Yet another retro source port?

[Teddipetzi]

While browsing the Doomworld forums I stumbled across this thead about a new classic source port.

I'm a bit puzzled. We already got Chocolate Doom, Crispy Doom, Doom Retro, PrBoom and possibly a few others. Why is that ultra-retro segment so crowded with ports that ultimately just cannibalize each others' user base but in the advanced segment nothing is happening - GZDoom is virtually alone, aside from a few more or less half-hearted and not particularly attractive attempts.

[Arctangent]

because humans are beings that exist beyond their utility to you

[Graf Zahl]

GZDoom is virtually alone, aside from a few more or less half-hearted and not particularly attractive attempts.

The same can be said about the retro ports. This new "So-Doom" will probably go down as some "half-hearted and not particularly attractive attempt" as well. You will always see people trying to do "their own thing", take another project and inject their own ideology. The only reason why there seems to be such an inflation is that Doomworld chose to promote 4 specific ports by stickying their development threads. But let's not delude ourselves. Aside from PrBoom+ it's all really just niche products, they are being made because their makers are a persistent bunch that do not look at the download numbers.

But the "advanced" landscape doesn't really look much different - the only major difference is that the Doomworld community does not promote these port in the same way as the classic segment so when looking at Doomworld you'll get an extremely distorted picture of the Doom source port landscape. But ports like k8vavoom very much exist, and I doubt that their download numbers are any worse than the true retro segment, it's just that on Doomworld they get drowned in the noise.

[drfrag]

There's also Russian Doom, Marshmallow and mine (RUDE). It was eclipsed by the "Can intelligent animals play Doom?" thread and well also it was some kind of a a joke project developed in a few days. But now i've done a lot of stuff and next version will be a major overhaul, i don't want to spoil the surprise but it will be great.
That's a Crispy clone without the intercepts overflow emulation, seems fabian didn't want to remove it. That thing is resposible for the all ghosts bug and seems there's an arcane demo abusing that effect but it desyncs at the end anyway. Crispy has an overflow guard with the laser aim but then you can't shoot beyond the intercepts.
Now that i think of it all this seems quite absurd but if you fix all the silly bugs then it's not "da real thing" and you break demos and if you fix them then you might say "there!, i've recreated Boom 20 years after".

[Enjay]

But the "advanced" landscape doesn't really look much different - the only major difference is that the Doomworld community does not promote these port in the same way as the classic segment so when looking at Doomworld you'll get an extremely distorted picture of the Doom source port landscape. But ports like k8vavoom very much exist, and I doubt that their download numbers are any worse than the true retro segment, it's just that on Doomworld they get drowned in the noise.

Less of a Doom World and more of a Doom Village?

[drfrag]

I thought Crispy was very popular, even Romero promoted it. I forgot to mention Russian Doom, it's pretty cool but only recently has been released in english. And there's even a DOS version.

[Graf Zahl]

I thought Crispy was very popular

Popularity is relative. I know a site with which I can look at download numbers of Github hosted projects, and all these ports are hosted there.

Both Doom Retro and Crispy hover at 2-3% of GZDoom's numbers. LZDoom currently has 15% of GZDoom so congratulations, you alone far eclipse the entire classic segment all alone with your port.
Here's the eternal misconception that a vocal subgroup appears to be larger than it really is. Even at Doomworld the majority uses more advanced ports for their daily Doom needs.

But if you asked me why that segment appears so crowded, it's actually quite simple: People who want a modernized experience are normally satisfied with the choices they get and won't invest any work into a time consuming side project unless the choice-to-go had some severe issues (think EDuke32.) but in the classic segment there's lots of people who have their own view of what they consider "classic" and tend to be put off even by slight deviations, so if they can they start to tinker themselves you got the next fork. Is this productive? Most certainly not - the entire segment might be served better by a group effort than this kind of fragmentation but it lies in the nature of things that even slight deviations will mean to cross a barrier for some of these people.

The currently active ports in the advanced segment are different - they aren't forks of each other but continuations of long-running projects - all of them derive from different predecessors of the early source port area.

