Still a little bit hand crafted and not fully automated to do the the calculation.
It won't be perfect. The correct conversion from cEV to $EV only works if the bubble is achieved. But at least in the end the final value will be correct.
With that luck I will end the challenge in 180 days on these highest stakes.
Phantom hand. That sounds like a good name for a bug!
The step from 12 to 13 going down 3.3 should not exist. But going down directly to -13.3 at 12 sounds wrong to me as well. I guess that's the reason nobody else plots it like this...
Found my next 3 hours of video to watch.
Always that though when I lose with the better hand that should put in more bad hands.
Friday and Saturday were lovely days. Just repeating them would already be a very good goal.
Wanted to make a critical comment about from zero to hero. But as I suffer myself, I could not provide any insights into how to leave those kind of challenges behind and evolve them into a hero to legend path. For example now that I learned how to write tests for my tool I mostly forgot the rest. Don't recall how to do it... It's not that it is more appealing to start from scratch it's very hard to get back into all again... Everything needs to be documented better.
Restart everything. Buy a new game on Monday. Create two new session in two other games.
Yes! In a satellite tournament, for which I won a ticket to in a satellite tournament I won a satellite tournament ticket to a satellite tournament where first place wins an entry to a live event!
Meanwhile for three days the AI disappointed me. The only thing I wanted was a simple example how to use the database sqlite in a new project. Didn't care what it does. The only thing that was required was the project should compile. Three days not a single AI output worked. When I asked it to fix it, the AI run for 15 or 20 minutes before I aborted it. For three days it was not able to write an example for the most used software. 1 trillion databases world wide in sqlite3 but the AI couldn't provide me a working example. Till today. One letter was missing - C.
Found some preflop ranges. Can I trust them? Better than mine because I have none... Still want to write my thinggy... Also have my old phone where two poker trainers are installed. Can I trust them? GTO didn't change since then right? From a time where you were able to pay a one time fee for a license.
Speaking of license. I asked an AI if it could provide me with a starting range for a certain situation. It refused because it can't just copy math solutions from other sources but is able to give me pointers.... But I will try to experiment a little bit more with that. Do not want to wast the full wisdom from poker reddit.
Couple a days later and I learned how to change add and structure a C++ project. I start to understand what those makefiles do. Just for fun I even added a the game engine to the project do see If I could. It feels like I was never so close to have the tools to tackle that big project. To help me I created one AI project. Instead that the AI will code my project it tells me what do do next. I report back and it keeps track what I did and tells me what I need to do next. It's only the second day but so far I like it. I don't worry which part I should do before the next. I understand everything I did. And I still have the joy of solving the problem for myself.
The only downside is that there is nothing to show. It's like in school when they told do for an assignment that no result is a result too. But how should I sell that nothing to you? Except trust me bro! I am able to code longer with less distractions. Just the second half of the day when getting back doesn't work yet. Don't know what the right approach is. Open the editor and hope I will works something. Ask the AI a lot of questions. Or just leisure time for the rest of the day...
Anyone can use it. The only confusion might come from when you are talking about the compiler.
Use GCC or Clang and you are save!
Avoid msvc compiler (Microsoft Visual C+). Which is free to use but when your company would get bigger than certain financial threshold you need to buy a license. Similar to game engines like unity or unreal.
That's very interesting. I once loved to build with C++ for Microsoft Windows OS, but nowadays nobody is using that OS over here... Everyone is travelling fast with their AndroidOS small touchscreens, which looks expensive and boring.
That's something I will need to do as well. Without a further step compile to an android system.
Don't know if it's easier to write my own GUI or mix it with with a native app.
Looked all a little boring i the terminal...
So I made a little side step!
And the number actually counted upwards!
Yes imagine the face of an Android user when they press a button on their phone and something like this loads Full Screen... Is very tuned down nowadays with only Google Store applications allowed natively and stuff like that.
Free form writing:
I have no clue what makes an app better than other. It's those small details. Over the years I installed a couple of poker apps on my phones. In general they all do pretty much all the same. But still everything feels so different. Right now I installed the next app. Maybe it's only buyers remorse. But so far I like it. Not as much detail as others but I am able to handle it without losing oversight. In total I spend $599 in a WSOP summer sale.
For almost a week I am studying daily quizzes. Instead of educated guesses I look everything up. The downside of doing it this way is that there are no wrongs or blunder to review later. For others might the fact that a hand takes 5 minutes to play through a negative as well. In my opinion I need to dig even deeper and tank way more when looking all up.
300 days do go through everything. Then either back to the other tool that was too complicated or have finished my personal trainer.
