Salesome

Hellas random map is waaaay too bright. It hurts my eyes. Tweaking the sun angle just a bit makes it much better. On a general note, the sun should not produce overrexposed maps when at the zenith (which it does on other maps, too). There is a "sun overbrightness" slider anyway (which is at zero in this map). IMHO, the install script (Windows) should default to creating a desktop shortcut (opt-out instead of opt-in). It is way easier to delete one shortcut too much than having to navigate through obscure folders to find the game's EXE file if you forgot to check it (like I did multiple times just now).

SVN Lobby has a weird text bug

I asked that question about the "GreatTower". It's probably the very same thing. In the German translation, I put a space character.

I am a big fan of the minus key, but someone in the past must have thought that ancient history needs something fancier... Patches are coming up. Who should I put as reviewers? Just Stan or other people, too? Maybe an English native speaker?

I ran languagetool over the English source text in Transifex and it came up with some errors. Most are easily fixable just by using the search function in the source files. Use of serial comma is inconsistent. Sometimes, there is a comma before "and", sometimes not. The minus sign is often an en-dash or a different dash that is too long. Also, the minus sign is often used where a dash should be put (between sentence parts). In proper names with adjectives like "Ancient Greece", both the adjective and the noun should be capitalized (this is often not the case): other examples Ancient Egypt English Channel Rubicon River Italian Peninsula Central Asia Indus River Many compound words are written with a space instead of a hyphen. All examples below require a hyphen well equipped full grown well balanced cost effective well trained decision making chock full up to date only (when used as an adjective) This list was created keeping in mind that the source language is US-English. The following US-GB-English specifics came up: center vs. centre unauthorized vs. unauthorised civilisation vs civilization many derived forms of verbs need a double consonant in British English, but do not in US-English: worship-worship(p)ing, label-label(l)ed, travel etc. Usage of quotation marks is inconsistent. "example" vs. “example” (I prefer the latter) Antialiasing should be written with a hypen Therefore and moreover require a comma afterwards Disclaimer: These are just suggestions. I'm not native in English and I might be wrong about some of them... I also found a few singular errors that I reported directly in Transifex.

Thanks for the imput. Concerning the "QA checks", I know that the system checks parantheses etc. It also regularly makes false alarms for glossary entries in German because we tend to inflect words, but that's something else entirely. I was talking about something different. There is a (relatively) new button in Transifex with "QA Checks". When you click on it, the message in the attachment pops up. I'm interested if you know in what that is and whether we can use this. I don't intend to throw out active translators. If the team thinks that it is not a good idea to remove inactive translators, then I won't do it - I was just curious. It's good to know that someone reads the raised issues in Transifex. Then I will use the feature more often.

During my activities as German translator, I came up with some questions that I would like to ask the team and the community: There are "QA checks" advertised in Transifex. Can we acitvate them or is this only for paid users? (assuming 0 A.D. does not have a paid subscription there) Is there someone in the official team who looks after Transifex in general? Back in the day, leper and Gallaecio did things like answering questions, processing issues (Transifex has a "raise issue" funtion) etc. They have both left, as far as I know. Is it possible to see statistics who translated how much? In the German translation team, there are literally hundreds of people and I'm sure many have never translated a single phrase. I would like to kick these and for this, I would need statistics. How are the "resources" in Transifex determinded by which strings are grouped? The distribution of strings looks a bit like it was made way back and then everyone forgot about it. What I'm trying to say is that maybe a rearrangement might make sense. There are some small groups that could be combined and large groups that could be split. How do I raise acutal issues with strings? A recent example is two strings "Great Tower" and "GreatTower". Both are in the file simulation/templates/structures/han/defense_tower_great.xml. It seems the script that pulls translations from the source code adds a second translation that was not supposed to be translated in the first place. There are many such cases where I suspect this. Should I compile a list and post it on the forum? I'm not a coder and although I can point at issues, I don't know how to solve them.

True indeed I thought, removable storage devices are uniteresting since there are no removable storage devices attached. I turned this option on, too. The error message persists.

I checked that, too, and the snap package has all the permissions it asks for (see attached). When I restarted the laptop just now, the game no longer crashes (restarts didn't work before on that issue...) The error message in the game menu is still there.

The 0 A.D. snap development package crashes when I try to start an actual game match. The snap is currenty on revision number 26557. After launching the game, in the menu, an error message pops up (see attached screenshot). When I try to start a single player match, the game crashes (see crashlog, also attached). I found the forum thread below where an identical error message is discussed. I double-checked and there are only standard ASCII characters in the paths for the installation on my laptop. The current alpha 25b snap does not show these problems. OS is Ubuntu 21.10. crashlog.txt

Ok, then maybe, choosing a player number greater than six should output a small warning that performance is expected to be low? Giant map size already has a warning message in its flavor text.

