What can spring engine/game devs learn from negative reviews of evo rts on steam?

[Super Mario]

Here are the list of common complaints from scanning the reviews

• Game crashes
• Seems to be in alpha version
• Fail to get it working
• Lobby is difficult to use
• Very Hard entry (Not user friendly,

Yes I am aware that some noticeable ones are made from 2014. However I do not know that some of the issues are revolved today, so please be patient with me if I didn't know that some of them have been fixed already.
If there would be a lesson from this it would be the following
-first impressions matter a-lot.
-do not assume that the users are power users when it comes to the interface, good usability design matters.
-if you advertise as multiplayer, then don't allow broken ai in you game and only list ai that works.
-user expect in game tutorial/documentation do not expect them to "google it"
-Just because your game has a high skill ceiling doesn't mean you couldn't have a low skill floor.
-it needs to "Just work"(TM)
-they expect the game itself not something else like a third party lobby.

You can learn a lot from reading negative reviews to see what needs improvement.

[Forboding Angel]

• Game crashes
• Seems to be in alpha version
• Fail to get it working
• Lobby is difficult to use
• Very Hard entry (Not user friendly,

1. The game never crashed. Chiliui widgets would cause the entire framework to crash. This was was resolved years ago. One of the best players in evo (Damgam) plays on the shittiest laptop and integrated gfx imaginable. Game is very stable.

2. Not real sure where to go with this. Most of these comments are from people who never actually made it ingame and SWL was still dojo back then and pretty hard for people to use.

3. Lobby.

4. Gajop is helping us get Chobby working with uberserver.

5. There are multiple tutorial videos, many of which are linked directly from the main lobby screen. The people who watch the first one that is just under 5 minutes don't have trouble. Lobby and autohosts still remain an issue because... chat commands. Ingame I have even added audiovisual prompts to tell people what to do when they're screwing up their resources.

It would be awesome to have some real help. But most people don't care to really work on a game that isn't their own. I don't blame them.

Lately a lot of the players have been helping out a lot. In fact Damgam overhauled the shard setup so that it actually plays a good and fun game. Tet has been trying to get dedicated server built and working for RPI, and we have been testing one of the latest engine versions and fixing bugs that come up (so far the only one that I know of that is left is minimap boxes widget craps itself).

The day evo came out on steam, 50,000+ people came into the lobby throughout the first 24 hours (according to google). 99.8% of those reviews are from around that time.


Wouldn't it be better to play it yourself to form an opinion? You're experienced enough to the point to where you would have a far more informed perspective than random steam reviews.

[Forboding Angel]

If there would be a lesson from this it would be the following
-first impressions matter a-lot.
-do not assume that the users are power users when it comes to the interface, good usability design matters.
-if you advertise as multiplayer, then don't allow broken ai in you game and only list ai that works.
-user expect in game tutorial/documentation do not expect them to "google it"
-Just because your game has a high skill ceiling doesn't mean you couldn't have a low skill floor.
-it needs to "Just work"(TM)
-they expect the game itself not something else like a third party lobby.

You can learn a lot from reading negative reviews to see what needs improvement.

You added this after the fact:
1. Yes, ofc, but I have no real control over the lobby situation as I am not a lobby dev. I have made customized lobby ui for swl that is very nice, but that's where my control ends.

2. I don't. But one the other hand, you can only hand hold so much. Anyone familiar with rts games won't have trouble. Those that are new to rts games will have trouble.

3. As stated on the forums and everywhere else, ai is a plus and ymmv.

4. There is an ingame "how to play". There is documentation in the wiki and there are tutorials on youtube. I can't imagine anyone would find any results on google?

5. The game has a very low skill floor. If you understand how the gameplay in evo works, that becomes very readily apparent. Evo is the most "comeback-able" rts game I've ever played. Additionally, mixed unit compositions work very well together. If a player gets in the game and builds 1 of everything on repeat, his composition won't really be that bad. Additionally, single unit spam isn't punished by the game, but is punishable by the other player's unit comp.

6. It does.

7. All the lobbies are 3rd party -_-

[gajop]

The first issue often stems from Spring crashing when used with ATI/Intel. No real solution there yet afaics.

