Man, these posts get so long, I’ll try to condense things, meaning I’ll skip past a lot of stuff.
Again, we don’t know the numbers for D3, but judging by sales numbers and normal player retention, we can safely assume that D3 had more seasonal players than D3 did. Much like we can do the same for D4. Both games have much more forgiving RNG (to the point where it’s negligible) than PoE. They’re more popular. Does that make them better?
ADOM vs ToMe is the same thing. ADOM appeals to more casual players of the genre, whereas ToMe appeals to the more hardcore ones.
This is a game identity issue, not “it’s more popular therefore it’s better”.
And yet there are still a lot of players that consider D2 to be the peak of this genre, with its unforgiving RNG. And you still get players crafting thousands of items even though crafting in D2 has the worse RNG of any ARPG ever, where you have to craft hundreds of items to get a single usable item. And you can’t keep reusing the same items.
That is where you’re wrong. There are no guarantees. Which is my whole point. You can farm hundreds of shades and an omnis might not fall for you ever. Reducing variance via your system means that you will be guaranteed one (even if your system is just for FP (for now) the same principle applies), because it affects rolls over time.
You do get an optimal result more than 50% of the time. That’s the effect that 2 rolls have. For example, if you have a 1d20 roll, you’re expected to get an average of 10.5 over time. If you use 2d20 and pick the highest, you’re expected to get an average of 13.5 over time. 2 rolls clearly favour your success chances. And when you add the glyph of hope (25% chance it won’t spend anything) and critical success (don’t know the odds of that one), it’s clear that the rolls are clearly in your favor.
Which is why you can pick an exalted and more often than not max the affixes. You just remember the failures more clearly, but that’s just perception bias.
People already demand it. Besides, it’s exactly the same thing. You have cases of people doing 15+ slams and not getting the affix they want. You want a normative formula that will eliminate these outliers. It’s obvious that the next step will be applying it to LP slamming and to affix roll ranges.
Bottom line, you’re just asking for reduced RNG. You want to cut off the high and low outliers. Well, that is reducing RNG.
It’s like having a system where if you keep flipping heads on a coin you get increasingly better odds of flipping tails until you’re guaranteed to hit it.
Whichever way you look at it, it’s reducing RNG. Both the bad and the good.
And, even more importantly, it’s something that only affects BiS and high-end item chasing. Because every other gear is already easy to find/make.
As for the programming, I’ll just condense all answers into 3 points:
1- One of the most common ways programmers waste time is by not analyzing in advance what the changes should be and what they affect. You start working on the code and then realize that what you’re doing can’t be done that way because it didn’t take into account something else that is affected by it.
There’s a reason why companies spend a lot of money on analyst teams and on system designers. A hint: it’s not because they like to spend money.
It’s also the same reason why you have so many people studying development systems like agile or scrum over the years.
2- I have no idea how their code is done, but I find it quite likely that there’s a single function for all the rolls. One that gets fed the min and max ranges, along with multipliers and gets called everywhere you need a roll. Which means that making changes to this means that you have to analyze the whole code to see the impact it can have on other systems. Otherwise you might be breaking the code elsewhere. Probably even in a place the dev that is doing the change isn’t even aware exists.
Because very large codes with very large teams means that individual programmers aren’t usually aware of most of the code. That’s why you get a software designer that is aware of the whole system and its interactions.
3- Even if we were to assume that this would just take 1h to code (let’s ignore the design and testing phases), there are already hundreds of other changes that need to be made that also take 1h. So things get prioritized. And many changes that only take 1h get pushed back for months or even years, because there are more important things to work on and too little time to do it in.
Which is why a modder can do something in a day that a team won’t do in months. It’s not because they’re slower or bad at their jobs. It’s because the modder is focused on a single issue and gives it their priority, whereas the team has hundreds of stuff to do so that single change gets pushed back to work on more important things.
A team having hundreds or even thousands of things to work on even often makes it so they actually ignore some things as not important. If something is working, it just isn’t working as well as it could, oftenly it gets completely ignored in favour of the important stuff.
5,8 million daily players at release, 2,3 million daily players 1 year after release.
That was given to us by Blizzard… how true those numbers are is another topic, we’re talking about Blizzard after all.
Afterwards it tapered off more and more.
My argument is that the setup of PoE prefers long-term engagement versus D3 which prefers short-term engagement.
And this leading to PoE being more successful in terms of outcome the longer that goes on versus D3 which cycles through the full market without the ability for more people to come in (Since basically everyone already did) and hence slowly dieing off, unable to reach the regular numbers PoE can reach.
LE in comparison is supposed to be in the middle, this means slower itemization then D3 (not hard) but faster itemization then PoE.
Difficulty is also between those two.
It’s a good market position as it’s not one which has been taken already.
But for itemization it doesn’t hold true. The curve is messed up which causes early game itemization to be far quicker then PoE and nigh similar to D3 (hence a bit too fast) but it flips over entirely to outpace the acquisition rate of PoE.
That shouldn’t happen, that goes counter to the positioning, that’s why I’m suggesting things to be done after all.
No, Blizzard works for investors, investors need raising revenue per quarter, this means if a game has many sales from the shelf it counts as a success, anything further on is secondary.
This is a outcome of how stock markets are set up, making investors seek out solely the most rewarding companies and not sustainable. Limited market size and all.
So Path of Exile in comparison is more successful since it is more stable, has vastly less growth but stays steadily on top with steadily high numbers and no major dips over time. You can count on it. Bad for investors… great for someone seeking out safety in investment.
For a pure capitalistic approach D3 is better but for a low-risk approach PoE is better. Capitalistic approaches are always short-term and never long-term, which causes the whole volatility after all, PoE’s approach is not a capitalistic one but one of sustainability. Which has proven to be one that’s relates more to success overall in history.
Other way around, ADOM enforces in-depth knowledge of areas, item acquisition and possible build-types to get through, ToMe less so since you’re reliant on RNG and have only miniscule amounts of fixed outcomes which often don’t affect your character in a severe way long-term.
Because it allowed you to gradually work towards a goal.
