PDA

View Full Version : Request to fix evline to $60 and $200 Hypers (6max)



vadiu
05-27-2012, 10:01 PM
Hi dear staff, recently stars added new buy ins at the hyperturbos, the $60 and $200...

as usual, the HM bug to this, so please fix both the redline and the 3rd place finish prize.

Thanks in advance.

Patvs
05-28-2012, 07:52 PM
We're aware of this bug.
To be save, can you also sent us some handhistory + summary files of these new buyins.

bettinglife
06-12-2012, 08:23 AM
When will be the update avbailable?

Patvs
06-13-2012, 12:53 AM
HM-4381. It's still an open bug.

huckleberry
06-17-2012, 11:11 PM
soooo looooong again
just multiply $28.83+$1.17 by 2 and release it

Sarek
06-18-2012, 11:39 AM
Bug 4381 is in queue to be fixed.
Thank you for suggestion, but how to do this - is up to developer's.

vadiu
06-22-2012, 02:13 PM
Bug 4381 is in queue to be fixed.
Thank you for suggestion, but how to do this - is up to developer's.


ETA for that?

huckleberry
06-22-2012, 07:50 PM
Bug 4381 is in queue to be fixed.
the point is that fixing this bug take less time than writing answer to this message. but we have to wait when you will fix all bugs in queue before this one. may be you are working on a very difficult bug that required about 2-3 week, and we have to wait when you will finish? is there any necessary for this? can you release something about internal version fixing this bug only?
PS you may think i'm just practicing in english, but it's not the quite true

Patvs
06-23-2012, 05:16 AM
HM1: detecting whether a game is a HYPER SNG or not..., what the prize distribution is, what the EV structure is:
everything is hard-coded based on the buyin + rake amount.

PokerStars often changes/tweaks the buyin/rake amounts. And add new stakes. (and new satellites)
HM1 isn't 'smart' enough to just read the summary file and detect these changes.
They have to be manually added to the program code.

We're currently looking into an alternative method: for example just having a .xml file on your computer with all the stakes:
if PokerStars adds a new stake or makes changes, all you have to do it download an updated .xml file... or make the changes yourself.

huckleberry
06-23-2012, 11:10 AM
HM1 isn't 'smart' enough to just read the summary file and detect these changes.

is HM2 smart enough? i had imported $60 hyper turbo sngs HH and TS and there was no ev line too :confused:

Sarek
06-23-2012, 11:39 AM
This is being worked on.
Currently - requires changes in code. Possibly in the future - XML will be introduced

huckleberry
06-23-2012, 02:20 PM
ok, thanks guys! but i still thinking that developers can fix this for 10 minutes, because they already have code "detecting whether a game is a HYPER SNG or not" and icm structure and everything except the prize distribution :p

huckleberry
06-28-2012, 02:29 AM
thanks a lot that inform me and TS that problem have solved

deafsheep
07-10-2012, 10:53 AM
SO HOW LONG CAN IT TAKE REALLY?
A century guys?
Add 60 and 200 support already.

Its easy and you know it. Just add support and release a fix , simple as that.

Patvs
07-10-2012, 05:06 PM
This is still an open ticket.
In fact it resulted in a new spinoff HMONE-127 ticket: ICM button EV Diff is different than EV Diff in reports


ICM button is accurate.
Reports is not.
The Luck Adjusted Winnings line is based on the values in reports.

hen
07-11-2012, 04:15 AM
This is still an open ticket.
In fact it resulted in a new spinoff HMONE-127 ticket: ICM button EV Diff is different than EV Diff in reports


ICM button is accurate.
Reports is not.
The Luck Adjusted Winnings line is based on the values in reports.

So is the fix you give at the end of this thread: http://forums.holdemmanager.com/manager-bugs/105891-hyperturbo-6max-ps-redline-not-working.html faulty? I don't understand exactly what you mean in reference to the ICM button? I personally care about Results/Reports tabs.

Thanks!

Patvs
07-11-2012, 07:45 PM
It's still faulty, affecting hyper turbo's: affecting both HM1 and HM2:


The EV line---> is the SUM of the EV Diffs (in the reports) + the outcome of the game.
The initial issue was the EV wasn't calculated at all... so the $EV Won would always match $Won, resulting in no 'luck adjusted winnings' line.

The new issue seems to be, the line is shown, but is faulty (the EV Diff values don't match what they should be according to ICM calculations). In the graph of a user who reported this, his EV line dropped from +$1995 to +$1407: a 29% drop. I wrote a detailed report on this as bugticket HMONE-127 (for HM1) the same issue exists as HM-5295 (for HM2)

huckleberry
07-13-2012, 05:12 PM
The new issue seems to be, the line is shown, but is faulty (the EV Diff values don't match what they should be according to ICM calculations).

76042

do you mean this issue? when opponents have same cards? if not - please fix this one too. by the way this issue exists for not only hyper turbo's

UPD sorry, for hyper-turbos only

UPD2 red frame have wrong position, i mean not value 5.15 of "Diff" but value 1.46 of "Luck"

Patvs
07-13-2012, 07:40 PM
Can you email us the handhistory of the AT vs AT STT hand ?

deafsheep
07-13-2012, 11:06 PM
This is being worked on.

Guys, really this is taking how long, 2 months almost since release of 60$ and 200$ games.
This fix literally requires 3 lines of code.
I am a developer myself so I won't lie.

This is some kind of sick joke, you guys so busy doing all other stuff, that you have no time to fix something which by anyones estimation cant take more than 1 hour.

Or please reveal the deep secret, how handling 60$ and 200 is so different from 30$ and 100$ which are just 2x smaller in buyin and require ABSOLUTELY SAME HANDLING.

huckleberry
07-14-2012, 03:14 AM
Can you email us the handhistory of the AT vs AT STT hand ?
done
one more question: in one location my StartEV value is $109.1 and in other location EV value is $112.79. why?

UPD although noone from support can't answer me, now i get the answer.

fozzy71
07-14-2012, 09:34 PM
Guys, really this is taking how long, 2 months almost since release of 60$ and 200$ games.
This fix literally requires 3 lines of code.
I am a developer myself so I won't lie.

This is some kind of sick joke, you guys so busy doing all other stuff, that you have no time to fix something which by anyones estimation cant take more than 1 hour.

Or please reveal the deep secret, how handling 60$ and 200 is so different from 30$ and 100$ which are just 2x smaller in buyin and require ABSOLUTELY SAME HANDLING.

The problem isn't actually the EV calcs. It is the tourney detection system recognizing those new games as different types of SNGs. If they are recognized as the wrong type then it will use the wrong EV calcs.

While it may seem like a simple fix that we should be able to fix right away, the fact of the matter is we have a large list of issues that we have to prioritize for our developers for the next release, and the major hybrid/datagrid update planned after that, and EV issues are not nearly as important as other bugs on that list. We can't put an EV issue at the top of the list and tell the developers to quit working on import errors and HUD problems instead. We must work on the problems that affect actual game play for the largest number of our customers first and then we can work on bugs like EV or incorrect results and new features.


done
one more question: in one location my StartEV value is $109.1 and in other location EV value is $112.72. why?

