Yes, most likely this is indeed because that option is ON. There is no limit to how large bets can be, so the bet being 145% should not be any problem. If you would like me to take a closer look, then please send a savefile to support. That being said, please consider just using a single size on the turn and river (assuming this is a flop file). It will not significantly affect the solution, while making the tree much smaller, solve faster and be easier to interpret. When going beyong the entry phase, there's really not much added value to making trees overly complex. Trees with single bet sizes are already hard enough to interpret; interpreting multiple bet sizes across 49 different runouts of the turn would be very hard to achieve; at the very least I personally would not be able to do that.

Ty for your fast response! Y i know it's impossible to interpret all those sizes and ev is similar just wanna see preffered sizes in different flops and runouts.

Please note that there isn't really any such thing as a preferred size. Any bet size will have a GTO strategy that will perform almost the same as any other bet size. You can check this for yourself by creating multiple trees, each for a single bet size. After that, check OOP's overall performance (go to the very first decision and check for the total EV under the table; see screenshot below). As you will notice, the overall EV will barely change, no matter which bet size is used. Even using multiple bet sizes instead of a single one does not lead to any significant improvement. So when doing analysis, it's best to just pick a size that you like.

Does anyone know if GTO+ supports multi-cpu systems e.g. dual xeon CPUs? I am trying to think of a cost effective solution to just run sims 24/7 and I was thinking old server parts but not sure how well it will work.

You should be able to test this with the trial version, which is already fully functional.
The only restriction is that the flop can only be Td9d6h.
However, this restriction should not affect the ability to test solving speeds.

Thanks but I already own GTO+ and not the hardware I was looking at using. Right now I am just using it on my home PC. I was more just wondering if the software would use two CPUs if available.

I have no personal experience with this particular setup, so unfortunately I don't have an answer for you here.

Hey Scylla I am trying to move beyond flop work and now into turn work with GTO+.

I am analyzing a Cut Off vs Big Blind situation in a 6max cash game.

The flop is T93

The flop goes - (BB) checks, CO cbets 75% sizing and BB calls.

The the turn is a K

How do I figure out if that turn card helps the Big Blind or the Cut Off more?

For this, look at the turn report (bottom button in the first screenshot).
The hotness will tell you who benefits most.
Here, yellow is neutral, green is to the player's advantage, and the red variants are to his disadvantage.

Scylla, consider adding a tool where you can appreciate the comparison of hand strength as seen in the photo?


https://imgur.com/a/8PH3joS

Is that from PIO or Flop Falcon? lol

Ok, I will consider it for later releases.

Is it possible to have more than 5 betsized besides all-in ?

How do you add betsizing smaller than 10% ?

No, at the moment the maximum is set at 5.

Bet sizes of less than 10% don't really make a lot of sense in practical spots, and for this reason the tree builder does not accept them. The tree editor however allows any value. So as a workaround, please consider using the editor for this.

Not a good design, it took some time before i realized there was a 10% minimum (it just got red with no explanation). Found a spot where the ideal flop size is close to 10%.

Being able to have more than 5 betsizes besides all-in would also be nice.

From your description I get the impression that you're using multiple bet sizes under the assumption that the preferred size will have the highest frequency? This approach actually does not work, given that the other sizes will have similar performance. The performances across the possible bet sizes are extremely close, and when using this approach, it's mostly just a coincidence which frequencies the solver lands on.

Contrary to popupar belief, there isn't really any such thing as a preferred size. Any bet size will have a GTO strategy that will perform almost the same as any other bet size. You can check this for yourself by creating multiple trees, each for a single bet size. After that, check OOP's overall performance (go to the very first decision and check for the total EV under the table; see screenshot below). As you will notice, the overall EV will barely change, no matter which bet size is used. Even using multiple bet sizes instead of a single one does not lead to any significant improvement. So when doing analysis, it's best to just pick a size that you like.

Hi Scylla,

I was trying to pot a variable in CREV and was getting an inaccuracy, I noticed the variables can only handle 1 percentage point and that was the reason of the inaccurate results.
Thanks

Ok i take a look at the performance impact later for flop, turn and river.

Maybe i can simply use fewer betsizes and not have to upgrade my computer, i usually use (1/4, 1/2, 1, 2) pot for turn and river (+ all in).

I don't believe that we round the values, but I'll check for later releases.

I would recommend only using single bet sizes on the turn and river. Your trees will be a lot smaller, and solve much faster, while not affecting the solution in any meaningful way. There's only very limited value in creating huge trees across the 49 possible turns and 2352 turn+river runouts, given that (at the very least speaking for myself) the human brain is not equiped for mass data analysis. The main purpose of the turn and river lines is to estimate the performance of all the hands after the flop play has been completed; for this purpose single bet sizes are sufficient.

Hi Scylla
Can we import a card ranking into GTO+? I have a card ranking I like in CREV, and would love to import it.
Thanks and regards
Rod

Yes, this is available.

The CREV ranges are stored in the file newdefs2.txt.
This file can be found in the same directory as CardRunnersEV.exe.
In most cases this will be C:\Program Files\CardRunnersEV\CardRunnersEV3.

GTO+ uses a different format, namely newdefs3.txt.
This file is stored in its subdirectory /config.

To import the CREV file, place the newdefs2.txt file in the same directory as GTO.exe.
The default directory is C:\Program Files\GTO
After that, start GTO+ and go to "Settings->Import preflop ranges" in its menu.
Select "newdefs2.txt" and click on "IMPORT RANGES".

See the screenshot below.

Thank Scylla. I am looking to import card RANKING though, not ranges. I will try something along the lines above and see if it works for rankings.

---
It worked, I just had to edit the read-write permissions on the card ranking file.