Every rune you dropped brought you closer, because runes often were a limiting factor.
The uniques were hard to find but not impossible, you found some, fairly regularly even for a longer term player.
And the other items were the chase items, which was fine.
Also the gameplay itself was top-notch for the time and still is enjoyable today.
LE in comparison has no gradual progression mechanics. No rune equivalency, no crafting currency to get closer step by step, no combination systems to increase value of gear or anything else.
Outsides of 3-4 uniques all of them are piss-easy to acquire, baseline uniques are not of value, the LP makes it, which enforces exalted items on top, which are hard to get in high quality as well.
Your chase hence are twofold, high LP uniques and rare affix exalted items.
Which then causes the issue since in D2 with your rune and your base item you had everything, in LE there’s 2 further layers of RNG beyond, crafting the jackpot drop of an exalted into one which actually has 4 usable affixes and slamming that item onto the LP item hoping for the right outcome.
This means the player is ‘deprived’ of their drop-reward. D2 made you excited for drops. LE doesn’t, it tries to make you excited for crafting and slamming, and that feels like gambling (while drops are too, but it’s a different feeling) and hence turns off players.
That’s the major setup difference between the systems. If we compare Last Epoch to D2, outside of the dated gameplay nowadays… D2 is superior. Few games managed to become as good or better then D2 because Blizzard stumbled by mistake over a masterpiece of psychologic engagement.
Let’s make it into a reality check there.
One of my personal BiS items to acquire, a pair of gloves.
Eternal Gauntlets with T7 missing health gained as ward… ok, that in itself is a unicorn drop a little bit, right base, right affix, miniscule access to get it. So it’s a one in a lifetime (for a common player) drop.
The game doesn’t let you get excited about it though.
Let’s say it has 40 FP even.
To make this base into the BiS I need it to have that mod as the 5th and get Int/Minion damage for prefixes and crit reduction/Hybrid health for suffixes.
We’ll ignore the hybrid and instead ‘only’ get health since it’s common.
My item now has int T1 already on it (lucky me) and I can put minion damage on directly since I seal the T7.
Suffixes I have T3 health regeneration and T2 fire resistance.
So, how’s the crafting progress here? Removal or chaos glyph? Removal costs 1-25 and chaos glyph 1-18, so the removal seems like the better option, our wanted affix is already save.
Ah yes, sealing that was 9 FP, so we have 31 left.
Removal 1 takes 11 FP, hits health regen, 20 FP left, removal 2 takes 3 FP, hits as well, 17 left. Lucky!
Now, even with every roll below the 50% outcome chance we already are at a point where it’s basically impossible to get the result, 17 is not enough to get 15 rolls in a row successful with a glyph of hope, even if we say 6 cost no FP (which is beyond the 25% chance given) we still have 9 rolls which each aren’t allowed to cost more then 2 FP at most, the nominal cost is between 5-9 FP per craft though.
Wanna take the chaos glyph route? Suffixes have 4 rare/very rare affixes, 5 uncommon and 14 common affixes. Even if we take my rolls to all count as ‘common’ the chance is still 1 in 22 to hit per try, which is below 5% chance to get the right one per roll, and a roll costs ~1-20 FP, which means depending on tier per try a 9 FP loss.
So even then, the outcome chance is vastly below 50% for the player.
Go ahead and math stuff out yourself and see your actual chances, even when putting everything in the player’s favor you’re still below 50%, actually, the realistic chance to craft a BiS item is somewhere around the 10% range.
Around… 10%… success chance.
Which means you need 10 unicorn drops to make your unicorn item commonly.
That’s insane. It’s beyond the range of the 50k+ exalted crafts of Path of Exile. It’s stupid… and we haven’t even taken actual BiS outcomes and roll-changes into consideration!
Good, put it with FP limits and amounts into consideration, make a full-scale math issue out of it and then tell me again.
You’re literally talking out of your ass there without taking the reality into consideration. It makes for a great example, I’ll give you that… but if you don’t take everything into consideration it’s just a fallacy you create for yourself.
For that I would need an item dropping with all 4 affixes with the ones I personally need and never having the need to actually exchange a singular affix to another one.
Or even better, have a free slot, which doesn’t happen too often, much less so 2 free slots.
Yes, because it feels like shit if your only solution to progression is a gamble mechanic which already is build into a gamble mechanic (drop-rate). The feeling is hence fairly understandable, and since it’s the easiest issue to see it’s the first to be demanded.
Such feedback gives you the knowledge that something isn’t quite optimal with drops yet, late-game players mention that Exalted bases are a problem, earlier ones that it’s the legendary system… so there’s an underlying cause for that and those feedbacks tell you to search for a way to make it at least be perceived better.
Or you copy the function itself, document it as the old and new version and let your one system now access the new one instead of the old one without affecting the other places it has any effect.
Then you gradually go through the list from where the function itself is accessed and decide if the old or new one are the better solution, exchanging access accordingly.
The example was ‘Starfield’ which had a myriad of complaints about clunky functionality, hence affecting 100% of customers while being accessed several dozen times per hour.
This is a top-tier priority change.
If not then the one setting priorities fucked up.
It’s more important then small-scale user-base not being able to run your game. It’s more important then fringe-case crashes, it’s more important then rare happening softlocks and it’s more important then 99% of the bugs.
It’s the primary interaction from user to program, which is unintuitive, lacking functionality and information which causes major secondary issues.
So don’t give me the crap about ‘we have hundreds more important things’ because you should be fired if you’re in that position and decide this way, plain and simple. That’s your job there and you failed it masterfully by pushing back such a top-tier problem.
And if your main return complaints are ‘This UI is shit and you can’t work with it’ then it’s obviously on the top of the list of things to be done first and foremost, not fixing lens flares, slight unintended blurs, hitching when using items, directly eating food from tables and so on.
And especially not when it takes 1 month and 3 days to implement a FoV slider which is a direct accessibility feature removing a large amount of people from getting vertigo.
So sorry… but really… you can go and shove that argumentation line somewhere, because it’s a stupid line of thought through and through.
