Another game...defeated by GS performance. UGH!
synthesis
Member Posts: 1,693
Once again...I got 80% complete with a GS game and the GS RAM allowances just won't allow it to happen (and I don't have any physics in this one...just actor counts). Where the heck is the update so I can get all these games I have on the shelf...out on the market.
Otherwise...Corona...here I come.
Otherwise...Corona...here I come.
Comments
EDIT: What about a game where you tap things that fall from the sky> HAHAHA... just kidding... I feel your pain. I'm waiting on the GS update so I can finish my current project, too.
Its obvious that actors are way over bloated in the rendering engine...even once with no physics and limited rules/behaviors.
Lord I hope this next update is even half of what they are promising...as I'm tired of making games that crash out from GS weight.
Then they will fit in memory. I bet they will still upsample to 600x450 without too much loss of quality.
Also, you could try adding a world and reducing the levels per world.
-Gamexcb
The game operates fine on preview and on the web play versions. The problem is...as you get about 10-15 minutes into playing it...the RAM craps out. I don't have an FPS issue...its purely a RAM issue...again!
Once again foiled by GameSalad bloat.
Here is a link to the game:
http://gamesalad.com/game/play/82038
I have about 100 hours invested it this. I am at the end of my rope with GS. If this next release doesn't fix the damn game engine...and doesn't come soon...I'm not sure if I can "waste" anymore time with it.
I've given it 8 months now and have little to show for it...as most of my product sits unfinished waiting for some GS maturity.
(Aside from the know leak culprits like spawn and timers) Are you guys finding that more issues are caused by heavy actors or too many actors on screen?
Until now I have no ram issues, but the game is jerky on the viewer. It didn't crash yet, but I am also not finished yet.
I try to make it as simple as possible. If it will be successful in some miracle way (I don'T count for success, it is just an experiment), I will redo it with a real programmer as a real iphone game for sure.
I think, GS is only good for prototyping an idea fast. If it is any good, you better get a real engine and do it with coding.
Sorry Gendai, but this is the last shot I give GS. Almost a year gone and nothing improved.
By the way I've got a couple codes to give out if willing to star or review.
http://itunes.apple.com/us/app/stick-climber/id392569079?mt=8
this forum post is sort of unsettling as ive purchased a mac with the idea of using GS to make some games/apps.. im not a coder at all but GS does get me some sort of result.. I am going to stay positive here for a bit till the new release any way :-)
I sure hope it works out for all you guys as you've all invested much more then i have at this point being it time or what ever else
Yes, it has its issues. I myself have held off the "actual" work and am only doing art on my current game until the update is released but GS is still in beta and the reason this update is taking so long can only be good for us.
Gendai promised massive performance increases and I think the wait will be well worth it!
Maybe the next update will bring many of your games life. Isn't it supposed to be more optimized? Although, Corona currently beats GameSalad with optimization. If that is your main concern, I can't really argue against it. One of the things Corona seems to do well is optimization. I don't know for sure... I saw some lines of code and went back to GameSalad.
Did you read my Textbook? There are lots of optimization tricks in there. It seems odd that you'd hit the same problem again.
However, I honestly believe that Bedazzled should not even have been attempted with GameSalad.
To pull something like this off - or any type of painting program - you really need to be able to manipulate the bitmap data.
The general logic that you have implemented in Bedazzled is fine. However, spawning all those actors is not the way to go.
All of those actors you have spawned are 'alive'. You currently aren't, but you 'could' set all of those actors to be draggable. All of those actors have weight to them. The have the capability of all the touch actions, the accelerometer behaviors, plus the 35 other behaviors.
What you really need to do is to be able to simply edit the pixels of a single bitmap - like you would do in Photoshop. Changing the color of a pixel does not increase the RAM usage.
But this of course is not currently possible in GameSalad.
Another example is a "bullet hell" shooter. All of those bullets flying around are not actors, or objects. They are simply pixels being drawn into a large 'bullets' bitmap. That bitmap gets redrawn every frame. Collision is done by comparing the bitmap of the player with the bitmap of the bullets.
Hopefully the next update will improve performance and allow us to add even more actors to the stage. However, there will still be an upper limit of how many actors you can add to the Scene.
QS
Edit: Oh, I meant to add that you can still make perfectly playable games (not taptard stuff) with GameSalad. I think that FMG and Tshirt, amongst others, have proved this.
Dr. Sam Beckett never returned home...
Twitter: https://twitter.com/Quantum_Sheep
Web: https://quantumsheep.itch.io