Favorite Source Port? (Multiple Choice Poll)

[HorseJockey]

GZDoom, PRBoom+ and Woof! are my mains. GZDoom for my projects, PRBoom+ and Woof! if I am doing more vanilla playthroughs.

20 hours ago, Graf Zahl said:

Indeed. The problem has long been solved. IMO the actual failing is that the launchers do not get sufficient promotion.

I use ZDL and Doom Launcher personally, has there ever been talk of "bundling" a launcher with GZDoom to help with said promotion or would that just muddy the waters? 

[Gez]

18 hours ago, segfault said:

My problem here is that the simple fact a launcher is necessary beyond "boot up the IWAD" is, in fact, a UX failure. I think having multiple different frontends (just like how we have an entire ecosystem of Doom source ports) is great, but it doesn't really solve the problem I have when I want to play Doom with friends who normally don't play Doom, don't know what IWADs are, and don't want to spend hours troubleshooting and figuring out what is, to them, arcane nonsense in order to play a video game.

Oh hey, let me tell you about Skyrim modding...

 

 

[Lippeth]

1 hour ago, Jizzwardo said:

Out of curiosity, what does DSDA take away that makes you prefer PrBoom+ over it? Far as I can tell DSDA is literally identical, except it has more features. You say you'd rather use fixed save points, but you can, rewind is just an extra option you have.

DSDA has more features but since they seem to be tailored for a speed running crowd, it can all feel like bloat to a more casual player. I'm all for more options but when I know that I don't want to use the options that one port offers, I opt for the option with less options.

 

1 hour ago, Jizzwardo said:

And from that, what does PrBoom+ have that DSDA does not have?

Colored blood from what I can see. I know it's a trivial feature but if given the option I usually tick it on.

[indigotyrian]

56 minutes ago, Gez said:

Oh hey, let me tell you about Skyrim modding...

 

Oh hey what's this? An in-game mod loader? With in-game mod browsing, downloading, and management? Exactly like I was talking about?

[Alex S.]

Proudly part of the minority who doesn't use source ports. I have nothing against them, just personal choice is all.

[ＤΞＬＴΛ]

GZDOOM is my favorite due to it's reliability and great versatility when it comes to modding.

[ducon]

Where are ZDoom and GZDoom? In Uranus?

[3t0]

20 hours ago, segfault said:

I'm going to ignore the factually incorrect claim that Windows is a "dead platform" as well as the incredibly condescending tone you're taking with me for daring to assert that maybe playing Doom WADs can be easier for other people who don't know a whole lot about computers to address something else: At no point have I ever mentioned any specific operating system. This is a problem whether the end user is running Windows, MacOS, or any flavor of Linux out there.

Sorry if I came off as condescending - it often happens to me when speaking/writing in English - and yeah I am grumpy guy. But that was not my intent. Trust me I mean you no harm. You don't need to be afraid of me either :). I love you, your avatar is beautiful, you are cute. I mean it, honestly. Warning, now I will follow in my usual tone (you can not learn old dog new tricks (that easy)), don't take it personally (I MEAN IT):

 

But yes, I do believe that "playing Doom WADs can be easier for other people who don't know a whole lot about computers" is unimportant metric. And I honestly believe that people "who don't know a whole lot about computers" really have no business playing Doom + WADs on a computer, period. They have consoles for that, there it is easy for "causal normies", as they are called, by us serial condescension offenders, to pick a Doom icon and play, you don't need anything to learn.

 

And also, by now, I am very strong believer in mandatory Computer Driving Permits worldwide.

 

Truth is that I don't know how Steam Doom is packaged (so maybe that changes point a tiny bit) as all my games are from GOG and some physicals, but... The fact that to actually play "original" Doom you have to either play it on retrocomp, play it in DOSBox or play it on VM, or if you gave up on "original" way, you have to use source port ... all of those choices immediately disqualify any pandering to "causual users", because all those things I listed are anything but casual level stuff to get into. While you are at it, you can easily learn whole lot (as I said, even on Windows, setting up shortcuts for your wad combinations is trivial (but not casual)) of other things and you will be much better off. There is this guy who actually did excellent video about his tiny launcher and the way he runs doom using batches. So it explains both concepts. In my eyes it could be easily followed by talented 6 or normal 8 year old (but maybe I am too off?).

