OpenTTD and patch packs

Forum for technical discussions regarding development. If you have a general suggestion, problem or comment, please use one of the other forums.

Moderator: OpenTTD Developers

KeldorKatarn
Transport Coordinator
Transport Coordinator
Posts: 274
Joined: 13 Apr 2010 21:31

OpenTTD and patch packs

Post by KeldorKatarn »

Quite frankly at this point I see no point in even discussion anything regarding development in this forum anymore. I've watched this literally for 5 years and everytime someone suggests a radical improvement or design change the response is overwhelmingly "Can't be done because XYZ" "Someone said blah X years ago so it won't work" "Nah too much work" "Nah need to remain compatible with the stuff from 10 years ago"

If I'm supposed to be super frank here: My version is crap since it's not made for compatibility or cross platform support and doesn't care ebout discarding entire features if I feel they are crap.

But JGR's version is by now so far superior and ahead of Vanilla that I'm frankly asking myself why Vanilla is even a thing anymore. In my book it lost it's entire reason for existing. It's WAY behind in code quality, performance, feature support and I'd even suspect player base at this point. There haven't been any major advances in years, the last probably five or so releases where bugfixes and translations if anything at all.

Regarding "we need to support all those newGRFs forever" well then get ready to die a slow death. What is being done in terms of newGRFs? New road types is MAYBE a light on the horizon but it's taking years because people have a hard time figuring out how to put it in the game, which already shows how incredibly complicated it is to add functionality to the NewGRF API.

What else is being done? There's 2 relevant industry sets of which one is only slightly improved, the other has been unchanged for years. There's a couple major vehicle sets who haven't been changed in years either. THere's maybe two or three terrain sets that are relevant and a couple of houses that haven't been worked on in five years. THe only recent development on anything really big is the Auz stuff recently. What else is there?

It's not exactly like NewGRF development is a super busy and attractive community. It's hard, it's awkward, it's something most people won't even take a look at unless they're complete OpenTTD fanatics. All the while other games have a thriving modding community because it's INREDIBLY easy to add some stuff to the games.

NewGRF development has been dying a slow death for the last 5 years if not longer and the vanilla game itself is borderline irrelevant at this point.

There's an athmosphere of "let's not change anything or it'll break", "can't be done" and "the way it was done 30 years ago is the way to go" here that just stalls everything.

When people suggested removing the 62 NewGRF limit 4 years ago!!! there was a massive counter strike telling them 50 different reasons why that couldn't be done. I finally looked into it and got rid of it in 15 minutes of work.

People here THINK something can't be done while 99% have absolutely no clue how modern games work or even how old games worked for that matter and just block off any kind of enthusiasm any dev has that comes to the forums.

Whenever there's someone working on a patch, instead of giving him support and fixing his bugs to take advantage of the rest of the work, it's all "Nah, your code style doesn't match our code base" and then the patch dies a slow death until MAYBE a patchpack picks it up.
And that literally because of a problem that should be a matter of a hotkey-press in an IDE to automatically style the code. That people are doing code style reviews in 2018 is MINDBLOWING to me. But what do I expect from people that I know for a fact often write their code in a text editor and try to make RUNTIME code run on 50 compilers at once, as if this was a code library, when for an executable it only matters that exactly ONE compiler per platform compiles it, like every single development company out there does it. Compiler compatibility matters for libraries that will be used for all kinds of things. This is a game that's supposed to run on 3 platforms, if one windows compiler works, why the heck does it matter if another one compiles it.

What's the result of this? Developers that complain about curly braces being in the wrong place while modern language features which would improve the readability and often also performance of the code in a major way are rejected because some obscure compiler nobody gives a crap about can't understand what a lambda is or doesn't support some STL feature or whatever the issues are.

