thermal vision It's time to upgrade Thermal Vision?

[mickeymen 322]

Good day BIS!

Over the past two months, I have seen a lot of changes for Thermal Vision Arma3 in the Development Branch Changelog.

This definitely makes me happy, but I would like to write to you so that you also try to eliminate the main Thermal omission of Arma3 that has been all this time! 

The fact is that Arma3 has never taken into account the current weather and time of day for Thermal Vision before, and this does not look good.  

In Arma3 (all versions including 2.08), at any time of the day/any weather, the thermal vision looked the same.

Unfortunately it was looked like a terrible flaw, like cheating/imbalance, because all targets looked like torches from huge distance on a dark background  and only the ACE3 mod some smoothed out this problem.

 

My Main message: In addition to temperature differences, the thermal imager must be affected by wind, rain or sun, which can heat up or cool buildings or the landscape, and this must affects the accuracy of the result.

For example, In sunny weather, on midday, when houses or landscapes are warmed by the sun, they must emit heat.This means that in such situation, the determination of targets should be difficult, because everything should radiate warmth - as target itself and its background also. In this case, the target should simply blend into its background. But if at this day time, with such weather we look at vanilla Thermal Arma3, then we will see flaming targets, as if everything else will not radiate heat! It doesn't look real...

 

Conversely, at night, when the entire landscape cools down, targets should be more clearly visible.

Please, due to the latest changes for Thermal Vision, try to also create at least the some influence of weather/time of day in the same way. It seems to me at the current stage of Arm`s life, this is not a super difficult task. 

Also, your recent Thermal Vision overhaul can help you do just that.

Thank you for Support our favorite game!

Edited August 18, 2022 by mickeymen
Added additional info

[mickeymen 322]

Yesterday I received update 2.10 and unfortunately I saw a complete disappointment in the changes in thermal vision. 

You can open reviews of SPOTREP #00105 in Steam and see a lot of dissatisfied with the new thermal system.

Many reviewers would prefer the old system over the new one. I think thermal update is a complete failure/

 

Some of my personal notes here:

 

1. The main problem is the complete lack of binding to the current day time/current weather.

This is complete mismatch. The game doesn't care if the sun is bright at noon or it's night and it's raining. Are the stones and buildings hot from the sun, or have they cooled down in the dead of night. No matter! 

For example, on a rough night we can see shining roads! They will burn as if heated by the sun at noon and will burn in deep night.  Wow! No difference!?

The player will see more or less shining objects regardless of any day time dependencies.

 

The simplest test - Set in the editor the time multiplier in the as X12 put yourself on the map and go into spectator mode in your game.

Switch to thermal vision and  you will see, how over time, the thermal vision will not have a cooling in the evening/night and will not again warm up to the maximum at afternoon. Instead, you will see burning stones and roads at night! You will see randomly changing color grass!  It just looks like uncontrollable chaos without reference to time of day without reference current temperature!

All this does not look natural in places, even silly.

 

2.  The thermal horizon will have clear ugly lines. New thermal view has problems with fog/haze and smoke FX.

 

On the horizon where the haze will always be strange lines: 

 

 

The distant parts of the landscape will be divided into lines!  What this in general?  What lines?

 

Also pay attention to the smoke from the engines/wheels/tracks of vehicles.

It looks the same as haze horizon above, like it's without anti-aliasing, no blending, no gradient, too noisy:

 

 

So the new thermal has problems with fog/haze/smoke.

 

 

 

 

3.  The same landscape with the sea horizon can look completely different with NVG and with thermal vision.

 

This is what the image looks like in NVG:

 

 

 

 

The same angle on the same position but in Titan Launcher thermal will look like two completely different places:

 

 

Why is this semi-island burning like a torch?

Is he warm under the sun?

In this screenshot a deep night, it should not glow, by night, this semi-island had already cooled down, for this reason he should not radiate heat, at least it should not be as maximum bright.

