I understand what you are saying but i wasn't trying to make it out to be a gimmick or something. But just a quick way to filter out any hands that are less than 0.1% edge or something like that, rather than having to locate the bottom of ranges every time. It would also be used to looking at a range that includes slightly -ev shoves as well and again rather than having to locate the hands manually. But not to worry!

Looking forward to the update!

Thing with edge I would also find useful, and I have idea to satisfy all of us
You already have I think 3 different colors of red and green based on how much +EV or -ev is the hand (can you tell as how are they given?). If we could manually set when these colors changes it would virtually mean we set an edge but there would be no words of edge, just call it
+0.5 EV hands = light green
+0.2 EV hands = green
...
And changing colors would also be nice, for example I want to have safe hands (lets say +0.20 EV) be colored green, and marginal +EV colored blue.

Also I would find useful to show us what "edge" did PSM gave to the hand comparing to normal ICM with same ranges like in PSM. I think this option would be indeed popular...

Hmm... okay... I think I see your guys points... its could be useful to be able to at-a-glace see the bottom of the range, etc....

Right now, there is a color gradient on the heat-map, but its based on absolute EV deltas... so, depending on the exact spot, you might not actually see a lot of variation in the colors.

Being able to compare ICM/PSM at-a-glance does sound like a pretty neat idea too...

I shall meditate on these things.

Yeah i think the heat color grid or w/e is distinct enough where you can mentally know that you need a slightly darker green in spots where opponent skill isnt being considered, but i can also see how being able to input a min edge could help.

I assume this has been brought up and/or considered by you, put being able to assign villain types to each opponent, in terms of aggressiveness in shoving and calling, woudl be pretty helpful. Say have everyone's base be 0 and then be able to go up to +5 and -5 for shoving and calling based on how far from eq theyll play future hands.

The sims can already take a while to run and i imagine this will make them even more complicated so might not be feasable.

SNG solver not working for me today, studied hands with it before I started today and was about to start up again and was gonna look at hands and when i open it snap get the 'sng solver has stopped working' error

Tried reDLing, run as admin, etc and cant get it to work

Does this happen right when you startup (before you see the main window)? Is the error message on one of those windows error dialogs where you're able to see a "details" about the crash? If so, can you get a screenshot of the details and either post here or send to support@sngsolver.com

Have you tried rebooting your system?

Found a weird thing, look at equil. overcalling ranges. This is with no PSM, with PSM overcall ranges looks more realistic but not completely.

Opponent modelling is something I've been doing a lot of experimenting with... and I'm happy to say that it doesnt necessarily lead to an increase in sim time...


Hmm.... those ranges do seem a little wonky... I'll take a closer look.

Hi Jason,

I assume this isn't available but it would be silly not to ask - is there a command line version of SNG Solver available? I create lots of push/fold charts for bubble play and changing small variables becomes tedious after a while. I Can probably create an AHK script or something to get the same effect but it'd be much slower.

As an example, say I want to study my Blind vs Blind Shoving ranges, it'd be cool to be able to set a stacksize range for myself and the BB, and a folding frequency range, and output something that contains appropriate shoves for every change in variable value. Perhaps setting ranges is possible in SNG solver already, if so I apologize, I've only just started using this great software.

AntiHer0,
I'm afraid I'm not likely to do any kind of command line version. I dont think I can justify the development of something like that right now given the current state of my to-do list.

I dont know if it helps at all... but the SnGSovler EXE does actually respond to
command line parameters in the form of: BB SB stack1 stack2 [... stack9]

Doing this will cause these values for the blinds/stacks to be entered into the appropriate fields on startup.

This is something I use for debug purposes and I dont think I've ever even mentioned this functionality before.... but there you go.


When you say "ranges" here, do you mean setting "range of stack sizes", or "range of opponent push/calling hands"? If the former, then no, there's no way to do that. But if you mean the latter, then yes, you just need to mouse-over and click on the opponents range that you'd like to edit. It will bring up a ranger-picker dialog box and you can edit away.

I meant range of stack sizes but ty for the answer!

One more thing - is there a way with this software to calculate non-allin situations eg hero facing a non-committing open and I can input his fold %?

Not at the moment... adding more kinds of villain/hero actions is one of the things I've been working on, but I dont have any kind of ETA

Hi Jason -- was having a conversation about simulation depth today and would like to be sure I understand what is going on. Could you please tell me whether a single future round consists of a single hand or an entire orbit?

Using 6-handed as an example, I understand at depth 0 it evaluates only the current hand. I was assuming at depth 1 it evaluates a total of 2 hands (the current hand and 1 future hand). Is that correct or does it consider an entire orbit, meaning the next 5 hands (6 hands total)?

Saw this answered by Jason in another thread -- 1 future round is 1 future hand.

Oh FTP parsing please. Such a pain not having it

Hi! I have the sng solver license installed in two computers, but I'd like to chance one of them to another computer, how can I handle that?

Coming real soon... I'm just trying to wrap up a few more things for the update.

Hi Jorge,

No problem. Just send an email to support@sngsolver.com and let me know which of your current installations you'd like to move.

im on trial period.

in details view, in the limited samples i have looked at; what is the last "push" for BB?

ex:
push (btn)
call(SB) ----------> when sb calls B push
overcall (BB) ---> when BB calls sb and B pushes
call(BB)----------> when BB calls a Button push but a SB fold
push (SB)---------> when hero folds and Sb pushes
call(BB) ---------> when hero folds and the BB calls the SB shove
push (BB)---------> what does this BB push mean? something when hero folds...........

Ah yes, that last "push" entry is mislabeled... clearly, its not possible for the BB to open-push. That last row should be called "all fold to BB", or something like that without a displayed range%. (The range% as currently displayed is meaningless).

i was just running a hand

65/35 structure 3 people on the bubble. 80/160 blinds

hero Button :730
sb :10
bb :1060

I was adjusting the calling range of the bb from .05% to 100% and the heros button push range did not change???? also, i could adjust the sb calling range when i should not be able to, he has 10 chips????

Because the SB's range is 100%, the BB's call range will never come into play and therefore have no effect on the Hero's recommended range. If you adjust the BB's overcall range, you'll see the effect on the Hero's range you're expecting.

And you're right, it doesnt make any sense to be able adjust the SB's range in this spot... the software doesnt prevent the user from setting "nonsense" ranges.

Point taken. My todo list now includes improving the UI to better handle spots like this.

I've landed myself with a Windows 8 PC (which is new and considerably more powerful than my Windows 7 laptop), yet the run times on SNG Solver are way longer. Might be one of those "it depends/unanswerable" questions, but is there something different about Windows 8 in general that makes it so much slower for Solver?

I dont think there should be anything about win8 that makes it slower... I've done a fair amount of testing with the win8 developer previews and I never noticed any difference in performance.

Can you tell me what the specs of your two systems are?

I'm not especially computer savvy, but I think these are the pertinent ones:

Laptop: Windows 7 Home Premium 64, Intel Core i5-2410M, 6GB RAM.
PC: Windows 8 64-bit, Intel Core i7-3770, 16GB RAM.

With SNG Solver having identical settings and running the same scenario, the PC has been a minimum of 4 times slower, which I found surprising. The PC is out the box and brand new, so maybe there are some things I can do to make it faster, but I was expecting it to be much quicker given more cores and memory.

As an example a 4 handed scenario took around 30 seconds to complete on the laptop and the same one 2 minutes on the PC (with CPU usage at 100%) using 3 complete runs for both.