It’s excuses over excuses over excuses. Several of the mentioned aspects being core-features which have just not been done. Thousands of hours focus on a non-functional building system but can’t be bothered to implement functions which for a decade are seen as ‘must have’ basic things? Any other job and you’re fired to hell and back if similar scaled problems arise, you never find a job in the segment again because you’re a liability, you’re a quack then, you’re a danger to any company. And that’s excused?
Save me the effort.
You are talking about BiS gear, so yeah, it should be hard. In fact, if it’s actually 10% then it should be lower.
Crafting, as it is right now, makes it very easy to get good gear. It just makes it very very hard to get BiS. And that is by design.
The curve progression is actually very good since it lets you keep getting better gear until you reach 300c and endgame. After that it get exponentially harder. Because you don’t need it. It’s a chase item and chase items should be really hard.
So, more than 1h then? You’re just moving goalposts where you duplicate code and shoot the actual final solution for “someday”.
And that inevitably leads to a bunch of duplicated code where you have to change a bunch of places at the same time when you want to change something, because “someday” will never come as there are more important things to do.
This is not a 1h fix, though.
This is also not a 1h fix. If you didn’t have a slider that meant you had a fixed value. Which means you now have to make it responsive and you have to make sure it behaves properly with all values along the range.
You said that a modder could make a change in a day when a team couldn’t make it in 1 month. But you provide no examples of a top priority change where this happened.
You certainly have cases where the community fixed your game issues, most notably in VtM:Bloodlines, but that was not a 1 day job.
Also not a 1 day change.
Seen by whom? If we take your suggestion as example, then it’s just some people that want easier RNG. That’s a design issue.
And if we take the instanced zones issue as an example, then that’s a very complex issue that is limited by the game engine.
So which “must have” basics does EHG not deliver that aren’t part of the game design or just limited by technical issues? They’re even working on the mouse customization issue, although that also is more than 1h.
It’s like saying that EHG is incompetent because they could change the XP requirements to reach level 100 in an hour but they don’t. When clearly they don’t want to and in fact are even pushing in the other direction.
That’s not to say there aren’t cases of development incompetence. VtM:Bloodlines, CP2077 and Wolcen are 3 of the most notable ones. But you don’t get easy fixes for those problems.
And the issues that are 1h to solve usually aren’t something that is top priority, they’re just player preference issues, like the one we’re discussing.
The issue with that is that we have a double-loaded RNG layer there which doesn’t need to be double-loaded.
The main limiting aspect is already the item drop, hence item drops on their own become worthless related to the crafting system. Only in conjunction it becomes valuable for the player, which is the main issue.
You either have a great drop-system or a great crafting system. If you try to make both equally important for the progression of core equipment then both will feel unrewarding.
A usually better option is to for example have rare item drops, hence bases, affixes and so on are all hard to acquire… but the crafting is fairly free-style, depending on resources and the access limitation to them, which is to the large degree how PoE for example handles it.
And another way is to go and have the crafting be a limiting factor, we don’t see diablo-clones usually doing that though is it demands for the crafting to have many fail-states and nigh no actual successes bringing you forward. For example the end-game crafting of ‘Black Desert Online’ is a prime example of that. Most items drop regularly and with ease, getting them up to a specific level though which you derive the majority of power from is nigh impossible.
If you use a system like LE has where you have extremely limited drops and then also add fail-states to the crafting system you double down on the rarity of acquisition, this means the acquisition curve raises exponentially higher then a single system. So you have the accumulate chances of two exponential curves which makes it vastly steeper… that’s bad if the curve isn’t handled to stay as flat as possible for as long as possible before ramping up. The ramp-up area is where people drop out as ‘Item A’ needs 5 hours but ‘Item B’ which has no step before it needs 50 hours and ‘Item C’ needs 500… but it ends at item ‘X’ for example, each step taking vastly longer then the one before.
By itself a normal happening, but when a singular upgrades goes beyond 100 hours the vast majority of long-term players have already mentally checked out. It’s ‘done’, it’s ‘over’. And this happens the second you go from common affix exalts to rare affix exalts, even in T6 form. Not to speak of BiS LP slams, those are a step beyond, ‘decent’ ones are those you usually see hence.
Also we have our game progression to put in mind. How long is the player supposed to play the game? 10 hours? 20? 100?
The T5 stage is done after roughly 15 hours by a decent player, maybe 20. Then the exalted stage itself also lasts until… maybe 60 hours into the game? And after you have no progress. It’s 'over.
The game cycle lasts for 3 months at least though, hence progression rate needs to be set at 2 hours a day for 90 days, hence at least ‘somewhat smooth’ progression for 180 hours of game-time. We can lower it a little bit to maybe 150 or maaaaaybe 120 hours (for a 2 month progression rate) and be fine.
60 though is half of the absolute minimum for long-term engagement. This means the game is in trouble long-term.
Not for the initial implementation of the first area which is all I talked about. It only needs to access this 1 area… everything else is up to discussion after all afterwards, different topic.
Refractoring it’s called. You refractor regularly, if you don’t refractor then you get the modern issues of ‘everything else is more important since this change in total needs a long time’ which overlooks simplistic changes completely while implementing code on the side.
For example take a look at how CDDA with their open-source setup handles it, allowing everyone to submit code but nonetheless the head of dev decides which direction the game goes and everything else is ignored.
That’s a good system set up, the actual developer base is vastly smaller then any AAA company is, solely for coders even… and nonetheless you get up to 3 merges daily for a variety of aspects in the code. Many of them are handled via refracturing as well which with proper documentation allows for new code to be implemented directly into the new functions while someone else can go ahead and move over the old code. It removes the ‘dead zone’ between people not being able to implement new code reliant on a system marked for change while also allowing the program to function despite the state.
And you can’t tell me that never in your career something similar to ‘I have to wait 30 minutes or so until approval for my code comes back’ have happened. Those 30 minutes hence can be used for refracturing. You have less downtime without interfering with ongoing processes.
