Poland Buying Korean Tanks?

[Chris Werb]

 

I would consider it prudent to protect it at least against medium caliber APFSDS (at least at longer ranges, at least frontally).

And I am still failing to see why is this not the case. If you hit mantlet - you will destroy breech. If you go farther to "cheeks" - you won't hit anything internal. How is that less protected than any other MBT turret face?

Frontal arc protection can be argued, but even then people are too willing to forget that there is barely anything besides breech inside turret. And that presumed turret size is somewhat bigger than just box around breech supposed to be.

 

 

In SB, IIRC the turret basket contains vulnerable electronics/electrics. You're going to tell us it's just a counterweight?

 

I want you to understand that I am NOT reflexively pro western or anti Russian/Soviet. A lot of Russian kit was/is excellent and I love playing Russian vehicles on SB (especially the T-62 and BMP-2). I just want the SB Armata model to be as accurate as it can possibly be.

[GARGEAN]

In SB, IIRC the turret basket contains vulnerable electronics/electrics. You're going to tell us it's just a counterweight?

 

It can be anything from non-mechanized ammorack to ZIP storage. I'm not gonna tell you what there specifically cuz I don't know. Just like anyone else on this forum.

 

 

 

I just want the SB Armata model to be as accurate as it can possibly be.

And this is beyond being just optimistic, considering noone here ever saw production version of tank. Hell, ever trial examples are not really known unlike for ex Kurg. All we seen is parade mockup. So... Want to properly model those?

[Ssnake]

I was about to say, while there can be educated guesses about some elements, other areas remain completely speculative for now, and probably for a good while. What can be said is that exposed sensors are inherently vulnerable, and an unmanned turret increases dependency on such sensors. At the same time one has to admit that near-field surrounding cameras can be made really small (as every smartphone owner knows) and still deliver good quality in lowlight conditions. I'd wager a bet and say that these are protected well. But that's just the near field. A manned turret offers more flexibility, but at the same time basic engineering principles dictate that a three-man armored crew capsule in the hull is way more mass efficient than the equivalent protection in a manned turret. So if you have adequate crew protection in mind for contemporary and near-future threats, I still think that the concept is valid. Whether the Armata is the best possible design we can but speculate.

[JasonJ]

70 tons seems to be the limit to how far designers seem willing to go. But 50 tons in Russia's case. Wonder if there is any new advantage to be had in making a 70 ton T-14. Maybe such a front hull could protect against even 135mm APFSDS.

[bojan]

70 tons seems to be the limit to how far designers seem willing to go. But 50 tons in Russia's case...

Armata stepped over 50t.

[JasonJ]

 

70 tons seems to be the limit to how far designers seem willing to go. But 50 tons in Russia's case...

 

Armata stepped over 50t.

Welcome back to 50 ton plus class of tanks

 

I'm assuming that's combat ready weight. Standard weight seems to be given as 48 tons.

[Ssnake]

Combat mass is the only useful metric for such discussions. Empty weight is something you can use for railway transport planning.

[Mighty_Zuk]

1)Oh, boy. We talked about the vulnerability of those electronic components because of their exposure and you began writing about panoramic sights. Yes, of course did I assume you meant the ordinary ones, for what you wrote would have made even less sense otherwise. Yet more poorly protected electronics stuff is no backup to poorly protected electronics stuff worth talking about. A proper alternative has to not share systemic risks with the main subsystem.

 

2)That delay is nowhere near "seconds". You should check the basics of external ballistics.

 

3)The current engagement sequence from target spotting till shot is about seven seconds with a good crew. You're behind the curve by decades.

 

4)Doesn't matter. A good IFV commander would likely pop a multispectral smoke wall right with the first salvo (kinda what you wrote earlier) if he sees an opportunity for reaching concealment and either withdrawing or using another firing position. There's enough confidence in the gun's ability to land hits where aimed from practice, and there would hardly be any directly observable effects anyway. The pellets don't have phosphorous coating or similar as impact indicator.

 

 

 

1)I honestly forgot what we were talking about here, so I'll leave it at that.

 

2)You're right. Apparently the velocity difference between autocannons and high caliber cannons is not that great as I thought.

 

3)Current. That's the key word. That is because we're still just preparing for the next generation of MBTs. As demonstrated, AI can independently detect and ID targets and classify them as friends or foes, either passively or reactively (i.e response to fire). First AFVs to fully utilize such tech are supposed to enter service in 2026/7, and first vehicles with limited aspects of said capabilities are set to enter service in 2021/2.