Without knowing exactly which locations you are referring to, or possibly seeing some screen shots, we can't really answer that. The best way for us to answer that is if you send us the detailed question and screen shots along with the problem hand history. Once our testers have had time to test the hand(s) and compare their results to your screen shots they will reply in detail.

huckleberry
07-15-2012, 10:11 AM
Without knowing exactly which locations you are referring to, or possibly seeing some screen shots, we can't really answer that. The best way for us to answer that is if you send us the detailed question and screen shots along with the problem hand history. Once our testers have had time to test the hand(s) and compare their results to your screen shots they will reply in detail.

thanks for such detailed reply, but what about look around? :p just one post earlier


76042

do you mean this issue? when opponents have same cards? if not - please fix this one too. by the way this issue exists for not only hyper turbo's

fozzy71
07-15-2012, 11:39 AM
thanks for such detailed reply, but what about look around? :p just one post earlier I am sorry, I didn't realize you were referring to a screen shot from earlier in the thread. It looks like it is possibly a bug and we will need you to send us the hand history like Pat asked so we can test it in detail.

hen
07-16-2012, 10:51 AM
Still waiting on this. Hypers are one of the most played format on Pokerstars and a huge contributor of rake to pokerstars due to the amount of games that go off. I'm sure that there are a lot of HEM 1 users who are waiting for this update, while nobody is requesting a major hybrid/datagrid update?!

Sorry, I like Foz and HEM, but I really need these EV numbers and this wait is really unacceptable especially with the "fixes" that have been released then been shown to be not working.

fozzy71
07-17-2012, 12:42 AM
Still waiting on this. Hypers are one of the most played format on Pokerstars and a huge contributor of rake to pokerstars due to the amount of games that go off. I'm sure that there are a lot of HEM 1 users who are waiting for this update, while nobody is requesting a major hybrid/datagrid update?!

Sorry, I like Foz and HEM, but I really need these EV numbers and this wait is really unacceptable especially with the "fixes" that have been released then been shown to be not working.

The EV and Tourney Detection issues will eventually be fixed but they are low priority when compared to to the bugs we are working on regarding importing/results, HUD, and HM2 Hybrid/Datagrid. Expected Value always takes a back seat to bugs that affect game play and actual results.

huckleberry
07-19-2012, 04:55 AM
I am sorry, I didn't realize you were referring to a screen shot from earlier in the thread. It looks like it is possibly a bug and we will need you to send us the hand history like Pat asked so we can test it in detail.
you don't need my hand history, you just need someone who knows what the value of StartEV in second location (the one that below) means and it doesn't matter what hand i have. why did graphics guru respond for technical question? why after three e-mails to support and many attachments i just have a link to the FAQ (there is no answer for my question) and another request to attach HH for answer the question that does not require HH. i send one more e-mail where i suggested that this value is absolutely useless thing that recalculates StartEV depending known hole cards and i attached HH again but received in response one more offer to send HH :confused:

Patvs
07-19-2012, 10:52 PM
You've only sent us a screenshot of the hand.
We need an actual handhistory txt file: something we can import into HM1.

huckleberry
07-20-2012, 05:02 AM
i hope you mean forum only, but i talked about emails to support

Patvs
07-20-2012, 10:33 PM
In your emails, I only saw a screenshot in two emails.
I couldn't find an actual handhistory.



you don't need my hand history, you just need someone who knows what the value of StartEV in second location (the one that below) means and it doesn't matter what hand i have.


I'd like the AT vs AT handhistory, so I can import the hand, click the ICM button, and calculate the EV Diff manually.
It's possible HoldemManager makes an entirely different mistake by assuming the 1.5% the AT has to win (so the hand not ending up in a splitpot) causes the difference.

TKO
07-24-2012, 02:46 PM
There is an EV-line at the 60s and 200s Hypers(6max) since this 25Jun2012a.exe update/fix, but as you said it`s still faulty.

Now my question: Is it only that EV diff. problem with two exact same hands at showdown?
When yes, that should not really effect the overall EV-line on a samplesize of 1k games or more any, right?

It is just important for me to know if my overall EV-line is maybe 0.2% off, or perhaps 2%???

Patvs
07-24-2012, 09:31 PM
The split pot issue looks like a different issue.



It's still faulty, affecting hyper turbo's: affecting both HM1 and HM2:


The EV line---> is the SUM of the EV Diffs (in the reports) + the outcome of the game.
The initial issue was the EV wasn't calculated at all... so the $EV Won would always match $Won, resulting in no 'luck adjusted winnings' line.

The new issue seems to be, the line is shown, but is faulty (the EV Diff values don't match what they should be according to ICM calculations). In the graph of a user who reported this, his EV line dropped from +$1995 to +$1407: a 29% drop. I wrote a detailed report on this as bugticket HMONE-127 (for HM1) the same issue exists as HM-5295 (for HM2)



The main issue is: don't trust the EV line. It's very likely your EV line should be ~30% higher than it shows now.

huckleberry
07-25-2012, 01:38 AM
The main issue is: don't trust the EV line. It's very likely your EV line should be ~30% higher than it shows now.

can you show an example what wrong with EV line? incorrect calculation of $Evdiff? i just checked a few hands manually - the values of "Luck" from ICM button was absolutly correct, but values of "Luck" and "$EV Diff" from reports everytime was a little different. it's obvious that one of them is incorrect. how does it work all the time?

in split pots this two values ("Luck" and "$EV Diff") just totally different, but it looks like the same issue.

or maybe there are some other problems with EV line?

huckleberry
07-25-2012, 02:42 AM
i checked a few turbo speed tournaments - "Luck" and "$EV Diff" values was identical every time. but if you want to get $EV Diff value in hyper-turbos you just need:
$EV Diff = Equity%/100 * icm_EV_if_you_win + (1-Equity%/100) * icm_EV_if_you_lose - result_icm_EV
and that is wrong, because you have to use not Equity, but probability of win, probability of lose and probability of tie

TKO
07-25-2012, 03:51 AM
The main issue is: don't trust the EV line. It's very likely your EV line should be ~30% higher than it shows now.


Ok, thanks.
You said only ~30% higher? I assume ~30% lower is also possible, or not?

hen
07-25-2012, 06:17 AM
Is winnings showing up correctly for the 60s and 200s in a working version? Or am I right in thinking that to get winnings line working properly you would mess up the EV line for 15s/30s/100s?

netsrak
07-25-2012, 09:06 AM
There were some Merge fixes in the latest builds, can you try this prerelease?
http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_22Jul2012.exe

hen
07-25-2012, 09:57 AM
There were some Merge fixes in the latest builds, can you try this prerelease?
http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_22Jul2012.exe

You know we have the problem on Pokerstars though right?