 

Funnily, even if anecdotal evidence, Doom always had such effect on people since it came in 1993 - drawin them "in" - why remove it? This UX (I hate that fad crap) would remove it. By this metric, Doomsday engine should be your go to port, which does not seem to be the case - it has everything you need in-engine (and sadly besides some few interesting features, it lacks everything else). It also shows how much adding this UX GUI créme stuff affected the project throughput, compared to adding these other features we might have wanted.

 

Then, these things need support code wise (embedded WAD pickers), and there is this one thing about code: each line not written is infinitely more valuable than every line existing. Non written code is the fastest, least buggy code in (non)existence. You don't have enough of that code, and you always want the most of it in your project. Ideally your project should be just couple of most necessary lines. Not more.

 

Last, I understand Graf's point looking at it from older dev view, he might not be that interested in immediate guis, and it would still require writing the directory traversal and picking layers. Might not be worth it the effort to get into, there are other features worth attention.

 

Finally, I think we all are continually derailing perfectly good thread about ports usage, and I am too lazy to locate the culprit who is responsible for this transgression (but I still believe this person did much worse offense than is my condescension). It would be cool if admin came and split the thread into two - "Embedded vs external Doom Launchers" and original "Ports usage". I remember it used to be done in the past in forums like these. This way we can have a cake and eat it too. @admin? good idea?

 

P.S.: Regarding the "windows being dead platform" I can present my thesis if you care, which I guess you don't. Still, if you want I will make a thread for that. It's not important for me to persuade you, but you might gain some insights which I believe you don't have. Of course, only if you care.

[3t0]

23 hours ago, Graf Zahl said:

The simple reason why this is not an option is that GZDoom does not use SDL on either Windows and macOS. Both platforms are served by native backends. 

I saw references to SDL in the code, but never dive/dove deep into GZD. I was not aware that windows and mac use native GUI either. Thanks and disregard my suggestion.

[aRottenKomquat]

I mostly stick to GZDoom. It works well and as a mapper I really like the engine improvements and new features. After that, I guess Crispy Doom would be my next choice, but really only to check vanilla compatibility without having to open up DOSBox.

 

On the cross-platform UI topic--which I only kind of skimmed--I recently found a decent cross-platform UI as part of my day job. It's called Avalonia UI, runs on .NET 6, and is inspired by WPF which means it's a mix of XAML and C# with databinding features.  I've had very good success using it on both Windows and Linux. It technically supports Mac but it sounds like Apple shenanigans somewhat handicap that, though I haven't tried as we have no Apple-using customers in our industry and I don't use Apple products in my private life. Every other cross-platform UI I've used professionally was such a frustrating experience that we gave up on cross-platform UIs. (I used WPF and Silverlight extensively over my career, which means I am quite biased towards that type of UI framework anyway, for what it's worth.)

 

Not sure if that info is useful to anyone, but I thought I'd give it a mention.

[Alfonso]

NEXTSTEP port by id Software and the other semi official by Omni Group like the devs intended.

Spoiler

Just kidding... DOSBox/Chocolate for pure vanilla and Crispy for limit removing vanilla wads.

Have yet to use the other source ports for advanced wads.

 

[Noiser]

2 hours ago, aRottenKomquat said:

I guess Crispy Doom would be my next choice, but really only to check vanilla compatibility without having to open up DOSBox.

You are not checking vanilla-compatibility with Crispy Doom (far from it). If you want to do that, you should use Chocolate Doom or Chocorender-limits.

[Kloki38]

Dosbox for casual play, GZDoom for mods. 

[Graf Zahl]

On 9/12/2022 at 8:55 PM, dasho said:

 

Very recent commits to FLTK 1.4 such as this imply that they are on the cusp of simultaneous X11/Wayland functionality. That being said, I understand not wanting to use the dev/non-stable version of a toolkit.

