| Pages:
1
..
12
13
14
15
16
..
98 |
Gregorian
Junior Member
Posts: 97
Registered: 23-5-2017
Member Is Offline
Mood: No Mood
|
|
Location of WF TS code
In 47.10, where do we find the WF'd code to be moved over to TS/MC? Under the Script tab, all I see is the Optimized code, not the WF'd code.
Likewise, sometimes but not all the time, after a WF run has been completed, the Walk-Forward tab under the Settings tab, and the Walk-Forward tab are
blank. It seems as though GSB doesn't always know that a WF run has been completed.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by Gregorian  | In 47.10, where do we find the WF'd code to be moved over to TS/MC? Under the Script tab, all I see is the Optimized code, not the WF'd code.
Likewise, sometimes but not all the time, after a WF run has been completed, the Walk-Forward tab under the Settings tab, and the Walk-Forward tab are
blank. It seems as though GSB doesn't always know that a WF run has been completed. |
Here it is.
|
|
|
Gregorian
Junior Member
Posts: 97
Registered: 23-5-2017
Member Is Offline
Mood: No Mood
|
|
In WF'ing around 8 strategies on three instances of GSB 47.10 thus far, I've not yet seen the - WF options appear in that drop down menu. Only the
original TS and MC options appear.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Gregorian | In WF'ing around 8 strategies on three instances of GSB 47.10 thus far, I've not yet seen the - WF options appear in that drop down menu. Only the
original TS and MC options appear.
|
try clciking on the row again.
if that doesnt work send me teamviwer details
|
|
|
zdenekt
Junior Member
Posts: 37
Registered: 1-10-2017
Location: Czech Republic
Member Is Offline
Mood: No Mood
|
|
I have a same bug all the time. No WFA code...and tried to click at least a hundred times.
No WFA code on 99% of strategies. Only sometimes I can see WFA code and when I click on different row and then click back, WFA code which was there,
is no longer there.
Perhaps it would be better to go back to the old method of displaying code in tabs to prevent such bugs.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
you can try .17. Must upgrade workers and managers. No idea if its fixed but if not will roll back WF code.
With Gregorian, it was very intermittent. Some systems work fine when you click on them, and others never work no matter how often you click on them.
.17 updates for sql server a little less often to reduce sql server load than earlier versions
Later note. Programmers says issue is fixed in this build
Attachment: Login to view the details
|
|
|
boothy
Junior Member
Posts: 54
Registered: 21-5-2018
Member Is Offline
Mood: No Mood
|
|
Hi Peter,
I have been running GSB with 3 workers for a few hours and then paused the process to start looking at the results. How long should it take for all
unique systems to load into Manager?
As you can see in attached screenshot, the workers have created about 100k unique systems, its been 45min since I paused processing and only 16k are
in the manager screen.
Cheers.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by boothy | Hi Peter,
I have been running GSB with 3 workers for a few hours and then paused the process to start looking at the results. How long should it take for all
unique systems to load into Manager?
As you can see in attached screenshot, the workers have created about 100k unique systems, its been 45min since I paused processing and only 16k are
in the manager screen.
Cheers. |
Lots of possible answers.
Today we upgraded sql, and I had to kill all workers. (sorry) but the main programmer had just gone to sleep when there was a problem.
if you build systems in the manger via a worker, and say get 500 systems. You then build 600 systems on a new manager, the new manager will show 600
systems, but the worker 1100 systems
Hope that helps, but are trying to improve the potential load of sql server for GSB. The work around was 2 minute polling which gives a lag. We want
to reduce this back to 30 seconds. Time for worker to respond is random time of 0 to polling time, + a bit of a additional delay
100k systems is likely too many. You should have tighter metrics. This will also reduce sql server load a bit too.
|
|
|
boothy
Junior Member
Posts: 54
Registered: 21-5-2018
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by admin | | Quote: Originally posted by boothy | Hi Peter,
I have been running GSB with 3 workers for a few hours and then paused the process to start looking at the results. How long should it take for all
unique systems to load into Manager?
As you can see in attached screenshot, the workers have created about 100k unique systems, its been 45min since I paused processing and only 16k are
in the manager screen.
Cheers. |
Lots of possible answers.
Today we upgraded sql, and I had to kill all workers. (sorry) but the main programmer had just gone to sleep when there was a problem.
if you build systems in the manger via a worker, and say get 500 systems. You then build 600 systems on a new manager, the new manager will show 600
systems, but the worker 1100 systems
Hope that helps, but are trying to improve the potential load of sql server for GSB. The work around was 2 minute polling which gives a lag. We want
to reduce this back to 30 seconds. Time for worker to respond is random time of 0 to polling time, + a bit of a additional delay
100k systems is likely too many. You should have tighter metrics. This will also reduce sql server load a bit too.
|
Point taken about tighter metrics, but have been using the ESsystemstight settings for starters trying to get a baseline of how many workers I can run
and not have problems.
ran another test this morning, only ran gsb for 5 to 10 min and got just over 2000 unique systems, then paused processing.
it has now been 3hrs since I paused and only 512 systems are showing in manager. I had exited out of gsb and re opened before starting new test.
