|
The last time I had to do that was back when the game was new and I was a junior in high school over summer break.
My biggest fear is that the game will be always online without the Open Battle.net and Direct TCP/IP multiplayer options so you HAVE to go through Battle.net servers to play with friends or even single player. |
|
How likely is it that Blizzard will offer “gear packs” or some other bullshit pay to win system with this? |
|
They really think I’m going to pre-order a digital re-release of a 21 year old game after what they did to Warcraft 3 🤣 |
|
Weird that Blizzard said that my banger of a PC is too weak to host a Diablo 3 session (so the game has to be hosted online) but the console versions all can host games offline. Really makes you think. |
|
Currently installing Diablo II from my release day CDs to scratch that itch. Going to mod it with https://github.com/bolrog/d2dx |
|
It’s wonderful. I have a stock Diablo 2 installation and then a separate directory containing the following mods:
Any of these would cause a ban on Battle.net, so having them in a separate directory means the game gets patched at runtime so I can hop on Battle.net by using the stock, untouched installation without mods easily. Might put up a PvPGN server for some ladder fun with friends. ❧ Edited by MasterOfMagic at 2021-09-27 14:28:042021-09-27 14:28 |
|
I am not entirely sure how it works, but here’s what I gather from my observations with d2dx:
Because it appears the network prediction was built with the intent to not interfere with game logic this works. The problem is that it causes the animations to appear different, and different enough to me that it is distracting, so I disabled it. The renderer still refreshes at a higher frame rate but the game update rate is unchanged. ❧ Edited by MasterOfMagic at 2021-09-27 15:26:332021-09-27 15:26 |
|
Update: took a quick tour through the motion prediction code - my guess was wrong. d2dx doesn’t look like it’s intercepting input and forcing changes to the game using network prediction as I had thought. What it’s doing is that the renderer, independent from the game state, is doing location prediction separate from the game state, so when the renderer renders a frame it interpolates where it thinks the units should be based on where they were during the last game update and where they were headed during the last game update. This gives a higher apparent refresh rate by interpolating locations between game ticks but doesn’t affect the underlying game logic. It can lead to units “snapping back” to their actual location of the prediction over or under predicts where they should be, similar to the snap backs you’d see for network prediction due to lag or packet loss. It’s actually a pretty reasonable solution because you only need to interpolate about three frames between updates to get to nearly 100 FPS (1 game tick and 3 interpolated frames = 4 displayed frames 24 times a second gives you 96 FPS. A fourth interpolated frame would give you 120 FPS.) ❧ Edited by MasterOfMagic at 2021-09-27 15:31:412021-09-27 15:31 |
|