 

 

That's actually good to hear. While there's still some obstacles with FLTK, most notably an extremely dirty CMake project that's hard to properly embed, without dragging along an extreme amount of baggage, it seems to be the only lightweight GUI toolkit solution that doesn't either take over control totally or add massive bloat.

 

On 9/12/2022 at 10:35 PM, segfault said:

 It still doesn't change the fact that getting into Doom modding is way more tedious and obtuse than it needs to be if you're not already knee-deep in PC game modding, and maybe the technical barrier to entry can be lowered a bit so more people can enjoy all the cool shit that's been and is being made. That's my point. 

 

I'll be blunt here: When considering effort vs return of investment here the result is negative. We're talking about a major effort on the developer's side vs doing some simple investigations on the user's side. If people are not willing to invest a little effort to make life easier for them, I really don't feel like devoting my time to helping them out. I simply got better stuff to do with my spare time.

 

It simply does not add up for me, regardless of how you think about it.

 

 

On 9/12/2022 at 10:35 PM, segfault said:

Why do I need to give my friends a crash course in managing files and folders, teach them how to set up and use several different source ports and launchers, and string everything together just so we can do co-op BTSX or Scythe 2?

 

Giving them that crash course will definitely benefit them more than yet another tool that helps them stay ignorant.

 

Aside from that, why don't you point them to a launcher instead that does these things for them?

With that there's two possible outcomes: If it works for them, it's win-win - if it doesn't, nothing will.

 

Here's also something to consider: If you want a launcher that works for people like you describe you need an experienced frontend designer, not some tech-nerd who knows their way around all the quirks and pitfalls of a computer. Those tend to write techy UIs that meet their own needs and will confuse less experienced users just as much as what's there now, just like beginner friendly UIs tend to put off the tech nerds.

 

 

 

 

[Graf Zahl]

3 hours ago, aRottenKomquat said:

On the cross-platform UI topic--which I only kind of skimmed--I recently found a decent cross-platform UI as part of my day job. It's called Avalonia UI, runs on .NET 6, and is inspired by WPF which means it's a mix of XAML and C# with databinding features.  I've had very good success using it on both Windows and Linux. It technically supports Mac but it sounds like Apple shenanigans somewhat handicap that, though I haven't tried as we have no Apple-using customers in our industry and I don't use Apple products in my private life. Every other cross-platform UI I've used professionally was such a frustrating experience that we gave up on cross-platform UIs. (I used WPF and Silverlight extensively over my career, which means I am quite biased towards that type of UI framework anyway, for what it's worth.)

 

Not sure if that info is useful to anyone, but I thought I'd give it a mention.

 

I missed this in my previous post but since it is somewhat relevant, here's my take.

I generally shared your experience with these things. Most flat out suck. I couldn't list a single cross platform GUI that doesn't come with its share of problems.

It's just that some suck more and others suck less, depending on what the task is the outcome may be different for each specific use case.

But for an integrated launcher a .NET tool is not going to work as it cannot be integrated. Poor Apple support is also an issue.

[Devalaous]

I tried out DSDA Doom for UMAPINFO testing, and its a neat port. Bit of an ass to get my controller working right though.

[kinker31]

For my daily driver, GZDoom (or it's fork, LZDoom, if need be); I've been spoiled rotten by gameplay mods and QoL enchancements, especially the new-fangled Gun Bonzai, to really consider using anything else as a daily driver.

 

For level-testing, Chocolate Doom for vanilla wads (Crispy if need be), ReBoom/DSDA for Boom wads, and RZDoom for UDMF maps, respectively.

For multiplayer, Zandro or Odamex, the latter has a new Horde mode I kinda like, even if I suck at it.

[SleepyVelvet]

Prboom+ 2.1.5.3 because it "works".  Though with the influx of MBF21 wads... I'm gonna have to make the jump eventually... :/

[ChopBlock223]

20 hours ago, Devalaous said:

I tried out DSDA Doom for UMAPINFO testing, and its a neat port. Bit of an ass to get my controller working right though.

I find that Woof! has better gamepad support.