If a dev-team has no refracturing processes going then it’s a shoddy company… unless it’s reliant on live-sustaining segments which needs to have other standards… but you don’t need medical standards in game design and you also don’t need it in the majority of applications in general.
1 day versus 1 month still is a bit of a difference, magnitude of 30. So any further argumentation without taking that into consideration is not a viable argumentation line. Which was the point I’m making.
Ok, lemme guide you through Bethesda’s issue list then. ‘Most requested features’ they called it, which was cherry-picking the ones they deemed easiest to do and not most important to do btw, because those weren’t the ‘most requested’ things when you check the feedback in Steam and Reddit over that course of time. Which was from 6. sep. to 13 sep. when the post came out officially. So probably 11 sep. for the decision, meaning in 5 days after release.
Brightness and contrast control, HDR calibration, FoV, DLSS support, ultrwide support, eat button for food.
The actual highest demanded things were: The menu is unintuitive, interactions cause characters to no-clip, my ship throws my decoration around when entering it… then the aforementioned ones.
Sep 25: Character location fix (softlock issue, ok), star-station labelling (nobody cared, little nice QoL though), Limiting access from vendors further (nobody asked, people wanted the opposite). Flare lens fix, blur fix, flickering issue when scrolling UI, hitching when using a tool, vanishing items on ships fixed (ok, good fix), another item deletion fix.
Wait a second, how many people work on Starfield again? 2 weeks and not a single demanded feature or unmentioned big higher demanded solution even started to be tackled.
Oct 9: FoV gets implemented
Great! Jack-shit was done again, once more nearly 2 weeks.
Need I to go on? Any more questions?
If you can’t see the issue there then I really recommend throwing the towel related to management or coding… because even a layman only dabbling in coding can see issues there.
Where’s the expected coding group which is focused on mass bug-fixing after release? Where’s the coding group focused on implementing oversights mentioned by players after release?
11 points of adjustments and fixes in 1 month? That’s the work-task of what? 2-3 people? Bethesda has over 500 people working on Starfield, compared to Skyrim which had 100… and they managed to fix 17 points on day one in comparison.
So once more, don’t give me this worthless BS. You’re actively lowering my respect for your quality in your job with that.
You mean the instanced zones which have been handled by both Skyrim and Oblivion in a vastly better manner without providing 3 factions that are only different through their little mark visible on their outfit while being functionally and reaction-wise same and also offering 15 unique different places to explore randomly for 1000+ planets when Morrowind had over 100+ different hand-crafted areas with a fraction of their dev-team while also taking Skyrim that also had FoV slider, ultrawide-support and so on and so forth all implemented already?
Oh, not to speak they had it already, used the same engine and didn’t implement it in their new product.
What are you talking about?
Yep, which caused a reduction in retention likely as well since the amount of complaints related to it compared to those which complained about being ‘done too early’ has raised. Nice test, opposite effect likely.
Wolcen is a scam, not incompetence.
Wolcen provided a demo with a great lighting engine and then released without that while giving no mention as to why or informing backers while providing it as a main aspect of their product, a headliner.
They should’ve been sued for fraudulent behavior. They deserved to fail.
Yes, I’m pretty sure that is by design, as I mentioned. Early gear is very easy to get (probably even easier than in D3), midgame gear is very easy to get (though not as easy as D3), endgame gear is still relatively easy to get. It’s just BiS gear that gets affected by the exponential curve and I’m pretty sure that is by design.
You’re not supposed to get BiS gear easily or even easier. They could make it easier in a variety of ways, but much like a 4LP red ring is pretty much impossible, the very best BiS gear is supposed to be as well. They are certainly aware of the odds of a 4LP red ring dropping and could have changed it at any time, and yet they chose to have it remain in the “will never happen” category.
The double RNG layer serves the purpose that for anything where you only need one of the layers or a limited amount of each (which is basically everything excluding BiS) you can easily get it.
This only works because no one is getting paid. Otherwise you have a bunch of people doing code that gets dropped, which for a company is the same as wasted money, since that person wasted hours developing stuff that isn’t used.
This is fine for a volunteer passion project where people don’t mind wasting their time for nothing.
There’s a reason why only fringe cases do this.
It didn’t. Changes are planned ahead. When they’re done they’re merged. They don’t need to be approved because they were already approved before.
If you plan things ahead, the code is already structured in a way that it doesn’t need refactoring. Refactoring is a waste of time. Yes, sometimes it’s necessary because you didn’t analyze things properly, but if you do, you rarely need refactoring.
How do you know that it didn’t start to be tackled? Are you not aware of how teams work? You have some working on the more easily fixable stuff, while you have some working on the UI problem, while you have some working on networking issues, while you have others working on other issues.
The fast ones get turned out more quickly, which is why they release sooner.
But changing your UI isn’t something you do in a day. Fixing the no-clip isn’t done in a day. So you have teams working separately on these issues until they are done. You don’t stop the whole team and make them all work in the same thing.
Yes, you do. You haven’t provided an example where a modder came along and fixed the UI or no-clipping in a single day.
I’m talking about the fact that EHG doesn’t have full control of the unity engine and are thus impaired by the engine limitations and have to find workarounds for those issues.
I’m pretty sure they weren’t concerned with player retention when they did that, but rather with the game identity. Not every change needs to be made with the sole purpose of increasing player counts.
Much like no respec is a game identity design choice. Much like the state of RNG is a game identity design choice.
Bad choice then.
There’s a reason why in all the decades it’s generally not done. It does lead for a select few outliers to have a enjoyable experience but you can’t build a functioning long-term community that stays loyal to the product for 10+ years, which a live-service game demands to have.
In a single-player game I would 100% agree with you, but it’s not a single-player game, the framework to work inside is a different one.
You’re not getting BiS gear at all, it doesn’t exist. A absolute BiS is a unicorn which starts to sprout unicorns from itself by then… a BiS is a quadruple T7 item with the right affixes or a triple T7 item with a T5 affix + T5 sealed affix.
And all of the roles on it being the absolute maximum too.
Proper design demands that they will be seen though, hence at least a triple T7 needs to drop every year once with the current amount of players we have, for MG, not for CoF, CoF needs to have 4 of those hence given the multipliers.