So at this point I'm gonna leave the conversation because it literally ends the same way everytime so what's the point.
User avatar
acs121
Tycoon
Tycoon
Posts: 1956
Joined: 03 Nov 2017 18:57
Location: Courbevoie, near Paris, France

Re: JGR's Patch Pack

Post by acs121 »

KeldorKatarn wrote:-snip-
For the "let's not change anything" part, it is actually a real thing. Andythenorth himseld said it on the first line of the NotRoadTypes topic :
"Some said it couldn't be done. Some said it wouldn't be done. Some said it shouldn't be done."
There's this kind of ambiance in about 99% of the people here. The 1% left is the fantastic world of users creating patches.
In the end, people really don't like new things, but when they are done, suddenly they like it. Same example : NotRoadTypes.
User avatar
andythenorth
Tycoon
Tycoon
Posts: 5656
Joined: 31 Mar 2007 14:23
Location: Lost in Music

Re: JGR's Patch Pack

Post by andythenorth »

KeldorKatarn wrote:But JGR's version is by now so far superior and ahead of Vanilla that I'm frankly asking myself why Vanilla is even a thing anymore. In my book it lost it's entire reason for existing. It's WAY behind in code quality, performance, feature support and I'd even suspect player base at this point.
Respectfully (to JGR, because the integrity of JGR Patchpack is a remarkable achievement), I dislike most of what's been added. I've got JGR PP checked out and compiled, and I find most of the new stuff confusing.

Specific to NewGRF and roadtypes, you seem to conflate the NewGRF spec with gameplay issues. The blockers to roadtypes have been, or are
  • the fork doesn't have clean git history and won't be accepted for review without rebase or refactor
  • designing a spec is very hard; I can't overstate how much that was a blocker to roadtype development
The NewGRF implementation of roadtypes is relatively simple, and mostly follows railtypes. The nml commits were not many lines (updating the example grfs took longer than updating nmlc). grfcodec 'just worked'.

You posted a lot of stuff about how posting stuff is a waste of your time. I'm sad that you feel that way. I'd like to have your contributions, but it's hard to discuss because you've very keen to point out devs don't know what they're doing. But you don't seem very familiar with what you're talking about inside OpenTTD. So it's kind of hard to see where to go from there.

"OpenTTD is dying" has never been a thing any devs are concerned about afaik. "Is it fun to work on?" is much more interesting. Right now, with the github move, new CF, and a move to accepting pull requests, yeah it's pretty fun to work on again.

Hope I can persuade you. If I fail, sorry for that :)
User avatar
acs121
Tycoon
Tycoon
Posts: 1956
Joined: 03 Nov 2017 18:57
Location: Courbevoie, near Paris, France

Re: JGR's Patch Pack

Post by acs121 »

The only other patchpack that makes vanilla losing any reason to exist too is Spring 2013 - though there are few things in this one too that bug me out.
User avatar
Erato
Chief Executive
Chief Executive
Posts: 740
Joined: 25 May 2015 09:09
Location: The Netherlands

Re: OpenTTD and patch packs

Post by Erato »

The reason vanilla is still relevant is because JGR and other patch packs just don't cut it for a lot of people. I can't stand how JGR behaves. Its way too buggy and laggy in my experience. But that might be just me.
KeldorKatarn wrote:What else is being done? There's 2 relevant industry sets of which one is only slightly improved, the other has been unchanged for years. There's a couple major vehicle sets who haven't been changed in years either. THere's maybe two or three terrain sets that are relevant and a couple of houses that haven't been worked on in five years. THe only recent development on anything really big is the Auz stuff recently. What else is there?
Lately there has been more newgrf development than ever before. With NML adding new vehicle sets has become so much easier; the last 2 years has brought about as many relevant newgrfs. MMT/RIMS, BRTrains/BORTS, NARS addon, all the smaller Japanese ones, Vactrains, etc. The difference between now and then is that now, most projects are done by 1 or 2 people and the communication for large projects like BORTS are no longer done through the forums. Only a few devs keep the forums up to date very often.

Development has actually sped up, not slowed down. NML projects the size of what used to be major newgrfs can be done by only one or a two developers, - although this might take a bit longer - which is what newer newgrf projects tend to have. Only BRTrains/BORTS seem to be an exception to this. The reason for this seems to be the large amount of british people in the NewGRF community.

The Auz series is exceptional of course in that the guy has been working on a set of 6 newgrfs in 6 different forum threads and posts regular updates to the forums.
KeldorKatarn wrote:It's not exactly like NewGRF development is a super busy and attractive community. It's hard, it's awkward, it's something most people won't even take a look at unless they're complete OpenTTD fanatics. All the while other games have a thriving modding community because it's INREDIBLY easy to add some stuff to the games.
NML is extremely easy so, especially lately, we see more and more people working on new newgrfs. Other games with larger modding communities are just newer and more popular than OpenTTD. It doesn't make much sense to compare these games with OpenTTD.

