Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

KSP2 is spamming the Windows Registry until the game stops working

forum.kerbalspaceprogram.com

Post reply on HN
Print view View original

131–140 of 306 posts

Re: KSP2 is spamming the Windows Registry until the game stops working

#131

Post by Aperocky »

ethbr1 wrote: post #20
SketchySeaBeast wrote: post #3

KSP2 is really becoming the premier example of "Nearly perfect labour of love's legacy ruined after being bought by a larger company." More on topic: I have no idea why one would want to use the registry to store this information.

It's a great example of "Quality doesn't produce market-leading returns" Of which the consequence should be "Don't turn labors of love into businesses that demands market-leading returns"

Are you suggesting KSP2 is quality? It is decidedly the opposite.

KSP1 however ARE and have created decent returns.

Re: KSP2 is spamming the Windows Registry until the game stops working

#132

Post by Brian_K_White »

rapind wrote: post #104
dataflow wrote: post #81

Earlier quoted context omitted.

> what is the windows registry and why do things need to write to it? It's a centralized, high-performance small key value store that's the alternative to writing a million config files in random places. It's arguably much better in my experience. E.g. one of the never-ending headaches I always have on Linux is updating config files when a package updates. There's no automatic file merge in general (hence .pacnew and…

What's always annoyed me with many operating systems is this need to organize based on function instead of domain. If I install an application named "Bob" then I want ALL of Bob's configuration to live within Bob. I don't want it split between different registry keys or /usr/local, /etc/, /var, /lib, etc. I'm fine with there being conventions and categorizations, but I'd like to root node to be the application itself…

That would be a directory called /bob right in the root dir, containing everything relating to tge bob app, including the firewall settings, both the distro defaults and the user layered, the executables, the image resources, the db libraries, the db store... never mind that the firewall is another package that needs those same settings...

No.

Or, if you really do want something so insane, well great news, you have snap and flatpak. If those are not complete enough examples of this shining vision, go ahead and show us how it's done.

Re: KSP2 is spamming the Windows Registry until the game stops working

#133

Post by Aperocky »

dotnet00 wrote: post #9

KSP2 had all the warning signs of being a disaster ever since the initial delays and controversies with the original studio. Unfortunately, as usual, the community put on hype blinders until it became impossible to ignore.

KSP1 is really complete, especially if you are using OKAN and common mods that people install.

I don't understand what they are trying to do with KSP2 to begin with. Skeptical from the start and it somehow turned out worse.

Re: KSP2 is spamming the Windows Registry until the game stops working

#134

Post by msla »

dataflow wrote: post #81
echelon wrote: post #76

Earlier quoted context omitted.

Coming from a Linux background, what is the windows registry and why do things need to write to it? All I ever read about it seem to be horror stories. Can't you store stuff alongside the install? Or in some user data location?

> what is the windows registry and why do things need to write to it? It's a centralized, high-performance small key value store that's the alternative to writing a million config files in random places. It's arguably much better in my experience. E.g. one of the never-ending headaches I always have on Linux is updating config files when a package updates. There's no automatic file merge in general (hence .pacnew and…

The performance part is questionable, as is the discoverability: Programs write binary blobs to impossible-to-find keys and, as we see here, there's nothing preventing a severe loss of performance if the program writes too much to the registry. It can only be examined with specialized tools, and, due to the binary values written to it, it might not be meaningfully editable at all.

Re: KSP2 is spamming the Windows Registry until the game stops working

#135

Post by Aperocky »

comprambler wrote: post #23

Everyone seems to love shitting on KSP2, when the same people forget what KSP1 was like in Alpha. No science, shitty graphics, excessively wobbly rockets, inconsistent orbits, (other than graphics) seems familiar? It took a looooong time for KSP1 to get where it is now, it will also take awhile for KSP2 to be as polished. Chill out.

But what's the end game for KSP2, where KSP1 is today?

Speaking form a software development perspective, I don't see it improving beyond that point, or if that's even possible at all. For all KSP2 promised to improve the physic engines of KSP1, it has seemingly came up with a foundation that is significantly worse, not sure what you can do when your underlying engine has severe limitations.

Re: KSP2 is spamming the Windows Registry until the game stops working

#136

Post by powercf »

ye-olde-sysrq wrote: post #22
Pannoniae wrote: post #6

Not surprised at all; KSP as a franchise is in the enshittification stage where they aim to invest the minimum amount of money and maximise returns. This is just simply inexcusable software engineering and quality control. Bit tangential but I'd like to remind everyone that this is the game where the lead designer (?) said that wobbly rockets and physics bugs are adding to the fun of the game, and they are implemente…

KSP2 isn't an enshittification play, imo. It's literally just bad project management. I'll list their sins from my armchair: - remaking a beloved, established game that has a prodigious base of features and extremely extensive modding support. you have a HUGE hill to climb just to get to your "MVP" - you have to supplant the existing game plus its modding community, and this is already a niche audience of ["people wh…

From my armchair I'll add that throwing away the KSP1 engine and replacing it with, based upon the number of bugs, a new implementation of the same basic idea seems like a terrible idea. My understanding is that most of the original Squad team (i.e. the only people in the world with experience building a successful orbital mechanics game) weren't kept on for KSP2. Take Two should have done everything possible to keep them as core KSP2 developers.