Patvs
07-26-2012, 12:35 AM
can you show an example what wrong with EV line? incorrect calculation of $Evdiff? i just checked a few hands manually - the values of "Luck" from ICM button was absolutly correct, but values of "Luck" and "$EV Diff" from reports everytime was a little different. it's obvious that one of them is incorrect. how does it work all the time?

in split pots this two values ("Luck" and "$EV Diff") just totally different, but it looks like the same issue.

or maybe there are some other problems with EV line?




The ICM values are correct.
The EV Diff (in reports, sessions) is wrong, but not by much. I don't know which formula it's using though.
If it's wrong by $0.25 each all-in hand, if you encounter 2 all-ins in each game, and play 1000 games, it will quickly result in being off by $500. The graph is currently using the EV Diff values.

Patvs
07-26-2012, 12:37 AM
Ok, thanks.
You said only ~30% higher? I assume ~30% lower is also possible, or not?

The only examples I've seen is where the EV Diff is always slighter lower than the correct value.
So the EV line is lower (and closer to the winnings line), where it should be higher.

Patvs
07-26-2012, 12:39 AM
Is winnings showing up correctly for the 60s and 200s in a working version? Or am I right in thinking that to get winnings line working properly you would mess up the EV line for 15s/30s/100s?


As long as you import the summaries, Read the tournament summary section @ http://faq.holdemmanager.com/questions/6/PokerStarsSetup
the winnings/ finish position, etc. will always be accurate.



Also the EV issue is not affecting regular STTs, turbo STTs, and other hyper-turbo stakes.

hen
07-26-2012, 06:24 AM
As long as you import the summaries, Read the tournament summary section @ http://faq.holdemmanager.com/questions/6/PokerStarsSetup
the winnings/ finish position, etc. will always be accurate.

Also the EV issue is not affecting regular STTs, turbo STTs, and other hyper-turbo stakes.

The winnings line for $60s is wrong for me in 1.12.10b - I importing both HHs and Tournament Summaries. It's well above what I am winning. It's paying 3rd $57.66, rather than $0. It also says 2nd is $121.09, when in the lobby it says 2nd $121.08.

Is there a simple fix for this at least please?

EV and winnings line are fine in 30s, if that's what you're saying in your second paragraph.

Patvs
07-26-2012, 11:11 PM
@hen email us a handhistory + summary file of a $60 STT where the winnings are not correct.


@huckleberry, thanks for your handhistories, I'll go over them today.

hen
07-27-2012, 04:09 AM
@hen email us a handhistory + summary file of a $60 STT where the winnings are not correct.

Sent to support@holdemmanager.com just now.

Thanks for looking at this...here's to hoping a fix comes soon :)

hen
07-29-2012, 01:39 PM
Any ETA on this now? Can you confirm that the winnings for the $60s isn't working for everyone, not just me ya?

Patvs
07-29-2012, 08:48 PM
Can you PM me from which email address you sent us the handhistory?
(I couldn't find them under gamedr********@hotmail.co.uk

hen
07-30-2012, 01:19 PM
I think the issue with 3rd place being recorded is still a problem in some of the games and we are working on a fix.

Please update to the latest version - http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_26Jul2012.exe

*Try creating a new UTF DB - http://faq.holdemmanager.com/questions/75/Create+%7B47%7D+Delete+Database
*Now import a small portion of your \HMArchive so you can see if the problem exists in the new DB - http://faq.holdemmanager.com/questions/98/How+to+Import+%26+Export+Hands


If that doesn't fix the issues, please reproduce the problem and email a screen shot - Capture a Screen Shot of your Desktop or the Active Window in Windows (http://graphicssoft.about.com/cs/general/ht/winscreenshot.htm)

Please zip and email the problem hand histories with the screen shot, with a link to this thread and your forum name, to support@holdemmanager.com

All original hands that are auto-imported get moved to an archive for performance reasons. Your \HMArchive is likely not in your \Holdem Manager directory. We normally suggest you use C:\HMArchive, but you may have put it anywhere. The archive should be organized by \Month\DayOfMonth (example: D:\HMArchive\2009\07\31).

Can you comment on if this build will have any affect on hyper winnings/ev? for 30s-200s?

huckleberry
07-30-2012, 09:05 PM
@huckleberry, thanks for your handhistories, I'll go over them today.
and what about me? can you give any information about existing tickets affecting EV in hyper-turbo? do you admit that the EV issue affect not only $60, but $30 buy-in too? and HU? what about Equity?

netsrak
07-31-2012, 07:28 AM
I'll forward this to Pat

deafsheep
08-01-2012, 02:10 AM
Good lord. 05-27-2012 is the start of this SUPER DIFFICULT BUG THREAD.

2 freaking months have passed.
You guys make huge $$ with HEM and HEM2.

But its 2 freaking months. And I am a software developer myself.

Should we beg for a fix? Gather some $$ so you hire a programmer who can fix bugs?
WTF is this really.

fozzy71
08-01-2012, 05:46 PM
Good lord. 05-27-2012 is the start of this SUPER DIFFICULT BUG THREAD.

2 freaking months have passed.
You guys make huge $$ with HEM and HEM2.

But its 2 freaking months. And I am a software developer myself.

Should we beg for a fix? Gather some $$ so you hire a programmer who can fix bugs?
WTF is this really.

Copying some of my replies on this topic from the 2+2 thread:

EV issues are generally lower priority than importing, HUD, and actual results bugs so these EV issues have had to take a back seat while. To be honest, some of these other bugs that are higher priority than EV bugs have also taken a back seat while we work on the database update process in our next release, and the upcoming performance update that will follow the database update release. Once we get this database update and performance update behind us these lower priority issues will be addressed. I hope you can understand why EV issues for one site/game-type have to be lower priority while we work on issues that affect our entire customer base.

I will try to bump the priority of this issue with the bug managers but there is only so much I can do to expedite it as they have to work this in to the current developer workload.

hen
08-01-2012, 06:14 PM
It's not just EV though, $60s winnings are off too. :/

fozzy71
08-02-2012, 12:05 AM
It's not just EV though, $60s winnings are off too. :/

Send us more hands. The more hands we get, the more likely it is we will get it fixed ASAP. Please zip and email the problem hand histories, with a link to this thread and your forum name, to support@holdemmanager.com

hen
08-02-2012, 08:28 AM
Send us more hands. The more hands we get, the more likely it is we will get it fixed ASAP. Please zip and email the problem hand histories, with a link to this thread and your forum name, to support@holdemmanager.com

Sent 8 more HH files with TSs. If you need more, tell me.

Emailed from hennnerz at my g-mail address.

huckleberry
08-02-2012, 09:58 AM
Send us more hands. The more hands we get, the more likely it is we will get it fixed ASAP.
lol?

huckleberry
08-02-2012, 09:59 AM
and what about me? can you give any information about existing tickets affecting EV in hyper-turbo? do you admit that the EV issue affect not only $60, but $30 buy-in too? and HU? what about Equity?
:confused:

Patvs
08-04-2012, 06:38 PM
The winnings line for $60s is wrong for me in 1.12.10b - I importing both HHs and Tournament Summaries. It's well above what I am winning. It's paying 3rd $57.66, rather than $0. It also says 2nd is $121.09, when in the lobby it says 2nd $121.08.

