Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Jimmy V.

Pages: 1 ... 3 4 [5] 6 7 ... 54
61
Could you be more accurate?

You place the shape, it goes somewhere. You go to click the validation button and by the time you do it, the shape has moved?

I am not completely sure how we could fix that. An extra "Are you sure?" seems too much. There is surely a better way to deal with the issue. Maybe we could disable the OK button if the shape has moved in the last 300ms or something.

In the short term, maybe zooming in may help lower the frequency of this happening? The bigger the grid, the more chance you have to be correctly placed with no rounding effects.

May I ask what phone you are playing on? Are they fast or old/slow?

62
Hi Tim,

Thanks for your nice words. The replay feature is indeed particularly beneficial for Cartographers.

Regarding the ratings, we thought we had fixed this but it does not look like it. Did you observe the issue lately?

Regarding the replay itself, the behaviour is intended. In a live game, you can select the pieces even if you are not looking at your own map (then the map switches to yours). So unless I am mistaken, you only see your option. I don't think it is a big problem as the original shapes are always visible. Sure enough it can be confusing sometimes but I think you can figure it out (and you did) when this happens.

Finally, I'd like to add a word on the replay feature. It was programmed in such a way that we have nothing extra to program once the game is developed. The archiving of the moves is automatic and the replay should work out of the box. But that does not mean it will always work exactly as you would like it to work. Some games will have specificities (like HexRoller that allows undos) that will make the replay sub-optimal. I don't think this will happen often, but keep this in mind. The replay feature may have a few glitches that we would correct if it was the main feature of the game. In a way, the same applies to the screenshot feature which can have a few glitches here and there depending on your browser, your computer, the game itself.

This said, if you see something wrong in a replay, please let us know! We'll do our best to fix it anyway.

Nicolas.

63
HexRoller / HexRoller minor bug
« on: 22/12/20, 11:26am »
There a minor bug in HexRoller right now.

If you select the "Place a 3rd group" bonus in a round and then undo it and finish the round with only 2 numbers as usual, the bonus is active at the beginning of the next round (it should only be after you have placed 2 numbers). If you "use" it at that moment, the game seems to continue normally but you only place 2 numbers, not 3. And your bonus seems to be considered used anyway.

So if the bonus is active too early as a result of the bug, make sure that you don't click on it before placing 2 numbers.

64
HexRoller / HexRoller feedback
« on: 21/12/20, 12:32pm »
Please use this thread to give feedback on HexRoller!

To react to the first things I was able to read about the game, we have to agree that some challenges require not only the correct strategy but also a bit (or sometimes a lot) of luck. That is the nature of the game. Some rolls are better than others and you will be more or less lucky each time. We have played all challenges ourselves, recorded all scores and seeded them accordingly. We can assure you that most of them can be done in a couple of tries. Getting 3 stars is of course significantly trickier, but even then, we believe that it is quite feasible to get them in most challenges without the need to try 100 times.

The challenges are designed so as to force you to look at the game differently each time. You should be learning new skills in each of them. If you still struggle with a challenge, I suggest trying the training or online mode and play a good number of games. You'll get experience and probably learn important skills for the challenges.

The question has been raised as to why the game is not 2-player rather than solo. It was a tough choice and personally, I was more rooting for the 2-player variant until I was convinced otherwise. It was a tough call. Given the randomness of the game, being able to compare scores with another player who got the same rolls makes perfect sense. But if you play the online version (for rating), the randomness will even out over a good number of games. So it is not so much of a problem. Your task remains to score as much as you can with what rolls you are given each time.

I want finally to tell you about the rating system. It works here exactly as for Raging Bulls. After each online game (and training too actually, but these games don't count), a performance is computed, translating your score into a rating-comparable number (=performance). If your performance on an individual game is higher than your current rating, your rating increases. You should have no problem increasing your rating in the first 20-30 games as the start rating (1400) is quite low.

