Yes. First thing is that you used "3,5x" as a raise size. We don't accept a comma as decimal separator, it just separates the terms (the same way you did use that in bet sizes). This mean that it's inerpreted as a raise size of 3% and a raise size of 5x. 3% is rounded to minimal allowed raise (so minraise) and 5x is a raise to 20.

You likely meant either 3x, 5x or 3.5x.

Another thing is that I strongly recommend using bigger blinds (just multiply starting pot and stack sizes by 10). Pio doesn't use fractions in bet sizes so if you want more precise rounding that will be beneficial. As an example: if a starting pot is 7 and the bet is 50% it will be rounded to 4 while with starting pot of 70 the bet will be 35.

Rerunning the simulation with 3.5x raise we still get 100% cbet (this time only bigger one). This is expected because IP has huge range advantage, equities on the flop are 69%-31% in favor of IP.
I've tried two different flops:

Jh 6d 4h - IP has 61%-39% equity advantage and is c-betting 91%
7d 5d 3h - IP has 56%-44% equity advantage and is c-betting 82%

The thing is that in your config IP's range is very tight so it's natural IP bets a lot.

Hello there. Tried to google it but theres no result for this. I keep getting this error "set_range incorrect or missing argument" when I try to run this script and I don't know whats wrong.

Here is the script: https://pastebin.com/QJNSeAiD

Can you shed some light on this please?

-i think it was 3.5x
-10xbb => got it
- yea i know its natural to have high bet freq because of range advantage, the weird thing to me is that pio like big sizing 2/3 instead of 1/3 when OOP range has a lot of snap folds... can you explain me why is this the case???

tx for help

Probably because for bet sizing of 1/3pot that range bets 100% of time on almost every possible flop.Final result is always the same ofc.

need a new laptop for pio solver / poker tracker 4 / poker // browsing web. Can anyone tell me difference in price + if neccessary good idea to get a gaming laptop.

$600 laptop
https://www.newegg.com/Product/Produ...77&ignorebbr=1

Acer Laptop Aspire E5-774G-52W1 Intel Core i5 7th Gen 7200U (2.50 GHz) 8 GB DDR4 Memory 256 GB SSD NVIDIA GeForce 940MX 17.3" Windows 10 Home 64-Bit

$1200 laptop
https://www.newegg.com/Product/Produ...04&ignorebbr=1

HP Laptop ProBook 470 G5 (2UA28UT#ABA) Intel Core i7 8th Gen 8550U (1.80 GHz) 16 GB Memory 256 GB SSD NVIDIA GeForce 930MX 17.3" Windows 10 Pro 64-Bit


----

anyone know of better online deals for laptops or have other suggestions? punter? thanks.

The second laptop has 2x RAM (16Gb instead of 8GB), this makes huge difference, I wouldn't buy a compute with 8GB of RAM today even if I wasn't running a solver (that is if you hope to use it for a few years).
2nd computer has much faster CPU as well (quad core instead of dual core).

Hello, I am new to Piosolver and really enjoying it. However, I was interested in a feature about rounding strategies. I would really like to round flop,turn and river into 20% intervals.
I noticed someone had asked this before and was advised to watch this video:
https://www.youtube.com/watch?v=5Ozjfnatwms

However it seems that the interface has changed since the video is recommending options in the menu that are no longer there?

If anyone could help me with this, it would be much appreciated.

Hi again,
I know you recommend rule GHz x cores for estimating and compare speed of solving for two cpus. What about RAM if it is still lineary? In my eyes should be as 3200 MHz RAM will be getting data 2x faster than 1600 MHz. Am I right?
What about latency?

Is there a way to input oop player checks 100% of flops on our chosen board? Obv I can click check for oop but then it affects oop's range if PIO thinks oop should be donking flop a significant percentage of the time.

As a workaround I have tried to just set the oop range as the IP and set the IP range as the oop, that way the first action is just a cbet but I fear that will disturb the tree more than it needs to be disturbed.

thanks

Can anyone recommend what percentage of the pot they typically solve down to?

Tried searching, "percentage" and "of the pot" but no luck.

Thanks!

0.5% is mega safe 1% is good enough.

Rounding is in Tree->Round Strategies now.

It's hard to say how much RAM speed matters because we rarely have a chance to test on very similar systems. Overall it doesn't matter that much, all you can hope for is maybe 5%-10% improvement at best.

Just leave OOP's bet size empty when building the tree. OOP will be forced to check then.