Is there a simple fix for this at least please?

EV and winnings line are fine in 30s, if that's what you're saying in your second paragraph.




I imported your handhistory + summary in both HM1 and HM2.
It was a Buy-In: $57.66/$2.34 USD - 6 players game where you finished in 3th place.
HM1 and HM2 both correctly show you won $0. (Net Winnings = -$60)

Patvs
08-04-2012, 06:44 PM
It also says 2nd is $121.09, when in the lobby it says 2nd $121.08.



Both in HM1 and HM2 I get the accurate: $121.08.


HM1 update to: http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_03Aug2012.exe and reimport your archive folder into a new database.

Patvs
08-04-2012, 06:56 PM
@hucklyberry:

The $30 games EV:

The ICM values are correct.
The EV Diff (in reports, sessions) is wrong, but not by much. And that's what the graph is based on.
So you're right the $30 hyper turbo's have the same EV issue as the $60 / $100 hyper turbos.



-In all the hands you sent us there was no AT vs AT splitpot hand, so I couldn't test the splitpot EV.





I'd like the AT vs AT handhistory, so I can import the hand, click the ICM button, and calculate the EV Diff manually.



and what about me? can you give any information about existing tickets affecting EV in hyper-turbo? do you admit that the EV issue affect not only $60, but $30 buy-in too? and HU? what about Equity?

hen
08-05-2012, 03:57 AM
I imported your handhistory + summary in both HM1 and HM2.
It was a Buy-In: $57.66/$2.34 USD - 6 players game where you finished in 3th place.
HM1 and HM2 both correctly show you won $0. (Net Winnings = -$60)

What are you saying? All I can do is use the latest HEM1 and report my findings. It's clearly not working for me and all I can do is report what I find with 1.12.10b...


Both in HM1 and HM2 I get the accurate: $121.08.


HM1 update to: http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_03Aug2012.exe and reimport your archive folder into a new database.

I've updated to this build and imported yesterdays hands however I am getting errors on the rebuys. Around 21 SNGs from ~250 are reporting rebuys of 57.66 and one is showing a rebuy of 172.98. I know it's straightfoward to fix this myself, just a tad annoying.

I have zipped and emailed these hands to support from hennnerz at my gmail address again.

What should work with this build? Winnings correct for all hypers? What about EV? Will EV for $30s/100s be messed up? Thanks.


Edit to add: The rebuy issue is also prevalent in the $15s and $30 hypers in this build, normally showing as having used 1 rebuy, sometimes 2. Obviously there should be 0 rebuys for these games.

Second Edit: No rebuys showing up in games played today so far...

huckleberry
08-05-2012, 11:45 AM
@hucklyberry:

-In all the hands you sent us there was no AT vs AT splitpot hand, so I couldn't test the splitpot EV.

why do you not listen me? now i'm not interesting about split pot issue - it's the same issue: the ICM values are correct and the EV Diff (in reports, sessions) is wrong. fix this and you will fix the split issue too. if you so need this hand you can find it in attachment to this my emails: first was with title "issue with EVdiff in 6max SNG" at July 14, 07:22 UTC and another at July 18, 18:53 UTC

but you said nothing about wrong EV Diff value in heads-up hyper-turbos and nothing about wrong "Equity %" value :confused:

Patvs
08-05-2012, 08:09 PM
@hen: did you reimport all your hands into a new empty database?

If the rebuys are wrong (for old games) email us a handhistory + summary file.

All winnings should be accurate. EV is not in hyper turbos.

Patvs
08-05-2012, 08:19 PM
why do you not listen me? now i'm not interesting about split pot issue - it's the same issue: the ICM values are correct and the EV Diff (in reports, sessions) is wrong. fix this and you will fix the split issue too. if you so need this hand you can find it in attachment to this my emails: first was with title "issue with EVdiff in 6max SNG" at July 14, 07:22 UTC and another at July 18, 18:53 UTC

but you said nothing about wrong EV Diff value in heads-up hyper-turbos and nothing about wrong "Equity %" value :confused:


I couldn't find handhistories attached to the emails: "issue with EVdiff in 6max SNG" at July 14, 07:22 UTC and another at July 18, 18:53 UTC. (one of these emails has a SCREENSHOT of the splitpot.... not the actual handhistory)

The handhistories you did sent were mostly 18man (2 table) SNGs where there is no EV since it's a MTT.. Also there were no heads up hyper turbos handhistories.
Sent us a proper handhistory file + summary file of these games.

hen
08-06-2012, 03:33 AM
I've updated to this build and imported yesterdays hands however I am getting errors on the rebuys. Around 21 SNGs from ~250 are reporting rebuys of 57.66 and one is showing a rebuy of 172.98. I know it's straightfoward to fix this myself, just a tad annoying.

I have zipped and emailed these hands to support from hennnerz at my gmail address again.


@hen: did you reimport all your hands into a new empty database?

If the rebuys are wrong (for old games) email us a handhistory + summary file.

All winnings should be accurate. EV is not in hyper turbos.

Yes - reimported into a brand new database from my HMArchive folder. I've already emailed some example hands to you guys.

huckleberry
08-06-2012, 05:50 AM
The handhistories you did sent were mostly 18man (2 table) SNGs where there is no EV since it's a MTT.. Also there were no heads up hyper turbos handhistories.
Sent us a proper handhistory file + summary file of these games.

my HH was *.txt files, may be this caused the problem. i sent it again in ZIP archive

18man handhistories was sended for other issues with HM2.

huckleberry
08-06-2012, 11:33 AM
79062
look at screenshot. this hand from $7 HU hyper-turbo sng. i sent HH of this hand by mail and i hope you have this hand now. i'm not talking about Luck and $EV diff in HU hyper-turbo (but they are different), i'm talking about equity.

equity = probability of win + 1/2 probability of tie

so, "Equity %" from reports is wrong

Patvs
08-07-2012, 09:14 PM
I wrote up the equity bug as HMONE-145.

For which issue did you sent the 18man SNGs handhistories?


---

I also wrote up the splitpot bug. This bug could be spinoff of HMONE-127, but I still wrote it up as a separate bug report anyway.
Bugticket on that is: HM-5469 (I wrote it up as a HM2 bug, since it also occurs in HM2, and HM2 bugs get higher priority)

huckleberry
08-07-2012, 10:40 PM
For which issue did you sent the 18man SNGs handhistories?


my mail about this was dated July 22, 10:48 UTC and title "few issues with HM2". it was rather HM2 issues, not 18man only. but i had 18man HH and sent it.



