Exness Trading Platforms — What Dates and What Does Not (United Arab Emirates)

Every comparison of trading software mixes two kinds of statement: one that survives the next release and one that stops being true when it ships. Sorting the rows by that alone answers most of the choice.

Open Exness Account →

100+ instruments · Founded 2008

Four platforms are available with Exness — MetaTrader 4, MetaTrader 5, the browser-based Exness Terminal and the Exness Trade app — and every comparison of them mixes two kinds of claim. Structural claims (who publishes the software, whether anything has to be installed, whether automation exists there as a category, whether one automation language runs on the other terminal) hold across releases. Versional claims (counts of timeframes, indicators and order types, panel names, which instruments appear in a list) describe one build and expire with it. A choice made on the structural rows survives the next release; a choice made on the counts has to be made again.

Two kinds of statement in one comparison

A platform comparison is a mixture of two claim types. Structural claims describe what a piece of software is — who publishes it, where it runs, what category of thing it can do at all. Versional claims describe the contents of the build in front of you.
The test is one question: would this sentence still be true if a newer build shipped tonight? If yes, it is structural. If a release note could falsify it, it is versional and it has a shelf life.
Structural: MetaTrader 4 and MetaTrader 5 are published by MetaQuotes, while the Exness Terminal and the Exness Trade app are published by Exness. Two publishers means two release calendars and two sets of decisions that the other publisher does not make.
Structural: MQL4 and MQL5 are separate languages and are not compatible with each other — code written for one does not run on the other. That is a property of two languages, not of a version, and no release closes it.
Structural: the Exness Terminal runs in a browser with nothing to install, and the desktop terminals do not. That is a property of where the software lives, not of what it currently contains.
Structural: expert advisors are a category the desktop terminals have and the browser Terminal and the mobile app do not. Absence of a category is a different kind of fact from absence of a feature.
Versional: counts. How many timeframes, how many built-in indicators, how many pending order types — every one of those numbers belongs to the build that produced it and is quoted with that build attached.
Versional: interface geography — where a panel sits, what it is called, which menu holds a setting. A comparison written on that is out of date the first time a layout moves.
Versional: which instruments appear in a list. The lists on MT4 and MT5 are not identical, and the list itself is read from the contract specifications for the account rather than from any comparison.
The practical consequence is to choose on the structural rows and to check the versional ones in the software itself at the moment they matter, rather than in any comparison — including this one.

Sorted by shelf life

Statement about a platform	Class	What could make it false	Where it is checked
MetaTrader 4 and MetaTrader 5 come from MetaQuotes; the Exness Terminal and the Exness Trade app come from Exness	Structural	A change of publisher, which is an announced event and not a release	Nothing to check between releases
MQL4 code does not run on MT5 and MQL5 code does not run on MT4	Structural	Nothing that appears in a release note; the two languages are separate	Nothing to check between releases
The Exness Terminal needs nothing installed; the desktop terminals are downloads	Structural	A change in how the product is delivered	Nothing to check between releases
Expert advisors run in the desktop terminals only, not in the browser Terminal or the mobile app	Structural	A whole category being added to a product that has none	Once, before the choice is made
MT5 carries 21 timeframes and 38 built-in indicators against 9 and 30 in MT4	Versional	Any release that adds or removes one of them	The terminal's own menus, at the moment the count matters
The MT5 strategy tester uses several CPU cores; the MT4 tester works on one instrument at a time	Versional	A rewritten tester in either terminal	The tester's own settings screen
Some instruments are listed on MT5 only	Versional	Any change to the instrument list for an account type	The contract specifications on the account
Where a panel sits and what it is called	Versional	Any interface release	The screen in front of you

Open Exness Account →

The one-question test

Any sentence in a platform comparison can be sorted with a single question: could a newer build make this false? If the answer is no, the sentence describes what the software is. If the answer is yes, it describes what the software currently contains, and the comparison is quietly quoting a date it does not print.

The two classes are easy to tell apart once the question is asked out loud. A publisher, a programming language and the difference between running in a browser and running from an installer are all properties nothing short of a redesign touches. A count of indicators, the name of a panel and the contents of a menu are all properties a routine release is allowed to change.

Nothing here says the versional rows are wrong. They are accurate about the build they were written against, which is the whole of their claim. The mistake is only in treating them as though they were the other class.

Why the durable rows are also the deciding rows

The rows that do not expire happen to be the ones that cannot be worked around. A language incompatibility cannot be handled by learning a menu, and software that requires nothing to be installed cannot be turned into software that does. A missing count, by contrast, is usually a matter of doing the same thing by another route.

This is why a choice anchored on the durable rows is stable in a way that a choice anchored on features is not. The durable rows are few, they are quick to check and they rarely conflict with each other, so the decision they produce is short.

Where a versional detail genuinely matters, it belongs to a different moment. It is verified in the software at the point of use, on the build actually installed, rather than carried around as part of a decision made months earlier against a build that has since moved.

Reading any platform comparison, in order

Take one row at a time and ask whether a newer build could make it false.
Put the rows that survive that question in one group: publisher, language compatibility, whether an installation is needed, whether a category such as automation exists there at all.
Put everything that a release note could change in a second group: counts, panel names, menu locations, the contents of a list.
Make the choice from the first group only, and notice how short the resulting list of real differences is.
Note which items from the second group actually matter, and check each one inside the software rather than inside a comparison.
Recheck the second group when a build changes, and leave the first group alone until a publisher or a product category does.

Every platform can be run on virtual funds first, which is also the quickest way to check a versional detail on the build actually installed.

Frequently asked questions

How can a statement about a platform be sorted into durable or dated?

Ask whether a newer build could make it false. A publisher, an incompatibility between two programming languages and the difference between a browser and an installer are not things a release changes. A count of indicators, the name of a panel or the position of a setting are exactly what a release changes.

Which differences between MT4 and MT5 are not going to change?

That both come from MetaQuotes, and that MQL4 and MQL5 are separate languages whose code does not cross between the two terminals. Those are properties of the products rather than of their current builds, which is why a comparison can rest on them.

Are the indicator and timeframe counts wrong, then?

No. They are accurate about the build they describe, and that is the whole of the claim: MT5 carries 21 timeframes and 38 built-in indicators against 9 and 30 in MT4. The mistake is reading a build-scoped number as a permanent property of the platform.

Does a newer build make an earlier platform choice wrong?

Only if the choice was made on the dated rows. A decision resting on the publisher, the automation language and whether anything has to be installed is untouched by a release. A decision resting on a feature count has to be made again whenever that count moves.

Where should a versional detail be checked?

In the software itself, on the build actually installed, at the moment the detail matters. The instrument list for an account is read from its contract specifications for the same reason, since the list changes independently of any page describing it.

Is there a charge for any of the four platforms?

No. MetaTrader 4, MetaTrader 5, the Exness Terminal and the Exness Trade app are provided at no charge, on demo and on live accounts. Costs come from trading itself — the spread, and commission on the account types that carry it — not from the software.