Otherwise you create a system which has a top-end that’s not used nor will ever be used, it’s ‘dead game space’. And guaranteed unobtainability is something which reduces engagement. No other game uses it, none does of the genre.
Ask yourself ‘why?’.
The game is also not marketed.
And the progress despite the vast amount of dropped code is still myriads faster then most comparable products on the market, it outpaces AAA development code-wise by miles.
So if despite this non-optimal, code-dropping environment it still produces more then the regulated environment we have our answer right then and there.
In a marketed product you instead have a person coming to the lead to swiftly tell them their ideas and then each viable one gets implemented, and a few core people work on the core concepts to improve them. Micromanaging in the modern situation for a gaming company which demands creativity is destroying the thriving force for productivity. It’s detrimental and has been proven repeatedly over the last 10 years with nigh every single bigger company.
Especially the part where several people work on several systems which need a different amount of time to be done and only get implemented when done leads to a increase in productivity alone. Why? Because you don’t need to heavily interact with other people and re-learn how the structure is done which is one of the biggest issues in game-dev nowadays.
Also this way you make use of the creative differences between your workers rather then enforcing them to be fixated on aspects of coding which they’re not shining in. Some people create new systems better then others, some optimize code better then others, some create stable foundations better then others, some find fringe-cases easily and implement coded measures to avoid then and so on. Usually you have dev teams strictly working on code which you better shut up and follow the guidelines completely, making them become copy-monkeys which don’t even know what they’re working on as a end-result. No surprise the outcome can’t be good if for example a designer works solely on eyebrows but never sees the faces which they should be put onto.
Yes, which is the issue! Changes are planned ahead… so no coding happens unless it’s ‘talked through’ which takes ages. Hours, days, weeks.
In that time a person can create 4 full systems of the same type all working in a different way where 1 does everything better then the proposed and ‘talked out’ one does.
Such measures are good to have, in moderation, fully hinging on them is a detriment.
And you realize you need for your work 6 months instead of 2 weeks, and both outcomes nonetheless work. One is spaghetti and needs tons of refracturing and one is clean and tidy. So you now have 5 1/2 months to refractor everything and add stuff on top while after those 6 months the other group still finds issues which take you ages to fix.
Yes, I know how that goes, I deem those systems commonly failed systems for coding environments.
As a customer I don’t care if it was in the background, tell me the progress
Nothing works like a charm then informing a customer about the thing they want to see actually being worked on, even if it’ll take a long time still.
You know what doesn’t work? Leaving a customer in the dark.
Look at how Factorio presents information to their community and then look at 98% of other developers and you see the difference.
Big teams are called ‘lazy’ when you don’t see results in a month… the Factorio devs? If you don’t see changes in a year nobody will ever call them ‘lazy’, the general notion is that they’re hard at work, doing all those great things and so on and so forth!
Involve your customers, perception is everything in marketing.
True, since it has not been done since a year by now
This mod was done in a day.
No-clip hasn’t been fixed since it demands individual character’s pathfinding not to intersect with any other one while speaking with them and hence walking through or in front of them (which is a design aspect Bethesda didn’t follow) as well as the characters simply turning towards the camera and the camera taking a fixed position (They can walk away, sink into the ground, fly through the roof… all still in the game) related to the face of the character in any given moment (which for some reason doesn’t work, and that’s rather basic stuff. Characters walk around while you talk to them or the camera takes a wrong position very regularly.)
Combine that with the issue that Starfield is extremely unfriendly to modders unlike what the creation engine initially has been designed for makes it impossible for most mods to even exist, something where Skyrim thrived and Starfield fails.
It had been fine for 5 years with the XP and suddenly it isn’t? Identity of the game is one of the core elements to uphold, hence my question then is why that hasn’t been done during those 5 years since it’s top priority after all, higher then new content implementation… and suddenly is done.
First of all, you’re talking about a game that has barely any graphics. The amount of work needed to make changes in a game like that isn’t comparable to something like LE.
Second, it still only works because people aren’t getting paid. Because if they did have to pay them, then most would have been fired by now because you can’t afford to be paying someone for working hours or days in something that isn’t used at all.
It produces more because the game engine is basic. It’s like saying I could make an RPG Maker game in a month, why the hell did BG3 take years to make?
You’re trying to compare apples with oranges.
Yes, because if you don’t, then someone is working hours, days, weeks on something and it turns out it can’t work that way and they have to start over.
So what happens is that the designers/analysts analyse what has to be done while the programmer is working on something else. This is not a team effort. The whole team doesn’t get to decide nor do they stop working until the designers/analysts decide it.
It’s called staggered development and is essential to any large enough team.
That’s a communication issue, not a programming issue, so I don’t know what that has to do with what we were discussing.
Can’t help but notice that no modder fixed it yet either. In a day or otherwise.
I don’t know where you get that. The mod was released before Starfield was. Which means it’s someone that had early access to the game. There is no mention on how long it took him either on the video or on the mod page.
Not only that, but they have been working on the mod fixing all the bugs. And adding improvements because apparently not everyone was happy with it. So overall they’ve been working on the mod for a long time.
So, according to your own standards, he releases something shoddy in 1 day (which was probably longer) which had a lot of bugs (which it still does), so he should have been fired.
I feel like you’re just trying to nitpick stuff just for the hell of it at this point.
It’s always been part of the game identity. Level 100 was always a long time. However, as more content gets added, they are allowed to adjust things. And in this case they’ve been adjusting it to take longer. Mostly because it’s easier to get XPs now than it was 4 years ago, so an adjustment is required.
Is that so hard to understand?
You know, the stuff which makes a game instead of graphics, animations, sound, shaders, cross-platform support, screen support and the whole auxiliary stuff.
And if you have the base systems to handle your graphics down then it’s solely leftover to work on the mechanics.
If your mechanics are lacking because the effort to get everything up otherwise then it’s a failed system.