I wrote up the equity bug as HMONE-145.


good. further.
look at screenshot again: "cEV Diff" is wrong too
940*0.30558 - 940 = -652.75, not 656

further. i found wrong values of Equity in ALL sng tournaments i had checked - 6max turbo, hyper-turbo, HU, 18max, 45 max, 180max. multi table sng tournaments have no EV, but they have wrong cEV Diff value, so they have wrong values of "$(EV adjusted)" and "EV bb/100" in the Reports tab - i checked it specially.

i have sent you one hand from CAP table: AdKc vs AcJh. wrong equity too.

i tested some hands with HM2. wrong again.

i can send you some HH to test, but better way you will first check any hands that i have sent you earlier

huckleberry
08-07-2012, 11:22 PM
by the way, what stakes HMONE-127 and HM-5295 affecting? what about HU and 9max hypers? i have no HH from $7, $15 stakes 6max, no HH from most stakes HU and 9max hypers, but it's more likely they all have the same issue: different values of "Luck" and "$EV diff".

DogNamedBluff
08-08-2012, 05:45 PM
my mail about this was dated July 22, 10:48 UTC and title "few issues with HM2". it was rather HM2 issues, not 18man only. but i had 18man HH and sent it.
I located the email with the hand history and forwarded it to Patvs.


by the way, what stakes HMONE-127 and HM-5295 affecting? what about HU and 9max hypers? i have no HH from $7, $15 stakes 6max, no HH from most stakes HU and 9max hypers, but it's more likely they all have the same issue: different values of "Luck" and "$EV diff".
The bugs are written as if they are affecting all stakes and will be looked into by the developers.

huckleberry
08-09-2012, 07:33 AM
I located the email with the hand history and forwarded it to Patvs.

i've already got the answer to this mail and i don't know why Patvs interested about it. but i'm still interesting about results of testing the bugs, which I described in that mail (dated July 22, 10:48).

Patvs
08-09-2012, 01:57 PM
I verified the bugs and I agree on the bugs you reported.
They occur in both HM1 and HM2.
On hyper turbo's (both HU and 6max)... possible also in other games.


Definately NOT just in the $60 and $200 games... but also with other stakes.
They don't seem to occur in "regular" STTs. (regular speed, regular stack)

I wrote up four bug tickets on this:
-For the equity
-For the EV Diff
-Splitpot (could be related -and the same as- the EV Diff
-To add the ICM button in HM2


Now two things will happen next:
-1 another tech member has to verify the bug
-2 programmer has to fix it, but I have no eta when these tickets will be addressed.

huckleberry
08-10-2012, 05:41 PM
thank you, Patvs, for the such detailed answer! i'll be very pleased if you will give me the number of "equity bug" for HM2.

Patvs
08-10-2012, 10:21 PM
Equity bug is HMONE-145

huckleberry
08-11-2012, 04:40 AM
you know i just have to ask you about so strange format for this HM2 ticket. before this one there was separate tickets with a unique number for the similar issues of HM1 and HM2 :confused:

Patvs
08-11-2012, 08:59 PM
For the big main ticket: Luck Adjusted Winnings from on PS,
there is HMONE-127 (HM1)
and HM-5295 (for HM2)


The other tickets mention INSIDE the ticket, it's a ticket affecting BOTH HM1 and HM2... although the tickets are separate JUST FOR
HM1 or HM2.

hen
08-16-2012, 03:22 AM
Any updates or ETA?

How far off is the EV in the latest version of HEM1 please? Is it consistently above/below? Do you know if 1.5s-15s are also messed up or is it just $30s+?

Patvs
08-16-2012, 09:57 PM
I think ALL hyper turbo's are affected.

The EV line should be up to ~30% higher than it shows now.

hen
08-21-2012, 11:41 AM
EV is still not accurate right? I think I am subscribed to the email but not certain, what's the code so I can subscribe again?

Some people are speaking as though current EV is working...are you sure that EV should be 30% higher, or is that just in a single case?

Patvs
08-21-2012, 12:55 PM
This is still unfixed.
There are several bug tickets on this issue. Your email address is also attached to them.
Though, I'll also update this thread, when I see the fixed is checked in.

EV is working... but the wrong values are used to make up the graph (and reports).
For 1 user who played a couple of thousands of games, the difference was 30%.

deafsheep
08-29-2012, 12:01 PM
Just great guys.

Nothing to add here.

udbrky
08-29-2012, 08:55 PM
We have bumped this up again to be looked at by the programmers.

hen
09-04-2012, 12:17 PM
Hi udbrky!

Any progress with this one? Or an ETA?

Thanks!

fozzy71
09-04-2012, 02:07 PM
Hi udbrky!

Any progress with this one? Or an ETA?

Thanks!

The developers pushed some fixes late last week and the testers are checking to see if the fixes are successful so I can't really give much info or any ETA until the testers have more time to investigate them. We are hoping we can get them fixed before our next official releases but it is just too early to be sure if they are indeed fixed or when they might be released.

huckleberry
09-04-2012, 02:16 PM
you can employ me as tester :cool:

hen
09-09-2012, 09:38 AM
Hate to bump this thread again, but is it correct that in HEM 2 there is correct EV in the hypers on Pokerstars?

Is it worth me trying the latest HEM 1 internal release too or not?

fozzy71
09-09-2012, 01:45 PM
Hate to bump this thread again, but is it correct that in HEM 2 there is correct EV in the hypers on Pokerstars?

Is it worth me trying the latest HEM 1 internal release too or not?

The EV seems to be working in HM2 but I can't say if it is 100% accurate or not as it is not my area of expertise. I will ask Pat to do some testing on this later to see if he can confirm if it is now completely accurate in HM2.

We are waiting for a special cfg file to be added to the HM1 build. This cfg file is what fixed the issue in HM2. Once it is added to HM1 and tested to see if it fixes all the hyper issues we should be able to finally get a new HM1 release out to address these issues.

Patvs
09-09-2012, 05:27 PM
It looks like: EV is accurate and fixed in the latest HM2 builds.
In HM1, I'm unsure if the reports show the correct value... or if you click on the ICM button that is the accurate value.

The programmers believe it's the ICM report that's showing the wrong value.
It's annoying since I would always rely on this HM1-ICM report to verify if the EV was accurate in HM2.
So this has resulted in several new bug tickets, and has changed the priority of existing tickets:
(we really should add a similar ICM report to HM2)

We're currently re-testing / re-verifying these tickets.

huckleberry
09-10-2012, 09:56 AM
It looks like: EV is accurate and fixed in the latest HM2 builds.
no



In HM1, I'm unsure if the reports show the correct value... or if you click on the ICM button that is the accurate value.
are you unsure? why? do you not know how to calculate it manually and verify it?



The programmers believe it's the ICM report that's showing the wrong value.

do they mean ICM button showing the wrong value? it's amazing. i want to help you, but it looks like you don't want to change anything.
you can use PokerStove, PokerStrategy Equilab (or PS Equilator) to check equity values from ICM button and they are identical up to the fourth decimal place. you can use ICM Poker Calculator (http://www.icmpoker.com/Calculator.aspx) or any other icm calculator to check the Prize Pool % in the ICM button.
then you can calculate ICM EV:
ICM EV% = probability_of_win * your_%_of_prize_pool_if_you_win + probability_of_loose * your_%_of_prize_pool_if_you_loose + probability_of_tie * your_%_of_prize_pool_if_tie

ICM EV$ = ICM EV% * prize_pool_$

and finally:
EV_diff = ICM EV$ - Result$

that's all. is it so hard?

fozzy71
09-10-2012, 03:42 PM
no

Which version of HM2 are you using? Can you give specific examples of incorrect EV calcs in HM2 and send us the details and hand histories via email for testing?