Besides NML there is NFO, and NFO is all HEX and stuff, which is super difficult. Things that can only be done in NFO, like stations, therefore take a super long time to develop. I'll give you that. But for vehicle sets this isn't true at all.

And if you're talking about the slow development of newgrfs like Dutch Trainset and Japanset; people come, people go. The devs have probably moved on already. If the original developer of Japanset was 20 at the time, they likely have kids old enough to play OpenTTD themselves now. They might not have the time to work on a project they might have already forgotten about.
No pics no clicks. Seriously.
ImageImageImageImageImageImage
ino
Traffic Manager
Traffic Manager
Posts: 152
Joined: 09 Apr 2017 14:58

Re: OpenTTD and patch packs

Post by ino »

As a happy JGR's Patch Pack user who also occasionally modify the source code to suite my play style, I actually kind of agree with the philosophy of OpenTTD devs here.

Having hacking around JGR's source code, I find that the entire things looks very fragile. I wholeheartedly agree with andythenorth comment that JGR's Patch Pack integrity is very amazing. For example, time handling is extremely complicated due day length, with too many variables storing time in various formats. I love JGR's, and I see myself doing so in foreseeable future, but that's because I like new features and I know that we are unlikely to see most of the features in trunk in its current state.
Eddi
Tycoon
Tycoon
Posts: 8258
Joined: 17 Jan 2007 00:14

Re: OpenTTD and patch packs

Post by Eddi »

KeldorKatarn wrote:But JGR's version is by now so far superior and ahead of Vanilla that I'm frankly asking myself why Vanilla is even a thing anymore. In my book it lost it's entire reason for existing.
i don't know what you're trying to say here, but if "vanilla" were to move as fast and careless as a patchpack, it would have died a long time ago. In fact, it almost did once, which is why the developers became so conservative. during that time, a dozen patchpacks have come and gone, yet "vanilla" keeps doing its core job, of staying a solid base for someone to come along, pick up the broken pieces of a patchpack that fell apart, and build a new one. leaving out patches that don't work anymore.

questioning the purpose of "vanilla" is like questioning the foundation of a house, when all you want to do is put fancy blinking lights on it
What's the result of this? Developers that complain about curly braces being in the wrong place while modern language features which would improve the readability and often also performance of the code in a major way are rejected because some obscure compiler nobody gives a crap about can't understand what a lambda is or doesn't support some STL feature or whatever the issues are.
that is just pure nonsense...

first, the purpose if the code styie is to weed out wannabe programmers that have no clue. if they cannot follow a code style (any code style), then likely something is wrong at the deeper core of the patch as well. and if they can follow a codestyle, correcting it for our code style is most likely trivial.

and second, openttd is very cross platfrom, and has been for years. dropping compilers for one reason or another has definitely been done, and very likely will be done in the future, but it's not a decision to be done lightly, as it potentially cuts out a large part of the community.
Last edited by Eddi on 25 Apr 2018 02:04, edited 1 time in total.
User avatar
wallyweb
Tycoon
Tycoon
Posts: 6102
Joined: 27 Nov 2004 15:05
Location: Canada

Re: OpenTTD and patch packs

Post by wallyweb »