AI will not pull the trigger, but the crew can take as much time as they need to assess the target if they feel a delay won't hurt them. Some of the fire may be done autonomously with a more advanced form of BMS (other users of the BMS pulling the trigger instead of you, although that feature is not yet confirmed).

 

4)The sensor fusion with radars should greatly minimize the capabilities of multispectral smoke, and may require the use of chaffs instead. SAR can easily become secondary eyes for the crew.

[Chris Werb]

First AFVs to fully utilize such tech are supposed to enter service in 2026/7, and first vehicles with limited aspects of said capabilities are set to enter service in 2021/2.

AI will not pull the trigger, but the crew can take as much time as they need to assess the target if they feel a delay won't hurt them. Some of the fire may be done autonomously with a more advanced form of BMS (other users of the BMS pulling the trigger instead of you, although that feature is not yet confirmed).

 

The sensor fusion with radars should greatly minimize the capabilities of multispectral smoke, and may require the use of chaffs instead. SAR can easily become secondary eyes for the crew.

 

 

I read that the Japanese Type 10 already has (for at least a decade) automatic target acquisition and tracking and cooperative engagement capability. I'm not sure if it can classify targets as friendly, hostile or civilian automatically though.

[Ssnake]

I'm sure, computers will come up with something. I doubt that it will be useful, but computers being computers, they will compute "a" result.

[JasonJ]

 

First AFVs to fully utilize such tech are supposed to enter service in 2026/7, and first vehicles with limited aspects of said capabilities are set to enter service in 2021/2.

AI will not pull the trigger, but the crew can take as much time as they need to assess the target if they feel a delay won't hurt them. Some of the fire may be done autonomously with a more advanced form of BMS (other users of the BMS pulling the trigger instead of you, although that feature is not yet confirmed).

 

The sensor fusion with radars should greatly minimize the capabilities of multispectral smoke, and may require the use of chaffs instead. SAR can easily become secondary eyes for the crew.

 

 

I read that the Japanese Type 10 already has (for at least a decade) automatic target acquisition and tracking and cooperative engagement capability. I'm not sure if it can classify targets as friendly, hostile or civilian automatically though.

 

Yeah, that's what is said about it. It can simultaneously acquire up to 8 targets, classify them as tank, IFV, soft skin, aircraft, fixed, or human, and automatically specify threat value so as to prioritize targets, and share that information with the other tanks in the platoon. The computer can also automatically assign different targets for each platoon tank and enable carrying out cooperative attacks so as to avoid things like avoid overkill. If its uncertain to the FCS that a target has been destroyed, the computer can recommend taking another shot. But this aspect of the FCS doesn't seem able to classify between enemy, ally, or civilian immediately but there is ally and enemy information sharing as part of C4I.

 

Well, its what's I can find. Can't say that everything about it is known. Even just inside pictures of the production version of the tank seem to still not be around.

[Mighty_Zuk]

 

 

First AFVs to fully utilize such tech are supposed to enter service in 2026/7, and first vehicles with limited aspects of said capabilities are set to enter service in 2021/2.

AI will not pull the trigger, but the crew can take as much time as they need to assess the target if they feel a delay won't hurt them. Some of the fire may be done autonomously with a more advanced form of BMS (other users of the BMS pulling the trigger instead of you, although that feature is not yet confirmed).

 

The sensor fusion with radars should greatly minimize the capabilities of multispectral smoke, and may require the use of chaffs instead. SAR can easily become secondary eyes for the crew.

 

I read that the Japanese Type 10 already has (for at least a decade) automatic target acquisition and tracking and cooperative engagement capability. I'm not sure if it can classify targets as friendly, hostile or civilian automatically though.

Yes, it does. However that is not in any way comparable to what I have described. The first tank with such a capability as you mentioned, entered service in 1995, IIRC. Specifically it was Elbit's Baz (Knight 3?) FCS it had developed for the Merkava 3.

 

Once you aim at a target, an AI identifies it as a mobile object and can track it automatically, which is how they achieved their advertised capability to shoot down helicopters with conventional non-guided munitions.

 

Most tanks today have this capability, as it is necessary in order to achieve true fire on the move capability, against a moving target (as opposed to firing on the move on a static target which is a lot easier).

 

Type 10 merely added a storage device dedicated to target data, and a limited analysis tool.

 

I am talking specifically about the Carmel demonstrators (there are 3, each from a different company), of which there is footage.

