OpenCL: internal implementation tests in MQL5 - page 41
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
Try running the test (see my post above). Is the screen frozen, is the mouse cursor moving?
lags (I mean screen and cursor)
Yes, exactly. I used to have one from AMD (very decent one, by the way), but decided to experiment.
But now I've pulled the Intel one for the same reason.
The mouse did not slow down, but the window "attach to graphics" stuck in a translucent form.
Joo you rascal, by the way, who calls it a Test Tast, I shook half the computer in search of the right file. You're a kidder.
Joo, by the way, you're an idiot, who calls it Test Tast, I shook half the computer looking for the right file. You're a funny guy.
Tast_Mand_ (EURUSD,H1) 22:07:32 4758 msec
i7-2700k + 590GTX
The mouse wasn't slowing down, but the "attach to graphic" window was stuck in a semi-transparent form.
->
Didn't notice the strain on the video card, it runs and clicks. Opened dispatcher during test, also all ok, no lags, only test time was 17 seconds.
Why is your result better? - Not bad for a budget card, by the way.
And even better than Wladimir's:
lags (I mean screen and cursor)
Urain:
Joo you rude bastard, who calls it a Test-test, I shook half the computer in search of the right file. You're a kidder.
Thanks for the laugh. :)
I didn't notice until after you told me.
Here, the driver hangs, and it doesn't work the same trick as in your test. Why not? Obviously, it has nothing to do with the graphics card memory.
Also. After the video card driver crashes (if the test fails) and resumes on its own, the test does not start again (it starts but terminates with an error). Apparently, the terminal continues to hold on to the vis-a-vis resource and won't let go until you restart the terminal.
->
1. Why are your results better? - Not bad for a budget card, by the way.
And even better than Vladimir's:
2. When video card driver crashes (if the test fails) and resumes on its own, the test after that again does not start (it starts but fails). Apparently, the terminal continues to hold on to the vis-a-vis resource and won't let go until you restart the terminal.
1. I deliberately repeated it with "more threads", I wonder what is the maximum number of threads you have with him (Nikolay).
2. probably something like this. I hope the signals are not beeping for nothing. Let them sort it out upstairs already. :)
Most likely yes. I've done tests with other codes, so on first run test takes ~1700ms, re-run ~120ms.
Apparently code hangs in processes and repeat code doesn't need to be initialized.
So that's the pie:
The driver never crashed.
During the "last" test, the table started to stick, although the terminal was showing new ticks without sticking.
It feels like 607 build selects faster device: 3% video load, 100% CPU load.