I'd say issues 2~5 are just due to the lobby and is something Chobby is trying to fix already.

[Forboding Angel]

The first issue often stems from Spring crashing when used with ATI/Intel. No real solution there yet afaics.

Yeah, that's a good point. Forgot about that. Evo steam used spring 96 until earlier this year. All engine related issues disappeared in 103 to my knowledge. 103 is this decade's 86b3.

[Silentwings]

ATI/AMD driver wtfness was long ago improved, iirc, and anyone trying to play todays rts on Intel

Lobby issues have been done to death elsewhere, no easy solution for Spring here -> this one takes time, progress is visible.

Tbh, can't see much of value here that goes beyond the usual generic stuff, and probably better not to conflate what modern issues do actually exist with a chaotic game launch from years ago. Reviews of ZK/LD are much more important these days.

[gajop]

This is one reason why I'd like to run a LD jam again.
The recent ones had issues with Intel/AMD where they couldn't support shaders/stuff properly or were just randomly crashing.

[Forboding Angel]

ATI/AMD driver wtfness was long ago improved, iirc, and anyone trying to play todays rts on Intel

True, I haven't had a gfx driver issue report that was an actual issue with the gfx driver since moving to 103. I had run across some ancient intel integrated cards that refused to run, but even that's rare at this point.

[Super Mario]

Is it possible for a game dev to enforce hardware requirements by giving errors if it fail to meet hardware requirements?

[Jools]

To the end user without experience in spring, it has no difference if chili-ui crashes or the game itself crashes. They would call both instances 'a crash'.

[sprunk]

Is it possible for a game dev to enforce hardware requirements by giving errors if it fail to meet hardware requirements?

The engine does provide some info about hardware. A lobby (such as Chobby) could kinda enforce it by not starting games in the first place but would need to have info from the game on what its requirements are. I don't think an actual game can do much once it is underway. People generally ignore warnings (and then scream bug when they see things actually fail to render). The only thing they won't ignore is when you actually prevent them from playing which I don't think does you good either.

[Forboding Angel]

To the end user without experience in spring, it has no difference if chili-ui crashes or the game itself crashes. They would call both instances 'a crash'.

Yes, that is correct, but the difference matters from a technical standpoint. If luarules is crashing or the engine itself is crashing, that is a huge issue. If it's a widget that is bringing down the whole UI framework, that's a different story.

[gajop]

We used ForceShaders=1 (springsettings.cfg) on some Ludum Dare entries, going by the thinking that it was better for the game to crash than to lack crucial visuals. As the games were shader heavy - to the point of using shader rendered visuals to describe mechanics, not only were they ugly without them, but also incomplete.

If you have no control over the lobby or user settings, you can't do much except abort ungracefully once the game has started. Ideally games would distribute their settings so lobbies knew what was valid/what could be set, instead of just modifying engine settings.

[ThinkSome]

I have no idea what lobbies these people used (SPADS doesen't log this *ring* *ring* bibim), but here are the most common obstacles for newbies to S44:

<Skaar604> !start
* WorldAtWarII_2 * Skaar604, you are not allowed to call command "start" in current context.

This person was probably a spectator. https://github.com/Yaribz/SPADS/issues/25

<comandanterojo> !start
* WorldAtWarII_2 * Unable to start game, not enough teams - use !forceStart to bypass

Self-explanatory, unfortunately they could not read.

<zypper> !start
* WorldAtWarII_2 * Unable to start game, zypper is not ready

This one actually could read, but proceeded to give up after putting a space in front of !forcestart, too bad:

<zypper> !start
* WorldAtWarII_2 * Unable to start game, start position type is set to "Choose in game" but no start box is set - use !forceStart to bypass

<arturobonin> !start
* WorldAtWarII_2 * Unable to start game, arturobonin is unsynced

Another unintelligible message, person was probably lacking game/map/engine or SL fubared and had to be restarted.

Suggestions?

[smoth]

ATI/AMD driver wtfness was long ago improved, iirc, and anyone trying to play todays rts on Intel

can we generate an error popup after crash telling the user that thier onboard video card being an intel cannot play this game because intels are garbage? saving the tedious, beg for an infolog, maybe get one after hours of fighting and then telling the user that their machine is piss poor?

[ThinkSome]

I play Spring:1944 on Intel HD 2500 and it works fine except the ammo resupply ranges are missing, something to do with GL.Points not working

On GMA 4500MHD it was actually bad as everything was rendered black and white haha

[Kloot]

The first issue often stems from Spring crashing when used with ATI/Intel. No real solution there yet afaics.

Sure there is, just spend 50 million dollars to hire 100+ FT engineers and setup giant automated test-labs like all the AAA studios do. Guaranteed 95% success rate, maybe host a kickstarter or gofundme campaign?

can we generate an error popup after crash telling the user that thier onboard video card being an intel cannot play this game because intels are garbage? saving the tedious, beg for an infolog, maybe get one after hours of fighting and then telling the user that their machine is piss poor?

You haven't been following recent mantis discussions.

Popups are not always technically feasible to show after a crash so they must be preemptive, which received unexpected opposition (https://springrts.com/mantis/view.php?id=5597) from within the ranks.

Raising the minimum GL requirement, which would among other things eliminate most garbage from running Spring entirely, would be a far better idea long overdue (https://springrts.com/mantis/view.php?id=5620) but was also opposed.

Quoting myself from here: 'aiming to include the extreme low-end and bending over backwards not to annoy anyone stuck there is *DETRIMENTAL TO DEVELOPMENT* in the long run, much like the forced support for ancient Intels has been.' If the decision to block those had been taken years ago (when the playerbase was larger), we would not still be talking about these issues in 2017.

[Super Mario]

There is a maintenance cost associated with supporting old hardware, even more so when that hardware is nearly a decade old. If zero-k want to keep cratering to users with outdated hardware, then I strongly suggest that zero-k devs should just fork the engine and just pull bug fixes from the main source line. Afaik there no paid dev working on this, as they are working on it during their spare time. Unless any devs are willing to set up a patreon account to work on the engine, then I think it's fair to kloot by taking maintenance cost into a strong consideration.

[Forboding Angel]

FWIW I agree with kloot and I also agree wholeheartedly with this reply in mantis:

Kloot (developer)
2017-06-24 00:46
Last edited: 2017-06-24 00:48
View 2 revisions
We might have reached a crossroads here, since I'm not interested in being a potato farmer.

You say we can't afford to lose any of the low-end people, I say we can't afford to keep catering to them. Old players might be able to look past the visuals of a particular Spring game, but they will eventually get bored and leave. Newcomers won't be so forgiving, appearance matters for first impressions. The massively declined level of forum activity is one *very strong* indication of this.

The short-term cost of raising requirements does not even begin to approach the long-term consequences (wrt community, technical debt, maintenance burden, ...) of being all-inclusive.

This coming from someone who's small playerbase consists mostly of potatos. They are a giant problem and sticking with the past is not the solution.

Edit: The decline of forum activity coincided with the wide adoption of discord. On discord in the various servers, activity is quite high.

[raaar]

we can try sorting the reviews by date

Most people who wrote negative reviews have less than an hour logged. The game needs to just work. If it doesn't people quit and never look back.

People have trouble setting up single player games, i see that in MF too. They start with no opponent, or choose the wrong AI.

There's some confusion about how EVO relates to the other spring RTS games. The fact that it's on a shared hub seems to be omitted completely and is something that players eventually stumble upon (not sure if it's related to the fact that some other games which share the hub have copyright issues, it's understandable either way).

There are several positive reviews, but many also have a very low number of hours played. It seems EVO's having trouble retaining players, a 1v1 ladder with rankings visible on the steam page and EVO website would probably lure more people into trying to play each other (not the AI).

I think one of the selling points of SpringRTS games should be working on low end machines. In terms of visual quality, they can be "decent", arguably "good", but they'll probably never be able to compete with AAA titles. They should work on the same mid range PCs they worked on half a dozen years ago. It should work wherever starcraft II works. It should degrade gracefully for low end clients...If it doesn't, both engine, lobby and game devs are probably doing something wrong...