 

They add independent scanning capabilities. By independent I mean you don't have to aim your sight at the target. It will use other sensors to do this job so you could get target data from many directions even if you are currently manually scanning an empty area.

 

Other than that, there is the fusion of multiple sensors - a radar with SAR function (SAR is a technology that creates human-interpretable images based on radar EM transmissions instead of light, similar to how LIDAR works with laser), possibly a LIDAR for short range, acoustics, varying wavelength static panoramic optics, and a data feed from BMS and unmanned systems.

 

The generation of targets with such a system is vastly increased in scope, speed, and reliability.

 

It's hard to compare any existing system with it, but both the US and Israel see a possible fielding by 2026/7, and right now Germany is also in the process of fielding the first fragments of that concept, with acquisition of the "glass battlefield" C2 system, which relies heavily on Rafael's Fire Weaver.

 

This whole concept, however, makes current layouts of 3/4 crewmen far less efficient than they once were, and vastly increases the relative efficiency of 2/3 man crews.

For this reason alone, a purchase of 3/4 man tanks with manned turrets is a bad strategy at the moment.

Edited February 8, 2020 by Mighty_Zuk

[JasonJ]

 

 

First AFVs to fully utilize such tech are supposed to enter service in 2026/7, and first vehicles with limited aspects of said capabilities are set to enter service in 2021/2.

AI will not pull the trigger, but the crew can take as much time as they need to assess the target if they feel a delay won't hurt them. Some of the fire may be done autonomously with a more advanced form of BMS (other users of the BMS pulling the trigger instead of you, although that feature is not yet confirmed).

 

The sensor fusion with radars should greatly minimize the capabilities of multispectral smoke, and may require the use of chaffs instead. SAR can easily become secondary eyes for the crew.

 

I read that the Japanese Type 10 already has (for at least a decade) automatic target acquisition and tracking and cooperative engagement capability. I'm not sure if it can classify targets as friendly, hostile or civilian automatically though.

Yes, it does. However that is not in any way comparable to what I have described. The first tank with such a capability as you mentioned, entered service in 1995, IIRC. Specifically it was Elbit's Baz (Knight 3?) FCS it had developed for the Merkava 3.

 

Once you aim at a target, an AI identifies it as a mobile object and can track it automatically, which is how they achieved their advertised capability to shoot down helicopters with conventional non-guided munitions.

 

Most tanks today have this capability, as it is necessary in order to achieve true fire on the move capability, against a moving target (as opposed to firing on the move on a static target which is a lot easier).

 

Type 10 merely added a storage device dedicated to target data, and a limited analysis tool.

 

I am talking specifically about the Carmel demonstrators (there are 3, each from a different company), of which there is footage.

 

They add independent scanning capabilities. By independent I mean you don't have to aim your sight at the target. It will use other sensors to do this job so you could get target data from many directions even if you are currently manually scanning an empty area.

 

Other than that, there is the fusion of multiple sensors - a radar with SAR function (SAR is a technology that creates human-interpretable images based on radar EM transmissions instead of light, similar to how LIDAR works with laser), possibly a LIDAR for short range, acoustics, varying wavelength static panoramic optics, and a data feed from BMS and unmanned systems.

 

The generation of targets with such a system is vastly increased in scope, speed, and reliability.

 

It's hard to compare any existing system with it, but both the US and Israel see a possible fielding by 2026/7, and right now Germany is also in the process of fielding the first fragments of that concept, with acquisition of the "glass battlefield" C2 system, which relies heavily on Rafael's Fire Weaver.

 

This whole concept, however, makes current layouts of 3/4 crewmen far less efficient than they once were, and vastly increases the relative efficiency of 2/3 man crews.

For this reason alone, a purchase of 3/4 man tanks with manned turrets is a bad strategy at the moment.

 

 

 

 

 

 

Yeah, that's what is said about it. It can simultaneously acquire up to 8 targets, classify them as tank, IFV, soft skin, aircraft, fixed, or human, and automatically specify threat value so as to prioritize targets, and share that information with the other tanks in the platoon. The computer can also automatically assign different targets for each platoon tank and enable carrying out cooperative attacks so as to avoid things like avoid overkill. If its uncertain to the FCS that a target has been destroyed, the computer can recommend taking another shot. But this aspect of the FCS doesn't seem able to classify between enemy, ally, or civilian immediately but there is ally and enemy information sharing as part of C4I.

 

Well, its what's I can find. Can't say that everything about it is known. Even just inside pictures of the production version of the tank seem to still not be around.

