Hi Punter, below is what I am referring to, there are rows where the turn card is blank

Can you send me the original .csv file? (either PM or support@piosolver.com).
We need that to see if it's just the program displaying it losing the cards or if the file doesn't contain the turns. It would be useful if you knew if the file was modified by your program as well (you can look at last modified date to see if it's very recent or if it corresponds to the time the report was created).

Hello all, I have posted the hand below in question between the brackets. This was Button vs SB SRP, 120 bb effective.

I am wondering as to why A8o (which I had in game), has such a high percent of checking range when making 2 pair, and in position? As well as even further as to why the different A8o combinations have such different percentages of check and bet (As8h checking 93% but As8c and Ac8s checking half and betting big half). The majority of A8 hands are predominantly check heavy though.

Am I doing something wrong in my setup? Why would this be? I figured having 2 pair here would almost always be a bet. I am very confused. Any Advice would be appreciated. Thank you.

https://ibb.co/R3Qd3nF (here is the flop)
https://ibb.co/qYTBvdF (here is the setup)

PS I ran this again giving river re raise sizes and I am still getting similar result.

[#TreeBuilding#V2
#Range0#AA:0.15,KK:0.1,QQ:0.05,JJ:0.25,TT:0.3,99:0 .4,88:0.5,77:0.4,66:0.4,55:0.4,44:0.75,33:0.75,22, AK:0.25,AQs:0.35,AQo,AJ,AT,A9,A8,A7,A6s,A6o:0.5,A5 s,A5o:0.55,A4s,A4o:0.25,A3s,A3o:0.2,A2s,A2o:0.2,KQ s:0.45,KQo,KJ,KT,K9,K8s,K7s,K6s:0.5,K5s:0.45,K4s:0 .15,QJ,QT,Q9s,Q8s,Q7s:0.5,JTs,JTo:0.5,J9s,J8s,J7s: 0.5,T9s,T8s,T7s,98s,97s,87s,86s:0.65,76s,75s:0.65, 65s,64s:0.2,54s:0.65
#Range1#AA,KK,QQ,JJ,TT,99,88,77,66,55,44,33,22,AK, AQ,AJ,AT,A9,A8,A7,A6,A5,A4,A3,A2,KQ,KJ,KT,K9,K8s,K 8o:0.4,K7s,K6s,K5s,K4s,K3s,K2s,QJ,QT,Q9,Q8s,Q7s,Q6 s,Q5s,Q4s,Q3s,JT,J9,J8s,J7s,J6s,T9,T8s,T7s,T6s,98, 97s,96s:0.4,87s,86s,85s:0.4,76s,75s,65s,64s,54s,53 s:0.4,43s:0.4
#Board#8d 9h Ad
#Pot#64
#EffectiveStacks#1300
#AllinThreshold#67
#AddAllinOnlyIfLessThanThisTimesThePot#500
#CapEnabled#True
#Cap#3
#CapMode#NoLimit
#FlopConfig.BetSize#30,60,90
#FlopConfig.RaiseSize#50
#TurnConfig.BetSize#40,80,120
#TurnConfig.RaiseSize#50,4.6x
#TurnConfig.DonkBetSize#33,90
#RiverConfig.BetSize#50,100,200
#RiverConfig.AddAllin#True
#RiverConfig.DonkBetSize#33,1000
#FlopConfigIP.BetSize#30,66,90
#FlopConfigIP.RaiseSize#50
#TurnConfigIP.BetSize#40,61,80
#TurnConfigIP.RaiseSize#50
#TurnConfigIP.AddAllin#True
#RiverConfigIP.BetSize#50,100,200
#RiverConfigIP.AddAllin#True]

In the future please use [ code ] [ /code ] tags (without spaces inside) so it doesn't take too much space.

To be honest I have no idea.
The why question is a difficult one to answer, sometimes it's hard to find intuitions about it. It's a job for poker coaches and content makers. I am sorry for not being able to offer more on it.

I can say a few things about the tree and your solution though:
-I've run it to very good accuracy with my dev version (0.05% of the pot) and I confirm A8 is a mix of bet/checks with checks being more frequent and some A8o combos being 100% check.

