MT5 and speed in action - page 15

 
Right, got it, thanks, it was a lame question. Something stuck in my head.
 
fxsaber:
b2579.

What is the CPU load when running this test?

 
Anton:

What is the CPU load when running this test?

0-2%, taking into account other tasks still running. Not reproducible?

 
fxsaber:

0-2%, taking into account other tasks still running. Not playing?

Closed all applications, leaving only Terminal with three charts. Saw that peaks were occurring.

It was up to 20%.

 
fxsaber:

Closed all applications, leaving only the Terminal with three charts. I saw that there were peaks.

It was as high as 20%.

I have a feeling that my lags are due to insufficient CPU power (i5-4670). I ordered 12 core Xeon (e5-2678v3) from Chinese, in a couple of weeks I'll rebuild my server and compare average latency and report my results.

 
Dmi3:

I have a feeling that my lags are due to insufficient CPU power (i5-4670). Ordered a 12 core Xeon (e5-2678v3) from the Chinese, will arrive in a couple of weeks, rebuild the server, be able to compare average execution delays, will write about the results.

The latency is due to excessive CPU load on the part of MT5. I'm not sure if multi-core will help me here. It will be interesting to see the results.

 
fxsaber:

The latency is due to excessive CPU load on the part of the MT5. Not sure if multi-core will help here. It will be interesting to see the results.

And I'm not sure, but it's the easiest thing I can do. And in principle it was about time for an upgrade.

 
Dmi3:

I have a feeling that my lags are due to insufficient CPU power (i5-4670). I ordered 12 core Xeon (e5-2678v3) from Chinese, in a couple of weeks I'll rebuild the server, compare average latency of execution, I'll write about results.

They are both Haswell, Xeon has much lower operating frequency, there will be performance degradation in runtime and single tests, only in multi-threaded optimization will be an advantage. The i3 of the latest models should be much faster to run

ask the developers about cache level effect on speed and in general the speed of Zen2 and latest intel


add

Ryzen 3700x I have, you can do tests with Intel

for example with this MQL5\Scripts\UnitTests\Stat\TestStatBenchmark.mq5 script

loop it several times with a timer

 

To reproduce the brakes, you need to run the script on multiple chars of the ONE character - get OnTick to be called at the same time. Then, the alerts will be pinged on every tick.

The CPU load graph shows that terminal64.exe loads up to 30% of eight logical cores. That's only four EURUSD charts with the script running. You can clearly see how much each chart is loaded at a time.

Where are all these resources going?

2020.08.27 12:57:19.299 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 36 ms.
2020.08.27 12:57:19.300 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 33 ms.
2020.08.27 12:57:19.314 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 5 ms.
2020.08.27 12:57:20.073 Alert: Time[Test6.mq5 22: HistorySelect(Tick.time,INT_MAX)] = 1 ms.
2020.08.27 12:57:20.080 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 1 ms.
2020.08.27 12:57:24.742 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 4 ms.
2020.08.27 12:57:24.748 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 6 ms.
2020.08.27 12:57:24.749 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 2 ms.
2020.08.27 12:57:27.836 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 9 ms.
2020.08.27 12:57:33.275 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 6 ms.
2020.08.27 12:57:33.292 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 2 ms.
2020.08.27 12:57:33.292 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 16 ms.
2020.08.27 12:57:33.306 Alert: Time[Test6.mq5 17: CopyTicks(Symb,Ticks,COPY_TICKS_ALL,0,1)] = 3 ms.
 

Setting a maximum priority doesn't help, of course.


Tests were done on b2582.

It would be interesting to compare the same script with other trading platforms.
Reason: