• Listen to a special audio message from Bill Roper to the Hive Workshop community (Bill is a former Vice President of Blizzard Entertainment, Producer, Designer, Musician, Voice Actor) 🔗Click here to hear his message!
  • Read Evilhog's interview with Gregory Alper, the original composer of the music for WarCraft: Orcs & Humans 🔗Click here to read the full interview.

3D Flying?

Status
Not open for further replies.
Level 11
Joined
Oct 9, 2015
Messages
721
Would it be possible to make that when the player clicks and moves the cursor down the flying unit goes down, and when moves up the unit goes up? Is there such a mouse detection system? (I already have an Up,Down, Left, Right ground movment system that would work in the directional movment, I need one that makes the unit increase and decrease flying height)
 

Dr Super Good

Spell Reviewer
Level 64
Joined
Jan 18, 2005
Messages
27,287
Or move to StarCraft II where full keyboard detection is possible. It still is laggy as a result of physical limitations but it far surpasses not having it like WC3 has.

WC3 is mostly a 2D game. Although it looks 3D and does have a concept of height, most mechanics operate in 2D with respect to the ground. You need to think carefully what a 3D flying system will achieve in a game such as WC3. Even Empire Earth, a RTS from 2000 which had aircraft, tried to keep them working in a 2D way.
 

Dr Super Good

Spell Reviewer
Level 64
Joined
Jan 18, 2005
Messages
27,287
Games from 2003 and even 2006 (and probably more recent as well) have their stuff in 2d calculations and sometimes even not in 3d space.
There is no 2D acceleration though. Hardware only offers 3D acceleration of which 2D is logically incorporated. Hence Direct2D is implemented as a wrapper for Direct3D.

I think they dropped 2D acceleration around 2000. Consoles like the Gameboy Advance were likely the last solely 2D accelerated consoles (even then they had some 3D features).
 

pyf

pyf

Level 32
Joined
Mar 21, 2016
Messages
2,985
There is no 2D acceleration though. Hardware only offers 3D acceleration of which 2D is logically incorporated. Hence Direct2D is implemented as a wrapper for Direct3D. I think they dropped 2D acceleration around 2000.

DirectX Control Panel for DX7 demonstrates the link between DirectDraw and D3D. Never disable DirectDraw hardware acceleration, you would lose Direct3D hardware acceleration in the process.

For so-called display "2D acceleration" in Windows, there is Blt, Transparency and Colorfill capabilities, plus the overlay stuff for video. Blt was first introduced in the optional WinG for Win3.1x / Win9x, if I remember correctly. Also, hardware mouse cursor was counted as 2D acceleration.

Attachment shows only a subset of the Intel i915GM chipset DirectDraw capabilities. The tool does not allow to export all the data to a text file.

Being on XP, I can not benefit from any Direct2D acceleration (which the i915GM chipset could never provide anyway).

In 1995/2000, hardware capabilities varied from video card to video card. For example, Matrox Millenium/Mystique cards had much more features in ROPS. S3 Virge DX had *a few* overlay capabilities. and so on...

The capabilities shown as attachment are driver-dependent. With Windows 2000/XP, many old video cards became *obsoleted*, because of poor W2K/XP drivers which did not *care* to use the old hardware's possibilities to the max (thinking of you, Matrox !)

Between MS-Dos and W95/DirectX, I heard some video card manufacturers had their own *proprietary* set of acceleration functions. It did not last long, and everyone ended up using DirectX with DX5.

In the MS-Dos days, there was display acceleration through Vesa video modes. But since Microsoft did not care about full Vesa 2.0 support in Win95, it died. VBE/AF and Scitech Nucleus came too late. Nucleus was an attempt to create some kind of DirectX for MS-Dos.
 

Attachments

  • i915GM - DX7 Control Panel.png
    i915GM - DX7 Control Panel.png
    87 KB · Views: 121
Level 24
Joined
Aug 1, 2013
Messages
4,658
That, plus that you won't really notice the difference.

My missile system can chase a moving point and can have a maximum turn speed defined.
However, if it has to change the maximum turn speed in horizontal movement and the maximum turn speed in vertical movement, then it turns slightly faster, but the additional calculation is pretty intense and it would only make an up to 2% difference in a case that would happen once in 5,000 times... aka not worth.

Some thing that are "not worth" are really useful... that is one of the main reasons that I write everything myself :D
 

pyf

pyf

Level 32
Joined
Mar 21, 2016
Messages
2,985
CPU efficiency has to be taken into account too.

Back in ye olde days, I perfectly remember testing Moto Racer on both an Intel PII, and an AMD K-6 of equivalent *real world* power (AMD was overrating his processors back then, it was the Mhz Wars era). On AMD hardware, the trajectory of your bike when turning/sliding was very different from the trajectory on Intel hardware. Thus, the game was much harder to play on AMD processors. yay !

A few years ago, I also remember testing PhysX on some AMD Dual Core processor, via 3DMark Vantage and Nvidia GeForce 8500 GT hardware, on Vista 32 bits. Results were strange (slow calculations, general sluggishness).

Some common CPU extensions (MMX, SSE and the likes) work slower on some CPUs (did I say AMD ?).

Optimization / code efficiency has to be thought out at every level. It includes CPU, GPU and compilers at the very least. phew !
 
Status
Not open for further replies.
Top