Crazy cache of testing agents - page 2
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
What is an amazing implementation?
Run a task in mcl5 cloud and be amazed that you can get the first hundreds of results in 100-300ms after the start button is pressed. And behind the scenes there is a huge task of transferring tasks, synchronization of historical data (including ticks), distribution of all this to thousands of computers around the world, etc.
None of our competitors (nor boinc) have created such a system, not to mention its fabulously low latency.
About gigabytes.
When people were clamouring for ticks and thought they were a strategist, we refused to introduce them for a long time because they really took up gigabytes (think of a lot of characters, not just one). We warned that it required resources on the trader side.
When the time came, the iron pulled up and the channels got wider, we turned on the ticks. And we spent a lot of time rewriting the packaging several times while transmitting over the network. Without this, millions of ticks could not be transmitted.
On the terminal side these deep minute bars and ticks of course have to be unwrapped, in order not to waste time on constant unpacking in the memory. And it's important that in MT5, the trader doesn't have to work and understand the history - everything is downloaded automatically and transparently.
The trader just "opens a few windows", and a detailed picture of the market is built in the background.
When the trader runs optimization on a dozen cores, he/she has to provide ten non-blocking copies of the market environment, with tick details and multi-symbol synchronization. The key here is exactly the non-blocking data, because without it it will be impossible to quickly scroll through the market changes. You will be stuck in the locks.
That's why depending on the task, multisymbols, etc. each trader has a huge amount of data in his/her tester. The more data, the less a trader understands his or her task.
Hence there is another conclusion - no 32-bit versions are out of the question. There is no more talk about using 32-bit versions.
The tester, as well as the terminal, independently monitors logs, files and caches. As soon as it becomes clear that the voltage has dropped, the disks are cleared. And this is a very good and tried-and-tested technique.
What is an amazing implementation?
Run a task in mcl5 cloud and be amazed that you can get the first hundreds of results in 100-300ms after the start button is pressed. And behind the scenes there is a huge task of transferring tasks, synchronization of historical data (including ticks), distribution of all this to thousands of computers around the world, etc.
None of our competitors (and boinc) have created such a system, not to mention its fabulously low latency.
I've heard a lot about Cloud - it's probably the only advantage compared to MT4 for me, but it's a hypothetical advantage. The thing is, it's not clear how much it costs to be serviced on this network - the formulas are not clear...
As part of advertising could make a table, for example a month PC with an AMD Phenom II x6 1090T processor will cost 15$ (150$?) and will take 15 minutes - to be clear how much I would have to pay for the result, which I can get in a month - to understand how much you charge for accelerated information.
I've heard a lot about Cloud - it's probably the only advantage for me compared to MT4, but the advantage is hypothetical. The point is that it is not clear how much it costs to be serviced on this network - the formulas are not clear...
As part of advertising could make a table, for example a month of PC with AMD Phenom II x6 1090T processor will cost 15$ (150$?) and will take 15 minutes - to be clear how much I will have to pay for the result, which I can get in a month - to understand how much you charge for expedited information.
The formula is in the faq, and the cost is obscenely low compared to the value of the savings achieved.
Give figures that are understandable to the consumer - my proposed representation of the benefits of the service will get potential customers interested. As it is, it's scary to use it...
Give figures that are understandable to the consumer - my proposed representation of the benefits of the service will get potential customers interested. Otherwise, it's scary to use it...
Is it scary to run something like moving average on 100 quick passes and spend half a cent?
Do you know that it is a titanic and hopeless task to get even one offer to the masses? Even repeating it many times to one particular person doesn't help.
And here you are offering generalized advice to someone to give numbers. As if there isn't a whole website dedicated to cludes, no documentation, no articles etc.
Is running something like a moving average on 100 quick passes and spending half a cent also scary?
Are you aware that getting even one sentence across to the masses is titanic and hopelessly difficult? Even repeating it many times to one particular person doesn't help.