-That being said even combos that check 100% are very close in EV between betting and checking

-in equilibrium the solver uses only one of your bet sizes on the flop (58 one) it takes a while to zero the other two though

-I feel using 3 donk sizes is a a bit of waste of RAM. I would also use bigger raise sizes (maybe 3x instead of 50% and then add all-in in all spots as well if it's less than say 3x pot). That is especially true on the turn when donk is rarely used at all.

-you are missing a river raise size, "add allin" won't always add it as there is a limit (5x pot by default) from a quick run though it won't change things by much though.

Btw, it's also surprising to me that A8 is not a bet with huge frequency. It's a very interesting hand!

Is there a way to do a solve where you want to have only one flop betsize, but want pio to choose the optimal size among several? (I know you can run multiple solves for this but I want to save time) I heard a friend mention this feature in simple postflop.

Hey, if i upgrade the basic version to the pro version, do I get 2 pro licenses or just one?

Thank you for your reply! Much appreciated, and I will include spaces next time sorry about that. Yes it is quite interesting. Id think you'd want to get value from other Ax, as well as drawing hands and 8x/9x. Maybe due to GTO theory you are so far ahead of a wide BB range that you can essentially let Villain catch hands and protect your range. Also the hand is hard to play later streets given BB range and the flush draw being out there and how deep.

There are a lot of scare cards as is what happened in game. I had bet, villain called. Turn was 7c, I bet again villain raised 4.5x on turn. I called. then villain donk shoved river 4s lol. I opted to fold, but could maybe be a bad play against worse 2 pairs or Ax overplayed, or busted flush draws. Still am not sure lol.

No way to do that in Pio.
We may implement in in the future but I am reluctant to do that because the idea of intentionally crippling yourself by choosing one bet size sounds just wrong to me so it's hard to find motivation to work on it.

You get a normal pro license which you can use on 2 computers. Your basic one stops working after a while.
Looking at a quick recalc of this hand with some simplifications I think the solver does not want to bet this turn and then after being raised it folds the the river quite often with A8 (maybe even 100% if we solve to 0). It thinks that AJ for example is a better bluff catcher (blocks a straight) and you still have some stronger hands there (AA/99/88/JT/T6 even sometimes).

Quite interesting, I did not pick up on that with AJ being a better calling hand but makes total sense! Thank you a ton for your input, really helpful!

Hello! I have a question relating to the use of blockers. My brain feels like it might explode trying to come to an explanation for this… so figured a decent question to ask! I know what I am about to ask is quite marginal too, but I’ve also heard understanding situations like this is advanced GTO strategy and really what separates good players from the experts. I will share below the hand for anyone who would like to recreate, and 2 screenshots of examples of the question I have relating to the calling range of 2 different hand/combos. Again what I am about to ask is marginal here in this example, but will help me understand on a wider scope of hands to be applied to when these situations occur, and what to look for/in future sims as I was unable to locate the reasoning on my own.

This is a BTN vs BB scenario, about 30 bb effective.

My question is why on the river here is T8c calling 62%, where T8h and T8d are calling 52-54%. I understand “because of blockers”, but can someone elaborate on that for me? I would think I understand T8s calling more because blocks hands that called flop with backdoor flush draw, but then why is T8d less? That is where I get confused. Something to do with unblocking bluffs?

Basically similar to the other combos, why is Q8h and Q8c calling more, but diamonds less?

I know these are very marginal but I am not able to wrap my head quite around why, or what even to look for. Thank you a ton and hope someone can help me out/or we are both able to come to a conclusion and learn something.