Many years ago I was bored. I discovered the internet and I had a CD that had a Transport Tycoon promo. I wanted more. Then I discovered search engines and I found this really neat article on something called TDDPatch, something that extended the capabilities of Transport Tycoon Deluxe which was proprietary and could not be directly altered. That was it! I was hooked. Then along comes the forums. I watched the patch grow. It was a bit cumbersome. A config file had to be modified to what one wanted the game to do. A newgrf file had to be cobbled together to invoke all those GRFs one wanted to use. But the fun was endless. Then, once upon a time, a new section opened in the forums. It was about something called OpenTTD. It was the beginning of an open source clone of TTD that adopted many of the features of the patch. It was slow to grow and it took several years to achieve a somewhat full functionality of the patch. Meanwhile the patch continued to grow, albeit somewhat slower. The achievement of OpenTTD was quite remarkable in that it supported the GRF concept and many of those old TTDPatch GRFs remained quite playable. That was the secret of the transition from one platform to the next. It was virtually seamless and yet the new kid on the block was very different behind the scenes. The OTTD developers had preserved the spirit of TTDPatch. Now we have new stuff. C and C++ have grown into other programing languages that offer efficiencies of scale and speed. I hope you are reading this KeldorKatarn because you are quite correct in many of your thoughts. History will repeat itself. It was difficult for several of those TTDPatch developers to transition to OpenTTD. Many of them moved on to other things in life such as new careers and family, but they took something with them ... knowledge and experience. It is very tempting to blame the current developers, but they are comfortable with what they have achieved and have no need to transition. You, on the other hand, represent the fresh blood, the new ideas that probably need to find a new path, and you have a new advantage ... You are drawing from an open source project. You don't have to look over your shoulder. Don't fight OpenTTD. Build on it. Blaze your own path. It will be slow at first, but along the way others will join you. It is because of folks such as yourself that all those years ago blazed their own path, but they kept the spirit of the original, the spirit that so ensnared all of us. Go for it lad. Show us what you can do. Just don't burn the bridges and be sure to keep the spirit. :D
User avatar
acs121
Tycoon
Tycoon
Posts: 1956
Joined: 03 Nov 2017 18:57
Location: Courbevoie, near Paris, France

Re: OpenTTD and patch packs

Post by acs121 »

wallyweb wrote:Many years ago I was bored. I discovered the internet and I had a CD that had a Transport Tycoon promo. I wanted more. Then I discovered search engines and I found this really neat article on something called TDDPatch, something that extended the capabilities of Transport Tycoon Deluxe which was proprietary and could not be directly altered. That was it! I was hooked. Then along comes the forums. I watched the patch grow. It was a bit cumbersome. A config file had to be modified to what one wanted the game to do. A newgrf file had to be cobbled together to invoke all those GRFs one wanted to use. But the fun was endless. Then, once upon a time, a new section opened in the forums. It was about something called OpenTTD. It was the beginning of an open source clone of TTD that adopted many of the features of the patch. It was slow to grow and it took several years to achieve a somewhat full functionality of the patch. Meanwhile the patch continued to grow, albeit somewhat slower. The achievement of OpenTTD was quite remarkable in that it supported the GRF concept and many of those old TTDPatch GRFs remained quite playable. That was the secret of the transition from one platform to the next. It was virtually seamless and yet the new kid on the block was very different behind the scenes. The OTTD developers had preserved the spirit of TTDPatch. Now we have new stuff. C and C++ have grown into other programing languages that offer efficiencies of scale and speed. I hope you are reading this KeldorKatarn because you are quite correct in many of your thoughts. History will repeat itself. It was difficult for several of those TTDPatch developers to transition to OpenTTD. Many of them moved on to other things in life such as new careers and family, but they took something with them ... knowledge and experience. It is very tempting to blame the current developers, but they are comfortable with what they have achieved and have no need to transition. You, on the other hand, represent the fresh blood, the new ideas that probably need to find a new path, and you have a new advantage ... You are drawing from an open source project. You don't have to look over your shoulder. Don't fight OpenTTD. Build on it. Blaze your own path. It will be slow at first, but along the way others will join you. It is because of folks such as yourself that all those years ago blazed their own path, but they kept the spirit of the original, the spirit that so ensnared all of us. Go for it lad. Show us what you can do. Just don't burn the bridges and be sure to keep the spirit. :D
People aren't destroying OpenTTD. They rather don't want to build on it, and it is the problem, and the origin of the creation of this topic.
User avatar
wallyweb
Tycoon
Tycoon
Posts: 6102
Joined: 27 Nov 2004 15:05
Location: Canada

Re: OpenTTD and patch packs

Post by wallyweb »

acs121 wrote:People aren't destroying OpenTTD. They rather don't want to build on it, and it is the problem, and the origin of the creation of this topic.
Nobody said anything about destroying OpenTTD. Sometimes there are feature requests that fall outside the developers' direction. This is why the concept of patches was improvised. A similar thing happened to TTDPatch where the Patch's developers found many new feature requests too difficult to implement. TTDPatch was not destroyed either. I continue to use it on occasion, but OpenTTD has evolved to offer more, and the day will come when OpenTTD itself will be passed on to something new. We have already seen some of this in cirdan's New Map Features. We used to live in caves and draw on the walls. Now we live in cities, but we still draw on the walls. :wink:
User avatar
JGR
Tycoon
Tycoon
Posts: 2557
Joined: 08 Aug 2005 13:46
Location: Ipswich

