Public discussion of the formula for calculating the cost of resources in the MQL5 Cloud Network - page 15
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
Okay, I don't get it. I thought they were talking about agents being able to count traffic. Where shall we look?
You will be able to watch directly in your profile on the website + we will add detailed daily reports on what you've earned.
I think that in a week the site will start displaying agent statistics.
You will be able to look directly in your profile on the website + we will add detailed daily reports on earnings.
I think that in a week's time, agents' statistics will start appearing on the website.
for 2 cents an hour per core ?
0,02×24×30×34≈ $490 per month
i don't know, i think the market will judge
I'm not ready to talk about a price yet!
I'm just building up the cores in the pool for now... and if it's profitable why not stay in the pool
I have 36 cores at the moment
I could use the passive income.
especially since all i have to do is keep the machine running
and i have a fairly strong server, i even offer it as a VPS. and the prices are lower than on the market
--
If it's no secret, what kind of devices are on the atoms, and how do they work when (fully) loaded by an agent?
I mean, do they allow you to do anything else on them, or do they slow down (connected only for agents and/or advisors)?
If it's no secret, on atoms, what kind of devices, and how do they work when (fully) loaded by an agent?
I mean, allow them to do something else or they slow down (connected only for agents and/or advisers)?
There are several atoms included in the pool ( there are some more atoms on the first page )
I like the asus 1201N, it's one of the best 12" devices (as reviewed on the net)
there is also a samsung 12 inches not bad either and a very small MSI 8 inches
well let's just say that under a load of even 2 or 4 cores, it is quite possible to work
They usually use these machines in internet cafes - students in the kitchens,
let's just say it's mostly to get on the internet,
these machines aren't usually used for applications where you need strong calculations
and they don't use super programs. They don't do complex processing.
i.e. it's ideal to use them for AGENT
since most of the time the processor on these little guys is bamboozled.
I see, thank you.
I am generally aware of the students. I'm interested specifically in device operation with agents + MT5 with advisors/indicators at peak load with agents.
Running agents on desktop 2-core, noticed that at quite a long load under 50 per core (in task manager, 0% system inactivity) even opera started suggesting to enable turbo, and one of the terminals started beeping that it loses/finds connection. A movie, for example, running at such a moment would go jerky. Seen this a few times during the week while at the computer. Will a more/less sophisticated EA with indicators handle such a situation, or will it pass the ticks just without having enough time to process them? If the connection is stable, of course. In particular, in a mobile device. Apparently the priority does not solve this, it is low as it is.
By next year it will definitely be relevant :)
I see, thank you.
I am generally aware of the students. I'm interested specifically in device operation with agents + MT5 with advisors/indicators at peak load with agents.
Running agents on desktop 2-core, noticed that at quite a long load under 50 per core (in task manager, 0% system inactivity) even opera started suggesting to enable turbo, and one of the terminals started beeping that it loses/finds connection. A movie, for example, running at such a moment would go jerky. Seen this a few times during the week while at the computer. Will a more/less sophisticated EA with indicators handle such a situation, or will it pass the ticks just without having enough time to process them? If the connection is stable, of course. In particular, in a mobile device. Apparently the priority does not solve this, it is low as it is.
By next year it will definitely be relevant :)
You could probably try to introduce a timetable here
If you actively use the car - say, watch a movie - well, turn off the agent for that time
I haven't noticed any disconnects.
The second option is for example not to run on two cores, run on one core!
You could probably try a timetable here
If the machine is in active use - say, watching a film - well, disable the agent for that time.
I haven't noticed that the connection is down.
The second option is for example not to run on two cores, run on one core!
It is not only and not so much a question of how else and whether the computer will be used at all. Look at the situation through the trader's eyes.
You have a single computer running on a real account. Would you risk connecting agents to it without being sure that they won't fail the system by making the Expert Advisor crash? And for mobile devices, the issue could be even more pressing.
And scheduling is not an option here. It could be used if I knew when the cloud would load my cores. And as it is, running the EA on a computer with agents, we run the risk of failures (or rather, unprocessed ticks) in its operation.
I cited the film as an example where you can really see the system slowing down.
On communication - no full certainty yet. Could be a coincidence, or the ISP "guessed", I'll follow up more in the week.
Some of the cores... well, maybe. On the other hand, the dispatcher seems to show good load distribution. But if they limited the total load of cores, or gave a choice of, say, not less than 50% and up to 100 - it would be great. Only the developers will again say that they are against extra efforts of users :)
Well, we'll see.
PS yes, and I still don't know how to find out if the EA is "slow".
Silent:
There is a single computer running on real. Would you risk plugging in agents on it without being sure that they won't let the system down by causing the EA to crash? And for mobile devices, the issue could be even more pressing.
And scheduling is not an option here. It could be used if I knew when the cloud would load my cores. As it is, running the EA on a computer with agents, we run the risk of failures (or rather, probably, raw ticks) in its operation.
I cited the movie as an example where you can really see the system slowing down.
On communication - no full certainty yet. Could be a coincidence, or the ISP "guessed", I'll follow up more in the week.
Some of the cores... well, maybe. On the other hand, the dispatcher seems to show good load distribution. But if they limited the total load of cores, or gave a choice of, say, not less than 50% and up to 100 - it would be great. Only the developers will again say that they are against extra efforts of users :)
Well, we'll see.
PS yes, and I don't yet know a way to know if an EA has "slowed down".
1. In my opinion, few people in their right mind would risk the com where the trade is done on the real account to hang something else (if we are talking about trading time). The other thing is outside working hours, then you can do anything and any way you want on your computer.
2. It depends on how you run it and what kind of computer it is. In such a case, it would be better to use half of the computer's cores or less. Scheduling can also help, because you can set a time when agents will not run at all (for example, on Monday - usually a slow day - you can use agents to the maximum, and on Wednesday and Friday it is desirable to switch them off).
3. about communication in any case it is necessary to have two options (or even three). You should also have a backup for electricity in case there are problems. 4.
4. It is not only about cores, there are enough bottlenecks there even without them. For example RAM and disks, the same Internet.