The player panels are more or less done: https://probabilism.wordpress.com/2016/ ... er-panels/

They're linked up to the game-state, too, and appear to be fully operational when testing on the live server. The clocks seem stable and don't jump about wildly due to network lag. They transition between time periods correctly and watching the little LCD segments that count Canadian overtime stones is a pleasure.

I now depart from the player panels and begin work on the end-of-game experience: handling resignation, results caused by timeout and scored culminations.

@Jhyn: I tweaked the seven-segment LCD styling a bit and faded the 'off' segments into the background. How's that for you?

@YeGO: Yes, I plan to open-source my work at some point.

Hi Charlie,

Thank you very much for all your work.

Wish list/suggestions:

• Wood grain board -- very nice. Please keep this option.
• Slate stones -- nice. Please keep this option.
• Please offer an option for "textbook" plain "flat", 2-D & stones.
• Please offer an option for "real", photo-realistic Snow Grade shell stones.
  ( Definitely a 'luxury' option; that's why the above option for 'flat' 2-D B & W circles, as a backup. )
• Please offer an option for non-LED clocks (plain, 'normal font' digits).

Thanks, again.

Great stuff!!!

A couple of questions about your last post. On your website it says:
"When spectating, the away panel (pictured above) belongs to the stronger of the two players, or the white player in the case that the players are of equal rank."
Does this mean that if a 2d takes white against a 3d, the away panel will show Black while the home panel will show White? If so, I think this will be a little confusing as most observers are used to White being in the away panel regardless of strength. Isn't this the way the KGS client works now?

In your landscape mode screenshot is that a list of observers on the right side? They seem to be unordered (except for the players first). Is this the way that the server delivers the info? Hopefully you will include sorting the list and persistence in the preferred sorting. Personally I think wms made a huge mistake in not fixing the persistence of sorting preference that was broken when he fixed the sound issue. It made KGS much less user friendly for the high dans, which may or may not have a contributing factor to the decline in their presence.

One of my favorite players to observe is twoeye. His normal playing conditions are 80 periods of 10 seconds each. His games might be good ones to use to test your display since in your current layout it will require three rows of dots to show the beginning status with all 80 periods available. How much space will this take up in portrait mode?

Finally, is there any chance that you could release a test version that only allows the user to observe games (does not support playing) before you get all the heavy lifting done? My fingers are itching to see what KGS is like on my iPad.

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

Regarding the possibility of a photo-realistic stones and board mentioned by @EdLee, here are some free resources for implementing that:

It looks like the board of KGS Leben is generated using WGo.js, which recently added a photo-realistic option. See https://github.com/waltheri/wgo.js/pull/71. The stone and board images were contributed by https://github.com/atarnowsky/go-assets.

There is also a nice shinkaya wood grain photo provided by the jgoboard package. However, I think the color balance makes that image a little overly yellow. Hence, I modified the color levels of that image for my own project, see the shinkaya[2-4].jpg images in this folder.

I put these photo-realistic ingredients together in my own SGF editor.

Here's what it looks like:

Hi YeGO,

Thanks for all the info.
'Photo-realistic' shell stones are very tricky:
for example, in your last screenshot,
the W stones look less like shell stones but
more like glass marbles with some 'wavy' internal streaks.
For good rendering of shell stones,
please see the current Panda clients,
and the (much older) gIgo, gGo clients.

@EdLee:

The screenshot that I showed is rendered using actual photos of real stones, which are from this repository. Maybe they aren't the best pictures, but at least they are a nice, free asset that may be useful. As I was mentioning, it may be pretty straightforward to use them in this project, since rendering with these stone images is already a feature of the latest WGo.js, which appears to be a library used by KGS Leben (based on the appearance of the current board rendering).

Perhaps, this close-up screenshot may show the details of the stones a bit better:


It is indeed a bit tricky to render realistic stones. With these photos, the stones were actually lit from two opposite directions, necessitating double shadows, which may need further tweaking.

By the way, it does not appear that any of the programs that you mentioned (GoPanda clients, old and new) are actually even rendering photo-realistic shell stones. Perhaps, I'm mistaken and didn't look at the appropriate ones? Could you share a screenshot to help clarify? I would be very interested to learn about more software with good stone rendering.

Hi YeGO, yes, the shadows are part of the problem, and they definitely need more work.
The inconsistent, double light sources -- sometimes rotated -- is another problem.

I put 'photo realistic' in quotes; let me rephrase:
By 'good rendering', I mean aesthetically pleasing, or pretty, or beautiful.
Please see this thread, post 23.

As you noted correctly, to make the stones beautiful is more than 'simply' taking photos of real stones, and cropping them --
more 'post-editing' is necessary.