Re: OpenTTD and patch packs

Post by JGR »

To clarify things as regards my patchpack, I am not trying to take over from trunk.
My patchpack isn't aimed at beginner players.
I don't expect any individual player to use all the extra features, however each of the extra features is intended to be useful to somebody.
I am extending a subset of the game functionality/gameplay in a way which is useful for players like me.
From my point of view it doesn't make sense to choose to do so in a non-robust or non-reliable way.

There are large chunks of the game/gameplay which I don't use personally or touch in the patchpack, these are still important for (separate subsets of) players.
Ex TTDPatch Coder
Patch Pack, Github
User avatar
acs121
Tycoon
Tycoon
Posts: 1956
Joined: 03 Nov 2017 18:57
Location: Courbevoie, near Paris, France

Re: OpenTTD and patch packs

Post by acs121 »

wallyweb wrote:
acs121 wrote:People aren't destroying OpenTTD. They rather don't want to build on it, and it is the problem, and the origin of the creation of this topic.
Nobody said anything about destroying OpenTTD. Sometimes there are feature requests that fall outside the developers' direction. This is why the concept of patches was improvised. A similar thing happened to TTDPatch where the Patch's developers found many new feature requests too difficult to implement. TTDPatch was not destroyed either. I continue to use it on occasion, but OpenTTD has evolved to offer more, and the day will come when OpenTTD itself will be passed on to something new. We have already seen some of this in cirdan's New Map Features. We used to live in caves and draw on the walls. Now we live in cities, but we still draw on the walls. :wink:
I am just saying that most people here actually don't want to draw on the walls. This is why KeldorKatarn posted the first message :
Quite frankly at this point I see no point in even discussion anything regarding development in this forum anymore. I've watched this literally for 5 years and everytime someone suggests a radical improvement or design change the response is overwhelmingly "Can't be done because XYZ" "Someone said blah X years ago so it won't work" "Nah too much work" "Nah need to remain compatible with the stuff from 10 years ago"

[...]

There's an athmosphere of "let's not change anything or it'll break", "can't be done" and "the way it was done 30 years ago is the way to go" here that just stalls everything.

When people suggested removing the 62 NewGRF limit 4 years ago!!! there was a massive counter strike telling them 50 different reasons why that couldn't be done. I finally looked into it and got rid of it in 15 minutes of work.

People here THINK something can't be done while 99% have absolutely no clue how modern games work or even how old games worked for that matter and just block off any kind of enthusiasm any dev has that comes to the forums.

Whenever there's someone working on a patch, instead of giving him support and fixing his bugs to take advantage of the rest of the work, it's all "Nah, your code style doesn't match our code base" and then the patch dies a slow death until MAYBE a patchpack picks it up.
Read those lines and read them again. It's exactly how people are acting generally. Diagonal roads, diagonal crossings, diagonal stations, sloped stations, Newstops, etc, has never been implemented because of this ambiance on the forums. If Wolf01, frosch and andythenorth hadn't created NotRoadTypes themselves, no one would have been here to create it. Just because people don't want to change from almost 25 years ago.
Last edited by acs121 on 25 Apr 2018 13:03, edited 1 time in total.
User avatar
andythenorth
Tycoon
Tycoon
Posts: 5656
Joined: 31 Mar 2007 14:23
Location: Lost in Music

Re: OpenTTD and patch packs

Post by andythenorth »

acs121 wrote:If andythenorth, Eddi and Alberth hadn't created NotRoadTypes themselves
Please stop spreading misinformation.

NRT was created primarily by Wolf and frosch. I was involved, and other people also helped.

The rest of your post I don't understand.
User avatar
Redirect Left
Tycoon
Tycoon
Posts: 7238
Joined: 22 Jan 2005 19:31
Location: Wakefield, West Yorkshire

Re: OpenTTD and patch packs

Post by Redirect Left »

I haven't played vanilla for a long time, easily in the 'years' sort of timescape, and always play JGRs. I also recommend JGRs patchpack to people who vocie an interest in the game. As its far more impressive than the standard vanilla OpenTTD at this point. I've introduced a few new players into OpenTTD directly through JGRs patchpack.

