but if the forums show a hint of *My Playground* theme, well… you can play there alone… I’m out…

cheers,

?:slight_smile: [import]uid: 3826 topic_id: 17492 reply_id: 66633[/import]

#8

@Peach,

Bug #3086

Case #9442

Example code demonstrating the bug was included. [import]uid: 16734 topic_id: 17492 reply_id: 66740[/import]

#9

Thanks Ken, most appreciated - you will be able to keep track of any progress via fogbugz if you wish.

Peach :slight_smile: [import]uid: 52491 topic_id: 17492 reply_id: 66850[/import]

#10

@Jayant, to respond to why it is useful to ask whether the platforms are behaving differently, it is because we have both common shared code and platform specific code. The problem with cross-platform code is that we have to build up from platform specific code first.

We strive to make Corona behave the same on all platforms. But it doesn’t magically happen by itself. We must make that magic happen by writing abstraction layers to make it seem like all the platforms behave the same way.

For example, platforms can’t even agree whether the y-position starts at the bottom and increases upwards or if y starts at the top and increases downward. They also can’t agree on the size of a point. (e.g. iPhone retina: 1 point=2 pixels). Even between Mac and iOS, some APIs like the movie players are very different.

Once we build up layers of abstraction, we can start writing shared code that is the same across the entire code base. One example is our audio engine. At the higher levels (ALmixer and above), our code is the same across all platforms. But lower layers like the platform specific sound decoders (which are optimized for battery life and performance) and the specific OpenAL implementations we use are very different. (Notice that Apple itself has different bugs in their OpenAL frameworks between Mac and iOS.)

So when we ask if the bug is common across all our platforms or specific to a particular platform, it helps us track down where the problem might be.

And this isn’t unique to Corona. All cross-platform frameworks have the same challenges. This is just how its done. And each cross-platform framework must make various trade-offs. And there are numerous trade offs. For example, since not all platforms give the same features and level of control, do we make simpler APIs that are common to all, or do we make complex APIs that don’t work consistently for all sub-features? Another example is performance. Do we try to support everything under the sun including a classic Commodore 64, or do we target more specifically. All these trade offs have an impact on the end API because they make things simpler or more complex.

[import]uid: 7563 topic_id: 17492 reply_id: 66856[/import]

#11

@Eric,

There was no need for Peach to be rude for asking “if that matters and does the functionality differ between platforms”

I understand that this is Ansca’s sandbox and playground, but

In any case, please don't post if you aren't contributing anything of value to the thread.

is not exactly a polite or acceptable way for Ansca staff to speak to a customer in any circumstance or is this the new Customer Service Charter from Ansca. This is a totally unacceptable behaviour at least officially from Ansca Staff.

@Carlos contemplate this, what kind of service/Official behaviour are you approving. Users using that is *not* your responsibility, but your staff actions, words *IS* Ansca’s responsibility.

Asking that question has evoked this response from you which is a very good addition which would otherwise never have been known unless this was asked.

Specially that there is a possibility that even a Commodore64 might be supported at some point in the future, could you look at ZX Spectrums, I was not much of a C=64 person.

cheers,

?:slight_smile: [import]uid: 3826 topic_id: 17492 reply_id: 66860[/import]

#12

I don’t see any way to add more info to the bug.

One other thing I noticed is that the load bar image that displays at the wrong position also displays on top of the mask image during that brief time, even though it is added to the display group prior to the mask image. [import]uid: 16734 topic_id: 17492 reply_id: 67025[/import]

#13

If you were sent an email from the bug, you can try replying to it and sometimes the response ends up in the bug report.
[import]uid: 7563 topic_id: 17492 reply_id: 67104[/import]