Comment by Night_Thastus
16 days ago
It would be one thing if it was a 20% increase in space usage, or if the whole game was smaller to start with, or if they had actually checked to see how much it assisted HDD users.
But over 6x the size with so little benefit for such a small segment of the players is very frustrating. Why wasn't this caught earlier? Why didn't anyone test? Why didn't anyone weigh the pros and cons?
It's kind of exemplary of HD2's technical state in general - which is a mix of poor performance and bugs. There was a period where almost every other mission became impossible to complete because it was bugged.
The negativity is frustration boiling over from years of a bad technical state for the game.
I do appreciate them making the right choice now though, of course.
It was a choice, not an oversight. They actively optimised for HDD users, because they believed that failing to do so could impact load times for both SSD and HDD users. There was no speed penalty in doing so for SSD users, just a disk usage penalty.
Helldivers II was also much smaller at launch than it is now. It was almost certainly a good choice at launch.
You make a million decisions in the beginning of every project. I'm certain they made the choice to do this "optimization" at an early point (or even incidentally copied the choice over from an earlier project) at a stage where the disk footprint was small (a game being 7GB when it could've been 1GB doesn't exactly set off alarm bells).
Then they just didn't reconsider the choice until, well, now.
Even at the end of development it’s a sensible choice. It’s the default strategy for catering to machines with slow disk access. The risk of some players experiencing slow load times is catastrophic at launch. In absence of solid user data, it’s a fine assumption to make.
2 replies →
>But over 6x the size with so little benefit for such a small segment of the players is very frustrating. Why wasn't this caught earlier? Why didn't anyone test? Why didn't anyone weigh the pros and cons?
Have you never worked in an organization that made software?
Damn near everything can be 10x as fast and using 1/10th the resources if someone bothered to take the time to find the optimizations. RARE is it that something is even in the same order of magnitude as its optimum implementation.
But this isn't an optimization. The 150+GB size is the "optimization", one that never actually helped with anything. The whole news here is "Helldivers 2 stopped intentionally screwing its customers".
I don't see why it's a surprise that people react "negatively", in the sense of being mad that (a) Helldivers 2 was intentionally screwing the customers before, and (b) everyone else is still doing it.
> The whole news here is "Helldivers 2 stopped intentionally screwing its customers".
That is an extremely disingenuous way to frame the issue.
1 reply →
I think what makes this a bit different from the usual "time/value tradeoff" discussion is bloating the size by 6x-7x was the result of unnecessary work in the name of optimization instead of lack of cycles to spend on optimization.
Eh probably not, it's probably handled by some automated system when making release builds of the game. Sure, implementing that initially was probably some work (or maybe it was just checking a checkbox in some tool), but there's probably not much manual work involved anymore to keep it going.
Reverting it now though, when the game is out there on a million systems, requires significant investigation to ensure they're not making things significantly worse for anyone, plus a lot of testing to make sure it doesn't outright break stuff.
1 reply →
This is not a reason for accepting it imo
Optimization takes up time, and often it takes up the time of an expert.
Given that, people need to accept higher costs, longer development times, or reduced scope if they want better optimized games.
But what is worse, is just trying to optimize software is not the same as successfully optimizing it. So time and money spent on optimization might yield no results because there might not be anymore efficiency to be gained, the person doing the work lacks the technical skill, the gains are part of a tradeoff that cannot be justified, or the person doing the work can't make a change (i.e., a 3rd party library is the problem).
The lack of technical skill is a big one, IMO. I'm personally terrible at optimizing code, but I'm pretty good at building functional software in a short amount of time. We have a person on our team who is really good at it and sometimes he'll come in after me to optimize work that I've done. But he'll spend several multiples of the time I took making it work and hammering out edge cases. Sometimes the savings is worth it.
7 replies →