Because there are 2 sheets, the performance is computed differently for each of them. Here are the formulas:
Grid 8: Perf=Score*40-500
Grid 7: Perf=Score*40-800
I write this from memory so this may be inaccurate, but in any case, for the same score, you will get 300 more for grid 7 as it is much easier to score with this one. The publisher told us 10 more points were scores on average on grid 7. Our own records showed 7. So we settled for a 7.5 advantage. Before leaving for the holiday, I looked at actual stats and the gap was only 2.5 points. It is still early to say though as the number of games was still quite low (about 65). In any case, we will adjust the gap according to the actual stats as soon as we get the change to access our machines again (unfortunately our remote access is now failing which means we can't update anything now).

Finally the figure that is displayed in the ranking page next to the rating of each player is the equivalent score computed from your rating, meaning the score you need to get to keep your rating stable. Another good way to look at it is to consider that this is your average score (which is basically correct if you have played a lot of games). I have just corrected it so that it takes into account the 2 sheets. The formula for computing is is now correctly this one:
Rating = equivalent score*40-650 (650 is the average of 500 and 800 as you many notice)

I hope this answers some of your questions. Please let us know how you are doing with the game!



65
HexRoller / Re: Link to Hexroller rules not working ?
« on: 21/12/20, 11:58am »
Thanks very much for the feedback. Very much appreciated. We can't improve if we are not aware of issues.

Unfortunately, we cannot access our machine remotely anymore. So we won't be able to change anything until the end of they year. Apologies for that.

66
HexRoller / Beta-testing HexRoller
« on: 04/12/20, 04:37pm »
Hello,

The time has come to beta-test HexRoller.

Please get in touch if you want to know how to do that! To do so, please use our contact form (little pen above the green dice or here)

If you have already done it for another game, the procedure is the same. You need to select "Tutorial + Challenges" and make sure you are in English language (otherwise, you'll get plenty warning messages as translations are not done yet).

Many thanks for your help!

67
Cartographers / Re: Wall of shame
« on: 18/11/20, 11:17pm »
Of course I did it.

What about a wall of honour for those who managed to always avoid the mountains honey trap?

68
I completely agree. One of the many things on the to do list unfortunately.

69
Hi Geraldine,

Does this happen at a particular time? Like 5:13 am English time?


70
Great to read Anne. Thanks!

Not it is your time to teach me something. How do stochastic decision processes work? What did you apply this on?

71
Cartographers / Re: Cartographers simulation v0
« on: 14/11/20, 02:47pm »
Thanks for your response Carl. You understood my question well.

... and luckily it is not that hard to identify the groups on the board.

If you know how to do it, it probably isn't that hard, but if you don't...
I must say I'm far from an expert at programming, I'm not a programmer by profession.
Still, I just had a look at the provided links. Now trying to translate that into working code.
Wish me luck, I'll need it...

Hi Lieven,

If you are not a professional programmer, then I can understand that this may be a bit tricky. It will be a good exercise for you in any case. Carl described it well, you need to be able to find clusters, their list of squares and their neighbouring clusters. Once this is done, most edicts can be written quickly.

What Carl means here is that AI is yet another level of complexity. It is a very open question as how you approach the problem. Some will give better results than others. The computing of edict scores on the other hand is pretty scripted. There are not many different ways to reach the goal, which is 100% accuracy. In the AI field, there is no such thing as 100% accuracy or perfect solution (unless the game is simple enough like Tictactoe or 4 in a row and even Reversi).

With the partial scoring of the edict, Carl gives a good example of what we did with most of our AIs here. Evaluating long-term potentials of moves is key to our AIs. We rely on these human-generated evaluations and never use any black box (Neural Network, Deep Learning, Monte-Carlo simulations) that would give a solution with little human intervention. We do it the (old) hard way, by building mathematical models of the games.




72
I guess she got confused for some reason and when asked whether to flip the column, she said no. Too bad, she would have won.

73
HexRoller / Re: HexRoller under construction?
« on: 11/11/20, 11:05pm »
Oops. Something must be wrong. Thanks for the heads up!

74
Cartographers / Re: Best Broken Road?
« on: 11/11/20, 05:13pm »
Love it. Funny that the missing one was the easiest one in the corner.

75
Tips, polls, updates / Re: How to change email address?
« on: 11/11/20, 05:12pm »
Please use our contact form while connected to send you your request. We will then proceed.

Pages: 1 ... 3 4 [5] 6 7 ... 54