Getting some "Connection timeout errors" atm when starting a flop calc.

A big thing on my wish list for the program, would be to be able to schedule multiple calculations and let it run when you're not near the pc.

For example:
I define a flop scenario to solve, and choose a file to save the solution to. I add this to a queue.
I do this for 10 other flop scenarios.

And then Simple Postflop can solve all scenarios in the queue and save each solution to a file on disk.

as far as i can tell, connection timeout errors usually signify a tree that is too large. try reducing available bets or making stack-to-pot smaller and see if it lets you run it - if it does then you know that's the issue.

We countinue to work on this issue. Think that we'll release update today or tommorow which will totally fix it.

Yes, this feature falls in our TODO list as one of the generals. Thank you for feedback and suggestions!

I don't have much technical/programming knowledge but it seems every similar program on the market uses CPUs to do this kind of calculations whereas GPGUP should be much faster I guess. What's the reason behind this. Is it hard to implement or some other reason(s)?

Hey!
We haven't tried yet to run our algorythm on GPU but we suppose that it will work slower due to specifics. Though it's not easy to make it run on GPU, we can do that but now we don't see any sense in it.

Hey, can anyone plz link me to the official commercial-trhead or the Hompeage of the developers, so that i can check out prices and so on...

also one question:
Does it take a lot of working-space (GBs) when you save trees?

http://simplepostflop.com/en/

My stored flop calcs are < 1 MB.

thx man!

and regards to stoed calc - yeah, thats good for me that it does not take a lot space....

btw, is this tool essentially the same like GTO_Bulider or piosolver in its approach?
I know that all of them have some little differences, but just to get an idea - is it like pokerrange vs Equilab?
Meaning, that the point behind the mentioned 3 tools are similar?

Hello!
Three of mentioned tools are doing the same stuff and the results of calculations are exactly similiar but we have differences in features.

Simple Postflop has Cloud and Standalone version which allows users to make calculations in the cloud if they have weak computer and on their own computer if it is powerful. In Simple Postflop it is also possible to edit and lock strategy for one player and find exploitative strategy for the other.
I also supposed that Simple Postflop is the leader among mentioned tools from the point of algorithm efficiency but we didn't have benchmarking.

Our target is to make our product simple, fast and functional.

Hello everyone!
Today we have upgraded to a new more power cloud infrastructure which will allow us to offer solutions of much more complex game trees. We believe the new cloud will also provide more stable service and ensure that Simple Postflop is always available when you need it.

thx a lot for answering^^

Getting more "connection timeout errors" than usual after yesterday's update when I start a flop calc on a standalone version. Even for simpler scenario's like 3b pots...

When I open task manager, I see memory getting allocated (3.5GB for the simple scenario) to Simple Postflop and than get an error, with only 50-60% of my memory in use. I have solved way more complicated trees like a BvB SRP with multiple betsizes, which used like 7GB RAM. So dno what the problem is at the moment...

Small update:
Now the problem is gone again.... I had the same thing yesterday: nothing works and than after trying some to resolve some old flop calcs, suddenly it works again. I don't see the pattern, but TeelXp is trying to help me on skype.

What justifies the much higher price point that your standalone product has compared to your direct competitor? Speed, specific features?

Hi,

I don't want to say anything negative about existing software so I will just focus on the advantages of Simple Postflop.

Regarding our prices relative to the prices of competitors I can say that our cheapest version is cheaper than the cheapest version of the competitor and our most expensive version is cheaper than the most expensive version of competitor.

With Simple Postflops SaaS version you can get started with 100 flop calcs for just $70 which is definitely the cheapest option on the market. The solutions you run are computed on our cloud servers so your desktop is free for other tasks and you can submit multiple calculations simultaneously. You can also save any of the cloud solutions to a local file on your PC and Simple Postflop solution files are very small, usually less than 1MB. Like our competitor we let you run turn and river calculations for free.