It’s fairly simple, especially since graphical fidelity has the highest cost nowadays and doesn’t at all compare to the achieved revenue through it. Otherwise we wouldn’t have games like Minecraft with crap graphics (but a nice style) being so ridiculously successful. We wouldn’t look at Stardew Valley, at Dwarf Fortress, at CDDA, at Caves of Qud, at BattleBit, at Factorio, at Terraria… you get the gist I hope.
Graphics are secondary, they’re worthless without good gameplay. You make a game, not a movie, not a interactive story, a game first and foremost. If you want a interactive Story then ‘Layers of Fear’ is great! If you want a movie then watch a movie.
Every medium has different needs to be fulfilled and for a game… it’s primarily mechanics.
So we should fire 90% of management in game companies?
Because they hold long meetings, standups which take vastly longer then supposed to and much more. That means they actively remove the working time of those which should code.
So that’s better then having some unusable code?
I would argue someone able to code more then 2 hours a day in a 8 hour work-day (people need to settle both after coming in, after meetings/stand-ups and after lunch, that takes between 15-45 minutes t get productive usually) is more viable then using such massive amounts of time talking topics into the ground like we do here.
Have someone code for a week straight without interrupting this person more then once in that week. Then evaluate and fix issues. You get 30 hours of active work instead of 8-10 in a week this way.
Because the sheer amount of code related to different outcomes and the tree of actions behind that is massive. Which is pure code, and hence sheer amount of replayable content.
Why did Starfield take so long when that’s not the case? They had their engine, they needed the assets (which take ages) and the mechanics. Just that the mechanics are nearly all shit. Shooting is bad, weapon variety is bad, bosses are badly designed, NPC interaction is badly designed, linear questlines, a non-functional base building system (no tracking of item production, limit setting, quick movement/picking up of items and so on), extremely basic and modular building and decoration in general. Very static itemization, utterly uninspired lackluster story, bad animations, lack of maps for hand-crafted areas, really bad space-fights… all bad, the only good system in the game was the modular ship building, everything else was worse then the games before it.
So yes, let’s compare solely mechanics and ignore the engine, because if you make a game with shit content in favor of making it shiny then you’ve failed to provide a proper product. It’s simple.
A week at best, and… so?
Aspects of the code can give optimization ideas and new ways to implement things better. It’s not lost time.
There’s sometimes whole projects where you go ‘Yes, delete it… all of it, we’re starting from scratch’ because a situation came up where something is not working with the intended engine. And if it’s integral you need to start from scratch in a framework which allows it.
‘A battle-plan only lasts until the battle’. You plan ahead and don’t run in wildly… yes… but you also don’t plan yourself to death. It’s a thousands years long known aspect relating to every situation in life, there’s not a ‘black and white’ situation, it’s a full-on gray area to find the golden ratio for it.
If you don’t have access to the code to adjust it you can hardly fix it.
Which Bethesda hasn’t provided with Starfield, which is why the modding community of Bethesda left.
Whoops I guess!
Yes, pre-loaded. I can’t find the source anymore for where it was talked about when the guy laoded the game and had access to the data versus his upload time, the initial upload was on 5th september on Nexus, that doesn’t mean it was finished or not, just accessible. I just remember that it was a big thing about him doing it ‘in a day’.
Nonetheless the biggest amount of time it could’ve taken was from 17th august - access to files - to 5th september, which is still only 19 days at worst, for a single person.
Nobody cared, the functionality was vastly better then it not being buggy. It did what it was supposed to so unless the one provided by Bethesda.
You can release a buggy product as long as it provides you with everything you can work with, you can fix bugs… but if you can’t even work with the given thing it has no value.
Well, nice way to pick up things I haven’t said
Now back to reality.
19 days of work to provide a functional UI.
Bethesda worked on their game for 7 years.
Let’s take away 4 years solely to set up the core systems… nah… let’s go further, 5.
So in 2 years their whole crew didn’t manage it? Nobody? Not a single person?
Who should be fired again?
Not even remotely.
PoE 100 is a longer time then LE takes, at least with the ‘normal route’ and ignoring the many ways since the side implementation of itemized breachstones. You got around 0,6% exp per map commonly at 99, which take 5-10 minutes per map.
How long does 99-100 take in LE currently?
How long did it before the last change?
If that were true, you wouldn’t get games like Dark Pictures or Life is Strange.
Gameplay is very important, but so are graphics. Some players don’t care about it, but others do and won’t play games with outdated graphics. I’ve personally met many people that wouldn’t try D2 (before D2R) because of the bad graphics.
In diablo-clones graphics are less important, but they still matter as well.
Much like if ADOM or ToMe still used the ascii graphics that the first ones used, they would have long been abandoned or played only by a handful of people.
That is because
1- programmers spend way more than 2h programming a day, especially since many work from home.
2- designer/analysts don’t take weeks to decide things.
3- this forum isn’t composed of analysts nor of anyone that actually knows the code the game is using.
If you simply tell a programmer to do something and he has no knowledge of all the underlying things behind the code (as is the case with programming teams where each programmer has a limited knowledge of the overall code), then you regularly run into 1 of 2 situations:
1- you spend a lot of time working on your changes, only to realize later that what you’re doing isn’t feasible, so you scrap all your work and start over, hopefully right this time.
2- you spend a lot of time working on your changes, only to introduce a bunch of bugs in other systems that interacted with the one you were changing.
The time it takes to analyze things is smaller than the time to code them, most of the time.
I’ll even give you an example:
-Recently I had to add a VISA payment with a different provider. That specific provider didn’t allow content in iframes and required a redirect to their site (as opposed to the other ones we already had implemented). Also, they didn’t send the response back synchronously.
I spent 1h analyzing the requirements (we now had to store the purchase somewhere in the database instead of doing it all on the fly), found a method that was already integrated (because it was previously analyzed properly and could be reused for this purpose), then spent about 4h implementing and testing the code.
If I hadn’t spent 1h analyzing it I would have spent 3-4h implementing the code for a solution that wouldn’t work, plus some more time to analyze how to accomplish that, plus another 4h to implement the new code and test it.