I also changed in worker app settings auto share in network to false, hoping that may reduce internet upload.
Do workers running on my computer send results to the sql server and then back to manager?
If this is correct, why?
cheers.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by boothy | | Quote: Originally posted by admin | | Quote: Originally posted by boothy | Hi Peter,
I have been running GSB with 3 workers for a few hours and then paused the process to start looking at the results. How long should it take for all
unique systems to load into Manager?
As you can see in attached screenshot, the workers have created about 100k unique systems, its been 45min since I paused processing and only 16k are
in the manager screen.
Cheers. |
Lots of possible answers.
Today we upgraded sql, and I had to kill all workers. (sorry) but the main programmer had just gone to sleep when there was a problem.
if you build systems in the manger via a worker, and say get 500 systems. You then build 600 systems on a new manager, the new manager will show 600
systems, but the worker 1100 systems
Hope that helps, but are trying to improve the potential load of sql server for GSB. The work around was 2 minute polling which gives a lag. We want
to reduce this back to 30 seconds. Time for worker to respond is random time of 0 to polling time, + a bit of a additional delay
100k systems is likely too many. You should have tighter metrics. This will also reduce sql server load a bit too.
|
Point taken about tighter metrics, but have been using the ESsystemstight settings for starters trying to get a baseline of how many workers I can run
and not have problems.
ran another test this morning, only ran gsb for 5 to 10 min and got just over 2000 unique systems, then paused processing.
it has now been 3hrs since I paused and only 512 systems are showing in manager. I had exited out of gsb and re opened before starting new test.
I also changed in worker app settings auto share in network to false, hoping that may reduce internet upload.
Do workers running on my computer send results to the sql server and then back to manager?
If this is correct, why?
cheers. |
One factor is, some systems might be duplicates, but I doubt that is the cause.
We are working on have a diagnostic duplicate system count.
changing autoshare in network wont change anything. All workers talk via sql server.
We are doing a lot of work in this area right now, included stress testing it.
do you have any exception files in the manager or worker folders in the exception folders?
I am going to talk to the lead programmer over this issue.
GSB used to share info via shared dropbox folder, but this was very problematic.
Big picture is we will resolve the issue, and cant support shared folder and sql server. Sorry for the loss of some systems. Issue needs to be fixed.
|
|
|
boothy
Junior Member
Posts: 54
Registered: 21-5-2018
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by admin | | Quote: Originally posted by boothy | | Quote: Originally posted by admin | | Quote: Originally posted by boothy | Hi Peter,
I have been running GSB with 3 workers for a few hours and then paused the process to start looking at the results. How long should it take for all
unique systems to load into Manager?
As you can see in attached screenshot, the workers have created about 100k unique systems, its been 45min since I paused processing and only 16k are
in the manager screen.
Cheers. |
Lots of possible answers.
Today we upgraded sql, and I had to kill all workers. (sorry) but the main programmer had just gone to sleep when there was a problem.
if you build systems in the manger via a worker, and say get 500 systems. You then build 600 systems on a new manager, the new manager will show 600
systems, but the worker 1100 systems
Hope that helps, but are trying to improve the potential load of sql server for GSB. The work around was 2 minute polling which gives a lag. We want
to reduce this back to 30 seconds. Time for worker to respond is random time of 0 to polling time, + a bit of a additional delay
100k systems is likely too many. You should have tighter metrics. This will also reduce sql server load a bit too.
|
Point taken about tighter metrics, but have been using the ESsystemstight settings for starters trying to get a baseline of how many workers I can run
and not have problems.
ran another test this morning, only ran gsb for 5 to 10 min and got just over 2000 unique systems, then paused processing.
it has now been 3hrs since I paused and only 512 systems are showing in manager. I had exited out of gsb and re opened before starting new test.
I also changed in worker app settings auto share in network to false, hoping that may reduce internet upload.
Do workers running on my computer send results to the sql server and then back to manager?
If this is correct, why?
cheers. |
One factor is, some systems might be duplicates, but I doubt that is the cause.
We are working on have a diagnostic duplicate system count.
changing autoshare in network wont change anything. All workers talk via sql server.
We are doing a lot of work in this area right now, included stress testing it.
do you have any exception files in the manager or worker folders in the exception folders?
I am going to talk to the lead programmer over this issue.
GSB used to share info via shared dropbox folder, but this was very problematic.
Big picture is we will resolve the issue, and cant support shared folder and sql server. Sorry for the loss of some systems. Issue needs to be fixed.
|
No exception files from this mornings test, and only a couple from yesterday when I was processing for a few hours. (related to not being able to
contact server and timing out)
I can understand why cloud workers would have to talk via sql server, but why do workers on my computer have to talk to the manager on my computer via
the server?