Our standalone version lets you run large calculations locally on your own PC and we believe it is the most memory efficient and the highest speed in most cases already and we expect to release a major speed upgrade in the coming weeks that will widen that gap significantly.

Finally I believe at the moment Simple Postflop is the only GTO solver that allows you to lock opponent strategies and run minimally exploitative calculations as shown in this video: https://www.youtube.com/watch?v=xfMmaWO42o8

We are highly dedicated to constantly improving our product both in terms of adding features, improving the interface and improving efficiency. We have extensive experience in making high quality poker software from our previous products (eg Simple Nash) and have been working to develop and improve Simple Postflop with some of the most well known GTO coaches.

We also have a lot more ideas that we are excited to implement and Simple Postflop is more than just a GTO calculator, it is a revolutionary product in poker that will grow rapidly forward. We love working on game theory algorithms and are just hoping to make a little bit of money as we pursue our passion.

is it possible to add an feature to "roundup/down" the check/bet frequencies of a single hand in the "strategy show" matrix?
it wouldnt be 100% GTO but it would be an approach which is more easy to learn.
for example:
- T8s hearts is 87.21% bet and 12.79% check u could round it up to 100% bet
- T7s herats is 38,21% bet and 61,79% check u could round it up to 100% check

Something like that can be really useful for learning and applying these GTO strats into one's game - if done right. You can't just round up or down like that, you need to at least make sure that the overall frequency stays the same (e.g. GTO c-bet freq is 36%, then with all the rounding you must not end up with 46% all of a sudden), and make sure that the difference in EV between GTO and your simplified strategy stays acceptable.

sounds good. would be nice to implement something like that to the tool!

Hello!

This is an applicable idea, we'll think how to could be implemented.

Today I have a good news for our customers! We have released an update of Simple Postflop to version 1.02. The program should be updated automatically - if it did not happen for some reason, you can download the new version from our website.

The update includes:

• Ability to view folders with solutions through a special Explorer.
• Ability to create a queue of calculations.
• We have significantly increased performance of Simple Postflop Engine without affecting on the calculation quality. The calculation in new version is performed 2-4 times faster depending on the parameters of configured situation.

As always all of your comments and suggestions are welcome.

Hi, I'm testing your software, it looks very promising, but kind of alpha stage at the moment.
Here are my remarks/doubts/questions:
- I wanted to try the desktop calculations for river/turn scenarios, but when I try I'm asked to enter a license. Does the free license only allow to perform cloud computations or am I doing something wrong?
- Cloud computations give connection timeout errors very often, it's like trying to win a flip...
- I tried to setup a very simple river spot to test everything out: I created an instance of the AKQ game by entering a 22223 board and assigning a single combo of AA/KK/QQ to both players ranges. I set up a pot of 20 bbs and effective stacks of 10 bbs, only allowing allin bets.
The solution seems consistent, but what really puzzles me is that hands which are mixed have a huge difference in ev between the mixed lines. For example after a bet of player 1, player 2 is mixing calls and folds with kings, but the EV of calling is reported as +2.509, whereas folding is 0. Shouldn't calling be 0 EV too or am I understanding it the wrong way?
If I consider the frequencies in the solution and calculate the ev of calling with kings manually, I get -0.018 rather than +2.509.

Edit: one more thing: to make it work I had to remove the files "msvcp100.dll" and "msvcr100.dll" or I would get a startup error

Hello, in the future plan to place the possibility to calculate spot with ICM?

Hello!
Here is my answers below the order:
- Now it's impossible to make local calculations for turn and river but I could exactly tell you that they are faster than Cloud calculations cause you don't need to spend time sending data over the network. So, all code for the local calculations is strongly cyphered before the license activation of standalone version that is why you couldn't run it.
- Yes, there was such problem but now we suppose that this issue has been resolved.
- The EV of fold for Kings is equals 0 because you wouldn't lose chips acting that way and the EV of call is positive bacause first player plays with Queens 31.5%.

Yes, we are going to add ICM model to the calculations in the nearest future.