[TheSlipgateStudios]

I chose DOSBox for this poll, but I mainly use ZDoom as my go-to Source-Port for Steam, and GZDoom for mods.

[Jacek Nowak (jacnowak)]

IMO this poll is flawed because it doesn't include GZDoom which is definitely the most popular Doom source port globally by a large margin. Even on this forum which is biased towards more conservative ports I would bet it's at least in the top in 3.

 

GZDoom is my favorite source port because it has the most editing features of any source port. And I'm in the Doom community BECAUSE of editing features. Because I can be creative and because I can see what cool stuff other creative people make. And yes, sometimes having limits boosts creativity but sometimes if you have less limitations you can create more interesting stuff. I enjoy mapping for vanilla, I enjoy mapping for Boom and I enjoy mapping in UDMF. Each format has its strenghts and simplicity is sometimes a virtue but I wonder where our community would be today if source ports didn't exist (and didn't evolve) and it would only be possible to make vanilla maps.

 

Having said that I have a soft spot for Eternity Engine and I use that port very often to play non-GZDoom content (particularly when software renderering is recommended by the mapper). It has a nice balance of features vs fidelity to the original game. I wish it was developed more actively - I think there are just a few missing features/compatibility problems to solve (on both sides) that, when resolved, would make creating UDMF maps that work in both Eternity and GZDoom a viable option. It would be great to have a "standard" that is universally accepted and allows to make more advanced stuff than Boom (not talking about MBF21 here as it's centered around modding, not mapping). Absolutely not saying that we should dismiss Boom. People still map in vanilla/limit-removing even though Boom is there. Just that it would be great if there was an alternative that would make your map appealing to the "conservative" crowd for whom GZDoom feels too far from the original and still allow you to use more advanced mapping features. I think it would benefit Eternity if there were more maps that require it - even if those maps ALSO work in GZDoom (let's be clear - making Eternity-exclusive map is not a great choice nowadays as, unless you are skillsaw, your audience will be extremely limited). It would also benefit GZDoom and community as a whole as more mappers will be willing to try out more advanced stuff (which means progress).

[dasho]

7 hours ago, jacnowak said:

(let's be clear - making Eternity-exclusive map is not a great choice nowadays as, unless you are skillsaw, your audience will be extremely limited). It would also benefit GZDoom and community as a whole as more mappers will be willing to try out more advanced stuff (which means progress).

 

This is backwards thinking. Why is the onus on mappers, modders and source port developers (and yes to a degree I am biased in this regard) to cripple themselves in an effort to ensure max compatibility with GZDoom? If you like Eternity and want to map for it, you shouldn't hamstring yourself by not leveraging all of its features. If you want the largest possible audience for a map, then map for Vanilla or Boom. 

[Gougaru]

Gz and eternity have the best controller support.. so much go2s. Recently gz because it's been feeling smoother 

[Jacek Nowak (jacnowak)]

48 minutes ago, dasho said:

Why is the onus on mappers, modders and source port developers (and yes to a degree I am biased in this regard) to cripple themselves in an effort to ensure max compatibility with GZDoom?

I'm not suggesting ensuring max compatibility with GZDoom anywhere. I'm suggesting coming up with a standard set of advanced mapping features that could be implemented by several advanced ports. Those that already support UDMF and already have a fair share of nice mapping features - GZDoom, Eternity, maybe EDGE? UDMF is a good starting point, maybe we could come up with a standard namespace that is something more than Doom/Boom of MBF. Of course I'm not influential enough in the community to lead such an endeavour and I understand that it would need a lot of effort and cooperation between source port developers and some promotion, but I'm just sharing this idea because why not, sharing ideas about Doom is one of the things this forum is for.

 

48 minutes ago, dasho said:

If you like Eternity and want to map for it, you shouldn't hamstring yourself by not leveraging all of its features.