And if the already implemented method hadn’t been analyzed previously and created properly to allow this, I would have also spent more time refactoring it.
So a small time analyzing the issue before you actually start working on it (whether you do it yourself in a smaller team or someone does it for you in a bigger team while you work on other things that were previously analyzed) saves development time all around.
You have a very skewed vision of how software companies work.
The decision tree is the easiest part of BG3. You can make it in RPG Maker in a few months, at most.
The thing that took the most time to make in BG3 was all graphics related, especially the cinematics and their decision to have all of them voice-acted with all variations. Because that is the hardest thing to do in a game these days.
You can’t, though. Because even in a case like LE where graphics aren’t extraordinary, you still have to have them and a lot of time has to be spent on it.
CDDA doesn’t spend any time on graphics because graphics don’t matter at all for that specific niche.
But most games have to balance both, because a game with good gameplay and crappy graphics isn’t as successful as a game with medium gameplay and medium graphics.
Yes, when they weren’t properly analyzed. Because if you do spend time analyzing that beforehand, you don’t actually have to waste months.
Your ideology for programming is what led to Duke Nukem Forever.
VtM players didn’t have access to the code either but they still fixed it. Most mods are done by altering the game dlls. Modders datamine the code and make their own changes. So you don’t need access to the code to mod games.
That is because the tolerance for mod bugs is much higher than for released products. If Bethesda had released the exact same solution but riddled with bugs, the community would all go up in arms.
However, since you don’t pay for a mod, it’s ok if it has bugs and you fix it later.
So as a modder you can release a buggy feature and fix it in your own time, whereas Bethesda has to release it working properly.
Are you just trying to deflect stuff at this point?
Before launch barely any build did 1k corruption. After launch, most builds did (and still do mcuh higher corruption than before launch). So obviously the 1-100 time decreased. They adjusted it in 1.0, found it wasn’t enough, adjusted it again. How is that against their game identity from the start?
And naming those 2 is especially underlining my argument.
Dark Pictures and Life is Strange are interactive stories. They provide ‘game elements’ but that’s it.
Their first and foremost aspect is to prove the story, they life and die through the story. They make it replay-able by offering branching storylines. In terms of game elements basically nothing is existing. At most you have time-based choices for answers or quick-time events. Most other elements don’t fit into the genre as it would avert from the storytelling.
Hence since it’s focus is not on the game but something else graphical fidelity has a bigger impact. Albeit we can argue that even then… the story is the most important aspect rather then that, if the story sucks nobody will care about how nice it looks.
I’ll argue that gameplay is vastly more important then graphics.
A shitty looking game with great mechanics will be played.
A fantastic looking game which is atrocious mechanically will not be played. And there’s no exceptions.
And then to not let any mistakes come up. A ‘Game’ and a ‘Interactive Story’ are 2 different mediums.
Game: Gameplay > Graphics > Story
Interactive Story: Story > Graphics > Gameplay
Simply other needs to focus on.
LE is first and foremost a game. The focus lies on the mechanics.
Starfield is first and foremost a game. The focus lies on the mechanics.
Hence lack of focus on making those great causes it to fail.
Repeatedly shown, give me exceptions if it doesn’t apply.
Which is why I said ‘Planning is important’ but ‘Overplanning is deadly’. The second is what many companies do nowadays.
Obviously you need to give someone a overview… but you don’t need to micro-manage it every step along the line.
So, you didn’t go to your team… proclaimed ‘this is the thing I need to do, let’s talk about it’ and then the whole team spent that hour (so 1 hour loss for each person of active coding) together to solve it before actually looking if it was able to be done in the first place?
Sounds like you had a task, saw an issue and solved it on your own.
Which is the whole point I’m speaking about, that this way often in gaming-development is highly hindered nowadays.
Someone did the task, it was documented by the person and you found it. You didn’t need to stop your work to have someone else decide if the code you wrote is even acceptable before you wrote it.
It’s the same with 2 devs, one guy working on a physics engine and one on the UI. The UI guy doesn’t need to care what the physics guy does and when… unless damage numbers are supposed to pop up and bounce off walls or some weird other interaction. He has nothing to do with it at first. Let both of em cook.
Assets are always the things needing the most time.
As mentioned.
And voice-acting is the most expensive besides that.
They are good graphics. They actually uphold to modern standards. They’re nearly as well made as those from Path of Exile… obviously a bit of disparity but… they are good.
Trust me… they do. If you switch to a tileset you know the difference.
I’m solely argumenting that the value of mechanics is higher then graphics. Not that one or the other is useless. Once more… not black and white, gray.
Unless the graphics are glaring or atrocious it’s no issue.
Which game can you provide as an example for great gameplay but crappy graphics which did badly? Marketing taken into consideration as well.
Duke Nukem Forever was people never talking to each other at all. No structure given, and each end result lackluster anyway. It had no charm like the original one and it was not even upkeeping the style properly.
So that’s quite some different issues.
For that you need to have access to the files? I mean… you can’t access the whole code (or otherwise someone could just re-make Starfield completely without DRM) in the first place, so if things are locked in the inaccessible parts how will you write code interacting with it? How will you let that aspect access the one you create?
Obviously it’s datamined… but unlike in Oblivion or Skyrim a huge chunk of formerly accessible code isn’t in Starfield, that’s the point why many things haven’t been modded at all… since people can’t access it.
First of all… Bethesda being a bug-festival is a given thing, nobody bats an eye anymore. Functionality though not being provided is another topic.
It’s expected that Bethesda gives people a buggy but enjoyable product which then gets fixed and improved manyfold by modders. If the modders can’t fix your faults though… well… kinda hard to argument about that then, especially if functionality also isn’t provided.
Yes, more builds reach 1k+… but what about the ones at the lower spectrum? They now need a ton more time.
A disaprity.
Many mediocre players don’t play 1k+, those are still the outliers.
This only works because it’s a small team and I’m in charge of this project. You can’t have programmers going rogue on their own in larger teams and projects, otherwise things start to break.
Cyberpunk 2077. It was actually fun to play, but a few bugs and the whole graphical performance killed it.