What is the status of my trainer thingy? Started to write something like a TUI engine. There would ncurses as available library but writing my own and keeping track of the cursor myself is fun as well.
Used the AI to write me an example how to read files in a zip file without unpacking. That will help to reduce the clutter on the disk.
Anyhow decided to deduct the expenses for the sale from the bankroll. I am not in a hurry to achieve the goal.
Summary Jun
37 tourneys 1623 cash game hands
$-524.82
$1,874.12
I have no clue what makes an app better than other. It's those small details. Over the years I installed a couple of poker apps on my phones. In general they all do pretty much all the same. But still everything feels so different. Right now I installed the next app.
Yes i think it is possible to make good software application. For example I love how Quotes on this forum are solved, while YouTube solves it by eliminating chat sometimes; and Meta solves Quotes by throwing these to your face if anyone has reacted or said anything under the same topic (they call it posts).
The last session I played was very pleasant. Maintained focus and was aggressive from the first till the last hand.
Still to learn a lot. I just don't know if I need to build intuition or understanding. No body definitely can say the solver used a line because of this or that reason. It's all hindsight and human pattern recognition that try to add some sense in the output.
Hopefully coding is solved in AI. Because I spent two hours chatting with a bot on how to make loops faster and rewrite masks.
The big question is now should I add what I learned to my equity calculator or start from scratch. Because there is a second issue. I need also to evaluate draws or distinguish between top two pairs vs bottom two pairs. All the debt piles up and I almost fear it's cleaner to do all over again...
Longtime no coding progress. Not a single line. But suddenly out of nowhere I now know how I need to store ranges in my uncreated solver. A quick function to adjust it each street. Fairly easy to get range vs range. Weighted ranges that is - after picking different mixed strategies.
It's so little code... which makes it beautiful but also makes me feel like an imposter as there isn't much created at the same time.
Asked the AI how to cut all down to thr smallest possible pieces. The idea is that those pieces can be replaced easier from someone with more experience in the future. And that I have quicker something that is useful.
Man, it's some many bits and pieces...
But to the AI stories another time...
Is the "AI" similar to the "Terminal" in desktop computers? Typing commands in a programmer-themed atmosphere? Lenovo and Samsung have recently tried to turn the more-fun-oriented chat robots towards only-fun; such as "Billy", a famous offline chat bot computer game from 1990s. Or is the main theme of "AI" accessing online libraries? I have heard of Bezi AI that can be plugged to your offline library to retrieve knowledge and fix a messed up offline game project.... Thanks!
The new AI models from the big companies. Many complained that they burned through their budget so fast. But people don't read. They wrote you should start with the weakest level of the model and look if you need more AI power. I request my AI to designe the GUI - and I must say I am impressed. Didn't know that creating something this beautiful was possible with that software. I guess I need to make several more suggestions but that first result looked professional to me.
And then I was trying to figure out how much I still remember of SQL. Spend the whole day on it. Approximately did half of it. Was worth doing study session instead of looking up what I need while designing. Second day will follow today.
Also fixed my equity calculator. Now I can start a clean setup again. August 3rd till September 2nd everything will be glued together and finished. I will feel like Frankenstein!
The first shot of the AI. (a part of the screen). Maybe somewhen on February 1st 2027 my tool will look similar this!
But my level of skill is not up to scratch. I tried to continue on the database part but I still have troubles to understand the advanced stuff. the only solution is to repeat each day the same conversation with the AI till I can do it on my own or at least understand everything the AI has produced. But I also realised that my expectation on myself is way too high. It's now the third day I dabble in SQL. A course at the university takes 15 weeks with two to four hours each week.
How close could I realistically be after two days light chatting about the topic?
Found a lovely new spot on a bench under a nice tree where I was playing a few hands while some runners and dogs passed till the sun glared too much.
Originally posted by Rhodriguez
A 2nd goal will be to achieve the activity badge from my phone 25 days each month.
In June I failed. Only a third of the required days I managed to succeed. Which is not too bad when considering that after struggling at the beginning because of my level I needed to abort. And despite life doing life things I crossed the needed thresholds.
This month - July - it took me 26 days to achieve the 25 days badge! Someday were effortless others I needed to add more and more to cross that line. Last Sunday I skipped today was difficult as well.
Next month I will try complete it again. Till then, expand my route and daily routine, so it will conclude naturally.
1st February is 190 days away. I mentioned it jokingly the other day. But maybe this should be a goal. Requested a small layout for an equity calculator. Not bad but has fields that made no sense.... but we will make it work!
Maybe next year also play on a different site. If not fully achieved the transition.
27 weeks and 1 day you say...
