GTO+/CardRunnersEV?

When multiple bet sizes are available, the performance of a hand at any bet size will be extremely close to the performance of the same hand at any other bet size.
Because of this, it's to be expected that two different GTO algorithms will arrive at a different distribution of the hands amongst the available actions.
Also, the trees are not identical, which is an additional reason why a different distribution might be found.

The trees are not the same, which means that the solution will be different.
It's not enough for the trees to be almost the same; for any comparison to be made, the trees need to be exactly the same.
And even then, it's often possible to find different solutions within the given Nash distance, with both solutions being correct.

Once again, unfortunately I don't understand the feature that you're describing, however, most likely we don't offer such a feature.

Hi Im having a problem with my flopzilla license
I tried to find out how send pm but did not find

How / where can we speak about it?

I have replied to your PM.
If you need further support, then it's probably easiest to contact us with the form here: https://www.flopzilla.com/contact/

Hi Scylla,
I've just upgraded to the beta version 1.7.4. Everything works fine, except now, when opening the app, it takes a little bit longer than usual.
How to revert back to 1.7.2?
I tried to download 1.7.2 and when I ran it, it showed like this:

Thank you in advance.

For this, uninstall v174 and then install v172.

Regarding your issue, do you mean that the software takes longer than usual to start up, after you double-clicked its icon?
You may want to try setting an exception in your virus scanner, given that it's a potential cause of the delay.

Thanks for the advice. Will do

Scy, how are you? I understand that sometimes you don't want to discuss many of the updates that are planned for the future, but I think it would be a good option for users to know the roadmap.

I mainly wanted to ask if you plan to update the trainer (mainly small things in the interface that become tedious when you want to go over the 'error' and do the hand review).

I think what the trainer is missing is precisely that—easy navigation when reviewing hands (currently, it's done like play/stop, and it's one of the most annoying things) instead of being able to skip streets or click to go directly to the exact point of the error, make notes, some filters etc.

Without a doubt, with the latest updates the software has had and improving small things in the trainer, what you're offering is among the best on the market.

thanks in advance!

Ok, thank you for the feedback.
I have made a note of your request.

Thanks Scy.. Another minor thing

In this tree, I applied negative incentives to a bet size on the turn (I applied it to all) so that it would only take one single sizing.

The only downside is that I can't see the new frequencies or the EV loss due to simplifying the strategy in the turn reports.

In other words, I have to look at each runout individually.

This is different in the river reports. (image 2)

Here im apply negative incentives to allin and other sizings, keeping only two.

In the aggregated river reports, we see a general "resolve" for all possible runouts, and this updates the strategy in the report itself (EV, frequencies, etc.).

Could this "resolve" be applied on the turn to all possible runouts?

The time required for resolving all turns would be roughly 49 times the required time for resolving a single turn.
For the river this works relatively smoothly, but for the turn (and depending on the tree) this feature might take a considerable amount of time to run.

https://gyazo.com/33ab6b31b85f3c93a30b4c...

When I want to export a flop from DB distribution, it doesn't open the selected flop when using alt+click r.

I have tried, and it seems to work over here.

Does it happen for other databases as well, or just this specific one?
And can you, for example, export a turn/river line in the editor?

As for causes, can you try setting an exception in your virus scanner?
Also, please check if the /tmp subdirectory is still present.
The full directory will most likely be C:\Program Files\GTO\tmp.

Should you have installed to a different directory, especially if it has Chinese or Russian characters in it, then try installing to C:/program files.

Does CREV export to Flopzilla Pro, similar to GTO+? If not, it would be a nice feature.

Hey!
I'm about to run a bunch of sims every night. I usually use the same flop subset and run it for different stacksizes/spots.
I want multiple the sims with different stacksizes to run at night so I can review them the next day. I will be doing this for approx 5 weeks.
Yesterday I set it up to run overnight but I didn't take into account that to run all instances of gto+ I need a lot more RAM. So I woke up to many flops being unsolved due to RAM implications.
What's the best way to get around this?
Do I need to merge all files and make 1 huge database and then export all flops separately and then merge them again? I'll try this tonight most likely but it seems excessive.
Is there a way to let a solve start after the RAM is available again?

CREV does not offer this feature, but I'll make a note of your request.

If possible, the easiest approach would to purchase additional RAM, given that RAM is both relatively cheap, and easy to expand.
This is of course assuming that your motherboard has RAM slots remaining.

Without additional RAM, an alternative would be to make a copy of the database.
Then, delete all the solved trees from the copied database, so that only a database with unsolved trees remains.
From the original database, delete the unsolved trees.
Solve the copied database, and merge the two databases upon solving.

Removing only unsolved/solved trees from the database can be done with a single click.
For this, press CTRL or ALT when clicking on "Clear database".
See the screenshot below.

When the entire database is solved, and only trees remain that could previously not be processed, when you click on "PROCESS DATABASE", GTO+ will again attempt to solve the unsolved trees. Should not enough RAM be available, then this will be detected prior to starting the solve. So, this would not be a time intensive process if there's still not enough RAM.

More RAM has been bought! However, I think I will eventually still be running into the same problem when solving blind vs blind solutions. In the best case scenario I want to run 4 instances at once so I think 64GB will not be enough. Either way I'll find a workaround. It could be a cool option to let the solver wait for the RAM to be available again but I'm not sure how hard this would be to implement.

How to convert a solved solution to a CSV file

Hi,
Few quick new user questions:
1)Should each database have the same tree so that aggregate statistics comparison across flops is meaningful?
2)Can i copy a database from one computer to another for playback (will buy the additional license).
3)Normally what level of precision will allow the see the impact of blockers and card removal? (.25%?)
4) I noticed oscillations on some extensive trees (what is a reasonable cutoff time for a 32 thread 1228 gb memory desktop)
Thank you very much. Great product.

A lot of data from trees and databases can be exported from the interface.
If you're looking for something specific that we don't offer yet, then we can consider adding it for future releases.
We would however need a description of which data would need to be contained in such a CSV file.

1. Aggregate data will only be meaningful if all the trees in the database are identical (except for their flops)
2. Yes, savefiles from one computer can be opened on any other computer
3. This does not depend so much on precision, as it does on tightness of ranges. Blocker and card removal effects will typically be almost imperceptible in wide ranges. It's only when ranges are extremely tight that card removal may have a noticable impact (for example after bet-raise-3bet). But even in tight ranges the effect is often small.
4. This depends on the size+complexity of the tree and the wideness of the ranges. However, even when knowing this, it's still difficult to give a predication.

Hi,
Twice the program has crashed while saving the large file after an extensive run with wide ranges and multiple bets.
The optimization finishes but it crashes while saving the solution.
Thanks.

Ok, can you send a savefile to support?
I'll see if we can reproduce the issue.

it crashes and doesn't leave a savefile. i downloaded beta to see if that works.

beta also crashed. where do i send the savefile?