Because even though I enjoy mapping as a fun activity in itself, I find it much more fun to have people play my maps, see some playthrough videos of my work, get some feedback, etc. than make maps just for myself that hardly anyone plays. Eternity is a very niche port (even looking at results of this poll, which of course are very innacurate but you can see how many votes it got). If you make a map for GZDoom - we can argue whether you lose 30%, 50% or 60% of the possible audience (I think the vast majority of Doom players have GZDoom at least installed) but if you map for Eternity you lose 95% of the cake. And unless your map is a masterpiece, nobody will download and configure a port they don't have installed just to play your map. I actually made a map that uses advanced features of Eternity but I needed to write a script that generates GZDoom version out of it and play around with EMAPINFO&ZMAPINFO that the right version is chosen based on port you use. This is an example of a situation that the same features are supported by multiple ports but defined in a different way in the map, so the map can't be played in multiple ports even though it uses features that multiple ports support.

 

 

Quote

If you want the largest possible audience for a map, then map for Vanilla or Boom.

I'm just saying that it would be nice to have something between Boom and GZDoom that is an universally accepted standard for people wanting to use advanved features in their maps. Of course it wouldn't work in all ports (in the same way Boom maps don't work in Crispy for example) but it would be an option to choose from. Currently the only real alternative if you want a larger featureset than Boom/MBF and still have a sizeable audience is to just target GZDoom.
 

[dasho]

25 minutes ago, jacnowak said:

 And unless your map is a masterpiece, nobody will download and configure a port they don't have installed just to play your map.

..

Currently the only real alternative if you want a larger featureset than Boom/MBF and still have a sizeable audience is to just target GZDoom.

 

This is the heart of the problem. There ARE alternatives, but if everything continues to converge on a homogenized standard for "an advanced source port", then what's the point of even having separate source ports at that point? If other ports' identities become so diluted that they are just regarded as "partial implementations of GZDoom with some other stuff I don't feel like learning", then how is that a better outcome? Someone downloading and launching an extra port is a pittance of time and effort compared to what it would take to fulfill requests we've gotten to either add DECORATE or make GZDoom versions of our mods, etc.

Edited September 15, 2022 by dasho

[Jacek Nowak (jacnowak)]

4 minutes ago, dasho said:

 

This is the heart of the problem. There ARE alternatives, but if everything continues to converge on a homogenized standard for "an advanced source port", then what's the point of even having separate source ports at that point? If other ports' identities become so diluted that they are just regarded as "partial implmentations of GZDoom with some other stuff I don't feel like learning", then how is that a better outcome? Someone downloading and launching an extra port is a pittance of time and effort compared to requests we've gotten to either add DECORATE or make GZDoom versions of our mods, etc.

 

I see it otherwise. DSDA-Doom, Doom Retro, Woof! and a few other "classic" ports implement MBF21 standard but I don't see their identities as diluted.

I wish there was a standard for advanced mapping features (for example things that can easily be done in UDMF like aligning flats or separate texture offsets for top/mid/bottom textures) that could be implemented by several advanced ports. I would happily choose such format for some of my maps. Of course I would still sometimes map for Boom, sometimes for vanilla and sometimes for GZDoom. But I think it would be a cool thing to have such a map format to choose from. Just something I wish it existed.

[dasho]

Quote

I see it otherwise. DSDA-Doom, Doom Retro, Woof! and a few other "classic" ports implement MBF21 standard but I don't see their identities as diluted.

 

True, but the matter becomes more complicated when it comes to scripting and content definition languages. For example, we're currently working on an "edge" namespace. If we wanted to maximize compatibility with the "zdoom" and "eternity" namespaces, we'd have to implement the ACS_* line of action specials. This would mean we'd have to implement ACS, which for our engine is completely redundant since we already have DDF/COAL/RTS in place. This is just an example, but the point is that extending Dehacked/MBF like with MBF21 is, in general, a much easier endeavor and doesn't really change the identity of any implementing port as they were almost certainly Dehacked and Boom/MBF compatible already.

 

EDIT: Just to avoid confusion, I don't necessarily disagree with everything you're saying. I'm just concerned that investing time and effort into implementing yet another 'universal' standard won't bear much fruit when people's messaging is already "that port has almost no userbase, don't bother".

[Jacek Nowak (jacnowak)]

@dasho I didn't want to sound that people shouldn't bother with ports that have small userbase. I really appreciate effort that is put in development of ports and I also understand that promoting stuff is hard. Also, I think it's great that we have many ports - this means we have diversity and innovation because various ports develop in different directions.