DISCLAIMER: I played a 4v4 online Game. I know this may become laggy. Basically, the problem was that for most of the late game, units were using a stop-motion technique, where they would move for a moment, then stop for a moment, then move for a moment, then stop for a moment etc. It was frustrating for everybody involved. There are some peculiarities, however. I could analyse stuff a bit because I was on audio connection with my teammates and we could discuss everything. The host (not me) was GPU-limited. Actually, he used software rendering on his 16-core CPU The lag didn't stop although three defeated players left the game and two were almost destroyed and did not have a unit worth mentioning In the end, it was 2v1 where the one opponent absolutely did not want to give up (no complaints here, he was just denying his imminent defeat ) My game crashed at some point and I dropped out. For my teammates, it continued and it ended normally (after 10 more minutes) Me and my teammates were in the same country in Central Europe, one opponent I know was from Russia (large ping?) There was never a notification that a particular player was lagging (which would normally be the case) None of my teammates' PCs were CPU-limited (Is pathfinding still the biggest lag problem?) I would like to discuss if this was "normal" lag or whether there is more to it. I attached my replay, the crashlog and the "interesting" one. commands.txt crashlog.txt interestinglog.html

This has got to be the most polished release ever The last two months seemed like a bit of a pain for the devs (lurking bugs jumping at them from the forums, trac and several other places), but many have been fixed and the effort will pay off for sure. I think the new workflow with the preview releases is a great change.

I don't know, tbh. You can look for the string ingame (it would be the campaign menu) or you look in the source code if the tag is explained in a code comment. Maybe one of the programmers could explain the systematics behind such replacement tags. I generally view them as black boxes

A campaign run translates exactly to German "Durchlauf". The system is that you can start a campaign several times (with different presets, difficulty whatever) at the same time. Every time you start a campaign, a separate save file is created. This is a "campaign run". This is a thing for languages other than German. when you have a string that says "4 players", there might be other languages that place the number after the noun, like "players 4". German and English are so closely related that you can translate these placeholders in an identical way in almost all cases. In your example, languages might want to have a different dash character (Chinese comes to my mind). This is not a plural "s", it is part of the tag that is replaced in the end. You know this because replaceable tags are highlighted in colour. There is a special category for pluralized strings. They are very rare among the 0 A.D. strings, so you might not have encountered them yet.

To give you an example: https://www.transifex.com/wildfire-games/0ad/translate/#de/public-gui-campaigns/324902396 "More Info" says it originates from "campaigns/new_maps.jsonDescription:" You browse the source and find https://trac.wildfiregames.com/browser/ps/trunk/binaries/data/mods/public/campaigns/new_maps.json Everything concerning game data is located in the "public mod", so the front part of the full source path is omitted on Transifex. If you follow the link, you find this string. If you want to change it, you have to write a patch that changes this text in this .json-file. Strings that are marked as translatable strings are fed into the Transifex system which processes them in a convenient way for localization. Transifex, in turn, converts the translated strings into the .po files. These are never changed manually because they get overwritten on a regular basis. In short: you find an improvable string in Transifex, look up the source path in the "More Info" panel, look that up in trac, you write a patch in Phabricator for that file.

Basically, strings are duplicated in the source sometimes. You need some basic text in many places and it will be always the same (e.g. "Cancel"). Transifex merges translations (not the source strings) when they are exactly identical, but if there is e.g. different capitalization in otherwise identical strings, it does not because technically, they are not the same. If you click on one of the examples, there is a "More Info" panel. In it, you can see the exact origin of a string in the source code. One of the "no campaign selected" strings is from a Javascript file and the other is from an XML file. IMHO, this is part of the optimization potential, because if the two strings get a unified spelling in the source code, the strings don't need to get translated twice (for all languages).

@Gurken Khan I don't think you can look at stuff without an account. I mean, if you really don't want a Transifex account, you can still browse the source on Trac (the *.po files). @Ceres The problem with the comments is that I didn't find a way to look for them, like "show me all the strings that were commented on in the last x days". You can, however, raise an issue in Transifex. These pop up in my notifications and can be searched for. We have to try out anyway which way communication works best, be it the forum, Transifex issues, direct messages or whatever.

Thanks for the offer. On Transifex, I am pretty much self-sufficient - I can promote people to reviewers, you were just faster with Ceres People can just ask me. It has also been a while since someone new applied as a translator. However, I do have some questions about Transifex in general, but I'll open a separate thread for that. Also, I will try to create a Phabricator account so I can submit patches (it will probably be text changes) and actually propose changes on the side of the source strings. It's true, some of the English strings (a patchwork from the past 10 years) may have potential for optimization.

