When I load my script it says
4 out of 184 done
0.205214
skip_if_done ok!
5 out of 184 done
0.206655
What do the numbers between "done" and "skip_if_done" mean?

I always recommended 2-3bb/100 for preflop spots.

I don't know, can I see the script?
(preferably paste it with [ code ] [ /code ] tags (without spaces inside) or pm it to me)

I suppose now that its just seconds spent on loading/solving flops.
I found my question pretty dumb but after your response I feel more comfortable

Right, there is solver_time command in the automatically generated scripts. That's more useful when the trees are solved and not skipped though

ty,yes, i was asking for flop subsets and how accurate they are. I will take a look at 2+2 thread.

You recommend we use these new subsets for preflop solver or both solvers?

how do i download these new files you have linked? never used pastebin.com before but it appears to me it isnt a fileshare website.

This is something not that easy to measure, there is a bit about the way we measured it in the blog post I linked.

Those aren't new. They are just selection of subsets we already made public (and ship with the edge version). You can download them all, here:
https://www.piosolver.com/blogs/news/piosolver-1-9-2 (point 3)

Pastebin just lets you paste text. It's useful when sharing long pieces of text. The link is just list of subsets we think are the best out of the ones we published.

okay, i have software version 1.9.2.4.
I guess I dont need to replace them.

This is not welcome here. Just in case anyone tries to buy Pio licenses second hand: you will be scammed it happened before and it will happen again. From our FAQ:

I havent done a preflop sim for a while and i cant remember some things

If i want to solve to 2bb / 100 accuracy do i have to write before i build the tree or i can do it before i press go?

If i want to see the bb response in a button open do i have to nodelock the range i want? what will happen if i put he fixed range i want button to open in the ip tab here? whats the difference



On this tree i would like to have 500 stacks with bb = 20 (25 big blind tree, with sb 10 and total antes 12) Do you see any mistake?

Thank you

Anytime before you click "go".

It's better to predefine IP range (BTN's) and force raise in the tree.

It looks correct. Now start the tree with a forced raise for IP to 40 or 50 (2bb or 2.5bb)

Dear Pio Team,
Is there a way to view Piopreflop Solutions with Pio Basic?
Thank you in advance

Yes.
First time you load the preflop save though you need to do the following:

1)Solver->Connect in the top menu and choose preflop-browser.exe
2)File->load tree and choose your preflop save

After that it will switch automatically between preflop-browser and normal solver as needed.

Having some stability issues with pio (preflop avx)
I've been pushing my overclock from 3.95ghz to 4.0ghz, and although the system is stable through 24 hour p95 torture test/aida64/etc it ends up crashing in piosolver after an hour or so of solving a large tree.

I've noticed that while piosolver is working, the cpu vcore is quite bouncy, jumping from ~1.38v to ~1.28v. I believe this might be the source of the instability. If vcore drops too low then it can cause momentary freeze/crash at that low point of voltage. Any thoughts on how to address this? I'm pretty much at the temp/voltage ceiling of my chip so I can't add more vcore for pio.

I am really not a hardware expert so I can't help here. It sounds more like memory issue (random bit flips may cause solver to crash although it shouldn't happen too often).

Is there anything from the software side/code that pio can be updated with to raise the minimum voltage used by the CPU during load? Or is voltage strictly controlled on the hardware side?

Just to give you an idea of my cpu voltage spikiness during pio:

There isn't. We have no way to control any of that.

Again, I don't know anything about hardware but it's natural that sometimes CPU has more to do at given point and then there is a little less (when waiting for memory reads or for threads to sum-up things in locked array). It looks like that something isn't stable on your system at this point but there isn't anything we can do about it from software side.

Ok, just wanted to make sure.

Haha you should compile a full stress test/benchmark version of pio that uses ~95% of ram, complete with scores. It is more difficult to stabilize than prime95 and I think alot of overclockers would use it just for fun/stability test/competion.

All right another bit of a tech question, hopefully we can figure this one out.

So I have a samsung 960pro ssd, which according to benchmarks I've run is capable of 3200mb/s read. Currently, I have a 120gb piosolver tree saved on the SSD.

Utilizing piosolver to load the tree from the SSD into my ram, I would have expected to see read speeds around 3200mb/s, especially since this is a sequential read. Unfortunately, windows shows the 960pro topping out around 1000mb/s (1gbp/s)

Is there something in piosolver that you are aware of that might be slowing down the read operation?

Yes, it needs to allocate memory which is very inefficient in Pio (because we use the default Windows allocator and not a custom one). Other than that there are some computations when loading the tree but not that much. Tree loading was never a priority to optimize: it's almost instant for small saves. Full trees are a rare usecase, most likely you want to perform more calculations on them and in that case the loading time will be a small % of that anyway.

Hi,
I'm a 100% noob when it comes to computer stuff but I do know I need to study my hand histories and that PioSolver might be the answer.
But where do I start? Will it be easy once I get the hang of it. how much is it and is there a trial version?

It's the way most of our users use the solver

You start with quick start video and a free version:
https://www.youtube.com/watch?v=JqGQoQKbCB8

There are some more videos/FAQs, here:
https://www.piosolver.com/pages/faq-videos

The prices are listed on the website: https://www.piosolver.com/

We don't and won't offer trials. I personally don't believe this is a good idea. We offer a free version which solves any turn/river as well as one flop (Qs Jh 2h). You can use it for as long as you want and evaluate the software.

Hey, been trying to run aggregation reports but the formatting is not working as intended. I changed the thousands seperator to a comma:"," and decimal seperator to a dot: "."

Any other advice? (even installed english language pack for excel)

https://gyazo.com/80daf74d40e26327a0f7874f4ef18972

Hey, what does solver connecting do or connecting to recently used solver?

Hey punter I had a question hoping you/anyone here could answer on laptop/computer specifications to run piosolver optimally i've read through most of the written guides but don't quite understand AMD quad core vs something like dual core i7/i5.

So the more # of cores for cpu is better for solving pio in a timely fashion/quicker, so anything quad core > duel core?

Looking at an AMD quad core, a12-Fx and another setup that is i7/i5 duel core. I hear that i7/i5 duel core is going to be a better/faster cpu then amd quad core, but thought more cores would be better in this circumstance for calcing stuff in something like piosolver.

So in general, AMD quad core A12-Fx > i7/i5 duel core in that it will be faster calcs?

Looking to do a budget piosolver build lol.

You can try removing "thousands separator" as commas are used to separate values not digits in .csv format.
There should be something like field separator as well which should be set to a comma.

I recommend downloading Libre Office or using Google Docs. Those programs handle standard .csv files without problems.
Why Excel can't open .csv (which stands for comma separated values) in their default format and keeps assuming there should be a tab or some other separator between values even though it's in the very name of the format that commas are used to separate values I honestly do not comprehend. If you have some time you could try emailing their support and making a rant about it.

Our architecture is: a separate solver (engine which does calculations) and graphical interface (PioViewer). You first need to connect the engine to the GUI. As of right now we have separate preflop solver/viewer and that's why it needs to connect to it when opening preflop saves. This will be dealt with in the next release.

As a general rule: anything AMD which is older than Ryzen is very slow and it's not recommended to run the solver on those CPUs.

It's number of physical cores * frequency they run on (base one, not turbo) but only if the CPUs are from similar generation/similarly good. As a general rule that means all Intel CPUs from around Sandy Bridge/Ivy Bridge generation and modern AMD CPUs (Ryzen and Thread Ripper). While older Intel cores are a bit slower than the newest one the difference isn't that big. With old AMD cores it's very big though.