Opinions vary. 1% is maybe good enough as crude approximation, with 0.5% it's already very close and won't change much although it still contains significant errors on some rivers. We usually recommend 0.25%-0.3% as a very good solution and 0.5% as good enough for most purposes. A lot of very smart people think that it's overkill already though.

I have tried to download it like 20 times but it keeps interrupting when it's almost done

After working on a handful of dedicated systems (128GB setups) I'd say ram speed influences less than 5% (at least for DDR4).

CPU speed/cores seems to the most significant factor to solving time.

Maybe it's something on your side then (like not enough space dedicated to downloads or temporary files). Let me know which exact file you have problems with, so far the ones I tried download without problems.

It could be a browser issue. I was having the same problem downloading with Chrome. I downloaded it with the Edge browser in Windows and it worked.

Hi guys. Im going to buy PioSolver but Im not sure wich one.I think the Edge's version is the best choice but I need a new pc and here its the problem because I have no idea how much I have to spend for it. For example I guess I need at least processor i7, 16Gb RAM and 1TB of memory right? If anyone can give me an advice about the pc that would be perfect guys. Thank you!

Hi punter. I've decided to give you a break from questions about hardware configs . Though maybe my question/feature request should be adressed to Kuba cos he is developing aggr. reports afaik.

The thing is I've been exploring population tendencies in different spots for the past few days. And unfortunately I've faced one significant problem: Some hands reach the showdown more often than others. So received ranges appear to be incorrect almost in every spot even with very large sample.
E.g. if EP raisers showed 22 1k times and AA 5k times, it doesn't necessarily mean 22 is being opened only with 20% frequency. 22 just reach the showdown way less often than AA. If oop players showed AQo 50% less frequently than AKo when faced a preflop 3bet, it doesn't mean AQo was folded to a 3bet 50% of the time. It was just called a lot and then check-folded a lot. And it is not possible to calculate this "a lot" manually.

So I was wondering if this problem could be solved with the help of Pio. It would be great to have a column in aggr. reports showing how often a hand went to showdown (in any way). Then we could solve a required spot on subset of flops and figure this stuff out.

Punter did I ask this question at the wrong place? Where can I get an answer to this? Could you guide me towards a direction at least? I dont want to bug you with it if its not your domain. Let me know please.

!

RAM is "memory". Pio needs RAM to build trees. 16GB is enough for postflop trees but for preflop ones you need 64GB and a fast modern CPU.

The faster CPU you get the faster the trees are solved. The more RAM you get the bigger trees you can build and solve. More here:

https://www.piosolver.com/pages/faq#hardware

This is a nice idea but it's not available right now.

I am sorry I forgot about. Did you generate this script yourself? Maybe something went wrong with copy-pasting? There is a random end of line in set_range IP line which causes there error.

I generated the script using the generate script button in the postflop tree building tab. I generated it and then I pressed copy to clipboard. This problem happened to me often so I thought Id ask.

Any chance you are clicking on generated script by accident? (thus possibly introducing a line break). Can you send me on of the tree configs (PM or email) which resulted in this? I will try to recreate.

Oh yeah you right. Sorry for the mistake. I'm not interested for preflop right now because I have pokersnowie for that and its enough (I know someone that said it and he is top one). Im going to buy the pro's version for this moment because I saw the computer already (desktop) and it will be with 16Gb RAM (optional for 32) and i7 (quad core). Maybe in the future I'll get a bigger one. I'm playing nl200 currently so I think its enough for now :P I'm reading the Piosolver's website for the details. Thank you buddy for your advices!

Im trying to check for EV for 1 bet size on river compared to 2 sizes. I get the higher EV for just 1 size, where my friends get the highest EV for 2 sizes.
I cant find what Im doing wrong.

For example. Ip bets flop, checks back turn. OOP can bet 70 %, or 70 % and 150 % on river.
I check for EV directly after clicking on the river card.

1 size
OOP 60.153
IP 39.847

2 sizes
OOP 59.451
IP 40.549

I also wonder for when Im checking for when IP can bet 2 sizes. Do I check directly after the river card also, or after OOP checks?

The "set_range incorrect or missing argument" problem is caused by the "word wrap" option enabled in notepad under the "Format" tab.

It breaks the line in order to make the lines visible. Or at least this is my strong theory at the moment. I have yet to try it but if I never come back on this thread it was because of this.

Thanks punter for the help in any case. You are awesome as hell!