Hey, I'm having troubles scripting lately, not sure what I'm doing wrong, I did some scripts in the past days w/o problems but yesterday 8/10 failed (stuck like this image).

I know that path is correct, what could I did wrong? As I said it's the first time I'm encountering this problem.

https://i.gyazo.com/a748740b5a1a6475...c041533fa9.png

Thank you for reply .

---

That makes sense thank you again, well there are no such laptops atm which contain amd ryzen or threadripper cpu's so if i were to go that route with AMD i'd have to buy a desktop.

So like you said in bottom paragraph old AMD cores that aren't ryzen/threadripper, like A12-Fx even though they are quad core won't be much faster then a i7 duel core when it comes to faster calcs in piosolver?

I am not sure what you are trying to accomplish. There is only one line in the script "load script" which doesn't make much sense to put into script generation Window.
If you want to run already existing script then go to Tools->load script. If you want to generate a new one maybe try following an example in our FAQ, here:

https://www.youtube.com/watch?v=VhG_YsEvfwU (question number 6, see the video description).

Old AMD quad core CPUs will definitely be significantly slower than Intel dual core ones.

Hey when node locking what does "lock all hands" vs "lock selected combos" do?

Do i just use "lock all hands" mainly when doing basic node locks?

Should I do preflop node-locking in the same way?

Yes, when doing basic node locking just use "lock all hands". The other option lets you lock only selected combos (you need to select them clicking the small rectangle next to the radio button). There is more about combo-locking here:

https://www.youtube.com/watch?v=gJEPGSIpIBM (point 12, see the video description).

Yes, the sooner you lock the better.

Thanks, cheers.

So with pio basic, i can only run single aggregated reports for run outs and single flops?

So if above is true and say i've solved like 10 flop spots, can i now run multiple aggregated reports and receive same compiled data like scripting the multiple subsets?

The problem are going to be weights for the flop which are contained within a script so multiple file report won't work unless you replicate the script structure yourself so the viewer can read the weights for the flop.

Really enjoying using PLOCalc!

It turns out it is very complementary with PPT.

In your work on the PIO PLO solver, is there indication that a fairly representative set of flops for the full game of PLO will be possible, like you have found for holdem?

That's nice to hear. We will have a small update in the near future with some options to import ranges from other tools (at least if they have more or less "raw" format).

It's hard to say. PLO preflop without abstractions, even working on a subset of flops will be tough. PLO postflop without abstractions is very possible (although still will be memory hungry). PLO is around 200x bigger than Holdem.

Getting some errors with pio. Usually computer just crashes about 1 hour into solve, but this time got this error code.


It's been kind of frustrating because everything in my computer works fine but piosolver now...


At least this time I got an error code though. Was wondering if you had any ideas what could be causing this? I've tried increasing vcore by egregious amounts from (over .4v+ extra) and it still crashes.

I don't know almost anything about hardware so I won't help you with the settings. It looks like the hardware issue though (probably memory is not stable) because:

1)it was much more stable at the beginning and started being unstable after a while
2)Pio is crashing on various trees while it's very stable overall, our customers run it weeks at a time and never crashes. It's true that about the only crash reports we've got which weren't explained by running a simple memtest came from new AMD CPUs. Feel free to send me the config of the tree it crashes on so I can run it for 20-24 hours to confirm it's not a software issue.

I recommend running some smaller trees with less threads. If that doesn't crash you can be sure it's a hardware issue and most likely one concerning the memory.
You can also try running non AVX compile and see if that crashes as well.

Why does Pio recommend -EV calls?

EG:
QdTh on the river facing a huge bet,
Raise = -780
Call = -122
Fold = 0

yet Pio recommends calling ~60%.

Hi,

I have a problem when running aggregate reports,

I get this error message
https://gyazo.com/efabf3e332743d5b8ed4ecf524569fe9

i have run dozens of report with success in the recent past
Any idea what could be the issue ?

thx

It's hard to comment without seeing the actual tree. The most likely reason is that it's not solver to good enough precision. The way to share the config is this:

1)Use "copy to clipboard" button
2)paste it to pastebin.com and link here or paste directly here (using [ code ] [ /code ] tags (without spaces) to prevent very long posts)
3)Describe the exact line and hand when it recommend a surprising action

I can then load that config and solve it myself and see what the problem is.

No idea for now but I will get back to you with that.
Are the saves in a separate folder?

How do I run a script and get the result?

I define ranges, pot, stack and bet sizes, hit "generate script" in the pop up I generate 1 flop, and hit generate script in this pop up, then I go to "run this script" and a command box pops up


PioSOLVER-basic 1.9.2 (Mar 26 2017, 14:25:17)
(c) Piotr Lopusiewicz
registered to ....@...

and then nothing happens?

In the folder Im saving to it has the script but not the tree result? Ive purposely done a easy tree with few bet options and If i run normally in PioViewer it solves it in under a minute.

[*img]http://imageshack.com/a/img922/3019/GjiGd4.png[/*img]

thank you punter,

I think at the origin I solved 49 trees and the folder was located at c/users/documents
then I moved the folver to a different location and I launched another script with 74 more flops (with same exact parameters , just new flops)

maybe it messed up something in the report path or so at one point ...

i tried to delete script info text file from the folder or to run aggregate reports on previous folders that I knew worked in the past.

I also tried to re-download piosolver zip file and to run it from there but whatever I try I get the same error each time.

btw , the error message is not unique , each time the name of the unfound .tmp file is different

https://gyazo.com/0530aaf517fa909ffb8793ee74971fd2

Hi,

I've had this error code a few times. Each time it occurs the tree dies and I have to rerun the sim. Any advice is appreciated!

Solver process died with error code -1073741819

Sorry I've just saw that HappyLuckBox posted the same problem. I, too, have the AMD TR, so it's possible that it's to do with AMD CPU's like you mentioned, Punter

will you post in this thread when the update is released?

It's really hard to guess. The only reports with crashes we got are from new AMD CPUs. It seems that for some people it works flawlessly, for some it works well for a while and then it starts crashing. In that case software issue is very unlikely (the software doesn't change so if it worked well and stopped it's almost surely a hardware issue).
That being said, I recommend waiting for 1.10 release which got a lot of things in the engine re-written anticipating future changes. Maybe whatever triggers the crashes is gone by then. Those things are very difficult to deal with because it's impossible to replicate the problem.

Yes, the schedule is 1.10 PioSOLVER release first and then small update for PLOCalc.

You may want to compare the memory you're running as well as the CPU - there are plenty of stories going round about memory compatibility issues with Ryzen/TR (certainly not just a Pio issue if it is that). Not sure what the rules are here about linking to other forums but a quick google for 'ryzen threadripper memory crashes stability' or some such combination should find plenty of examples.

If your memory is run above speeds officially supported by TR (2666 Mhz I think?) you could try dropping it way down (maybe even 2133 or 2400) and see if the issue persists.

Thanks, this is very useful.
From Pio perspective I can say that the kind of error which was reported with those crashes strongly suggests the memory is unstable.

Hello, are there any recommendations about what amount of seconds break after each flop should be taken when scripting?