( Another aesthetic matter: in the screenshot in post 24,
the grid lines are too thick -- heavy & distracting.
One will never see thick lines like that on a good board.
At least the line width could be an adjustable option;
let each user pick their favorite thickness.
The aesthetic aspect of a good client is non-trivial: it runs in parallel to
good engineering. In general, I find the IGS clients -- gGo, gIGo, Panda, etc. -- very beautiful. )

Thanks.

From gIGo, a 2003 client -- note the consistent, single light source (top left direction):Another 2007 editor, with real stone photos (non-randomized) -- note the elegant, single light source (top left direction):From gGo, a 2002 client -- same, elegant stones & lines:

People, we really should let Charlie finish the basic functionalities before overloading him with our own design preferences... Btw I find the stones in WGo.js player already very realistic and even better than glGo's.

yishn, nobody is overloading anyone, with anything.
Of course, the basic functions are important.

It's fantastic that Charlie is even doing ALL this work.
We're very grateful.

These are suggestions; based on past experience (10+ years of Go clients usage).
From passionate people. Posted here for future reference.
If any is implemented, great; if not, it's not the end of the world.Different people will have various preferences (different from yours, different from mine);
thus the suggestion to at least add the most basic, no-frills 2D flat B & W circles -- as a fall-back, basic option.
Because some people like iOS, others Windows, still others Android, etc.
All very different eyes and tastes.

It is much better; the readability is greatly improved. I might have some additional suggestions but nothing of that importance, as yishn says, this is a bit early for that and it might be better to wait for a final overall design before tweaking the details. Again this is great work.

_________________
La victoire est un hasard, la défaite une nécessité.

Thanks, everyone, for all the interest and continuing positivity.

@<most of you>: I promise there will be options to tweak the appearance of the board and stones and whatever you like. Don't expect that, soon, however. If I've put up a screenshot of it, that probably means I've called it “good enough” for now and moved on to other things – otherwise, I risk building an enormously aesthetic chocolate tea-pot and that's not useful to anyone. Visuals are terribly dangerous – it is too easy to get caught up, tweaking and polishing them and they can easily consume infinite time.

@ez4u: Yes, you understand the feature correctly and, yes, I am still in two minds about the stronger-player-is-the-away-team idea. It does seem a bit confusing. It was an extension of an idea that I implemented in the game-list view because, when testing, I found it much easier to find intriguing games by sorting the game-list by “stronger” or “weaker” instead of “white” or “black”. It's all of two lines to change it so it might not survive too many more revisions. In the end, it works out that white is nearly always 'away', in practice.

@ez4u: The order of the list of users in a channel (either a game or room) is currently arbitrary but will certainly be sortable at some point.

@ez4u: Lastly, if the number of overtime periods or stones is greater than 30 (two rows of three groups of five; five more than the most common Canadian overtime setting), the dots will simply be hidden and only the numeric counter will be displayed to avoid a very large and unwieldy hoard of the little critters.

This is looking great! From the video it looks like it is planned to have automatch functionality, though from the original post it sounds like that isn't implemented yet. Will it be difficult to implement? Automatch is one of my favorite KGS features.

Automatch actually looks like one of the easiest features to implement and is also one of my favourite KGS features.

Earlier this afternoon, with my development environment still open and my code yet uncommitted, I connected to the live KGS server with one of my own ranked accounts and played the first games ever to be played via KGS Leben.

https://probabilism.wordpress.com/2016/ ... uto-match/

TL;DR:

• Auto-match working and operational - played automatch games
• Scoring (marking stones, agreeing upon the result and viewing the score-summary)
• Markers on the most recently placed stone on the board
• Some philosophy about feedback for training mental counting skills

More coming soon... ish. Depends how much "testing" I feel like getting done. It's been a while since I actually played games of Go, myself.

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

Hi Charlie,

Thank you for the good looking circular current move marker.

@EdLee: Those kudos belong to my wife. I just implemented a whole bunch and asked her which variation on the theme struck the best balance between helpful and jarring. (She's a bit annoyed with me because I haven't set up a test environment for her to play with the client, yet.)

Personally, I found them to be a little too subtle at first but after I stopped trying to critique them and let them do their job, I found that they really do draw the eye to the stone as it is played, even without sounds.

Hi Charlie,

You could try to increase the diameter of the marker slightly,
say, to, slightly larger than (half of the stone's diameter).

Please see also your messages.

Hi Charlie,

Dev of https://weiqi.gs here. Is there any update on open-sourcing this? I am interested in learning from other go-frontends.

Anyway, looks very nice!

_________________
https://weiqi.gs - Open source go server

Ja. Expect an announcement before the end of this month. I am hoping to have the code-base in a semi-stable state by then and to have set up project-management related stuff like roadmaps and issue trackers and readmes. Also, the automatch feature I mentioned in my last post will be ready for use so there will actually be a point to using the client.