3D Rad - Free 3D game maker - Forum

This forum is now archived!

This forum is locked, and is a read-only version. A new community-ran forum can be found at classdev.net

News:

The 3DRad community can be found at classdev.net.

Pages: 1 2 3 [4] 5 6

Author Topic: Sailing/exploration type thing - ["UPDATE" 13/04/14]  (Read 2844 times)

« Reply #45 on: December 09, 2013, 08:59:15 AM »
the screenies look vert inviting... but again, like in the last demo, i could not get it to run on this antiquated machine...

something is going on, but it is going on at a slide show frame rate...  i can't see anything except an occasional flash of the deck...

--Mike
« Reply #46 on: December 09, 2013, 09:52:59 AM »
How about some video footage of it? For those whos pc is too old..? Looks rrally good hope to have time to try it soon :)
« Reply #47 on: December 09, 2013, 02:38:08 PM »
The framerate on the boat was 55-60. On foot, on the other hand, the framerate dropped to about 30-40. 40 fps is noticeable and slightly annoying.

Lower resolution helped a few frames per second.

It does start.

Intel Core i5
ATI Radeon HD 5750 512 mb
8 gb RAM
Rocket Rumble, a 3D Rad puzzle game:
http://www.3drad.com/forum/index.php?topic=9896.0
« Reply #48 on: December 09, 2013, 02:58:59 PM »
I gave it a try, but I forgot to look at the instructions u put up and ended up splashing around the water as the only key that did anything was F. ;D

The Frames did slow even though i have:
AMDFX8350 (8 Core)
nVidea Gefore GTX 760 Graphics
16gb Ram

So u may need to optimise it as there is not much to do yet and hopefully more will come from this which means more lag.
New 3DFoundry:3DFoundry.tk
« Reply #49 on: December 09, 2013, 06:12:23 PM »
Hmm, okay. I don't know what the problem is here.

I'm running it fine on my computer, which is nowhere near as powerful as PlayDasher's. I also ran the compiled project fine on another modest laptop with integrated graphics (though it had to be windowed to even work).

I guess one thing to mention is that running the game in 800x600 resolution is much slower for me than 640x480 as well as 1024x768. So if people are having trouble, trying out a few different resolutions may prove useful.

Also, since I forgot to mention it in my last post: the game's map only works properly in 4:3 screen resolutions.

I will continue with my attempts to optimise it. However, I really don't know why its working so poorly on particular systems.

Also, sorry to those who are unable to play this for whatever reason. I'm trying my best to get things working.

Cheers guys,

George.
 
« Reply #50 on: December 09, 2013, 07:31:25 PM »
Well I've tried it.  I like the sound and environment.  I figured out that there was an enemy I had to blast with my cannons and how to leave the ship and run around the island.

Like the others I had framerate issues even at 800x600. 

From a design standpoint:  It sure would be nice if I could actually see what's in front of me.  The sails pretty much completely obscure your view forward.

My best guess is the water.  It looks really good but you might have a lot of shaders and/or particle effects going on (there's no way that little of geometry would cause such a slow framerate. 

This could be a really cool game if you get the performance issues worked out.  Great work otherwise!
« Reply #51 on: December 09, 2013, 07:33:23 PM »
Oh and my system specs are:

i5 core duo
6gb ram
ATI Raedon HD 5450 1ghz

Sorry for double post but SOMEBODY reset my posting privileges so they have to be approved...
« Reply #52 on: December 10, 2013, 11:42:02 AM »
Looks good! Going to check out!
« Reply #53 on: December 15, 2013, 02:44:24 AM »
Hi Guys.

So, this is pretty much the same deal as before. I just want to find out how this runs on other people's systems. I've taken the last "demo" and made a few changes which should speed things up for a few people.

Once I can get this working on peoples systems, I'll resume normal updates where I'll focus on things like game play.

Controls are included in the project. Just hold down left-control to view them.

http://www.mediafire.com/download/vlgbxzch6rebwq0/SailboatTest-15-12-2013_20131215191534.zip

Thanks to everyone for their patience,

George.
« Last Edit: December 15, 2013, 02:52:16 AM by georgio123 »
« Reply #54 on: December 15, 2013, 03:55:16 AM »
Full HD, solid 60 fps.
Rocket Rumble, a 3D Rad puzzle game:
http://www.3drad.com/forum/index.php?topic=9896.0
« Reply #55 on: December 15, 2013, 04:13:34 AM »
Quote
Full HD, solid 60 fps.