We have a new internal HM1 release we would like your feedback on. The results and EV seem to be inline with HM2 now but we obviously need to do a lot of detailed testing for both HM1 and HM2 to verify if they are in fact accurate according to those EV tools you mention.

HM1 - http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_10Sep2012.exe

**Stars - You will want to create a new database for testing this and import a small portion of your archive folder. This has had limited internal testing at this point so please report any problems you discover in the HM1 forums.


are you unsure? why? do you not know how to calculate it manually and verify it?


do they mean ICM button showing the wrong value? it's amazing. i want to help you, but it looks like you don't want to change anything.
you can use PokerStove, PokesStrategy Equilab (or PS Equilator) to check equity values from ICM button and they are identical up to the fourth decimal place. you can use ICM Poker Calculator (http://www.icmpoker.com/Calculator.aspx) or any other icm calculator to check the Prize Pool % in the ICM button.
then you can calculate ICM EV:
ICM EV = probability_of_win * your_%_of_prize_pool_if_you_win + probability_of_loose * your_%_of_prize_pool_if_you_loose + probability_of_tie * your_%_of_prize_pool_if_tie

and finally:
EV_diff = EV$ - Result$

that's all. is it so hard?

Thank you for all of this info. Typically the developers do the bulk of the EV calculation verification but we are going to use the info you provided to help us develop a new detailed EV testing process for EV fixes and test cases when releasing new builds. This should add a level of EV verification we were previously lacking and lighten the load for our developers.

hen
09-10-2012, 05:21 PM
We have a new internal HM1 release we would like your feedback on. The results and EV seem to be inline with HM2 now but we obviously need to do a lot of detailed testing for both HM1 and HM2 to verify if they are in fact accurate according to those EV tools you mention.

HM1 - http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_10Sep2012.exe

**Stars - You will want to create a new database for testing this and import a small portion of your archive folder. This has had limited internal testing at this point so please report any problems you discover in the HM1 forums.

I don't have much time right now, but I did the above and imported today's games/hands and the graph produced was identical. I didn't check EVs in the reports tab, but the before/after update graphs were the same.

fozzy71
09-11-2012, 04:55 PM
I don't have much time right now, but I did the above and imported today's games/hands and the graph produced was identical. I didn't check EVs in the reports tab, but the before/after update graphs were the same.

I noticed drastic changes in the results and EV when comparing my results imported to the 26Aug build vs the results imported to the 10Sep build. Can you send us some screen shots and hand histories?

hen
09-12-2012, 03:01 AM
I just downloaded HEM2 for the first time yesterday, then upgraded to the latest version I could find: 7318_HM2AutoUpdate.exe

Is this not the latest file? I looked for a 10Sep build but I'm struggling? I checked your recent posts but no luck either.

huckleberry
09-12-2012, 03:13 AM
Which version of HM2 are you using? Can you give specific examples of incorrect EV calcs in HM2 and send us the details and hand histories via email for testing?

HM2 .7318. i created a new db and i again imported this (http://forums.holdemmanager.com/manager-general/247931-request-fix-evline-60-200-hypers-6max-7.html#post1363582) hand and HM2 again shows this values: equity = 30.2%, $EV diff = -8.99, cEV diff = -656.39. this values are wrong.



Thank you for all of this info. Typically the developers do the bulk of the EV calculation verification but we are going to use the info you provided to help us develop a new detailed EV testing process for EV fixes and test cases when releasing new builds. This should add a level of EV verification we were previously lacking and lighten the load for our developers.

you can also watch this EV Explained FAQ (http://faq.holdemmanager.com/questions/120/EV+Explained), 13th minute



We have a new internal HM1 release we would like your feedback on. The results and EV seem to be inline with HM2 now but we obviously need to do a lot of detailed testing for both HM1 and HM2 to verify if they are in fact accurate according to those EV tools you mention.


i have the problem with 10sep2012 release. replayer does not open with some hands


I noticed drastic changes in the results and EV when comparing my results imported to the 26Aug build vs the results imported to the 10Sep build.
"in the results"? why? as i know it was alright with results.
i do the same thing: imported some $30 hyper sng to the 26aug build and to the 10sep build (created new DB every time) - there is no difference.

it's very annoying, guys. i waste so much time for all this: to explain, to prove, it looks like you and programmers have very poor knowledge about how it works, but you don't trust me and ask more and more hands, explanations and proofs. then you release new build without any changes and again ask about testing and hands and proofs that it doesn't work right. it's very annoying, guys.

Patvs
09-12-2012, 08:27 AM
HM1 sept-10 build is: http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_10Sep2012.exe


Note: EV is NOT working properly in this release for hyper turbo's.
-1 you can't replay the hands (this is a known bug)
-2 it will show EV Diff = 0 for most hands.

The only good news is, I've now convinced the programmer the EV is wrong in reports, so he now knows what should be fixed.

---

PS you don't have to sent more hands.

hen
09-12-2012, 09:52 AM
Thanks for clearing that up.

To clarify my findings are that there was no difference in the graphs produced between HEM1 (with and without 10-Sept update) and HEM 2 with latest update.

TKO
10-05-2012, 08:18 PM
Any chance on getting this hyper turbo EV running correct?
And of course the hand replayer would be sweet!

Sarek
10-06-2012, 04:49 AM
Please try this build
http://forums.holdemmanager.com/manager-general/announcements.html
but you need to do fresh import - create new database and test with it

TryingHard
10-06-2012, 12:46 PM
Would be nice if someone told us if it fixes the EV line problem instead of every single person trying it out...

So let us know, anyone :) (HM support :D)

huckleberry
10-06-2012, 12:53 PM
02 oct release - ev line is incorrect, replayer doesn't work with some hands

Patvs
10-06-2012, 02:11 PM
HMONE-127 and HM-5295 are still unresolved issues.

huckleberry
10-19-2012, 09:42 PM
just noticed Hold'em Equity widget in HM2. the values of equity from this widget and the values in reports are different - another proof of equity bug.
this bug affecting the cash games too. by the way, wrong equity in cash game = wrong "$ (EV adjusted)" stat, are you interesting about this?

fozzy71
10-20-2012, 02:50 PM
There are some bugs in the EV calculation in both HM1 and HM2 currently depending on which site/games you are playing. Part of the holdup on the EV issues is related to the fact that we use a lookup table for EV which isn’t as accurate as full enumeration. Full enumeration seems to hurt import speeds pretty significantly, so were looking into a couple of options. We are considering a more accurate lookup table but this is a lot of work due to the combinations involved and it will be a while before the developers can implement it. Another option is giving the user the option of full enumeration or lookup tables but changing this would require a reimport and choosing the full enumeration, while it would be the most accurate, could slow import speeds significantly which would affect HM1 and HM2 both. Before we can implement this we first want to explore how accurate the new lookup table is.

huckleberry
10-25-2012, 08:27 PM
thank you for such interesting information (seriously), but it's absolutely unrelated to my words.

ok. if it's so hard for you.
1st step.

this cash hand:

Poker Stars, $0.25/$0.50 No Limit Hold'em Cash, 6 Players
Poker Tools Powered By Holdem Manager - The Ultimate Poker Software Suite.

