Hey, quit picking on my cpu! I'm on an Intel duo, 3.2 ghz, 2gigs physical and 2gigs virtual memory.. only kidding about picking on me. I did run some more files from past files I hadn't bothered with, about 8k and it got 28/sec. Still way slower than you've discussed. I can't see it being my machine. I even have a ram monitor and my physical ram was never lower than 1.3gigs at any time during the operation. CPU is fast enough, got enough memory, that only leave either the dos or FPDB, I am running Vista.

dog

One question: I'm running the "mucked" module again, and without any problems, other than a small annoyance that I'm sure is an easy fix...

The mucked cards are too delayed. The next hand is dealt and starting play, in most cases, when they pop up. I did see a "timeout" in that part of the config file, with a "5". Is that the delay before it displays? How can I shorten the delay between end of hand and displaying the mucked cards.

One other things, again minor. Is it in a future snapshot that when we move the mucked cards with the mouse that they will hold. I hate having to make minor adjustments, then sit in a game, then hope the seat you're adjusting plays a hand. Too time consuming. Thanks

Dog

No idea if this is the case in this situation but if this is an I/O bound process then hard drive speed could make a big difference.

I'm not sure what is a "good" speed. I ran a simple speed test on my hardrive and it was rated at 58.3mg/sec, how is that related to your speed?

Dog

Different hard drives spin at different speeds, usually something like 5600rpm, 7200rpm or 10000 rpm. On my last computer I upgraded from a 5600 rpm disk drive to a 10 000 rpm disk drive and my HEM import speed skyrocketed.

edit: "skyrocketed" might be a bit ott, but iiirc it was something like a 50-60% improvement in speed.

Yeah, those were the exact lines. I just tried installing on my work machine (also ubuntu), and all went smoothly and took all of 8 seconds. I'll just clean out my home machine's databases and start over.

Thanks for the help -- and for keeping this project going. A while back I was fiddling around with the "python scripting for computational science" book", which is probably right up the alley of what y'all are doing. If you haven't already checked it out its worth a look, there may be some tidbits in there for you.

The time between imports is controlling the delay of the mucked cards popping up. Adjust that with the very top widget on the autoimport screen. I set it to 2--since you have a speedy pc, 1 or 2 is fine.

Adjust it in the config here:
To save your mucked layout, just select save HUD layout from the FPDB Menu thingie.

Thanks for the tips. I also changed the "timeout" to 20 secs which allowed me more time to reposition during the reorganizing of the mucked cards, much better. When it was all in the places I wanted, I saved. Then I replaced the timeout back to "5" and made the delay 3 to speed that up. Total success.

To anyone interested, here's the mucked layout for a 10-man table in case you want a fairly good starting point in the config file.

thx

Dog

Bump, can anyone help?

This is an error from a bulk import. Do you get this every time you try to BI?/Is it preventing you from importing? (you can delete the fpdb-errors file and try a BI to see if you get a new file with that error.)

It looks like you have tried to run a BI with an empty Site filter: drop down. (It should have the name of the site you are importing from.) I can't mine to be blank--maybe if the config file was corrupt.

I can't get the site filter to work. It looks like a graphical error.

I assumed the site filter wasn't needed anymore because of this "graphical error". I will get a screenshot posted soon to explain.

It is interesting, I'm running on a Dell mini 12 and if the fpdb window exceeds the screen size it screws up and won't allow you to scroll up and down to view the screen correctly.

Screenshot of what I see when I click the drop down arrow to select the site filter:



Also, it seems like there are no options in the list. If I click the arrow I can't use the keys the move through any options and hit enter to choose.

It looks like you are missing the <hhcs sectoin o fthe config

This is what is in the file HUD_config.xml in the directory c:\fpdb\pyfpdb

You can change the the fpdb window size at line 880 of fpdb.py in the pyfpdb folder:
You are the 2nd person I've heard of that is using a netbook. We should probably change that to a netbook-friendly size, like 800x400 or so.

Anyway, I deleted the <hhc section of my HUD_config and tried a BI--I got the same message you reported above. Try: 1) rename your HUD_config to something like BU_HUD_config. 2) Run fpdb, it should make a new HUD_config from the example. 3) See what you have in the BI screen, you will have to navigate to your HH folder.