 

I'm terribly sorry if this may come off anal, I do not mean too, and have no qualms about whether or not the Type 10 has or doesn't have such FCS but even in just strictly factual and syntax basis.. doesn't the blue describe the red? And doesn't the green misrepresent the blue?

[Chris Werb]

Sorry, Zuk, in addition, this bit is just plain wrong:

 

Most tanks today have this capability, as it is necessary in order to achieve true fire on the move capability, against a moving target (as opposed to firing on the move on a static target which is a lot easier).

 

I accept this is true of the Merkava 3B onwards, but it is not true of many other tanks. Certainly no in service Abrams, Leclerc or Leopard of which I am aware has it and no T series tank up to at least the T-14 - the Korean K2 is the only non Japanese or Israeli MBT I'm aware of that does. It is also not necessary in order to hit a moving target from a moving tank - this has been done reliably without auto-tracking for decades. I have done it myself in SB many times - this is the tank gunnery simulator used by the Danish Army among others, so you can take it up with them if you wish. If you wish to try out autotracking - two vehicles in SB have it - the Piranha Mk3C DF30 and the Pizarro. It's not entirely a blessing vs manual tracking, as you will discover.

[Ssnake]

Also the ERCWS remote weapon station on the Iveco LMV. The DF30's fire control system is, actually, a derivative of the ERCWS, both being from Elbit.

[bojan]

...and no T series tank up to at least the T-14 ...

T-72B3M and T-90MS.

BMD-4M also IIRC.

[Chris Werb]

 

...and no T series tank up to at least the T-14 ...

T-72B3M and T-90MS.

BMD-4M also IIRC.

 

 

Aaargh! I almost fell into the "There aint no such thing as" criterion of how to spot a fake expert My statement that it's not "most tanks today" still stands though.

[Mighty_Zuk]

 

First AFVs to fully utilize such tech are supposed to enter service in 2026/7, and first vehicles with limited aspects of said capabilities are set to enter service in 2021/2.

Type 10 merely added a storage device dedicated to target data, and a limited analysis tool.

 

It can simultaneously acquire up to 8 targets, classify them as tank, IFV, soft skin, aircraft, fixed, or human, and automatically specify threat value so as to prioritize targets, and share that information with the other tanks in the platoon. The computer can also automatically assign different targets for each platoon tank and enable carrying out cooperative attacks so as to avoid things like avoid overkill. If its uncertain to the FCS that a target has been destroyed, the computer can recommend taking another shot. But this aspect of the FCS doesn't seem able to classify between enemy, ally, or civilian immediately but there is ally and enemy information sharing as part of C4I.

I'm terribly sorry if this may come off anal, I do not mean too, and have no qualms about whether or not the Type 10 has or doesn't have such FCS but even in just strictly factual and syntax basis.. doesn't the blue describe the red? And doesn't the green misrepresent the blue?

 

 

Good question, but no. Neither is in conflict. You just have to try and imagine the system I am talking about.

To implement said system, you have to have the following:

 

1)Sensory:

 

a)Non-static optics - rotating and elevating panoramic sights as we see on basically every modern western tank. Will give each crewman an independent view for long range.

b)Static optics - a mesh of less powerful and cheaper optics to provide 360° picture for close range situational awareness, plus a powerful main gun sight.

c)Radar - already part of an APS, but adds SAR function.

d)LIDAR - for autonomous driving and short range object classification.

 

2)Analysis:

 

a)Next gen BMS - Adds fire source selections and EO-enabled coordinates generation.

b)Sensor fusion - overlays all received data and enables active sensor switching based on current reliability rate (e.g optics less reliable in smoke, LIDAR cannot pierce rubble).

c)AI - independently activates sensors for scanning where crew isn't looking, and goes up to friend or foe classification, and slew to cue for human analysis.

 

A demonstration was conducted in 2019, with 3 companies showing their different approaches to the project. Although showcasing only Israeli companies, American companies are doing parallel work for the OMFV and we may hear of partnerships in the coming years. It is likely KMW and Nexter will start similar developments in preparation for their joint MBT project.

 

https://www.youtube.com/watch?v=f2OK3fVUHOo

https://www.youtube.com/watch?v=si8tpgSHErw

https://www.youtube.com/watch?v=cCVLGAU_26c

 

Rafael's long demonstration video:

https://www.youtube.com/watch?v=wLcJ0DDcXHM

 

A lot of the capabilities are actually hard to show on camera, because they happen internally in the system and cannot be visualized, but it gives a good idea of how the crew will be operating the next generation of AFVs.

