Like here

Also all the ranges are easily accessible and possible to copy individually or in bunches. They are stored in Ranges folder as individual txt files. You can put some of them in a folder and then just send to someone who put that folder in their Ranges and they will see ranges_from_chinagambler in their range selector.

It should be very easy for authors of those tools to translate our ranges. The format is self-explanatary and they are all in one place so it's easy to for example write a converter for the whole folder.

Answering one more question I missed:
We will not make ranges exportable to CREV or anything else. That is responsibility of authors of those tools. We are not going to deal with some propriety closed not very well designed format.
We made all our ranges easily accessible in text files in the simplest format possible. Our range selector will accept CREV format in next release. If there is any other popular tool which people have ranges in and they want that tool -> PioSOVLER direction we will look into it as long as the format isn't brain dead (was borderline in case of CREV).
The other direction really is on authors of other programs. Ask them to do it. We made it as easy as possible from our side to export/import/share the ranges and we don't hide them in some obscure binary blob as some other tools do.

Is the 'Exploitable for' in terms of chips, percentage of the pot or something else?

Chips. Longer explanation is here: https://piosolver.myshopify.com/pages/technical-details
Paragraph titled "Solution trees", 3rd bullet point.

Tldr: It means for how much perfect adversary can exploit us per hand

My laptop's battery goes down extremely fast when running a flop solution even though it's plugged in. The battery icon says (plugged in, not charging) and when I quit the software it starts charging again; is that normal?

Very not normal.
Your laptop should be able to run without functioning battery when connected to the grid. Unfortunately this will be very hard to diagnose over internet. You can try emailing some details (laptop model, OS, etc.) but I think here google does better job than I would.

If I already ran a solution on a certain flop and want to remove one hand from one range to see how it changes the solution given all the other parameters stay constant, do I have to run the new solution from scratch or is there a faster way?

You have to run it from scratch.
This however shouldn't be the case and there are for sure ways to give the solver a head start for this kind of situation. Unfortunately it's not implemented yet and with amount of stuff we have planned as higher priority it won't be in the nearest future.

Thanks for mentioning it. The more suggestions like that we have the easier it will be to design some general solutions to cater for all use cases like that.

Is there a way to set the bet sizings in such a way:

Bet half pot on the flop, if it checks back bet 75%

Not yet but reorganizing betsize input to make it both easy for quick usage and flexible for advanced stuff tops our to-do list.

This is so important, imo, and am glad its on the top of your list. I think its also important to have different sizings for flop raises and turn bets when flop gets called. Something like 1/2p cbet OTF, 2/3 cbet OTT, 2/3 cbet OTR - 3x flop raise, etc, etc. along w/ multiple bet sizes like 2/3p or 1.5x pot OTR to see when and what likes to be overbet without having to have overbet be the only option. (just wish listing!)

Yeah I agree with all of this. We could do "walk the whole tree and adjust manually" thing to start with but we thought it's better/more friendly to start with some simple tree generation so our early customers get used to it and when they start asking questions we would add the whole functionality.
It is a challenge though to design it well so give us some time for few test/feedback iterations before we add it. I agree with you that once you start using it on regular basis to do serious analysis it becomes a necessity. We just didn't expect so many people being at this point 5 days into release
This is btw the reason there is not marketing of our product outside this 2p2 thread, we wanted to get small group of customers and fix stuff for them first; the plan backfired though as the group is not that small anymore.

fwiw betsizes don't make much difference

how do you calculate the solution? brute force method?

Yeah it seems everybody I talked to who tested it in their solvers confirms. Still there might be some differences and if you can model the game more realistically then that's good even if the results won't be much different.

Yup. Just with a bit more force than solvers published so far.

Do you mean as far as EV goes? Because I think different bet sizes should have an impact on the strategy that gets played even if it doesn't change the overall EV of the hand much. Knowing as the IP player you never face a large bet after checking back would seem to have a decent influence on the types of hands you would decide to check back or bet. I would think this goes for a lot of situations when the threat of future action is drastically reduced because large bets are impossible. If I am wrong about this could you explain why. I have ran a few solutions through another solver and the EV is rarely affected (much) by differing bet sizes but the strategies were (I think).

Luckily you will be able to test that in not so distant future

One of the best ways to study the solutions computed from a software like yours is to input them into CREV. Pretty much every user is asking for a way to convert the ranges into CREV ? You seriously are unwilling to do this :/ ? I'm not too keen on paying 500$+ for a software if the developer isn't willing to add a feature that virtually every user is asking for.