I'm not very techy, but to me that seams to be a lot of info flying around over the internet for no reason.
Am I the only one with this issue??
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Im not sure of others have the issue, but there is a lot work going on in GSB data base at the moment. Lots more programing work to use sql and to use
communciation from workers to managers. Will let you know when I hear from the programer about the issue
|
|
|
Gregorian
Junior Member
Posts: 97
Registered: 23-5-2017
Member Is Offline
Mood: No Mood
|
|
Missing slippage and commission fields
In 47.40, even with GUI-Advanced set to True, the Fitness and Report Slippage and Commission fields do not display. Have they been moved elsewhere?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Gregorian | | In 47.40, even with GUI-Advanced set to True, the Fitness and Report Slippage and Commission fields do not display. Have they been moved elsewhere?
|
I removed them as many users use different settings for reports and fitness. This slows gsb down a fair bit. Adding slippage and commission in fitness
likely will do nothing if fitness = np*at
If you strongly want it back, I will add if for gsb in admin mode only
|
|
|
Gregorian
Junior Member
Posts: 97
Registered: 23-5-2017
Member Is Offline
Mood: No Mood
|
|
I do not strongly want it back. It was a convenience, a feature long present in TSL, AT, and SQ, but not, strictly speaking, necessary.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Gregorian | | I do not strongly want it back. It was a convenience, a feature long present in TSL, AT, and SQ, but not, strictly speaking, necessary.
|
Try with GSB in admin mode.
See this post.
http://trademaid.info/forum/viewthread.php?tid=92
If that doesnt fix it and you want it back, I will add it in admin mode only.
Overall GSB uses are better without it. I think it was 15% or so degradation in speed if slippage reports<>slippage fitness
|
|
|
Carl
Member
Posts: 342
Registered: 10-5-2017
Member Is Offline
Mood: No Mood
|
|
Hi Peter,
I do like the possibility to be able to add the slippage figures in GSB.
So if it can be re-introduced, please do.
Thanks.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Carl | Hi Peter,
I do like the possibility to be able to add the slippage figures in GSB.
So if it can be re-introduced, please do.
Thanks. |
Minor bug, yes s&c for reports and fitness is gone. 
Its supposed to be gone for fitness only.
It will be back soon.
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
Hi Peter, your email keeps getting bounced back!
The response was:
550 Unknown user
Final-Recipient: rfc822; info@trademaid.info
Action: failed
Status: 5.0.0
Remote-MTA: dns; mail.trademaid.info. (162.218.228.242, the server for the
domain trademaid.info.)
Diagnostic-Code: smtp; 550 Unknown user
Last-Attempt-Date: Wed, 25 Jul 2018 03:24:01 -0700 (PDT)
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by TradingRails |
Hi Peter, your email keeps getting bounced back!
The response was:
550 Unknown user
Final-Recipient: rfc822; info@trademaid.info
Action: failed
Status: 5.0.0
Remote-MTA: dns; mail.trademaid.info. (162.218.228.242, the server for the
domain trademaid.info.)
Diagnostic-Code: smtp; 550 Unknown user
Last-Attempt-Date: Wed, 25 Jul 2018 03:24:01 -0700 (PDT)
|
Thanks for that, as I didnt know. Ive been reviving the email on this domain just recently
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
Hi Peter,
I appear to have corrupted the data file structure in some way, as I go to check the data and find that this files appear to be missing in the shared
folder. I then attempt to use the one in the data file under the C:\GSB\GSB (Managers)\Price Data and I'm instructed to use the shared dropbox folder.
What have I done?
I hope the attached screenshot helps
Thanks.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by TradingRails |
Hi Peter,
I appear to have corrupted the data file structure in some way, as I go to check the data and find that this files appear to be missing in the shared
folder. I then attempt to use the one in the data file under the C:\GSB\GSB (Managers)\Price Data and I'm instructed to use the shared dropbox folder.
What have I done?
I hope the attached screenshot helps
Thanks.
|
go to app settings data and set public data path here
C:\GSB\GSB (Managers)
Worse case re-install should over write any config files.
Let me know if it works.
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
Hi Peter,
I couldn't send via the help as the file format isn't supported, so I hope via the forum works for you.
Thanks.
Attachment: Login to view the details
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
I got the file thanks. zip it up likely would have worked.
Will be looking at what you sent
|
|
|
eatrade
Junior Member
Posts: 2
Registered: 23-7-2018
Member Is Offline
Mood: No Mood
|
|
I just installed the trial file GSB 47.44 build. July 24th 2018 and am surprise,d that the duration ist set to only 4 days left from the beginning. I
thought that the trial period is 14 days. Is this a bug or indeed only 4 days to try the software?
Thank you very much in advance,
|
|
|
| Pages:
1
..
12
13
14
15
16
..
98 |
![[*]](images/davis/default_icon.gif){kind=icon}
