Did I miss something ? It's very quite here or in other words here is no activity visible ?
So - what do you have in mind posting about "pretty damn impressive" things ?
AGT » Wed Feb 18, 2015 10:53 am wrote:
Did I miss something ? It's very quite here or in other words here is no activity visible ?
So - what do you have in mind posting about "pretty damn impressive" things ?
Yes it has been quiet, just waiting till the bugs get ironed out of HGB before starting a long term forward test.
Sorry, just meant all the persons who agree to forward test this in a structured and systematic method. I cant say that I have ever seen this before from so many different persons and do find it (the forward testing, not the results yet) impressive that so many have collaborated.
Hey guys,
from here http://www.myfxbook.com/members/HGBforwardtest2
we can conclude that HGB runs worse in VPS than on PCs - in my personal tests was the same with twice the trades number - and I think it is important to understand why : in order to improve the implementation of HGI in an automatic system.
almartins » Tue Feb 24, 2015 7:52 am wrote:Hey guys,
from here http://www.myfxbook.com/members/HGBforwardtest2
we can conclude that HGB runs worse in VPS than on PCs - in my personal tests was the same with twice the trades number - and I think it is important to understand why : in order to improve the implementation of HGI in an automatic system.
I noticed that in my VPS account I get a message that the indicator is too slow and needs to be re-programmed...
almartins » Tue Feb 24, 2015 1:52 pm wrote:Hey guys,
from here http://www.myfxbook.com/members/HGBforwardtest2
we can conclude that HGB runs worse in VPS than on PCs - in my personal tests was the same with twice the trades number - and I think it is important to understand why : in order to improve the implementation of HGI in an automatic system.
normally VPS have reduced prestations(often only 1GB RAM disponible ecc), that imo will cause this with a relative resource hungry EA like HGB
almartins » Tue Feb 24, 2015 7:52 am wrote:Hey guys,
from here http://www.myfxbook.com/members/HGBforwardtest2
we can conclude that HGB runs worse in VPS than on PCs - in my personal tests was the same with twice the trades number - and I think it is important to understand why : in order to improve the implementation of HGI in an automatic system.
I noticed that in my VPS account I get a message that the indicator is too slow and needs to be re-programmed...
no indicator problem, but problem of the vps computer..., using the hgi lib instead of the iCustom call for the indicator would probably resolve this problems.., maybe Steve will include this in one of the next updates..
almartins » Tue Feb 24, 2015 1:52 pm wrote:Hey guys,
from here http://www.myfxbook.com/members/HGBforwardtest2
we can conclude that HGB runs worse in VPS than on PCs - in my personal tests was the same with twice the trades number - and I think it is important to understand why : in order to improve the implementation of HGI in an automatic system.
normally VPS have reduced prestations(often only 1GB RAM disponible ecc), that imo will cause this with a relative resource hungry EA like HGB
Cheers
Tommaso
Hi Tommaso,
I think it's not really the explanation, then why in VPS occur twice the trades? In my case the VPS system has 2GB and the CPU rarely reaches 30% and only a Empty4 platform running .
Perhaps the greatest frequency of ticks in VPS, give rise to the occurrence of more signs which in PCs do not occur. :
I think it's not really the explanation, then why in VPS occur twice the trades? In my case the VPS system has 2GB and the CPU rarely reaches 30% and only a Empty4 platform running .
Perhaps the greatest frequency of ticks in VPS, give rise to the occurrence of more signs which in PCs do not occur. :
Cheers
Almartins
I agree with that, my VPS never used more than 50% of CPU. Now that the code for opening one trade per bar and closing on opposite wavies has been corrected, perhaps we can run another test with more similar results.
Click here to go to the BoardKnowledgeBase