https://ibb.co/VM0cWTt

https://ibb.co/bN4zRbL

[#TreeBuilding#V2
#Range0#AA:0.15,KK:0.1,QQ:0.05,JJ:0.25,TT:0.3,99:0 .4,88:0.5,77:0.4,66:0.4,55:0.4,44:0.75,33:0.75,22, AK:0.25,AQs:0.35,AQo,AJ,AT,A9,A8,A7,A6,A5,A4,A3,A2 ,KQs:0.45,KQo,KJ,KT,K9,K8,K7,K6s,K5s,K4s,K3s,QJ,QT ,Q9,Q8,Q7s,Q6s,Q5s,Q4s,JT,J9,J8,J7s,J6s,J5s,J4s,T9 ,T8,T7s,T6s,T5s,98,97s,96s,95s,87,86s,85s,76,75s,7 4s,65,64s,54,53s,43s
#Range1#AA,KK,QQ,JJ,TT,99,88,77,66,55,44,33,22,AK, AQ,AJ,AT,A9,A8,A7,A6,A5,A4,A3,A2,KQ,KJ,KT,K9,K8s,K 8o:0.4,K7s,K6s,K5s,K4s,K3s,K2s,QJ,QT,Q9,Q8s,Q7s,Q6 s,Q5s,Q4s,Q3s,JT,J9,J8s,J7s,J6s,T9,T8s,T7s,T6s,98, 97s,96s:0.4,87s,86s,85s:0.4,76s,75s,65s,64s,54s,53 s:0.4,43s:0.4
#Board#8s 5d 6c
#Pot#64
#EffectiveStacks#295
#AllinThreshold#67
#AddAllinOnlyIfLessThanThisTimesThePot#500
#CapEnabled#True
#Cap#3
#CapMode#NoLimit
#FlopConfig.BetSize#30,60,90
#FlopConfig.RaiseSize#50
#TurnConfig.BetSize#40,80,120
#TurnConfig.RaiseSize#50
#TurnConfig.DonkBetSize#33,90
#RiverConfig.BetSize#50,100,200
#RiverConfig.RaiseSize#50
#RiverConfig.AddAllin#True
#RiverConfig.DonkBetSize#33,1000
#FlopConfigIP.BetSize#30,66,90
#FlopConfigIP.RaiseSize#50
#TurnConfigIP.BetSize#43,61,80
#TurnConfigIP.RaiseSize#50
#TurnConfigIP.AddAllin#True
#RiverConfigIP.BetSize#50,100,200
#RiverConfigIP.RaiseSize#50
#RiverConfigIP.AddAllin#True]

PS I hope I posted Pio Solver hand above properly this time in brackets and without spaces, please let me know if I still did improperly.

It should be in [ code ] [ /code ] tags. Not in brackets. Like this:
I will answer your question tomorrow. I am quite tired right now. In general the why question though is something you will have to figure out yourself (or with coaches/content makers help). We are just programmers. We may spot something but we don't have either resources nor high enough poker level to answer those.

Hi punter,
You mentioned that users who want to upgrade to pio2 would get a discount and the discount will be based around the purchase date of the previous version. Are you also going to take into consideration the type(basic, pro, edge) of pio previously bought?

We are happy to announce PioSOLVER 2.0!

This is our first major version upgrade and it's a big one. There are many new features and improvements:

-Big performance improvements, the solver is 2x faster on average and even up to 5x faster on some trees using new algorithm (especially simple trees)
-GTO trainer
-Job queue (much more convenient way to schedule tasks)
-Redesigned and much easier to use node-locking interface
-New tree-browsing features like Strategy+React, Hotness for one combo, Strategy Distribution Map
-A lot of improvements for configuring the behavior of the solver and displayed information

You can find a video with more details on our blog:
https://www.piosolver.com/blogs/news

The plan for coming weeks/months is to:

-fix inevitable bugs
-add features/functionality suggested by our 2.0 users
-work on improvements of the preflop solver (it's much faster now but still not as good as I would like)

Feel free to share feedback/suggestions.

Hi punter,
Qeustion for 2.0
1,What the difference between Pro 2.0 and edge 2.0?
2,Should I upgrade my pro to edge 2.0 with this link?https://www.piosolver.com/products/u...olver-edge-old
3,The new user to buy form this link:https://www.piosolver.com/collection.../piosolver-pro
Is the pro1.0 or 2.0?

The same as in 1.0 version, the edge license gives you:
1)The preflop solver
2)Support for big CPUs
3)Ability to move it from computer to computer without limits (useful for cloud instances and other cases when you need to move the license on regular basis)

Depends how old is your license you will get the correct link by following the link in the announcement (to licenses website).

It's all 2.0 from now on.

Is there an option to lock all hands for one side of the tree for all turns and rivers? I would like to see eV changes with strategy adjustments without the solver finding new equilibrium.

Would also be nice to use the new node lock interface in v 2.0 to lock strategy by hands types across multiple turns and/or rivers.

