News: 5th digit in quotes - page 5

 
Mathemat писал(а) >>
Why? As it was 15,000 ticks per day, it will remain so.

The DC filter is a PID regulator that moves the quote from the volume of flows in bilateral constraints.
If the digit capacity (sampling) of the regulator is +1, then the number of acts of regulation is also x10.
P.S.
PID controller is Proportional Integro Differential.
there is also
PIDD
PID
PIDD regulators

 
Mathemat писал(а) >>
>> Why? As it was 15 thousand ticks a day, it will remain so.

Compare how many there are with how many there were. Even visually, you can see that the number of ticks has increased. They have become more accurate in their quotes. The old tick is now +-10. And there are no such jumps, it goes smoother.

 

Yeah, thanks for the face on the table. It's actually too early to judge an unambiguous increase in volumes: the volume probability distribution is also very thick-tailed. But that seems to be the case. Hardly an order of magnitude more, but the increase in sampling has played its part.

P.S. Tick registers when the price changes. It seems that before changes much smaller than 0.0001 were just skipped, while now they are registered.

Well, it's a damn good care of the poor trader (write in red and bold for everyone to see): there will be less mysterious requote and old quote now! Isn't it so, colleagues? And yet, and yet: who benefits from it?

 

OK, so a lot of people see the five digits as a boon...

On the one hand: the coding

but what about the handhelds???

*

From my point of view, well, no change other than the fact that previously written about.

On the 4x: opened position at 1.2502, closed it at 1.2555.

On 5x: opened a position at 1.25015, closed it at 1.25555

So? Once I got half a pips hot, the other time I got the same amount...

* that's if you take into account the fact that these operations were performed on both types of accounts at the same time.

The benefits are 0, and the inconvenience of "reading" the last three digits, with the last one discarded

to understand the real and familiar price level of 1.2555 is incomparable...

*

Stops again, talk about losses in pips in NON-denominated,

about the horror of the 10,000 spread on the quid and other exotics... :)))))))))))))))))))))))

*

As for "smoothness", YES! there is more jerking, but not as poppy as one might

As for the "smoothness", YES, there's more twitching, but not as pippy as one might have originally thought... but 2-3-5 pips of the latter...

And the "double", i.e. jumps of at least 2 pips are much rarer...

*

EURUSD;2009.01.26;02:00:56;1.29235; diff: 0.00006; spr: 19
EURUSD;2009.01.26;02:00:56;1.29231; diff: 0.00004; spr: 38
EURUSD;2009.01.26;02:00:56;1.29241; diff: 0.00010; spr: 28
EURUSD;2009.01.26;02:01:00;1.29239; diff: 0.00002; spr: 22
EURUSD;2009.01.26;02:01:02;1.29231; diff: 0.00008; spr: 28
EURUSD;2009.01.26;02:01:08;1.29229; diff: 0.00002; spr: 22
EURUSD;2009.01.26;02:01:15;1.29235; diff: 0.00006; spr: 19
EURUSD;2009.01.26;02:01:17;1.29245; diff: 0.00010; spr: 19
EURUSD;2009.01.26;02:01:18;1.29241; diff: 0.00004; spr: 39
EURUSD;2009.01.26;02:01:18;1.29232; diff: 0.00009; spr: 48
EURUSD;2009.01.26;02:01:18;1.29224; diff: 0.00008; spr: 22
EURUSD;2009.01.26;02:01:19;1.29251; diff: 0.00027; spr: 19

It is supposed to be possible to collect statistics for example for a day and calculate

arithmetic mean value of a tick in pips...

For example in the above stump 96/12=8 pips (0.8 standard)

 
StSpirit >> :

The volumes of tick data do increase, I don't know, I haven't checked them specifically, but testing of Expert Advisors using all ticks will definitely become slower. :))

Haven't tried testing yet, but it seems to me that the testing should go at the same speed, i.e. it shouldn't slow down.

After all there is no ticking history. The tester emulates ticks as before in minutes. And here, whether it is four decimal places or ten, the tick emulation algorithm in MT4 is the same and the speed of testing should be the same.

It would be interesting to hear the developer's comments.

 
kombat >> :

EURUSD;2009.01.26;02:00:56;1.29235; diff: 0.00006; spr: 19
EURUSD;2009.01.26;02:00:56;1.29231; diff: 0.00004; spr: 38
EURUSD;2009.01.26;02:00:56;1.29241; diff: 0.00010; spr: 28
EURUSD;2009.01.26;02:01:00;1.29239; diff: 0.00002; spr: 22
EURUSD;2009.01.26;02:01:02;1.29231; diff: 0.00008; spr: 28
EURUSD;2009.01.26;02:01:08;1.29229; diff: 0.00002; spr: 22
EURUSD;2009.01.26;02:01:15;1.29235;diff: 0.00006; spr: 19
EURUSD;2009.01.26;02:01:17;1.29245; diff: 0.00010; spr: 19
EURUSD;2009.01.26;02:01:18;1.29241; diff: 0.00004; spr: 39
EURUSD;2009.01.26;02:01:18;1.29232; diff: 0.00009; spr: 48
EURUSD;2009.01.26;02:01:18;1.29224; diff: 0.00008; spr: 22
EURUSD;2009.01.26;02:01:19;1.29251; diff: 0.00027; spr: 19

The idea is to gather statistics for example for a day and calculate

average tick value in pips...

For example, in this example 96/12=8 pips (0.8 standard)

Now think about the accuracy of the test.... :(

 

Who can tell me where this fifth sign came from?

 
Integer >> :

Who can tell me where that fifth sign came from?

I don't think anyone can say for sure. I think it was a marketing ploy.

 

It's all a waste of time, it's just fooling people. The four digits sometimes make our eyes glaze over, and five makes the number of mistakes when setting the order price several times larger.

Moreover, with the current volatility it is time to reduce the number of decimal places, not to increase them.

 
BigeR >> :
the number of order price errors may well increase manifold...

This is what is being sought.