I just looked, you already are a reviewer. Although everything is translated, there are still German strings with room for improvement. By the way, I actually need more active German reviewers, since I did a lot of translations myself and I cannot review my own strings, even as a language coordinator. In case you want to improve the source strings - Transifex has a "Add Issue" function that lets you comment on strings and this can be discussed there. I don't know how actively this is followed by the core developer team, though.

Thanks for bringing this to the forum - I thought this would be an obvious improvement, but I‘m ready to discuss this. Just one question - in which way does the shadow cutoff influence shadow quality? These are two different options that can be set independently... Don‘t get me wrong, I think the shadow cutoff is a good idea in general, I only want to change the default value.

I did some research and found out that the Opus (Wikipedia) codec might be worth a try for the game's sound files. Currently, all 0 A.D. music files are encoded with 160 kbit/s using OGG Vorbis. With Opus, it would be possible to encode them with 96 kbit/s, where the threshold for transparency is for this format (Source). Maybe they could even be encoded with 64 kbit/s (At least, I don't hear the difference). Several ambient sounds might be encoded with even lower bit rates (for speech, 32 kbit/s is sufficient using Opus (see link above)). The sound files currently make up about 150 MB of the installer. The savings could amount to about 50-60 % by encoding with Opus at a lower bitrate, which would result in a ~75 MB smaller download. Doing a quick re-encode of all music files with 96 kbit/s VBR, I obtained 51 MB (as opposed to originally 109 MB). One big advantage of the Opus format, apart from the better encoding quality, is a very low latency (~25 ms). It can also be packed into OGG containers (so the configuration files don't need to be re-written). A possible downside is that decoding Opus files takes more CPU power than OGG Vorbis (I am not a programmer though, so I cannot quantify that). I attached a few 0 A.D. music files in Opus format as samples (encoded from the .wav files in trac). Ammon-Ra_96_kbits.opus Honor_Bound_2014_96kbits.opus Ammon-Ra 64 kbits.opus Honor_Bound_2014 64 kbits.opus Edit: VLC Player can open the files.

I absolutely agree with you that it is silly to have a forest growing into your settlement. The cleverest way is to let new trees grow near to old ones (like a cellular automaton - if there are enough trees around, a new one will grow). That would mean that a big forest that has almost been chopped down would slowly regenerate itself, but it would stop when it has reached its original size. With that configuration, there should be no problems with forests overgrowing. Regenerating fish is quite easy, as the number of food in one fishing patch would slowly increase over time and when it's empty, it would not disappear. Instead, it has 0 food and is regenerating. Animals shouldn't be a big deal either. E.g. in Empire Earth, there are small groups of animals with a defined amount of individuals standing around on the map. When their number is dropping, new individuals are spawned until the original number has been reached again. Generally speaking, I would be really excited to have these features in the game, as it would make it more alive and (that's a strong advantage) more realistic.

I've got a suggestion for simplifying the combat system and making it more transparent without any random effects. What do you think about it? New Damage/armour calculation

I saw that the calculation of damages dealt to an entity is not very transparent and armour bonuses are just given by an abstract number. My suggestion would be to make a percentage-based damage system. Concerning the armour, there would be basic manipulators for every damage type (e.g. Buildings: Crush 150% Pierce 10% Hack 100%) These percentages would appear when hovering over the armour-icon. (The manipulators that are 100% would not be shown, of course) In addition to that, the "classic" armour would still exist, but it would be only one number and it takes effect after the basic manipulators. The damage reduction could be something like 2% per point. Small example: (the unit stats are not the ones from the game, but they are irrelevant for now) A hoplite attacks a hippeus The hoplite has a damage of 20 spear The hippeus' armour has a 200% basic manipulator for spear damage an an armour count of 3 The basic damage dealt to the hippeus is 20 spear. The spear manipulator says 200% --> 40 damage 3 Armour points is 3*2 = 6% reduction --> 37,6 damage --> rounded 38 damage dealt to the hippeus Research - when it's finally implemented - could then modify either the basic manipulators or the overall armour. (Let's say iron ringmail --> +3 armour for infantry; horse armour: spear damage for cavalry reduced by 10%) The idea comes from Armies Of Exigo, a fantasy RTS-game from 2004. In the game, the whole thing was a bit more complicated, classifying damage into hack, pierce, slash, fire, ice, lightning and all the other fantasy stuff. I think this system is way less complicated than having to deal with three seperate types of armour and giving artificial bonuses against certain units.