Multi-player proof of concept

[LTGH]

This is really amazing.
I would be happy only to get able to run around in town co-op.
Hopefully we will see a alpha release in the near future, that way i will be able to trade stuff with my friends (thats more fun then just cheating all your gear and destroy the game )
maybe just an alpha-alpha version where we only have multi-player access in the main town? that would be more then anough for a couple of weeks.

Keep it up!

[Karjalan]

bump

[OlaNordmann]

Still going strong?

[DarkSynopsis]

Glad Co-op is getting added in This game seems very enjoyable but I personally would only buy it to play Multiplayer so I hope you guys can get it done

[Looter]

Would be amazing if, after the dev's see the success of this mod (upon its alpha/beta) they patch in 4-player Co-op for their players..

[Skumeth]

Great job thus far guys, keep up the amazing work!

[nasarius]

If you don't want to post your source yet, could you share your IDA files? Or whatever you've used to make notes about the disassembly?

I was already very familiar with API hooking, and I know just enough x86 assembly to be dangerous, so figuring out internal subroutine hooking was a fun challenge that took me about a day. My code is here.

But the really tedious part is documenting an opaque codebase. Especially when at least the Steam version seems allergic to debuggers (a DRM measure or me doing something wrong, I'm not sure). So if you'd like another hand in wading through the ~2MB of raw code, let me know.

[drivehappy]

If you don't want to post your source yet, could you share your IDA files? Or whatever you've used to make notes about the disassembly?

I was already very familiar with API hooking, and I know just enough x86 assembly to be dangerous, so figuring out internal subroutine hooking was a fun challenge that took me about a day. My code is here.

But the really tedious part is documenting an opaque codebase. Especially when at least the Steam version seems allergic to debuggers (a DRM measure or me doing something wrong, I'm not sure). So if you'd like another hand in wading through the ~2MB of raw code, let me know.

We're probably going to keep the source close to the chest until we do a rewrite, most of the core hooking is done by dengus and he's expressed interest in releasing it. The Steam is pretty light on DRM - it starts another process after launching so you need to simply attach to it.

We've got a lot of the major functionality done, it's simply wading through obscure stuff to get the game to play nice (i.e. lever-bridge interaction doesn't update the collision).

[Vis]

If you don't want to post your source yet, could you share your IDA files? Or whatever you've used to make notes about the disassembly?

I was already very familiar with API hooking, and I know just enough x86 assembly to be dangerous, so figuring out internal subroutine hooking was a fun challenge that took me about a day. My code is here.

But the really tedious part is documenting an opaque codebase. Especially when at least the Steam version seems allergic to debuggers (a DRM measure or me doing something wrong, I'm not sure). So if you'd like another hand in wading through the ~2MB of raw code, let me know.

We're probably going to keep the source close to the chest until we do a rewrite, most of the core hooking is done by dengus and he's expressed interest in releasing it. The Steam is pretty light on DRM - it starts another process after launching so you need to simply attach to it.

We've got a lot of the major functionality done, it's simply wading through obscure stuff to get the game to play nice (i.e. lever-bridge interaction doesn't update the collision).

Any chance of giving us another teaser video showing off some new hotness?

[chebuoi]

man i love teaser

[Vis]

man i love teaser

Teaser, tazer, just give it to me baby, give it to me hard!!

[Blue]

good to see this is still moving forward

[mesasone]

We're probably going to keep the source close to the chest until we do a rewrite, most of the core hooking is done by dengus and he's expressed interest in releasing it. The Steam is pretty light on DRM - it starts another process after launching so you need to simply attach to it.

We've got a lot of the major functionality done, it's simply wading through obscure stuff to get the game to play nice (i.e. lever-bridge interaction doesn't update the collision).

So have you guys managed to get dungeons synced as far as monsters and drops go? That'd be really awesome if you had. I'm looking forward to seeing your progress, it sounds like you're getting pretty close to having an almost fully functioning multiplayer already.

[nasarius]

most of the core hooking is done by dengus and he's expressed interest in releasing it.

Yeah, just a list of addresses and names would be fantastic...

The Steam is pretty light on DRM - it starts another process after launching so you need to simply attach to it.

Good to know. I tried that, and IDA stalled forever. I'll figure out what's wrong.

get dungeons synced as far as monsters and drops go

The clean way to do it, the way it's kinda-sorta done in most multiplayer games, would be to slave the client to the host, so all the real work of running the game model is done by the host and the results are mimicked by the clients. Plus I'm guessing you can cheat a lot by syncing up the PRNG seeds.

[Crucestech]

free bump but see this ain't gonna happen and if it does the mmo will be out by then lol

[chebuoi]

I still prefer a coop to a mmo

[Vagrant]

free bump but see this ain't gonna happen and if it does the mmo will be out by then lol

Do you have that first hand from the two chaps working on it or is that just a baseless assumption?

Also, bump!

[drivehappy]

I'll give a quick update for those following this thread (I know you like to keep somewhat informed), of major stuff not working. This is partially a list for myself as well:

• Clients aren't handing off their inventory items to the host to create.
• Gold/HP/Mana potions aren't triggering item pickups.
• Bridges aren't updating collision information on the client when triggered on either end.
• Client monster models die in the air.
• Host doesn't handle client resurrection properly.
• Client needs to be initially forced to load into host level.
• Object interaction needs a sync check (perhaps through a GUID - can we rely on level loads to be deterministic?).
• Stability problems on level transfers.

Obviously some are higher priorities than others.

Again, we're working on this in our free time and I believe both of us haven't had much of that in the past week.

[OlaNordmann]

Thanks for the update!

Again, we're working on this in our free time and I believe both of us haven't had much of that in the past week.

That's ok, we just want to know that the project is still active.

Keep up the great work!

[Looter]

Thanks for the update drivehappy, hope you can iron out those problems fairly easily (too optimistic i presume).