Excellent. It looks like I may be getting somewhere  ;D

soulfly

« Reply #56 on: December 15, 2013, 04:18:16 AM »
0.1 FPS here...  ::)
« Reply #57 on: December 15, 2013, 08:43:05 AM »
i managed to get a single frame of a scene where the boat is in a inlet near the shore, and the bow of the boat is pointing towards a part of the terrain bounded by two hills... 

this single frame never changed... but it did finally give me a look at the game...

i hate to burst your bubble, but something is verrrrrry wrong here... there is nothing i see in this scene that should slow things down soooooo much on even this old setup i have here...

take a look at this video G...  http://www.youtube.com/watch?v=ZN7ukbcB264

the scene in the video is built to stress the capabilities of this machine... it has a large terrrain (20000 x 2000)... the terrain is shader lit... there are 3 different sets of particles making up the clouds, each with large semi transparent billboards... the infinite ocean is running with 9  imposters of an animated mesh... and the chopper has at least 2 concurrent running sounds along with the ambient ocean sounds...

and it still runs in this antiquated system of mine... so, with the scene i saw, yours should run fine...

lets see if we can find out why it's not...

1 - the night day sequence... how are you doing that again...

2 - are you using any full screen post processing shaders... glow (hdr), blur, stuff like that...

those are the first things that come to mind...

--Mike
« Last Edit: December 15, 2013, 09:06:00 AM by Mike Hense »
« Reply #58 on: December 15, 2013, 04:47:45 PM »
Quote
i managed to get a single frame of a scene where the boat is in a inlet near the shore, and the bow of the boat is pointing towards a part of the terrain bounded by two hills... 

this single frame never changed... but it did finally give me a look at the game...

i hate to burst your bubble, but something is verrrrrry wrong here... there is nothing i see in this scene that should slow things down soooooo much on even this old setup i have here...

take a look at this video G...  http://www.youtube.com/watch?v=ZN7ukbcB264

the scene in the video is built to stress the capabilities of this machine... it has a large terrrain (20000 x 2000)... the terrain is shader lit... there are 3 different sets of particles making up the clouds, each with large semi transparent billboards... the infinite ocean is running with 9  imposters of an animated mesh... and the chopper has at least 2 concurrent running sounds along with the ambient ocean sounds...

and it still runs in this antiquated system of mine... so, with the scene i saw, yours should run fine...

lets see if we can find out why it's not...

1 - the night day sequence... how are you doing that again...

2 - are you using any full screen post processing shaders... glow (hdr), blur, stuff like that...

those are the first things that come to mind...

--Mike

Thanks for taking another look at this Mike.

1 - The day night sequence is made up of an object which rotates constantly around the selected camera. That object basically functions like the skinmesh from the lens flare demo. The second, and perhaps more processor intensive aspect of the day/night cycle is the sunlight object, which rotates back and forward constantly during the day and night.

2 - I'm not using any full screen post processing shaders.

Some other things I thought about are these:

- Bumpmapping/terrain shader on the Island skinmesh (though I suppose you were doing a similar thing in the video you posted)

- for the on-foot character, there is a short scanner object which is constantly run to see if there is an intractable object in front of it. (I know that running too many scanners, or running one too often/in too many locations can crash my system. However, I tried taking this out to see if it sped things up, but I couldn't notice anything different)

- large numbers of imposter skinmeshes for the trees and food (if an object is outside of the viewdistance, would there be anything to gain, performance-wise, from an iObjectImposterHide command?)

- the island being reflected in the water (In the demo I posted most recently, the refresh rate was left at 30 since I was initially using fbf animated water and forgot to change it. However, now that I'm using a flat water plane, and only reflecting static objects, I've set that value much lower)

I can't really think of much else.

I've managed to get this running smoothly on a fairly modest laptop, so I'm really confused as to what is causing difficulties on other systems. Hopefully I'll work out what the problem is before too long!

Cheers,

George
« Reply #59 on: December 15, 2013, 06:50:22 PM »
well... sounds like you've looked at every possibility... i wouldn't suggest making make any changes as everything seems to be working ok on higher spec machines...

when you get further down the path with this maybe you could make it an option for the reflections and the day/night sequence...  maybe...   ;D

i'll settle for video clips till i get one of my higher spec machines fixed  :'(


--Mike

Pages: 1 2 3 [4] 5 6