One thing that is easy to miss, is that every target they've engaged, was generated before the human crew got a visual.

 

So now that I've got this behind us, I'll explain away the confusion.

 

Red - The Merkava 4 Barak ("Lightning") is due to be received by March this year, and the first battalion will be operational with it by March 2021. It will use the IronVision system by Elbit, a new APS, and the Fire Weaver BMS. The Carmel uses new sensors, and multiplies the number of pre-existing sensors. If there was 1 panoramic sight for the TC, then the new AFV will have 2-3. If there was 1 radar array for every sector, then now there are more, some movable.

The Merkava 4 Barak adds the new sensors, but does not multiply the pre-existing ones.

This in turn also limits the potential of the AI, as it has fewer sensors to fuse and analyze from. In the end, the Barak will get significantly better situational awareness with some automatic environment scanning, it will get better slew to cue, and much better automatic target generation that is passive as well and not just reactive, but it will not quite be the AFV shown in the Carmel demonstrators. For that, it has to be completely redesigned.

 

Green - Type 10 can automatically track targets, and can assign targets, but it does not automatically scan the environment. It does not provide the crew with a proper 360° picture. It does not add any of the above mentioned capabilities, other than perhaps some analysis of the target. But it's a far cry from what's being done now. It improves upon an old technology, but does not add any new technology.

 

Sorry, Zuk, in addition, this bit is just plain wrong:

 

Most tanks today have this capability, as it is necessary in order to achieve true fire on the move capability, against a moving target (as opposed to firing on the move on a static target which is a lot easier).

 

I accept this is true of the Merkava 3B onwards, but it is not true of many other tanks. Certainly no in service Abrams, Leclerc or Leopard of which I am aware has it and no T series tank up to at least the T-14 - the Korean K2 is the only non Japanese or Israeli MBT I'm aware of that does. It is also not necessary in order to hit a moving target from a moving tank - this has been done reliably without auto-tracking for decades. I have done it myself in SB many times - this is the tank gunnery simulator used by the Danish Army among others, so you can take it up with them if you wish. If you wish to try out autotracking - two vehicles in SB have it - the Piranha Mk3C DF30 and the Pizarro. It's not entirely a blessing vs manual tracking, as you will discover.

 

I could have sworn I have read about the Leopard, Leclerc, and Abrams having automatic target tracking in late models. It appears I was wrong then.

EDIT: Seems the Type 99 has it as well, and subsequent NORINCO export tanks.

Edited February 8, 2020 by Mighty_Zuk

[JasonJ]

You said "limited" in reference to AI system for initial implementation by "2021/22". So that shouldn't mean the full package of self-driving or the increase in situational awareness with use of auto scaning 360 view cameras and sensors. The point of the Type 10 wasto show elements of "AI" already being fielded. You initially responded to it by down playing it as just "storage" and "limited analysis" but obviously the information sharing and what you later conceded "can assign targets" is more than just data storage. But why does it have to be like pulling teeth to get proper acknowledgment?

 

And one aspect that you seem to be overlooking is the probably the most effective feature which is the information with other tanks in a platoon and the auto assignment of targets to create "cooperative attack" to prevent overkill but also means improving engagement efficiency. So even your late concession of "can assign targets" still over looks that the target assignment is being spread throughout the platoon of tanks, not just an activity happen within a single tank that's not in sync with the other platoon tanks.

 

But well Israel STRONK! yeah :)

 

Nice videos and general update on the R&D of things

Edited February 8, 2020 by JasonJ

[Chris Werb]

 

 

Green - Type 10 can automatically track targets, and can assign targets, but it does not automatically scan the environment. It does not provide the crew with a proper 360° picture. It does not add any of the above mentioned capabilities, other than perhaps some analysis of the target. But it's a far cry from what's being done now. It improves upon an old technology, but does not add any new technology.

 

 

 

 

So tanks that could identify prioritise, track and engage targets and automatically assign targets between tanks in a platoon already existed before the Type 10. Which ones were those then?

[Chris Werb]

. But why does it have to be like pulling teeth to get proper acknowledgment?

 

Because it's not...

