Zico's development blog 2
indeed. I dont think DXX or you is lame. Thanks for doing DXX. One of the reasons I had to stop playing for 2 or 3 years because Descent wouldnt play on WinXP/Win2000. Heck it barely played on WinME. Me and my friends are extremely glad u are doing this. Descent and Descent II are my favorite games.  :-D
Packe loss prevention? That is interesting. I had such code in D2X-XL, and I have disabled it because this just can't work imo. Games like Descent send a lot of data per second (the pps parameter actually is a lower and not an upper limit to what is sent), and recovering lost packets can slow down the network pipeline or even mess it up by completely de-syncing object states. You cannot e.g. send single object sync'ing packets out of order - you will have to keep them strictly in order (or probably rewrite that part of the program).

Either people have a good connection and experience little or no packet loss, or they have a bad connection and trying to make sure they get every (significant) packet will probably rather make things worse for them.

The idea with prioritizing certain packets is however a neat one.

If you want loss free packet transmission, you should go for TCP/IP anyway which can do exactly that for you (there's however a reason online shooters do not use TCP/IP).

Good luck with your feature.
You are right that the PPS is only the lower limit. The question is which part of the packets you want to keep and which ones you can safely toss away.
I admit that the Packet-loss-prevention has not been tested too much but results will show up so we can see what works and what not. I don't even expect it working flawlessly after introduction.

However the fact is: Losing a packet can also de-sync the game same than getting certain ones out of order. The most critical ones are definetly object-related data as you already said. And as long as the whole game netcode is P2P based, it's very difficult to deal with such problems. And no matter how you look at it, TCP/IP would also not solve these problems.
The greatest pleasure in life is to do what people say you cannot do.
Uhm... Honey, there's a head in the toilet!
Hey guys,

Some of you might wonder: "What the hell is the problem in DXX-Rebirth? Why are there new versions after all? The DOS version was just fine.", while others may ask: "When do we finally get feature XXX?".
And recently I often hear more about adding more custom content.

Guys I would like to make all your wishes come true, but in fact I am very exhaustet right now. It's not that there's something in general - buggin me, or something that limits my spare time to work on DXX - while I have to add, that my life BESIDES DXX-Rebirth needs work, too - but it's just the fact I am burned out atm. As for a project I gain NOTHING - except kudos I am very proud of, it's time to reset my priorities.

Even if you might not notice, there's always a huge work going on in the background of the program - mostly rewritings to make the program more stable, failsafe and/or better maintainable. About custom content I can generally say: "NOT NOW!" Sorry.
I have not managed to get as far as I want before I can design simple concepts to make something actually new with this engine. So all I *can* do is begging for patience. The next release will be mostly (but not entirely) focused on Multiplayer and some "background parts". However this isn't easy.
You read about my Packet-Loss prevention code and I am still far away from being done with all tests that should be done as early as possible and the more I think about it, the more ideas I have to make this system even safer.
However this needs a good quality control (as every feature/fix needs actually) and it's not the only Multiplayer improvement happening - even if you won't maybe notice the most of them.

I know DXX-Rebirth has not the fastest development cycle and I respect everyone developing faster than me and working on different places (Tracker, Modding, and all other stuff) at the same time, but I cannot offer that. On the contrary this is just very stressful and therefore I need a break.

This hopefully also gives me the time to think about further developments. This project with two sources which almost have two seperate communities, with different demands and also different expectations in concerns of Multiplayer and Singleplayer, opinions about what SHOULD be done (hell, what do I know?) while *I* am also concerned about consistency actually now totally confused me. It's time I try to get back closer to my original orientation about the project. Currently I am just headless.

I don't know how long I will exactly stall my own development, but be sure I'll continue it!

Stay tuned ... or not - as you wish
The greatest pleasure in life is to do what people say you cannot do.
Uhm... Honey, there's a head in the toilet!
No problems, matey. I will stay tuned.
zif: Why haven't they fixed that yet :S
Ralathar: Bug report it :p
zif: I'll bug report you
Dude, you NEED a rest.  frankly, you've given us more than I was hoping for. It's way more stable than 1.43 (my ol' faithful) and there some neet-o features ta-boot!! It's still good ol' Descent with all the prettiness we REALLY wanted with the Voodoo cards. Wink

Thank you!!!!

Oh, and thanks for being straight forward, good natured, and staying approachable.
Have a break, Zico!

You've done so much until now, and the most important thing is that you've delivered us a port that is actually *working*.

I've seen plenty of attempts for old DOS game ports, and most of them are hopelessly inadequate, or so unstable that the number of crashes goes astronimical.

Get some rest and enjoy the fruits of your labor Smile
Gib mir Benzin!
Indeed  :-D
Zico you need to take a break have plenty and then some. But great blog anyways
I've heard rumors that Zico may actually be a... robot!

He wouldn't need rest in this case Big Grin
Gib mir Benzin!

Forum Jump:

Users browsing this thread: 1 Guest(s)