All that said, I don't think the game will be left unfinished. All costs are sunk and Take Two has a (reportedly) somewhat functional, nicer looking copy of KSP1 with, presumably, at least base elements of interstellar travel and colony systems in place. It's probably worthwhile trying to get the project over the last few hurdles, as it's a potential goldmine if they can pull it off.

Re: KSP2 is spamming the Windows Registry until the game stops working

#137

Post by londons_explore »

ilyt wrote: post #125
londons_explore wrote: post #35

Adding 322 mb of data to the registry isn't going to be healthy for windows either. Registry hives expand but don't contract again. The registry is effectively held entirely in RAM. You are therefore effectively wasting 322 mb of RAM even when KSP isn't running.

I'm surprised they didn't just shoved it in SQLite and called it a day. It's public domain. NIH I guess. Or maybe some old fart in company doesn't want to let go off their abomination.

You're surprised Microsoft doesn't replace the registry with SQLite in Windows?

If so, you might not be aware quite how many millions of entries are in the hierarchical key value database that is the registry, and quite how critical read-write latency is.

sqlite can do clever things like multiple indexes and full text search, but won't beat the windows registry on the performance front at the narrow usecase it was designed for.

Re: KSP2 is spamming the Windows Registry until the game stops working

#138

Post by digging »

Dylan16807 wrote: post #89
digging wrote: post #48

Earlier quoted context omitted.

> KSP1, for example, or No Man's Sky Those are two very different examples, though. NMS was improved heavily but in many ways never approached the features and qualities that people were expecting. KSP2 feels more like NMS than KSP1 in that regard - people have expectations. They've been sold a specific vision which doesn't look like the game. Further, unless it's great ... why not just keep playing KSP1 with extensi…

> NMS was improved heavily but in many ways never approached the features and qualities that people were expecting. Asking who? From what I've seen, the typical sentiment is that they've gone far past the expectations they set. For KSP2 I think the main expectations were eventual new content (starting with a good chunk less than KSP1 and adding more later) and better performance. And they sure haven't delivered perfo…

Well I really hope neither of us spends too much time worrying about this today but I will respond with some points in addition to the other reply you got.

> Asking who? From what I've seen, the typical sentiment is that they've gone far past the expectations they set.

I'll note that anybody still playing the game probably likes it, and anybody who doesn't like it probably stopped playing years ago (except to perhaps check out new updates and continue to feel disappointed).

Some specific things:

- planets feel lifeless and very samey: there's nothing to really explore because the depth of geology, biology, ecology is not there. I suspect that's why they added so much base building, which was never a significant part of the early hype/marketing.

- even features they implemented feel incomplete and shallow, such as player customization and base building, compared to many other games which do those things better.

- the game was described as allowing the player to navigate an actual star system, and instead it's just a skybox densely littered with pirates and asteroids. If I want that, X4 is a hundred times better. For exploring stellar systems, Elite Dangerous or just SpaceEngine show that it's possible to do well.

Re: KSP2 is spamming the Windows Registry until the game stops working

#139

Post by akira2501 »

msla wrote: post #134
dataflow wrote: post #81

Earlier quoted context omitted.

> what is the windows registry and why do things need to write to it? It's a centralized, high-performance small key value store that's the alternative to writing a million config files in random places. It's arguably much better in my experience. E.g. one of the never-ending headaches I always have on Linux is updating config files when a package updates. There's no automatic file merge in general (hence .pacnew and…

The performance part is questionable, as is the discoverability: Programs write binary blobs to impossible-to-find keys and, as we see here, there's nothing preventing a severe loss of performance if the program writes too much to the registry. It can only be examined with specialized tools, and, due to the binary values written to it, it might not be meaningfully editable at all.

It's also really a pain once you realize that many strings are stored in raw UTF16 form. It makes using any tools or doing any automation on the values much harder than it should be.

Re: KSP2 is spamming the Windows Registry until the game stops working

#140

Post by spoiler »

Brian_K_White wrote: post #132
rapind wrote: post #104

Earlier quoted context omitted.

What's always annoyed me with many operating systems is this need to organize based on function instead of domain. If I install an application named "Bob" then I want ALL of Bob's configuration to live within Bob. I don't want it split between different registry keys or /usr/local, /etc/, /var, /lib, etc. I'm fine with there being conventions and categorizations, but I'd like to root node to be the application itself…

That would be a directory called /bob right in the root dir, containing everything relating to tge bob app, including the firewall settings, both the distro defaults and the user layered, the executables, the image resources, the db libraries, the db store... never mind that the firewall is another package that needs those same settings... No. Or, if you really do want something so insane, well great news, you have s…

You're constructing an argument the person never made.

The binary for my app could live in a hypothetical /apps/bob.

Its config could be in there too.

There's no need to duplicate dynamically linked system libraries in there.

If the app need a different version of a library than the one provided by the distro/os, it could vendor it (or link statically). Optionally it can vendor and still try to use OS version if they version is satisfied, but this is just a memory optimisation (disk isn't that expensive).

There's also no need to place firewall rules in there. I'm not sure where you got that from. Firewalls are beyond the scope of a single application?

As for making thing system-wide available, there's already a few solutions for this (symlinks being a very backwards compatible way of doing this on nixes).

For all its faults and poor UX execution (albeit, maybe things improved since I last used them) flat pack and snap have some good ideas!

Post reply on HN