This is mostly due to the quick development and fixing turnaround. I and the majority of people I know would much rather play a game that moves, upgrades and fixes swiftly with potential bugs, than something that changes often minorly and fairly slowly but (probably) bug free.
Image
Need some good tested AI? - Unofficial AI Tester, list of good stuff & thread is here.
User avatar
acs121
Tycoon
Tycoon
Posts: 1956
Joined: 03 Nov 2017 18:57
Location: Courbevoie, near Paris, France

Re: OpenTTD and patch packs

Post by acs121 »

Redirect Left wrote:I haven't played vanilla for a long time, easily in the 'years' sort of timescape, and always play JGRs. I also recommend JGRs patchpack to people who vocie an interest in the game. As its far more impressive than the standard vanilla OpenTTD at this point. I've introduced a few new players into OpenTTD directly through JGRs patchpack.

This is mostly due to the quick development and fixing turnaround. I and the majority of people I know would much rather play a game that moves, upgrades and fixes swiftly with potential bugs, than something that changes often minorly and fairly slowly but (probably) bug free.
Vanilla is not bug free, but it has way less bugs than JGR - though it's been a lot of time i haven't played vanilla again. IIRC in 1.8.0 they added a feature from JGR's : total profit this year and last year for a vehicle type, a group, etc.
User avatar
wallyweb
Tycoon
Tycoon
Posts: 6102
Joined: 27 Nov 2004 15:05
Location: Canada

Re: OpenTTD and patch packs

Post by wallyweb »

acs121 wrote:Read those lines and read them again. It's exactly how people are acting generally. Diagonal roads, diagonal crossings, diagonal stations, sloped stations, Newstops, etc, has never been implemented because of this ambiance on the forums. If Wolf01, frosch and andythenorth hadn't created NotRoadTypes themselves, no one would have been here to create it. Just because people don't want to change from almost 25 years ago.
I did read those lines ... several times beginning from when they were first posted by KeldorKatarn.

All those features that you cited are excellent suggestions. Yes, there has been resistance from the developers, but the fault does not lie with only them. The underlying problem is that the ideas have never been properly sold to them. All they have heard from the community is "You should do this and you should do that". I know this because I was the first one to propose newobjects for TTDPatch. I had to be well reasoned in my thoughts and eventually eis-os relented and took up the challenge and after much trial and error he was able to make it work. Note that he was not an OTTD developer. That was another challenge. The OTTD devs were reluctant at first. Eventually there were enough requests from the community and after seeing from TTDPatch how it was possible Rubidium took it on and he did and excellent job. He actually was able to improve upon the TTDPatch version. Note that I was never confrontational. I concentrated on being logical in my presentations. Today we have an advantage in that we have an excellent OTTD patch process so that those desiring new features can create a patch to work out any bugs and demonstrate the values of their suggestions. Not all of them make it into the nightlies and then into trunk, but they do often find a home with the likes of JGR and cirdan.

Before I retired I was a salesman. The secret is not to demand. It is to persuade. If you want something to happen, you have show value to the person who can make it happen. This takes time ... Sometimes a very long time. By the way, I should ask ... Have you tried Bridges Over Stations in cirdan's New Map Features?
User avatar
acs121
Tycoon
Tycoon
Posts: 1956
Joined: 03 Nov 2017 18:57
Location: Courbevoie, near Paris, France

Re: OpenTTD and patch packs

Post by acs121 »

wallyweb wrote:
acs121 wrote:Read those lines and read them again. It's exactly how people are acting generally. Diagonal roads, diagonal crossings, diagonal stations, sloped stations, Newstops, etc, has never been implemented because of this ambiance on the forums. If Wolf01, frosch and andythenorth hadn't created NotRoadTypes themselves, no one would have been here to create it. Just because people don't want to change from almost 25 years ago.
I did read those lines ... several times beginning from when they were first posted by KeldorKatarn.

