You can add the same tree multiple times, by clicking on "Add current tree", with no content in the field after it.

Ok, I'll see if we can manage to make this work.

Sorry i found the button "add current tree" and works fine , before i used to do with the "+" button. Thanks!

hi, I wanted to ask you to add the possibility to change the display of the interface of the size of bets from the big blind to the percentage of the pot! circled in red! I will be very grateful)

Ok, I'll see if we can add customization here.

Yeh, that actually would be helpful. I miss that sometimes.

Something like "1.5 (33%)" instead of one or another, or even better, an option to chose whatever we want. One or another, or both.

Enviado do meu RMX3370 através do Tapatalk

hey just wondering if it's being touted in the future to be able to (latest pio version can do it) scroll and progressively nodelock a range? so rather than individually nodelock each combo, scroll a bar and it will gradually change the bottom/top % of the range the further u scroll it?

thanks

I can't seem to find such a feature; if possible, please send a mail to support with a further description, and we'll take a look at it. Should you be referring to the ability to lock the top/bottom/other part of a range to a certain action, then this can already be done with the large bar below the starting hand matrix (this bar contains all hands in the range, ranked according to equity).

scy how are you?

I think that here you can see what several users are requesting regarding visualization and features:

In the gif 1:

1. display of horizontal bars.
2. Possibility to select several parts of the range at the same time, while it is shown in the matrix and to be able to visualize the overall frequency of said group of hands.
3. The number that appears next to each bet size is the frequency difference (higher or lower) with the general frequency of the node.

https://gyazo.com/a478ba5e5d3ed55fee6ee9b55d7897cb


Gif 2.

1.It is similar to the interface of the new range comparison module we have, but the matrix is ​​much friendlier this way and depending on who has to act, it shows you the strategy (the GTO+ ​​matrix in this feature is a bit ineffective in the current mode IMHO)

https://gyazo.com/b9873cff59c25d8f8c7100e15a7479d1

could you do something about it?

Ok, we will take it into consideration.

hey, encountered some bug latley. after solving the tree is not updated and still has its intial values. i have v1.5.2. is this bug known? probably has something to do with extensive storage, which i just turned on lately. also this happens (just) when i cancel the solve manually.

I'm not aware of such an issue in v152.
Can you send a savefile to support?
To bring down its size, store it with "File->Save: Convert to Basic storage".

i dont have the file anymore, i'll keep an eye on it and send a file when it happens again

Is there a possibility to drill a current decision vs database? Ex. I want to train facing check-raise on different boards without having to manually changing them.

At the moment this is not available, given that there would be spots where this feature would not work entirely smoothly.
The problem is, that it's not always possible to predict in advance whether a spot in a tree is suitable; in rare occasions this could lead to long loading times between hands.

Ok, having had some time to think about it, I believe that one scenario where the solver stops is when either NAN (not-a-number) or INF (infinite) is detected.
When this happens, we set a random distribution in the tree, so that we're sure that the user notices that the tree was not solved.
The presence of NAN or INF could indicate an issue in the solver, although this is the only report that we have received about this so far.
An alternative explanation would be that there's an issue on your computer (for example with your RAM), due to which the calculations are not performed correctly.
Should you have customized your BIOS settings (such as overclocking), then please consider resetting the BIOS to factory defaults.
In the event of the issue occurring again, please send a savefile to support.
We have made sure that all trees are processed in the exact identical way on different computers, so if this occurs on your computer, then we should be able to replicate it.

I was just running a sim for a 3bet pot and a thing came up that I couldn't get my head around. I forced the OOP to check by putting 0 in flop 1st bet and put a 50% cbet option for IP. Since the board was quite good for OOP, the cbet % for IP was very low, around 15% of the time. Because of that I tried to simplify the strategy by forcing the IP to always check flop. And to my surprise, the IP's EV slightly grew, and even more surprisingly OOP's EV also grew a little bit. I assumed it was due to rake so I ran a no-rake sim and finally giving IP an option to cbet actually increased its EV slightly, at the same time decreasing OOP's EV.

My question is: if checking the whole range would generate higher overall EV than implementing some cbets(even if it's only due to rake), then why does solver implement some cbets when it's only optional and not a mandatory action?

I don't think the results were skewed because of sim inaccuracy as I set Target dEV to 0.01%

When using rake, there's an infinite number of solutions within the target dEV (with no rake, and a dEV of 0, there will only be a single solution).
However, the GTO search algorithm is a black box, and it's not possible for us to determine why the solver landed on any particular solution.
The solution that was found will definitely be within the stated dEV though.

To elaborate on the above request, ability to drill X node vs various selected trees from a database (e.g. facing XR on A-high boards) would be immensely useful. This is the only feature I can see that the trainer is "missing".

We can consider it for future releases. A drawback of this feature is that we don't always know in advance whether a spot in a tree is suitable for playing against it. As a result, some spots may exist where we may need to try several random spots before finding a suitable one. So, depending on the spot, it can happen for there to be long loading times between hands. This will be a rare occurrence, but it's something that we first would need to find a solution for.

Are you able to play against the solution on multiple tables simultaneously in GTO+? I know Pio allows users to play up to 4 tables. Cheers

Within our approach for the interface, I don't think that it's necessary to have multiple tables.
There's no thinking time involved for villain, so you can play through the hands as quickly as you like.
I you really want multiple tables, then I suppose that you could open up multiple instances of GTO+.

Judging by the description on the site, an excellent update 153 has been released! But I can't find a link on the download site 153, there are only 152(

Sorry about that, we placed the release data a bit too early.
It should be available either Friday or Monday.

Scy Hello, on the new mode for bars, the data show us is incorrect. here an example