all the sounds are very good, except for the “real” stone placement. It’s good that it is a mute sound, but it should resemble the sound of an object placed on soft wood.
i really like the “tower go” sounds - the atari one is briliant -, but the capture one could use a small retouch - it’s too sharp/acute and too long
But I do not think it is good to have sounds to place on wood when there clearly is not wood.
I think they mean for the real board case which is more or less at the end of the video. It does look like a kind of wooden board in the background.
This week I implemented the stone removal phase interaction for online matches, which can be completed manually, or auto-scored by the AI. Additionally, I added a score breakdown once the match is completed, which should help new players understand how the match score was computed.
https://twitter.com/i/status/1309914171077779457
Thanks for the feedback! I’ll take another look at those in particular.
This week was mostly small features, bug fixes, and code cleanup, but I did add two new sets of achievements for tracking territories conquered on the campaign map and dragons slain during Go matches.
To help bridge the skill gap between GNUGo and KataGo, a new hybrid AI option has been created which takes data from GNUGo and KataGo to create an intermediate opponent. It uses GNUGo’s generated move as the baseline and KataGo’s analysis scoreLead data to determine which moves are available that are estimated to be at least as good as GNUGo’s. The chosen AI level in the UI determines where the chosen move is selected within the range of GNUGo’s move to the top pick from KataGo (lower level closer to GNUGo, higher level closer to KataGo). I’m still working the algorithm for the hybrid AI, but that’s the basics of it.
Another week primarily filled with bug fixes and code cleanup in order to get the game ready for release in November. This week I posted about how the main menu plays through notable matches from Go’s history. Each time you get to the main menu it will load a random match from the set of notable matches. In this case it was the game of the century.
This week I posted about the automatch feature for online play. I’m trying to keep some new content updates coming along while I continue to polish up existing features.
While polishing up existing features this week, I revisited the glossary system and added some new terms and board representations. It now has 47 terms. Along with going to the glossary directly as shown in this clip, text within tutorials are automatically bolded allowing you to hover over the term where a tooltip will display the details of the term.
The presentation looks incredible. In fact, the whole project is seriously good work. The glossary terms look 95% there, too. I don’t know if you’re looking for help with the polish, but if so, I’m a willing volunteer. Feel free to send me a message privately.
A few revisions worth considering:
Dame. Doesn’t exist in all rulesets. Perhaps add a note, offset by commas or parentheses: “In territory-scoring rulesets like Japanese rules.”
Eye. Famously hard to define. Calling it an empty point “surrounded by stones of the same color” is good but technically vague. That definition includes false eyes, for example. Perhaps clarify that only an empty point that the opponent can’t occupy (unless the surrounding stones are captured in the process) counts as an eye.
Ladder. I’m not sure it matters for definitional purposes if the ladder is properly set up. Alternating ataris that don’t capture the target stone(s) are commonly called a “broken ladder.” Or we might say the “ladder” either works or it doesn’t work. But in either case I’ve always known it as a ladder.
Seki. This definition mentions that two eyes are required for life. I recommend adding the word “typically.” After all, seki is itself alive without two eyes (mutual life; thank you @Vsotvep).
I believe the Chinese term 双活 literally means mutual life, the Japanese term seki does not “literally” mean mutual life, as far as I’m aware. I don’t really know what its etymology is. According to this source it’s written with the kanji 持, which is usually pronounced ji when encountered on its own and means draw / tie. The pronunciation seki is unusual, though, thus can’t be the etymology…
Eye: the definition is not only technically vague.but wrong too. One of the surrounding intersections can be empty (not occupied by a stone but by another eye)
Mark, great feedback, and thanks for the offer to help! I’ll make those clarifications you pointed out.
I thought the concept of dame exists regardless of ruleset, but how they’re scored changes depending on the ruleset. For Chinese rules, are you saying that they don’t refer to those points as dame? I understand that they increase the player’s score in Chinese scoring, and if there are an odd number of dame points, the last player to fill gets the advantage, but I thought they still called them dame. I tried to leave the definition ruleset agnostic by just saying that the dame points don’t increase the player’s territory, then using that you can apply the logic to whatever ruleset you want (territorial scoring or not).
Are you just referring to the fact that an eye doesn’t need to be a single empty intersection, but can be multiple empty intersections, or even include intersections occupied by dead opponent stones? Along with Mark’s suggestion I plan to also clarify that it can be more than an individual intersection.
" an empty point “surrounded by stones of the same color”. >> That’s wrong definition one of the delta is not a stone but another eye.
I think the difference is the interpretation of “surround” in this case. When I think of surrounding intersections, I consider them the direct north, south, east, west coordinates to a specific intersection, the same as how we treat liberties, and do not consider diagonal intersections. Perhaps I should clarify further and just say surrounding adjacent intersections. Combine that with Mark’s suggestions as well and I think it should cover the bases fairly well.
It’s a bit difficult to give a short definition of an eye but let’s try:
For X an empty intersection to be an eye
You need the 4 stones on the connected intersections(A to D) and you need to “secure” 3 of the 4 other intersections marked O
Secure meaning that the opponent cannot have access to these intersections (because there is one of your stones already which cannot be captured or because if he play there he will suicide)
This will cover almost all lives (besides seki) because there is still the famous case where you live with 2 false eyes:
Well I mean for the understanding of a beginner, a eye is not a false eye. The O intersections are as important as the A to D to define an eye.
For a quick access and understanding, the case where one of the O is not occupied by a stone but by an eye is often omitted (see the general simple definition in sensei library for example). I can understand that in a pedagogic view, don’t make things complicated; but this case is very common on the board so it can’t be omitted too long or will create confusion.
Another definition which helps more Cartesian minds is to not really talk about eyes but to say:
If all your chains have access to two liberties, and these two liberties cannot be forced to be filled (optional: because then your opponent need to play two moves at the same time to capture and this is not allowed by the rule) then your chains will never be taken, it’s called life (optional: the two liberties are called eyes)
Not to mention that if you only consider the A to D surrounding stones, you may push your beginner into the ko question a bit too quickly.
That’s a good point. And this may well be. The trouble comes when defining dame as a neutral pount that doesn’t increase either player’s territory. Under Chinese rules, the point isn’t really neutral, and it does increase the player’s score (though perhaps not territory, depending on how you define that). It would be clearer, I think, to define it as an empty point next to and between opposing stones, often of little or no interest to either player.
Yeah, I agree, the “neutral” part is where it starts to get misleading. I’ll remove/reword that.
I appreciate your write-up. When trying to cover all of the edge cases like this, things get rather detailed and complex. In my opinion, it gets too complex for beginners, who are the main target for the glossary. Maybe the more detailed explanations for eyes are better served in tutorial form than the glossary.
I get that. Maybe then use a gross adaptation of the Cartesian explaination.
Not sure I would even put “eyes” in the glossary…
Given the recent misclick event during the 2020 Samsung Cup, I took the opportunity to post about the different input methods in The Conquest of Go. These include single click, double click, and confirmation, which are shown in the video clip. Similar to OGS, the latter two input methods help avoid misclicks.