 

[Mighty_Zuk]

You said "limited" in reference to AI system for initial implementation by "2021/22". So that shouldn't mean the full package of self-driving or the increase in situational awareness with use of auto scaning 360 view cameras and sensors. The point of the Type 10 wasto show elements of "AI" already being fielded. You initially responded to it by down playing it as just "storage" and "limited analysis" but obviously the information sharing and what you later conceded "can assign targets" is more than just data storage. But why does it have to be like pulling teeth to get proper acknowledgment?

 

And one aspect that you seem to be overlooking is the probably the most effective feature which is the information with other tanks in a platoon and the auto assignment of targets to create "cooperative attack" to prevent overkill but also means improving engagement efficiency. So even your late concession of "can assign targets" still over looks that the target assignment is being spread throughout the platoon of tanks, not just an activity happen within a single tank that's not in sync with the other platoon tanks.

 

But well Israel STRONK! yeah :)

 

Nice videos and general update on the R&D of things

The core of the system will be fielded next year, but will not reach its full potential until 2026/7.

What is described about the Type 10 does show an AI, but I defined some clear goals for the AI of next gen AFVs. AI in itself is not just one thing. It can be defined in many ways. In a way, existing FCS and BMS already show limited AI, but it does not provide those capabilities we expect to see fielded soon.

 

The sharing of information between tanks in a platoon is not the unique aspect. The unique aspect of the Type 10 is the allocation of targets itself, between tanks of a platoon. But one might ask, where does the infantry come into the picture? Does the Type 10 platoon get fed from independently sourced data, or does it get data from other sources like infantry and aerial assets? Can it allow target allocation to other elements?

Where does the company coordination come into play? Or the battalion?

Don't get me wrong. The idea of sharing targets between platoon mates is great. But if existing interfaces and data buses allow, and within permissible latency, I think the framework should be expanded.

 

And of course, the main point here is that even those Type 10 tanks in the same platoon are fed data that a human gunner or TC acquired, and sent to the other tanks. The AI I was talking about, is one whose purpose is to allow tanks to operate independently if needed, by automatically and independently acquiring targets that the human crew is unable to detect. Everything else, including target allocation and sharing, is in the realm of BMS which I intentionally separated.

 

 

 

 

Green - Type 10 can automatically track targets, and can assign targets, but it does not automatically scan the environment. It does not provide the crew with a proper 360° picture. It does not add any of the above mentioned capabilities, other than perhaps some analysis of the target. But it's a far cry from what's being done now. It improves upon an old technology, but does not add any new technology.

 

 

 

 

So tanks that could identify prioritise, track and engage targets and automatically assign targets between tanks in a platoon already existed before the Type 10. Which ones were those then?

 

I do not know. My intention however was that it does not add any new technology that does not exist in what I described earlier. Not that all tanks were already equipped with such systems. And the system that exists on the Type 10, if it is as described, is not one that would fit the needs of all potential users. Depending on the situation, the BMS should be flexible enough to prioritize targets on anything from the individual level, to brigade level. I do not understand the fixation on the platoon level, and especially the exclusion of other maneuvering elements from the equation.

Are you sure this is an accurate description of the system? I'd like to read more.

[Mighty_Zuk]

 

. But why does it have to be like pulling teeth to get proper acknowledgment?

 

Because it's not...

 

 

You know, shoving nationality into debates out of frustration, really detracts from your credibility. If you want to have a healthy debate, please refrain from doing so.

[GARGEAN]

Sorry, Zuk, in addition, this bit is just plain wrong:

 

Most tanks today have this capability, as it is necessary in order to achieve true fire on the move capability, against a moving target (as opposed to firing on the move on a static target which is a lot easier).

 

I accept this is true of the Merkava 3B onwards, but it is not true of many other tanks. Certainly no in service Abrams, Leclerc or Leopard of which I am aware has it and no T series tank up to at least the T-14 - the Korean K2 is the only non Japanese or Israeli MBT I'm aware of that does. It is also not necessary in order to hit a moving target from a moving tank - this has been done reliably without auto-tracking for decades. I have done it myself in SB many times - this is the tank gunnery simulator used by the Danish Army among others, so you can take it up with them if you wish. If you wish to try out autotracking - two vehicles in SB have it - the Piranha Mk3C DF30 and the Pizarro. It's not entirely a blessing vs manual tracking, as you will discover.

Quite a few tanks have autotrackers integrated into their sights. T-72B3, 90SM/M to name few. First one to use it was Type 90 btw, not Merk.

[GARGEAN]

T-72B3M and T-90MS.

BMD-4M also IIRC.

Not B3M, just plain B3 has it (Sosna-U has it to be more precise). And you're right, BMD-4M too.