We will add (in next release so in days) option to read CREV ranges.
The reason it's not trivial to the other way around is that CREV use brain-dead way of inserting postflop ranges (you can't just copy paste) and keep them in some close binary blob so you can't for example write a converter (as you could easily do with the we keep our ranges as they are public and in self-explanatory format).
I would rather implement every feature CREV provides than deal with software which display those user hostile behaviors.
Again, ask the author, it's his responsibility. We made it very simple to do anything with our software:
-ranges are kept in human readable .txt files you can copy-paste w/e you want; it's also dead simple to write a converter from our software to anything else (other than CREV because there you can't copy-paste and their format is closed)
-same goes for tree configs
-everything solver does is accessible in console and possible to log to file or connect with other software

All those were made this way so it's as easy as possible for people to use their results/ranges and for authors of other software to make it possible to read them.

For now if you would like to insert ranges from the solver to CREV you would need to write an AHK script to click every single hand because it's not possible to just copy-paste. The way the author of GTORB did: https://www.youtube.com/watch?v=buIXTjwfgXY

It's CREV author responsibility to deal with that and fix his code.
We are not charity to support and fix legacy code of other poker tools, especially those which were planned to make importing/exporting results difficult.

Let's make it clear: if CREV had open format available for pasting postflop ranges then even despite my opinion about that software we would still consider making a converter for our users. The thing is though that CREV was specifically designed so you can not do that.

Thanks for the good work!

Could you explain how the betting structure is used?
If you enter 40 80 160, will it be used on each tree level or will 40 be used after a check/check?
I guess it is the later as 40 80 160 can be used on first street, then I would prefer a betting fraction of the pot to be able to be consistent on future streets.

Also is the syntax 3e1000 a geometric growth of betting on 3 streets with effective stack 1000 ?

How would you do if you want to test two strategies, one with betting 60% of the pot or checking at the river, and one where your options would be pushing or not (but on the river only) with normal 60% betting beforehand ?

I tried to use for second case where the game went check/check, bet 60%/call, push: pot 100, bet: 60 940 and compare with first betting structure, but I'm not sure it makes sense.

40 will used for the first bet made so also after check/check.

3e1000 means that the effective stack of 1000 will be distributed to make the bets equal in relation to the pot (every bet is the same % of the current pot) so yeah, it's a geometric growth.

As of now you would need to make separate trees and calculate them independently.
If you wonder how to build those just give me some example spot and I will tell you.

I am not sure if I understand it. If you want to make river bet a push no matter what came before (but to have the bets before being standard size) this is currently not possible (but it will be along with all other things you may want to adjust concerning bet sizes)

Not sure I get it.

Take a pot of 80 and a betting structure for half-pot 40 80 160 320...

If it goes bet/raise/call, so 40/80/80, the pot is now 80+160=240, what is the betting size on next street, 40 again or 160 which is not half of 240 ?

Wouldn't it be easier to describe the betting here by 0.5,0.5,0.5,0.5 as a fraction of the pot when you bet?
That would allow you to test cases where you increase the pressure when called with 0.5,0.8,1.0,1.0...

You did not explain what the 3 was in 3e1000.

Those are cumulative amounts, so with starting pot of 80 to get half pot bets you need to do:
40 120 280 etc.

It's easier to input using "Choose Pot and Bets" button and just follow the instructions.

Yeah, you can do:
50p 50p 50p 50p which means bet 50 percent at every opportunity. Or you can do:

50p 50p 50p 975 which means bet 50 percent at every opportunity but go allin for 975 total after that (so say with 97.5bb effective stacks at 5/10 blinds)

Sure you can do that in either of 4 ways (cumulative amounts, percentages, percentages with last number being cumulative amount or just using Choose Pot and Bets button)

Right. XeN means that N will be split into X equally sized bets (in relation to the pot) so 3e1000 means that it will be split into 3 bets (perfect for 3bet pot for example).

You can import preflop ranges into CREV in text input, as a whole textblock, with weights. I can't try from PioSolver right now because I dont have CREV on this computer, but if the ranges from PioSolver are exportable into a format which are accepted by CREV preflop range, that should be easier than having to deal with the postflop scenario. Some of the GTORB ranges are exportable into the CREV preflop condition text input (I say some because for a reason I'm not sure yet, it sometimes works and sometimes doesn't, depending on the range).