[ckawahara]

Personally opinion here, but the word "unstable" gives off bad vibes. I use the Monogame Raildriver version almost exclusively for 'playtime' activies and to include it in yet another version with yet another name was not a clarification. I understand 'stable; and 'testing'. I understand a fully separate raildriver iteration within the monogame development. But 'unstable?' I'd like to see a single monogame/raildriver version, perhaps called 'development' or alpha, or pre-test which will eventually become 'testing' then ORTS version x.x. As I stated in another post, to me there are five versions to me, 3 .ORG and the two wildcat monogame versions.

Like I said, just my opinion.

[Csantucci]

James Ross, on 12 August 2019 - 08:50 AM, said:

perpetualKid chose to exclude it with the "not-for-unstable" label 4 days ago because of potential issues raised on the forums.

This is exactly what the label is for, but perhaps we need higher visibility on what things are being excluded from the unstable version?

That is a decision which will be hard to understand by users.

[James Ross]

ckawahara, on 12 August 2019 - 09:09 AM, said:

Personally opinion here, but the word "unstable" gives off bad vibes.

That's the intention - you should not use it unless you are prepared to deal with a more buggy program and unexpected changes in behaviour.

ckawahara, on 12 August 2019 - 09:09 AM, said:

I use the Monogame Raildriver version almost exclusively for 'playtime' activies and to include it in yet another version with yet another name was not a clarification. I understand 'stable; and 'testing'. I understand a fully separate raildriver iteration within the monogame development. But 'unstable?'

There are 3 official and 2 unofficial versions that I am aware of. There have been no new versions nor name changes recently, so I do not know what "yet another version with yet another name" refers to.

ckawahara, on 12 August 2019 - 09:09 AM, said:

I'd like to see a single monogame/raildriver version, perhaps called 'development' or alpha, or pre-test which will eventually become 'testing' then ORTS version x.x. As I stated in another post, to me there are five versions to me, 3 .ORG and the two wildcat monogame versions.

The MonoGame RailDriver version contains (I believe) everything in the Unstable Version, plus all the MonoGame changes, plus all the RailDriver changes, so that would be the version you want here.

If you want all of those changes in an official version, you're going to have to wait until the developers of each set of changes makes them available in a pull request, at which point they'll be included in the Unstable Version.

[Genma Saotome]

WRT the use of the word unstrable. I agree it is wise ti give a warning but unstable suggests something that'll produce a BSD. How about untested instead? Still a warning but it put sthe focus on features that might not work rather than the whole program crashing. We havn't had many of the later over the years.

[EricF]

When I first saw the "Unstable" label, I made the assumption that since this was a branch focused initially toward Monogame adoption and related features, that it needed to be named similarly to the "Experimental" branch. In fact, "Experimental" is somewhat less common in naming for most of the software projects I've been around -- "Unstable" is commonly used to describe the ongoing testing branch. If there are to be two parallel pre-release branches for testing, then "Unstable" seems like a very typical name.

To be fair, the Experimental branch has always carried the cautionary description that it's a testing branch with the possibility of bugs, both minor and serious.

It might serve to re-iterate the same cautions applied to the Experimental branch with the additional explanation that in the current development path of OR, that Experimental may be closer to late-Beta or "Release Candidate" quality, and "Unstable is more like early- and mid-development Beta.

A lot of us have gotten very accustomed to using the Experimental versions, particularly when they're halfway or more to the next Release version and the major bugs are squashed. Rather than scare people away with too dire of a warning, it's probably sufficient to remind them that "Unstable" releases are similar to the earliest versions of the "Experimental" branch in terms of reliability. Explain that it's intended to get more eyes and hands on testing new features, but not necessarily stable enough to be used as the basis for route development, etc.

[James Ross]

EricF, on 16 August 2019 - 02:57 AM, said:

When I first saw the "Unstable" label, I made the assumption that since this was a branch focused initially toward Monogame adoption and related features, that it needed to be named similarly to the "Experimental" branch. In fact, "Experimental" is somewhat less common in naming for most of the software projects I've been around -- "Unstable" is commonly used to describe the ongoing testing branch. If there are to be two parallel pre-release branches for testing, then "Unstable" seems like a very typical name.

It's been a while since we had an "Experimental" version, although it was much like the new "Unstable" version (in that trusted developers can change the code at any time). The main improvement we have over before is that the "Testing" version is actually distinct and should be more stable.

EricF, on 16 August 2019 - 02:57 AM, said:

To be fair, the Experimental branch has always carried the cautionary description that it's a testing branch with the possibility of bugs, both minor and serious.

EricF, on 16 August 2019 - 02:57 AM, said:

It might serve to re-iterate the same cautions applied to the Experimental branch with the additional explanation that in the current development path of OR, that Experimental may be closer to late-Beta or "Release Candidate" quality, and "Unstable is more like early- and mid-development Beta.

The website prominently lists only the "Stable" and "Testing" versions, and has just a small link to the "Unstable" version with the warning:

Quote

The unstable versions are more likely to contain serious bugs and are only recommended for users wishing to help with Open Rails development.

Similarly, the options screen lists the "Unstable" version as:

Quote

Daily updates which may contain serious defects. For developers only.

EricF, on 16 August 2019 - 02:57 AM, said:

A lot of us have gotten very accustomed to using the Experimental versions, particularly when they're halfway or more to the next Release version and the major bugs are squashed. Rather than scare people away with too dire of a warning, it's probably sufficient to remind them that "Unstable" releases are similar to the earliest versions of the "Experimental" branch in terms of reliability. Explain that it's intended to get more eyes and hands on testing new features, but not necessarily stable enough to be used as the basis for route development, etc.

I think the current descriptions convey this reasonably well, but I'm happy to update the website and options with new text if anyone has a suggestion on how to describe these different versions.

[X3D]

I was about to try out ReShade again as it can enhance the overall look of Open Rails, after installing ORTS-MG 1.3.1 RD ReShade doesn't seem to function anymore?
I wondered if it was my configuration or if this was a general change in the programming of the monogame branch? The latest 'Ultimate' update also removes runactivity.exe for some reason?

Thanks, Graeme

[R H Steele]

X3D, on 21 August 2019 - 03:54 AM, said:

I was about to try out ReShade again as it can enhance the overall look of Open Rails, after installing ORTS-MG 1.3.1 RD ReShade doesn't seem to function anymore?
I wondered if it was my configuration or if this was a general change in the programming of the monogame branch? The latest 'Ultimate' update also removes runactivity.exe for some reason?

Thanks, Graeme

I've never gotten ReShade to work with any version of the Raildriver branch of OR that perpetualkid is working on ( Thank you, sir http://www.elvastower.com/forums/public/style_emoticons/default/hi.gif) ... I don't think ReShade has been implemented in that version. Probably will be in the future, when all the version merge.

[Csantucci]

Info to the OR users: in these days the OR official unstable version is really... unstable. Due to this stuttering issue http://www.elvastowe...post__p__250102 tests are under way to understand the cause of the stuttering. Therefore the various additional features present in the unstable version have been partly removed and will probably re-added one by one to see if it is one of them which causes the stuttering.

[Genma Saotome]

Carlo, thank you for the important update.