And here you are offering generalized advice to someone to give numbers. As if there isn't a whole website dedicated to claud, no documentation, no articles, etc.
Look, I'm a consumer, I know that something out there somehow exists and it can improve my life - call it a claud - questions arise:
1. How will my life improve;
2. how much will this improvement cost.
And I go looking for clear and concise information - knowing that I need to load a PC for a year - costs a PC 1000 c.u. - will I be able to save by using Cloud?
I look at resources - read articles and forum - I learn that everything works cool, but it is not clear what is measured - its own unit, and about the price I learn - people lose all the money and do not know until the end of the optimization, how much will cost optimization in advance - ie find out after the fact.
Received information scares - and I decide not to waste energy on the proceedings with the settings, payment and other things - as I learned that all is not certain, and the uncertainty scares.
I am aware that people use what is really convenient for them - so I offered to think about convenience for the consumer - including a clear example of the benefit of the product. I'm not saying that this example should be limited, but the fact is that there are no even banners advertising this service (and if there are - I can't see them - they do not reach the consumer).
At one time I was working for an IT company related to phone services - mid-2000s, and I was asked to implement services into applications - from buying food and making restaurant reservations, to ordering personal services. It was a bold idea at the time and no one was willing to invest money in it, plus I wasn't an expert in the field. My point is that there are people like you who can build a team and there are people who have flair but no entrepreneurial spirit. I just want to see you as an adequate person who admits he may be wrong.
Look, I am a consumer, I know that something out there somehow exists and it can improve my life - call it a clud - questions arise:
And even you don't want to check anything and find a reason not to use it. All you have to do is to type MQL5 cloud network into a search engine. And the prices have been discussed many times. But after 3.5 years on this forum you still don't believe it...
You don't even believe in technical discussions, but you give advice and blame me:
I have, after all, given the hint "it is impossible to spread the offer to the masses". Although that is exactly what we have been doing massively for many years. I guess I do understand enough about it after all.
You can lead a horse to a watering hole, but you can't make it drink. We work hard and bring traders to the services. We will continue to encourage it.
You have a small sample - only you.
Well, share the statistics of the service (this one little https://cloud.mql5.com/ru/stats)? Any information on the number of users and their growth?
Who are the masses? Are they the consumers of the market who are taught to shop, but who also need to be taught how to evaluate the product correctly?
I don't want to check anything, but you find a reason not to use it. You just type MQL5 cloud network into search engine. And prices have been discussed many times. But after 3.5 years on this forum you still don't believe me...
I tried to do it more than once and wrote my first impressions that may arise among potential users of the service. Especially about prices - the formula is not convenient - more practical to evaluate the cost of the computer desktop and virtual - purely subjective - clear and understandable.
Point me to the formula - so that I can answer my own question - which is more profitable with a cloud or a desktop PC.
You don't even believe a lot of technical discussions, but you give advice and blame me:
I have spent more than 16 years to develop and spread Metatrader technologies, not you.
Which discussions do I not believe? I don't blame you, on the contrary, I am glad you exist and do what you can!
Are you not confusing, advice with guidance? Advice does not compel or condemn - it simply communicates the message that the adviser thinks is right for the situation to which the advice applies.
I gave a hint "it's impossible to extend the offer to the masses". Although this is exactly what we are doing massively for many years. I guess I do understand it enough.
You can lead a horse to a watering hole, but you can't make it drink. We work hard and lead traders to services. We will continue to lead and stimulate.
It's good that you see your way and intend to achieve new goals!
At one time you suggested using cloud for tasks not related to trading - how do you develop this idea?
This is the underlying explanation for the difficulty of selling. But instead people come up with supposedly rational reasons :)
Yesterday I found out that it turns out we're not doing well enough. And in order to be better, we have to take over full technical support for traders from brokers. It's like the whole world. It makes so much sense. And the man really advised it, trying to help us.
And the reality is that he just wants to save money and to impose the costs on us. That is the real reason for the man's behaviour.