BTN: $43.59 (87.2 bb)
SB: $94.15 (188.3 bb)
Hero (BB): $51.38 (102.8 bb)
UTG: $53.29 (106.6 bb)
MP: $48.25 (96.5 bb)
CO: $16.88 (33.8 bb)

Preflop: Hero is BB with Qd As
2 folds, CO raises to $1.50, 2 folds, Hero raises to $5.50, CO raises to $16.88, Hero calls $11.38

Flop: ($34.01) 6h 4c 2c (2 players)
Turn: ($34.01) 8s (2 players)
River: ($34.01) 9d (2 players)

Results: $34.01 pot ($1.53 rake)
Final Board: 6h 4c 2c 8s 9d
Hero showed Qd As and lost (-$16.88 net)
CO showed Kc Ad and won $32.48 ($15.60 net)





87741

in the report AsQd vs AdKc have 25.6% equity. Hold'em Equity widget shows AsQd vs AdKc have 25.3% equity. the correct value is 25.25%. could anyone from support admit that AsQd vs AdKc have 25.25% equity? and i will go to the step 2.

Patvs
10-25-2012, 08:49 PM
Email us an actual original handhistory; something we can import.

Widget shows 25.3 because it rounds off the numbers to only 1 decimal.

huckleberry
10-25-2012, 09:06 PM
Email us an actual original handhistory; something we can import.


done.



Widget shows 25.3 because it rounds off the numbers to only 1 decimal.

i know it. i also know that widget uses something like Monte Carlo Simulations, so widget can shows 25.3, and next time 25,2 and may be 25,1 or 25,4 if you will refresh it again and again. so if you want to know the correct value of equity you have to use HM1 ICM button (or third party software: Poker Stove, PokerStrategy Equilab, etc).

huckleberry
10-28-2012, 08:47 AM
ok. i will help you. icm button isn't active in cash replayer. but i hope you know that there is no difference between cash and tourneys if we're talking about equity. so i found a similar hand:
87901

so could anyone from support admit that AsQd vs AdKc have 25.25% equity?

PS please don't ask me about something you can import, it's not photoshop, be sure

udbrky
10-28-2012, 10:05 PM
I'm sorry to hear you're having issues with this and I apologize for the delay. I have escalated this to our 2nd level support techs and have explained the issues you are having and they will be in touch shortly to help resolve the problem.

huckleberry
10-28-2012, 10:13 PM
no, no, no. what did you escalate to 2nd level? i just want anyone from support to admit that AsQd vs AdKc have 25.25% equity. and then i will describe the problem

Patvs
10-29-2012, 02:32 PM
Confirmed, I get 25.249% in Pokerstove

huckleberry
10-29-2012, 10:04 PM
Confirmed, I get 25.249% in Pokerstove
do you know there is no way to round off 25.249% to 25.6%?

ok. lets calculate "$EV diff" and "$ usd (EV adjusted)"
88051



Results: $34.01 pot ($1.53 rake)
Final Board: 6h 4c 2c 8s 9d
Hero showed Qd As and lost (-$16.88 net)
CO showed Kc Ad and won $32.48 ($15.60 net)


total pot i can win = $32.48
i have about 25.25% equity, so $32.48*0.2525 = $8.2012 - it's "my piece" of the pot. but actually i won $0, so "$EV diff" = $8.2012. it's correct value. HM2 shows $8.32. it's wrong.
by the way $32.48*0.256 is not equal to $8.32. because 25.6% is just rounded value. if you will open evLog.txt (RVG Software\Holdem Manager\Logs) you can see the exact value. HM1 thinks AsQd vs AdKc have 25.6061% equity.

now "$ usd (EV adjusted)". the simplest way to calculate it is: "$ usd (EV adjusted)" = Net Won + $EV diff

correct value:
"$ usd (EV adjusted)" = -16.88 + 8.2012 = -8.6788.

HM2 shows -$8.56. it's wrong.

this stat (EV adjusted) one of the four main stats that are displayed on the cash graph in HM2. and it's wrong. and no one cares. i'm appalled. may be i'm too meticulous? may be EV adjusted line is not important for cash games?

finally. do you admit that HM2 shows wrong values of Equity in cash and tourneys games, that is the reason of wrong values of "$EV adjusted", evbb/100, "$EV diff" stats for cash games and the similar stats for tourneys (EV adjusted chips, evbb/100, cEV diff)?

Patvs
10-31-2012, 10:45 PM
I'll reply to this post + your email (with the AQ vs AK handhistory) when I have more time to also make the calculations.
We're already currently discussing the EV issues.


Morny already replied (via internal email):



it’s not that our calculations are wrong, it’s the method. HoldemManager has to find a balance between import speeds (making the EV calculation during the import) and EV accuracy.

We need to think through what were going to do, it’s a pretty big change and we have to decide if we either do our own more accurate lookup table or give users the option of choosing their method i.e monte carlo, full etc. resulting in slower import speeds when they want a more accurate EV calculation.

huckleberry
11-01-2012, 01:16 AM
now i get it. HoldemManager use the table with equity values and there is only the AQo vs AKo equity value in this table. so there is no difference between AsQd vs AdKc and AsQd vs AhKc for example. and if you will create all cases (AsQd vs AcKh, AsQd vs AcKd, AdQs vs AcKs, etc) instead 1 total case it will resulting in slower import. ok. but may be it will not? now HoldemManager knows difference between AQo vs AKo and AQo vs AKs (25.6% and 24.6%), so HM have to analyze the suits while importing and then find required row in the table. so if you will have full indexed table it could be very quickly to find the required value. may be faster than now, without any analysis. i know, there is no way to create full tables with equity values for situation when 3 or more players go all-in, but such hands is not very often, just calculate this hands - it will not really slow down import

i'm not sure in all this, but may be it will help you somehow

Patvs
11-01-2012, 11:42 AM
I'll forward your last posts via email to Morny.

TryingHard
11-10-2012, 06:52 PM
is the EV line for hyperturbos fixed yet?

fozzy71
11-10-2012, 10:05 PM
is the EV line for hyperturbos fixed yet?

No, not yet. Please refer to my earlier post about why it is taking so long to fix this - http://forums.holdemmanager.com/manager-general/247931-request-fix-evline-60-200-hypers-6max-3.html#post1503912

TryingHard
11-13-2012, 04:09 PM
Do we now know for sure weather the true EV line should be higher or lower?

udbrky
11-14-2012, 03:53 AM
I have escalated this to our 2nd level support techs and have explained the issues you are having and they will be in touch shortly to help resolve the problem.

nitur
11-14-2012, 09:40 AM
is the EV line for hyperturbos fixed yet?

