MechanicalLens
New member
At 4:02 in the latest horde night video, the 3 icons in the forge... I see an anvil, crucible, and... a bellows? Is that a bellows? Where does the Tool and Die set go now then? I'm so confused...
PC Alpha 18 Dev Diary!!
- Thread starter madmole
- Start date
- Status
Roland
Community Moderator
Won’t help. They’re complaining that Madmole is talking about non-A18 stuff. The dev tracker won’t help them avoid the MMA, school, body building, etc topics they don’t care about.
They are all about sticking to business. They don’t want the personal or the fun or the socializing and they can’t understand why it is allowed to talk about Hellcats and nutrition but not about LBD and avoiding blood moons.
Dev tracker won’t fix any of that.
beerfly
Colony Rounder
Used to use my whole base at horde nights before, where I was scared as ♥♥♥♥ not to lose my chest full of callipers, or engines, or any valuables. Back then a separate horde base is out of my plans and I love how you did similar thing here.
You did great @madmole, another good one.
You did great @madmole, another good one.
Odetta
Tester
Something is definitely up with military zombies. I thought the health might be bugged or they get knocked down to easily, but when watching further it looks like a majority of the bullets pass right through them. I dunno, just seems off.
Or it might be the type of ammo he is using is super weak against armor.
Or it might be the type of ammo he is using is super weak against armor.
MechanicalLens
New member
Either that, or armor piercing rounds really are required for them. They went down FAST when madmole was firing them.
MechanicalLens
New member
But I did also notice that the darts appeared to be travelling through some zombies when madmole first showed off his new gauntlet design.
MechanicalLens
New member
Watch from 0:27 to 0:34. Notice how the zombie casually walks through the gauntlet. (Unless those dart traps didn't have darts in at the time?)
Edit: Nope, they were definitely armed.
Edit: Nope, they were definitely armed.
Last edited by a moderator:
Zombie_Jam
New member
I saw a nice video on the potential of graphene.
TundraWolf42
New member
I love hand gliders, I want it.Roland, is your weird ass hint about this?
Mechanimal
RangerRedd
Simulating a virtual world in greater detail is a lot to ask from the type of hardware we use to begin with, especially one with as much detail as a voxel engine. The real problem is trying to do so much concurrently using a von neumann machine, which is a fundamentally serial architecture, which all modern computers are. As with many subjects most people do not realize there are other ways to do it, but everything went this route back in the 80s, mostly, so alternative architectures has gotten little love, until recently.
The few gray area exceptions, and what led me to discover this 20 years ago, was graphics cards. There is a reason they are highly used in other fields like advanced machine learning, scientific simulations, and even bit coin mining... basically any task that requires many separate calculations to occur simultaneously, which serial architectures try to overcome by just doing things faster, which also causes a ton of heat and thus wasted energy. What's really silly, and drives this point home, is super computer anatomy. The meat of them is focused on essentially turning a serial machine into a parallel machine. Still. I've been face-palming this subject for decades, but I work in robotics so it's probably for the best that this realization happens slowly, and doesn't take off until civilization gets it's ♥♥♥♥ together.
As much as new materials will help increase speed and continue to shrink things will be helpful, the real jump will occur once we get more chip structures to work with. Good news is gaming cards have been the best at pushing this need, so far. Serial is horrible at these kinds of tasks so most computer software is cursed with having to try and work around this inherent bottleneck, it's ridiculous, and is still very painful to watch. We will likely continue to get the best results from new chips aimed at games and Machine Learning, but not because of their material science, but because those systems will continue to become more parallelized at the hardware level, because that's where the problem itself stems from.
Anyway, it is finally catching on at a high rate, pushed mainly by industry shifts I keep a close eye on. It shouldn't be too long before we can simulate worlds with a 1 inch or smaller voxelization, but we'll also have a bunch of smart bots to deal with, and that will either be awesome, or the final horde night for the human species. Hard to say, still. But the games, if we make it, will be ♥♥♥♥ing nuts. It'll be awhile before we reach atomic(near really) level sims, for example, but we already live in one, so... The atom is the voxel, and only has 3 ingredients, 2 quarks and a lepton. That's it, that's everything. And I didn't even get into Quantum Computers, that is another story, not going there, yet.
The few gray area exceptions, and what led me to discover this 20 years ago, was graphics cards. There is a reason they are highly used in other fields like advanced machine learning, scientific simulations, and even bit coin mining... basically any task that requires many separate calculations to occur simultaneously, which serial architectures try to overcome by just doing things faster, which also causes a ton of heat and thus wasted energy. What's really silly, and drives this point home, is super computer anatomy. The meat of them is focused on essentially turning a serial machine into a parallel machine. Still. I've been face-palming this subject for decades, but I work in robotics so it's probably for the best that this realization happens slowly, and doesn't take off until civilization gets it's ♥♥♥♥ together.
As much as new materials will help increase speed and continue to shrink things will be helpful, the real jump will occur once we get more chip structures to work with. Good news is gaming cards have been the best at pushing this need, so far. Serial is horrible at these kinds of tasks so most computer software is cursed with having to try and work around this inherent bottleneck, it's ridiculous, and is still very painful to watch. We will likely continue to get the best results from new chips aimed at games and Machine Learning, but not because of their material science, but because those systems will continue to become more parallelized at the hardware level, because that's where the problem itself stems from.
Anyway, it is finally catching on at a high rate, pushed mainly by industry shifts I keep a close eye on. It shouldn't be too long before we can simulate worlds with a 1 inch or smaller voxelization, but we'll also have a bunch of smart bots to deal with, and that will either be awesome, or the final horde night for the human species. Hard to say, still. But the games, if we make it, will be ♥♥♥♥ing nuts. It'll be awhile before we reach atomic(near really) level sims, for example, but we already live in one, so... The atom is the voxel, and only has 3 ingredients, 2 quarks and a lepton. That's it, that's everything. And I didn't even get into Quantum Computers, that is another story, not going there, yet.
Drithyl
New member
Yes, older generations always think that new generations are weak and new generations always think that old generations are out of touch with reality. Funny how people still think they are right when they make such remarks on one or the otherThat's pretty much what Horace wrote some 2000 years ago so it's been going on for a bit.(that, incidentally, is something history teaches us)
- Status