I need to reinstall PioSolver 1.0 Pro, but I'm not able t. The only options showing are Free and 2.0.

Anywhere we can download/install the old version of Pio?

There is no reason to do that. You can node-lock and then just click "calculate results" in treebuilding tab. This is very fast and shows EVs calculated with strategy changes you have made.

That's now possible. The reason we haven't implemented is that I don't see a way to make it usefu. Something like "top pair" is very strong hand on some rivers and very weak one on others. Can you describe a practical use case you would like to solve?

Options->Show old products in the top menu.

To calculate how overfolds or under folds on future streets affect current actions. A specific example would be locking folds for anything less than top pair on river to overbet.

I own Piosolver pro 1.0 and want to upgrade to Piosolver Edge 2.0, the link will only get me to Piosolver pro 2.0, do I need to purchase Piosolver pro 2.0 then piosolver edge after?

I have just watched the video for V2, it looks very good with some great new features, and a major upgrade on V1. I'm excited to get it shortly. I appreciate you are still ironing out bugs etc, but I have some initial feedback for you on some minor improvements that could be made, and also a few questions.

In the Generate Script menu on V1 there is an option to load flops from a file, please could you add this to the new Job Queue section?

When a script is running, the visual in the Job Queue section does not show how many boards have completed, e.g. 20 of 50 flops completed, as would happen in the output you get whilst running a script in V1.

The new GTO trainer looks very useful, some ideas to improve it slightly:

Prevent the trainer from loading hands in which Villain acts first and simply open folds if possible.

After selecting an action, the trainer could show you how often you would take this action if this is a mixed frequency spot with other actions, e.g. bet 60% of the time, check 40% of the time with this hand combination. Perhaps also show the EV of all possible actions. An additional graphic of the square you would see in the solution in the PIO browser for this hand combination at this decision node coloured with 60% bet, 40% check would be good. Additionally, showing the strategy with other suit combination of this hand would be beneficial (both suited an offsuit).

Including a random number generator in the GTO trainer would be of much use, as this is commonly used to make decisions in mixed frequency spots. If so, please include the random numbers in the output table generated after using the trainer next to the relevant decision.

Can PIO files generated on V1 be loaded into V2? Will they work with the GTO trainer too?

How much more RAM is required, roughly, on average for the faster algorithm compared to the previous one?

Looking forward to using V2,
Mat

Hi ! I'm currently running a big script with the new Pio 2.0 Pro and yesterday, it took around 3 mins per flop and now 24 hours later , it's around 8-10minutes per flop. Also , I've notice some rollback in exploitability. I'm currently solving to .350% pot and I've notice the rollback starting around 0.6% pot. By rollback I mean , it is solve to 0.6% and 30 seconds later it's 0.7%. I'm using the new algorithm. Should I give the solver some kind of a break ?

If I need to give the solver a break , will I be able to jump right back into the script where I left it ?

Do you still have a Skype ?

Thanks !

perfect, thanks!

Those will be added in coming days.

This was an oversight. Will be added back in coming days.

It does show it. Sadly I missed the slider when doing the video. Normally it looks like this:


We will take that into account although it's pretty fast anyway and maybe it's useful to get some feel for how the spot is played (if you just ignore those hands you may get the impression that villain "always has it". Still, surely a useful option to have.

Both are good suggestions. We are collecting feedback on it to see how to solve frequency issue. There are some ideas I especially like (like random generator).

Yes, saves are backward compatible but not forward compatible (old saves work in the new version, new saves don't work in the old one) as it always was with Pio.

2.5x
Please notice that the old algorithm in 2.0 is optimized as well making it about 2x faster than it was in the 1.10 version.

A few comments:
-there is a bug causing the trees to be awfully slow to build if you use rake or ICM, the fix will be available tomorrow
-rollback in exploitability is expected and nothing to worry about, new alg jumps around quite a bit
-we don't use Skype, we use Discord and it's pretty popular with thousands of members! Sadly we can't link to it from our registration email because that causes email servers to freak out and block our emails or mark them as spam; you can find the discord link on our blog, I will repost it as well on the starting page

Do you stop scripts in pioviewer 2.0 with Script execution queue>Advanced>abort calculation and save currently solved tree? Then next time you open jobs you can start it again?