 

To be fair - I find it sad that most people won't download a source port to play a map exclusive to that source port. They are all free after all. But that's unfortunately how the world seems to be, people are lazy. I find it weird when some people haven't played Heartland, which is a fantastic mapset, just because they didn't bother to install a source port it requires (I remember posts like that here on Doomworld). So if people don't want to download a source port to play a (deservedly) cacoward-winning mapset, I can't count on them doing that for my humble map which is probably average (even though I put a lot of effort into it but the amount of quality content here means people are spoiled for choice). This is just a mapper's perspective. Might be just mine but I think a lot of mappers share it. Personally I would download EDGE if I come across an EDGE-exclusive map that sparks my interest but that's just me and unfortunately it looks that I'm in the minority.

 

Now, I can see that the selling point of EDGE-Classic is running on less powerful hardware like Raspberry Pi and still offer advanced features (correct me if I'm wrong). Imagine a hypothetical situation that there is an UDMF namespace, let's call it "Doom-advanced" that you support and it has a subset of EDGE features. Let's say this namespace is supported by GZDoom, Eternity and EDGE. There is a sizeable amount of mappers that will go for this compatibility level because they want to use advanced features but they don't want to limit their audience to just to one port. If a sizeable amount of such maps appear in the wild, then people with less powerful hardware will get interested in EDGE because they want to play all these maps and it turns our that EDGE is the best choice for their hardware because other ports run too slow. It could give EDGE more exposure and some people might still get interested in this port's more advanced features and because the port will be more known some of them will be inclined to make a map exclusive for it. I think this situation would benefit the port and benefit the community. That's how I view it but it might be just my wishful thinking.

 

I understand though that the situation with advanced ports and features they support is more complicated than with classic ports. Both GZDoom and Eternity use ACS but EDGE has a different scripting language - I get your point here and I see it's not a trivial problem to address. EDGE and GZDoom both have 3D floors, whereas Eternity and GZDoom have portals. But there might still be a subset of features that all these ports could support. Definitely some basic UDMF features like freedom in aligning/scaling of floors/walls that I mentioned earlier. Polyobjects? I think all 3 ports have some support for them. There is probably more. Whether it's worth it? I don't know. This is just something that came to my mind, I know that implementing anything is always much more effort than it seems. And I know we're all (mappers, moders, source port developers) doing it for fun so we can't be forced to do anything we don't like.

[aRottenKomquat]

[dasho]

Quote

Now, I can see that the selling point of EDGE-Classic is running on less powerful hardware like Raspberry Pi and still offer advanced features (correct me if I'm wrong). Imagine a hypothetical situation that there is an UDMF namespace, let's call it "Doom-advanced" that you support and it has a subset of EDGE features. Let's say this namespace is supported by GZDoom, Eternity and EDGE. There is a sizeable amount of mappers that will go for this compatibility level because they want to use advanced features but they don't want to limit their audience to just to one port. If a sizeable amount of such maps appear in the wild, then people with less powerful hardware will get interested in EDGE because they want to play all these maps and it turns our that EDGE is the best choice for their hardware because other ports run too slow. It could give EDGE more exposure and some people might still get interested in this port's more advanced features and because the port will be more known some of them will be inclined to make a map exclusive for it. I think this situation would benefit the port and benefit the community. That's how I view it but it might be just my wishful thinking.

 

This is a good point; I wish I could be as optimistic about such an outcome. I fear it will end up like UMAPINFO where ports implement it, but there's still a lot of fragmentation because developers don't want to throw away or deprecate their other efforts (see ZMAPINFO, EMAPINFO, RMAPINFO, DMAPINFO (not really anyone's fault on this one), and even our own LEVELS.DDF). I wouldn't expect mappers to have to continually include 4-5 specialized lumps to address all ports, so naturally it will gravitate to being UMAPINFO and probably ZMAPINFO, which will (if the mapper chooses to use ZDoom-specific properties) make it look more capable in comparison.