No, it is planned for summer 2015, sorry.

Patvs
11-14-2012, 03:00 PM
Do we now know for sure weather the true EV line should be higher or lower?



From all the reports I've seen, the EV in the reports is always slightly lower than what it should be (compared to the ICM button). .
The difference is 10-30 cents in each allin hand. Assuming you have 3-5 allin hands in a single game, and you play thousands of games, the EV difference becomes quite big.

For one user his EV line dropped from +$1995 to +$1407: a 29% drop.
So the true EV line should be higher.

TryingHard
11-16-2012, 06:49 AM
Does that also mean that the EV ROI value should be aprox. 30% higher?

Patvs
11-17-2012, 12:06 AM
Yes, so if ihe ROI is 5% it should be 6.5%. (=30% higher)
Though this is an estimate based on players that play $50+ buyins with many allins.
I'm not sure how this would effect microstakes/lowstakes players.... and players, that are extremely tight that don't bully the table when they have a BIG stack, resulting in less allins.

hen
11-17-2012, 10:06 AM
Is there a version of HEM 1 that reads the $37.50 satellites properly? As in, results in HEM are correct and EV is different to winnings (but obv EV is still a little out).

I used to have one, I think it was an internal release, but I have since had to get a new laptop.

fozzy71
11-17-2012, 03:05 PM
Is there a version of HEM 1 that reads the $37.50 satellites properly? As in, results in HEM are correct and EV is different to winnings (but obv EV is still a little out).

I used to have one, I think it was an internal release, but I have since had to get a new laptop.

I am not sure off the top of my head. I am also not positive if this is a known bug so it might be a good idea to send us some hand history and tournament summary files for these games so we can have our testers make sure it is properly documented.

You can download previous official releases from our FAQ - http://faq.holdemmanager.com/questions/426/Download+Links+For+Past+Builds

There are a couple of previous internal releases in our forum announcements from Sept, Oct, and Nov - http://forums.holdemmanager.com/general-support/announcements.html

The official HM1 1.12.10b build is on our downloads page - Downloads and Manuals - Hold'Em Manager, Omaha Manager & HM Apps (http://www.holdemmanager.com/store/downloads-manuals.php)

If none of those work I can search through our downloads page for other internal builds between the official .10b and the Sept internal build but it is a little tedious so hopefully one of those builds I linked to will give you the results you are looking for.

hen
11-17-2012, 05:54 PM
I am not sure off the top of my head. I am also not positive if this is a known bug so it might be a good idea to send us some hand history and tournament summary files for these games so we can have our testers make sure it is properly documented.

OK I'll do my best. I have played only $37.50, $55.08 and $74 satellites in the sample I am importing.


There are a couple of previous internal releases in our forum announcements from Sept, Oct, and Nov - http://forums.holdemmanager.com/general-support/announcements.html

Nov = No EV line for $37.5s, $55.08s or $74s. Winnings is correct.
Oct = Shows EV and results look fine for the satellites.

I feel like I lost some games with the Oct version from today, maybe 20 games. Pretty weird, can't figure it out. I'll email PS and get them to send me all HHs + TSs to make sure.

I didn't check any others or compare with older versions.

TryingHard
12-09-2012, 04:45 PM
any news on fixing the EV line for hypers?

fozzy71
12-09-2012, 10:17 PM
any news on fixing the EV line for hypers?

HM2 is our focus right now because of the issues that were introduced in our last HM2 release. Once we do start working on HM1 again I believe we have other, more important, issues to deal with before we will have time to work on the EV issues: Pacific cash game HUD, FTP Rush HUD, FTP MTT issues, etc. Anything that affects the live gam e experience, such as the HUD or actual results, will take priority over the expected value issues.


There are some bugs in the EV calculation in both HM1 and HM2 currently depending on which site/games you are playing. Part of the holdup on the EV issues is related to the fact that we use a lookup table for EV which isn’t as accurate as full enumeration. Full enumeration seems to hurt import speeds pretty significantly, so were looking into a couple of options. We are considering a more accurate lookup table but this is a lot of work due to the combinations involved and it will be a while before the developers can implement it. Another option is giving the user the option of full enumeration or lookup tables but changing this would require a reimport and choosing the full enumeration, while it would be the most accurate, could slow import speeds significantly which would affect HM1 and HM2 both. Before we can implement this we first want to explore how accurate the new lookup table is.

huckleberry
12-10-2012, 07:19 AM
looks like you not going to fix anything in HM1. by the way not only HyperTurbo sngs have wrong EV now, even turbo 1table sngs have such issues, but they works correctly before you start to use this table


There are some bugs in the EV calculation in both HM1 and HM2
what about HM2?

by the way, don't forget - for 1table sngs you have to use table with such values: percentage of win, percentage of tie, percentage of lose. not only total Equity value (Equity = %_of_win + 1/2*_%_of_tie) like it works now

Patvs
12-12-2012, 02:10 AM
These are still open tickets.
HM1: HMONE-12, HMONE-127 (both have a medium priority status)
HM2: HM-5295 (high priority status)


All "critical priority" status bugs are being fixed first.
Once those are fixed, the HM-5295 will be addressed.




looks like you not going to fix anything in HM1.



HM1 also still has critical priority tickets, which will still be fixed. Unfortunately none of those tickets are EV-related.
They are: adding support for RUN IT TWICE tables, fixing the HUD for 888 poker, etc

TryingHard
12-19-2012, 08:39 PM
Hey, I just heard that for some versions of HM1 the EV line is working fine and for some its not. Can you say which version it works fine with so that I could upgrade back to that version? And maybe post a download link. thx

Patvs
12-19-2012, 09:41 PM
There are two HM1 versions:
versions where EV is NOT calculated at all (so EV will always match $Won)
and versions where EV is calculated (but it's wrongly calculated) (wrongly for hyper turbo's).

Best version where it is being (wrongly) calculated is: http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_03Aug2012.exe

http://www.holdemmanager.com/Downloads/HmUpdate_Release_1.12.11_10Sep2012.exe

SandmanNess
03-14-2013, 10:18 PM
Please tell me SOMETHING is in the works for this by now. This has been a problem for almost 9 months now

fozzy71
03-15-2013, 02:11 PM
Please tell me SOMETHING is in the works for this by now. This has been a problem for almost 9 months now

We hope to have it fixed in one of our next 2 official HM1 releases but I do not have any sort of ETA for how long that will take.

coeckie
03-15-2013, 04:57 PM
We hope to have it fixed in one of our next 2 official HM1 releases but I do not have any sort of ETA for how long that will take.

What about HM2?

Still in private build testing or w/e?

Thx

fozzy71
03-15-2013, 05:02 PM
What about HM2?

Still in private build testing or w/e?

Thx

I am not sure if it is completely fixed in HM2 yet (I believe we still have some open EV tickets for HM2) but it is much better than HM1 and seems to match very closely to PT4.

New Playlist(66) - fozzy71's library (http://www.screencast.com/t/rZdvLMCPgAJ)