Hey thanks a bunch for the effort. As much as I agree with your premise, I was about to say that it was so bigger that become "unfit" in the screen. At least on PC.

I just emailed you now with some PRT SCR comparing my pgnweb and Olga view from the comp.

You can either email them to chess@chessgames.com or upload them to a public repository such as http://imgur.com.

Also, there's a very special edge condition I'd like to witness: a game in which a three time repetition occurs, but is not claimed, and one player ends up winning. If somebody can find such a game please share it here.

<SQ> found a game where the notes are off-by-one; I'm sure that's easy to fix once I examine how it parsed the notes out. I am pretty sure there's a bug in my code there. It's been tested and usually works fine but this game at least proves one exception.

So we have a small change list already. Keep 'em coming.

I forgot about that game. What a tour-de-force of creative writing! I'm glad we didn't delete it as a hoax, for it's actually proved to be useful in our testing.

After some coding changes I think I identified the problem of the misplaced notes (but please don't stop checking that and other games, I have fears that the misplaced notes bug is liable to rear its ugly head again in another manner. I may have to tackle the problem with a different algorithm so that it's not as sensitive to PGN variances.)

However I'm not quite sure what you mean by <confusing �sub-variation�> in the notes. I don't see broken characters on my screen in that game; so perhaps it's platform dependent, or perhaps it was already fixed with a previous change, or maybe you are referring to other games? Please clarify, and thank you.

This leaves us with only a few options.

1. Modify chess.js to fix the problem, and donate the fix to the chess.js community, thereby solving the problem for us and countless other applications.

2. Leave things exactly the way they are now (buggy) and when the new version of chess.js comes out the problem will solve itself, assuming that issue is addressed.

3. Strip out all notifications from Olga of when a 3-time-rep is claimable whatsoever; just pretend we know nothing about it. (It's sad to lose a feature, but a feature combined with a bug is kind of ugly.)

If time and money were no object, I'd opt for #1. But the reality is that #1 may be a good deal of effort, and I'd hate to focus all our energy on such a minor detail when we have bigger issues such as the overall look and sizing to tackle.

What we did:

You can now change your board ratio from 20%-210%. Yes we now allow you to increase the size even more! This will then "stick" like the color settings if you wait the required 10 seconds to update the server.

With that change comes scaling labels. If you do use the labels/coords (rank and file), after deployment you will notice they scale appropriately with the board.

A note on size - it seems most people (that have commented) would prefer a smaller board. One thing I did not mention was that the size of the board is independent of your browser scaling %. We designed OLGA to look very good at 100% scale. if you increase the scale in your browser (ctrl + '-' or ctrl + '+' on most desktop browsers) or pinch to zoom on your mobile. The board will still function and allow you to make variations.

It is easy to say "it just doesn't need to be that big, ever". But you are not properly thinking of the many use cases for OLGA.

For instance, on a large projector screen an instructor can "upscale" OLGA to take up the entire screen and use the arrow keys to move through the PGN. Therefore, hiding the score from the audience. As they work each ply, they might stop and pose a question to the class as to what the next move or a variation. There are many more use cases of this nature.

To just reiterate on fonts -
You don't need Helvetica. Just 1 of any of these fonts. Helvetica, Helvetica Neue, Verdana, Aria
For the most part they are fairly standard. I for instance don't have a single platform, windows, linux, mac osx, android, or blackberry OS that does not have at least one of these fonts. If you are using english as a second language however, you might not. Rather than add everyone's font to the list, simply installing your favorite out of the existing list will ensure you have a font that can display the characters. Helvetica, Helvetica Neue, Verdana, Arial

Bug reports on fonts, also please report what font you are using at the time of the bug.

Font: "Arial" - for example.

If you don't know how to do this follow the directions in this link (look for the answer with 75+ upvotes) https://stackoverflow.com/questions...

Anyhow, I'll let Daniel do the actual announcement for deploy as he will likely have some comments as well.

Basically, any character that isn't in 7-bit ASCII shows up as �; in this case, the quotation marks around the word “sub-variation” are the left double quotation mark and right double quotation mark.

Even the ö in Ulf Hammarström's name (in the notes to Leko vs Kramnik, 2004) shows up as �; and of course, there's no hope for chessy characters like the ellipsis.

Look at this game first with Olga, then with pgn4web: McDonnell vs La Bourdonnais, 1834

Olga completely ignores variations and comments within variations, while with pgn4web you can click on variation moves and show those on the chessboard.

This might work as a short term fix... but don't you think it would be much better to have proper pgn with variations that the user can actually click through? The added value from "reading" the alternative to White 12th move (12. e5 c6! 13. exf6 fxg2+ 14. Kxg2 Bh3+! 15. Nxh3 Qg4+ 16. Kf1 (16. Kf2 Qf3+ 17. Kg1 Ne2#) 16... Qe2+ 17. Kg1 Nf3#) to actually "seeing on the chessboard" is massive.

The reasons for this are related to legacy: there was a day when some PGN viewers choked on RAV, others didn't support it, and some attempted to support it but did so in such a very awkward way you'd wish they just spelled out the notes in the straightforward {braces} method. MyChess, which used to be our default viewer, was the main culprit.

Another issue is that we generally don't like annotation unless it's attributed; and 90+% of the RAV we received was done by a mystery author with incredibly deep sub-sub-subvariations, so complicated that you wished you could just see the moves of the game.

Because of these factors, from an early stage, we decided to encourage games with annotations using braces but to avoid (although not 100% outlaw) RAV annotations.

Olga does not currently support RAV, although it should function properly on RAV games other than the fact that it suppresses the RAV. You pointed out a problem with the notes that was a side effect of RAV; I wanted to check that thesis by removing the RAV on the La Bourdonnais game, sure enough, that's what was confusing it.

Will Olga support RAV in the future? It should, otherwise we can never call it a "fully functional PGN viewer", but the real answer to that question is one of user demand and funding. Against the backdrop of CG containing very few RAV games, it may very well be postponed for a while.

Chris found a simple way to make a side-by-side presentation with a minimum of changes. A drawback is that for particularly long games it could look a little odd as the moves flow below the diagram, but I tested it on the infamous I Nikolic vs G Arsovic, 1989 and even there it was acceptable.