![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Now I've looked at the thread - mum's the word - 68 pages. And they still haven't decided what to test and how to do it! They've gone into all sorts of things not interesting for a trader - CPU architecture, specifics of memory usage, etc. Personally, I like the result we have now.
Personally, I like the result we have now.
There is no definite leader.
If you have any other additions to what I have written about the Expert Advisor with opening on each 15-minute bar and calling indicators - welcome! Or you have radically different ideas.
But, IMHO, the answer to the question "with what to optimize - Intel or AMD" will not change.
Personally, I like the result we have now.
There is no clear leader.
If you have any other additions to what I have written about the Expert Advisor with opening on every 15-minute bar and calling indicators - welcome! Or you have radically different ideas.
But, IMHO, the answer to the question "whether to optimize by Intel or AMD" will not change.
I don't see any difference whatsoever.
to gain 10-15% is nothing
If the difference is several times, then it's different.
on the one hand what to take amd or intel
I've always preferred Intel myself
--
on an old laptop 15:49 is already more substantial wait 3 times longer of course does not want to
but then again 15 min and 6 min is nothing - you can wait
by the way, sometimes good thoughts come out of the pause
and 15 days and 6 days is essential
--
in general I don't care what to test on
The i7 920 has Turbo Boost. Well even if the frequency is the same as begemot61's (3.47), still the comparison is clearly in favour of the i7 920. Maybe that's the beauty of virtualization?
2 Docent: there is very likely a different, real frequency, higher than 2.66. It's just that the virtual app doesn't show it.
The 920 and my Xeon are twins. The turbo boost on the younger models is indeed more aggressive, but not to that extent. It can give 10-15% on one core if the others are not very busy.
The difference between Bloomfield and Xeon is that Bloomfield usually allows overclocking. Although about 920 and overclocking - not sure, maybe only in echtrims. Also the 920 has a slightly slower QPI link. But we shouldn't care about that.
Interesting results. In general, Jura's config is competent. That's exactly what 9 series should do better than 8 and 7.
......Aha, while writing I saw the new results. Seems correct to me now.
Personally, I like the result we have now.
There is no clear leader.
If you have any other additions to what I have written about the Expert Advisor with opening on every 15-minute bar and calling indicators - welcome! Or you have radically different ideas.
But, IMHO, the answer to the question "with what to optimize - Intel or AMD" will not change.
Yes, I agree. ))) There is already something to be understood. But we cannot turn it into a branch or an article with a title something like: Test your computer in MT4.
I support the idea of 15 min bars. Perhaps that would be optimal. We can make a check on currency pair, correct start deposit, start and end date of testing with error output via alert (print may not be seen). That's my suggestion/addition.
I don't see much difference in what you optimise with
It's no big deal to win 10-15%.
If the difference is several times, then it's different.
on the one hand what to take amd or intel
myself always preferred intel
Agreed. I.e. a multi-threaded optimiser in the tester - and a gain of 2-3 times
By the way. Even if I continue to use 4, I will be testing on 5. I will rewrite indicators and EAs for it anyway.
MQL5 itself gives at least x2 speed plus multithreading. If it were written on the OpenCL platform with distributed computing... Vaughn, the new Radeon has 1600 floating-point ALUs. One run, one ALU. That's 1600 times faster. And in optimisation, it is possible to get a linear dependence on the number of computational cores due to the algorithm perfection for paralleling.
Agreed. I.e. multi-threaded optimizer in tester -- and gain 2-3 times
once tried to write my own tester - that was the effect in speed
it was designed for a specific strategy, i.e. it is not functional
didn't do all the necessary checks
C++ console application
input is simply csv
--
I know a lot of people write too
here's an open source project - it's not at the tinker level anymore
http://gordago.ru/opensource/forex-optimizer/
gorgado there's a terminal and a tester with sources - put some svn and get access to the sources
Programmer:
Forex Optimizer has recently become an Open Source project.
You can get the source code of the current stable version 2.7 via subversion:
svn checkout http://gordago.googlecode.com/svn/tags/2.7/ gordago-read-only
You can get the 2.8 version which is in deep development here:
svn checkout http://gordago.googlecode.com/svn/trunk/ gordago-read-only
Now I do! :)
I've put the result in the table.
may be added for frequency 3.8
didn't overclock the 3.8 stone too much (didn't go higher than 4)
228 * 3.8 = 866.4.
Wonderland: the efficiency increases with frequency and is now considerably higher than the W5590. Can anyone explain it? Yura, didn't you mess up something else there?
228 * 3.8 = 866.4.
Wonderland: the efficiency increases with frequency and is now considerably higher than W5590. Can anyone explain it? Yura, didn't you mess up something else there?
Yes, it's interesting. And it's not very clear to me yet.
How was the overclocking procedure carried out?
What is inconvenient is that the test is being run under a non native operating system and it's impossible to view CPU-Z with full information about the processor and memory.
The only thing is that calculation of t*f for Core ich makes no sense since only base frequency is taken into account while turbo boost is not. That said, it's quite possible that the memory is also overclocked (to DDR3-1600 type mode) and the Core i7-920 @3.8 + turbo boost will run faster than the Xeon @3.47 + turbo boost but with slower DDR2-1333.
The margin of error is not that big, and is probably determined by this.