[dpJudas]

there's lots of people who have their own view of what they consider "classic" and tend to be put off even by slight deviations, so if they can they start to tinker themselves you got the next fork. Is this productive? Most certainly not - the entire segment might be served better by a group effort than this kind of fragmentation but it lies in the nature of things that even slight deviations will mean to cross a barrier for some of these people.

Hehe, I learned this one the hard way. Nothing been more unrewarding than improve things for this segment as no matter what improvement you make they'll still complain.

[Graf Zahl]

Yeah, I did so too when working on the UMAPINFO proof of concept. Instead of getting encouragement to continue, the entire discussion soon devolved into technicalities. There's a handful of people who like it but the rest will stick to the same old megawad formula and churn out one overinflated project after the other.

Unfortunately Eternity suffers from some major PR issues preventing it from stepping into PrBoom's place - but nobody seems to be willing to improve that.

[dpJudas]

Hehe, yep, that UMAPINFO thread sure was a train wreck. Flooded by armchair devs and totally absent of anyone actually running the other ports.

[Graf Zahl]

Except Fabian Greffrath from Crispy. He has been the only one showing some interest.

[Gez]

I know a site with which I can look at download numbers of Github hosted projects, and all these ports are hosted there.

For anyone curious, here's one.
https://www.somsubhra.com/github-release-stats/

A few Doom projects on Github:

• Chocolate Doom
• cnDoom
• Crispy Doom
• Doom Builder X
• Doom Retro
• Eternity
• Freedoom
• RUDE
• SLADE
• So-Doom
• UMAPINFO fork of PrBoom+

To be fair, one shouldn't compare the total downloads between them, but look at the kind of download figures they get on good releases. It's extremely variable. GZDoom reached 49K downloads with 4.2.1, 4.2.4 is at 10K so far. Crispy got 2.8K for 5.5.2. Doom Retro also got 2.8K for 2.9.3. Chocolate Doom got 2.1K for 3.0.0. Eternity got 2.2K for 4.0.0.

SLADE got 11K for 3.1.1.4; but all the 2019 releases together get less than 1K in total.

[Graf Zahl]

SLADE does not count - its main download site is elsewhere.
And when looking at the numbers, always consider the time span the download was active. Eternity needed 1.5 years to get to its 2200 downloads, whily both Crispy and Retro need two months for the same number (if a release lasts long enough.)

Chocolate Doom also does not tell the full story. Not only are there multiple independent projects but I believe it also got a separate download location. Github numbers are only representative if it's the only or the preferred download loaction.

[Blzut3]

Possibly worth considering that at the retro end of the spectrum normal users don't have any real incentive to upgrade from one version to the next. I know there are people not really involved with the Doom community that use old versions of GZDoom for extended periods of time until they stumble across a mod that doesn't work, so I'm not going to say that makes up the difference. But thought it should be mentioned.

GZDoom is definitely not alone on the advanced port front: Eternity Engine, k8vavoom, Doomsday, Risen3D, EDGE, and DelphiDoom are all there in some form or another (you can technically add Doom Legacy and ZDaemon on that list). However GZDoom has set the bar for "advanced source port" so high that unless you fork from GZDoom it's an uphill battle to get people to even take a second look at the project. If you're looking for attention and starting a Doom source port it's much easier to throw a gimmick on vanilla doom then to create a gimmick version of GZDoom which probably could have just been a mod anyway. Pretty much the last time we've seen that category of source port in the advanced space was ScoreDoom.

Eternity Engine is a funny one, they have so much potential to upset the status quo. In theory the port should be able to fully replace PrBoom+ and attract many GZDoom users. I mean there's no reason for the subset of features that Eternity and GZDoom both support to not become the new standard after Boom, but it seems some of the people involved (not all of them to be clear) seem to believe that GZDoom is actively holding them back or something. It certainly doesn't help that they have basically no way for people outside of the community (which by the way is where most of the players are) to discover the port. (Guessing that's what Graf means by the PR problem.)

[Kinsie]

The largest issue facing Eternity is that the lion's share of the development team is busy working on remakes and remasters at Night Dive Studios.