All those features that you cited are excellent suggestions. Yes, there has been resistance from the developers, but the fault does not lie with only them. The underlying problem is that the ideas have never been properly sold to them. All they have heard from the community is "You should do this and you should do that". I know this because I was the first one to propose newobjects for TTDPatch. I had to be well reasoned in my thoughts and eventually eis-os relented and took up the challenge and after much trial and error he was able to make it work. Note that he was not an OTTD developer. That was another challenge. The OTTD devs were reluctant at first. Eventually there were enough requests from the community and after seeing from TTDPatch how it was possible Rubidium took it on and he did and excellent job. He actually was able to improve upon the TTDPatch version. Note that I was never confrontational. I concentrated on being logical in my presentations. Today we have an advantage in that we have an excellent OTTD patch process so that those desiring new features can create a patch to work out any bugs and demonstrate the values of their suggestions. Not all of them make it into the nightlies and then into trunk, but they do often find a home with the likes of JGR and cirdan.

Before I retired I was a salesman. The secret is not to demand. It is to persuade. If you want something to happen, you have show value to the person who can make it happen. This takes time ... Sometimes a very long time. By the way, I should ask ... Have you tried Bridges Over Stations in cirdan's New Map Features?
They not only had resistance only from the developers, but also from the patchers. I did see many, many discussions about all the features i cited, even some from 2003 or 2004. There was someone that proposed diagonal crossings some years ago now and again, he got a massive counter-strike telling him "No needs to be compatible with this roadset blahblah" "No it would cause this this and that" "Kill yourself". And later on, a patcher did make code for diagonal crossings, that worked !

There are plenty of great patches (Infrastructure Sharing, New Map Features, etc) that never made it into the nightlies, even for one of the features though they work perfectly (Infrasharing is a great example).
User avatar
JGR
Tycoon
Tycoon
Posts: 2557
Joined: 08 Aug 2005 13:46
Location: Ipswich

Re: OpenTTD and patch packs

Post by JGR »

On the issue of bug rates in patchpacks and trunk. The rate at which bugs are added and the number of unfixed bugs in any given release are both significantly higher in patchpacks than in trunk, this can be around an order of magnitude difference. My patchpack is no exception.

Other than the obvious reasons of lots of experimental and not fully polished features being added, a key problem with patchpacks that I observe is that there is negligible testing by non-developers between releases.
As the developer of my patchpack I obviously test intermediary changes between releases, but I can only test a very small space in the total state space of feature combinations, playing styles, GRFs, etc.
Often I get bug reports immediately after doing a release as a player has used a feature in a way which I would never do myself, or which doesn't really make sense but still needs to be handled gracefully.
With trunk, there is a non-zero set of users who use nightlies and can notice some of these issues before the next release.
acs121 wrote:There are plenty of great patches (Infrastructure Sharing, New Map Features, etc) that never made it into the nightlies, even for one of the features though they work perfectly (Infrasharing is a great example).
Infrastructure sharing as implemented in the patches in the appropriate thread does not work "perfectly".
Every non-trivial 3rd-party patch which I've merged has required further work in some form or another.

It is worth noting that adding a feature to trunk effectively implies a commitment to indefinite future support and backwards savegame compatibility. This is a significant disincentive to merging potentially problematic features.
Patchpacks can and do drop features (though I try to avoid this).

Most users only use a straightforward subset of the features in trunk, and aren't going to have much use for the fancier things in patchpacks which are intended for advanced/experienced players.
wallyweb wrote:...All those features that you cited are excellent suggestions. Yes, there has been resistance from the developers, but the fault does not lie with only them. The underlying problem is that the ideas have never been properly sold to them. All they have heard from the community is "You should do this and you should do that"...
In terms of selling ideas, working, debugged code has a currency value orders of magnitude higher than suggestion posts.
Developing and integrating features takes significant time and effort, neither of which are "free", even if the developers aren't being paid in fiat currency for it.
Ex TTDPatch Coder
Patch Pack, Github
Eddi
Tycoon
Tycoon
Posts: 8258
Joined: 17 Jan 2007 00:14

Re: OpenTTD and patch packs

Post by Eddi »

acs121 wrote:And later on, a patcher did make code for diagonal crossings, that worked !
you must have really strange definitions of "worked". Because the original implementation of diagonal crossings (from like 10 years ago) had significant gameplay problems (like spurious pieces of crossings staying blocked while the train had already left).

and i myself once tried to solve that problem, which "worked" for my personal use, but was definitely not ready for the general public
Post Reply

Return to “OpenTTD Development”

Who is online

Users browsing this forum: Argus and 15 guests