No, the biggest problem it had was that it had to switch engines 2 or 3 times. Because it wasn’t planned properly.
That’s already available. How do you think hackers create their hacks? Even LE is hacked. Even a bunch of server-based games are hacked by making a custom server run them.
Starfield can be played offline, so all the files are already in your computer. So there’s no chunk of code missing.
So? Most of the players will never reach level 100, either in LE, PoE or D4. How does that change anything?
The progression rate is based on the people that do reach it. Which is why it’s reasonably smooth until level 80, still okay until level 90 and then it starts to really slow down.
Since level 100 isn’t required for anything, I’m pretty sure that’s the whole point behind the design.
So you can do it but they people underneath can’t?
The framework is there. It’s documented, so they should be able to do it. Obviously depending on effort.
Imagine your worker sitting there and having that situation. Should that person then come to you, ask you how to solve it first before even looking for the answer on his own, then have you sit down and see what the issue is, how to solve it and send him off again? If that person gets the freedom to find the solution on their own, put it in and then simply come by and say (depending on complexity, a 2 day coding endeavor for this aspect obviously is another topic) ‘Hey, I had this issue there and found this solution. Can we do it this way?’
Upon which you only need to check it rather then one person talking or solely waiting while you also search around and 2 peoples time is bound at once.
Yeah, we won’t take into consideration that the studio dropped several promised features without notification before release, the implant mechanic was lazy and boring and only the story itself kept it afloat since it had a good one.
Without the overhaul for the core mechanics with the DLC the game would’ve been ignored. Why? Because it was mechanically not at best ‘ok’ at the time, in no way good.
Baffling enough given that most 3D engines definitely have the ability to support that rather… mediocre shooter.
I imagine there was more of a reason behind that, because that would be mismanagement beyond mismanagement, that should’ve become obvious for people really early. I imagine people mentioning it were likely ignored and it went along instead.
Depends, which ones?
Wall hacks? By reading out the RAM, which is external from the program. You can make a informed decision and hence create a program running reading that out and making it visual for the player.
Aim-bot? Same.
Different from accessing code which isn’t commonly accessible.
Because a mediocre player which was playing regularly was in place to reach 100 before but now will in the same time reach 95 or so. Which is a difference obviously.
For pushing players it feels rather similar though since they can reach higher and need roughly the same amount of time.
Not speaking about requirement or not, it just changed the status quo.
Yes, the people underneath can’t. And that is because they only work with small parts of the project and aren’t aware of how everything is linked. So if they start just doing stuff on their own they will likely break something else in another module of the suite.
The person does what he is told after it’s been analyzed. Unless something is analyzed there isn’t even a task created.
You’re talking about small localized fixes. Those obviously don’t need analyzing. Much like if I tell him that there’s a bug while editing details of something, that has no other impact on things outside of it.
What I’m referring to are things that impact other systems. Which is the relevant thing we’re discussing. Obviously no one believes you need to analyze everything just to correct a typo. That’s unnecessary hyperbole.
You still don’t seem to understand (or want to). NO ONE IS WAITING AROUND!!!. There are hundreds of stuff to do already. If a specific issue hasn’t been analyzed yet, there are lots of stuff to do in the meantime.
That is your opinion. Plenty of people enjoyed the gameplay on launch, myself included. There were a few bugs, but nothing really game-breaking and something to be expected at launch.
The biggest issue it had was the clusterf*** of its graphics on console. It was the main reason why it was removed from the Sony store. And everything that happened afterwards was a consequence of that. If the game had released only on PC it would have had time to fix the bugs and establish itself.
But since it released for consoles at the same time and it had very bad graphics, it got slammed.
It turned out to be a mediocre shooter because it never became what they wanted it to be. After a bunch of engine changes and a bunch of team changes, when it finally landed on Gearbox they just wanted to get it out there without most of the mechanics intended for the game originally.
So if it had been analyzed properly, they would have either started on a better engine or they would have cut down all of the stuff that ended being cut out anyway.
I’m clearly talking about game hacks as in removing DRM.
IT really didn’t, because mediocre players didn’t reach level 100 before either.
But they don’t? You have to check in with your boss one way or another. You make a plan, plan is executed, instead of taking care of every eventuality (only the important ones hence) you then go ahead and to work. Issues arise even if you over-plan after all, over-planning is lost time. So in both cases you check, if the check is good then implement it.
If it’s not… both the coder learned something about the overall structure (good for the company as that person can autonomically do those things next time) and re-do it.
Don’t go this route and you get people not understanding what they’re working on, got a great plan, issues arise, time to solve them rises as they don’t understand why they even arose since they didn’t understand the full construct better.
Exactly!
Finally
We’re getting slowly onto the same page.
Some things don’t interact with others, they’re localized. Not everything is integrated into the deepest depths of the engine’s hell.
And nonetheless… sadly… it’s done to a vast degree beyond the needed measures in several companies nowadays, needing to get your actions admissed to then have them getting admissed from another space which regulates all the admissible changes and has no overview of the product which causes massive waiting times.
Car not showing up when called at all, car flying away and unable to be re-called, several softlocks by auto-saving in conversations which stuck in progression (I had that personally even, 5 times on release).
Then missing content promised for release, like the metro and real-time moving to get an overview of the city for example.
Yep, for some it works, many also do that through changing values in RAM which gets send to the respective segment though, tricking it as a successful check.
Not for me it didn’t, the graphics were awesome. Though I do have sympathy for people on the previous gen consoles 'cause that did look legitimately bad. But on my 2070 super? It looked good & ran decently.
It is, the devs have said so.
If people could have travelled back in time to let everyone know, sure, I guess? But I’m pretty sure it wasn’t ignored, at all. Unless you’re using that term in a very non-standard way 'cause I really don’t get what you mean. For me, Cyberpunk on release was a good game, looked pretty, great story & I enjoyed playing it through several times.
Yes, that was my point. I did specifically say that if they had released only on PC they would have had time to fix the bugs and wouldn’t have tanked so badly. It was the crappy graphics in consoles that really sank the game.