Thanks for your help, I got BI to work by moving the hud_config.xml from c:\fpdb to the application data\fpdb directory.

No joy with the screen issues, I even changed the settings to 640 x 480 as well as 800 x 400 and nothing appeared to happen.

First of all, Kudo's for the developers of this tracker. You came a long way since last time I checked it out.

I'm an active user on another dutch poker forum, and a user over there ask me to help him out installing FPDB. As I think FPDB is a nice solution for beginners to start with as they usually don't have the bankroll for buying HeM or PT3, I decided to write a proper step by step howto on my blog.

While I got it all running, i still have some issues with matplot and the graphing functions not working. While digging throught this thread for the solution I actually realized a lot of other people over here would love to have a detailed step by step install instruction as well. So in the spirit of opensource I decided to take the time to translate it in english (or at least tried it) and to give it back to the community.

You can find the detailed step by step howto install on my blog at the link below.
http://pokerstart.web44.net/poker/fr...-installation/

It is just a simpel blog which I decided to create a place to record the start of my poker carreer, so no commercial hidden agenda overhere.


Let me know, what you think and offcourse if you have the answer to my plotting issues I would love to hear it.

I tried this but a new HUD_Config is not created. I don't know what "from the example" means, though, so I probably did something wrong.

I also tried what Banzai did, to no avail.

I figured it out and thought I'd share what I did to get it working. It's basically what Eratosthenes and Banzai said to do.

I copied the example HUD_Config file in my c:/fpdb folder into the ApplicationData/fpdb folder. I then deleted the HUD_Config.xml file in the ApplicationData/fpdb folder and restarted the program. It worked like a charm after that.

I don't have a HUD_Config.xml file in my c:/fpdb folder. Not sure why, or if anyone else does.

Thank you very much for all of your help.

Quick question, does this tracker work on the new Euro cash tables at Everleaf? (So can it import the new € handhistories?)

Looks great. I was considering doing one of these when I finally get around to installing fpdb in Windows but maybe I'll just follow yours instead

All-in-one executable now available for Windows.

Hi all--

There is new way to get and install fpdb for Windows users. Just download and run the self extracting 7-zip file from our download page. This should be the default download if you are using windows--the file is named:
fpdb-0.12-20100213.exe (16 MB)

What you get: fpdb plus all of the libraries that are often trouble install. So there is no need to download and install gtk, python, etc. The SQLite database manager is included and enabled by default, so you don't need to download, install and configure PostgreSQL or MySQL (see below for more on this). Except for the method of delivery, this is no different from the tar.gz file from the previous snapshot, so if you already have fpdb running, then don't bother with this file.

How to get it running: Download the file to some convenient location, like your desktop. Double click--it makes a folder named fpdb-0.12-20100213. Open that folder and click on the run_fpdb icon. This will start fpdb with an empty database and using the generic config file. First thing to do is to go to the Import->Bulk Import menu and import some files. Since you have not set up the config yet you will have to navigate to your HH folders. After you do the bulk import you should be able to make graphs and see player stats.

SQLite: As the name implies it is not as heavy duty as PostgreSQL or MySQL. I have tested it with about 50k hands and performance is fine. So SQLite is a good choice if you just want to try fpdb or if you don't have a large number of HHs to import. If you are going to build a database of several hundred k hands, you'll want to use one of the heavier duty database managers.

We've tested this on Windows XP and Vista. So if someone trys this on W7, let us know how it does.

I'll try and take a look at this over the weekend. I was the one that changed it originally - i seem to remember setting an 'auto resize' flag somewhere there, so that fpdb would always resize to something more appropriate(!).

Nice work j1nx, very well written - and in better english than mine.

There are some working installations of matplotlib for python 2.6, but i'm not sure exactly which package they used.

Now we have a .exe version of fpdb, you might find using that with your config file is easier.

Sorrow

I believe the answer is yes - if not it shouldn't be too hard to get some level of support happening - I just need a couple of hand histories. Ekdikeo will be able to answer more specifically.

Sorrow.