 

 

 

 

3. In the new thermal, the grass color is significantly lighter than the its ground color, but actual, within a gameplay this is one same surface

 

Due to the fact that grass is a limited render distance from the player, we see light grass near us and in the distance we see a black strip of ground:

 

 

In fact, dividing the same surface (grass) into two different colour lines, but in reality this imposible because it's the same surface whether the player sees the grass or not. Thus, land without grass should not differ from grass in terms of heat release. So way the player should not get two strupes with wild contrast for one surface.

 

 

I hope that the developers will see this post and make thermal vision in Arma3 adequate.

I would just take the image configs from the @A3 Thermal Improvement mod and bind them to the current time of day. 

The picture in this mod looks awesome, but they do not have turn on the engine, do not have cooling/heating vehicles and there is no other cooling/heating objects.

I also noticed that their corpses cool down in 10 seconds, which looks stupid. Their mod needs some improvements.

 

I sure, such formula - Image configs from @A3 Thermal Improvement + daytime binding + heating/cooling objects =  Arma3 would get excellent thermal vision.

 

[jgaz-uk 130]

I totally agree, the update  to thermal has ruined it. I posted on here Myself.

 

It has to be obvious to everyone, inc Bis, that the thermal update is  NOT an improvement but just the opposite, the question is Why ruin it? There must be a reason behind it, which is what???

 

 

https://forums.bohemia.net/forums/topic/239862-update-23082022-thermal/

 

[c.710 3]

BOHEMIA!!! Please fix this wildly inaccurate representation of thermal scopes.

 

Nausea inducing headache machines now... At least apply a little more research into the scopes. (i/e contrast settings, brightness settings, thermal trim....) the list goes on... They just slapped some psychedelic color pattern over what used to be a smooth easy on the eyes night scope. That thermal pattern (or sensor) they copied is one for scientific purposes. NOT a rifles scope!!!!!! That is what firefighting helicopters use and firefighters use when adressing the thermal signature of a FIRE or scientists observing the thermal behavior of something burning, and exhausts from an engine etc. Its not for long time viewing in a sniper scope. (nightstalker)  

 

And why does the smoke only paint terrain and becomes invisible over the horizon?

 

A burning tanks plume of smoke is as hot as a family home in the middle of the night. Along with volcanic trees or something. Their official servers are DEAD today too.... 

 

CAS is subject to weird warping of thermal temps rendering the use of the targeting pod useless at night. 

 

Giant grey oval bands when seeking enemy naval craft. 

 

Its like they see the bugs but are optimistic thinking it wont happen for the thousands of players... yeah right. I even asked the devs about a bug on candidate branch

 

"oh no its not like this.... it was a bug... its fine" 

 

I tried it within 5 minutes, replied with my findings being exactly what I thought, buggy... and got no response thereafter when voicing my opinion of the obvious bugs that were just mentioned to not exist. But was instead forced with this update a week later.  

 

I D K at a loss for words on this one. Night missions were so fun! Loved sniping the unsuspected. Now I need to grab some Advil and sunglasses if I intend to use the nightstalker at night ever again. Which I don't. 

 

(fingers crossed, at least a Modder can maybe mitigate this problem if they haven't moved on to Reforger already)

 

And YES.. I signed up just to say that! 😛

 

 

 

[Reticuli 14]

I couldn't successfully install the release branch of 105 spotrep and had to resort for the first time to trying the dev branch, which it updated to perfectly on the first try.

I am saying the following based on evaluating all of the TI options using Pilot Helmet (Integrated Vision) from Thermal Vision Goggles by Rad (OP Edition), not just one or two TI modes.

I think it's partly an improvement on the foot of the contrast curve, but the light portions of the screen are too bright and eye-straining.  It needs to be a gentler highlight dynamic range and not blasting out at your eyeballs as max white like it currently is. 

The 208 RC branch needed both the brightest and darkest portions of the contrast curve (shoulder and foot) to be a bit more gradual, with a little more nuance certainly, and the darkest and brightest portions especially not actually using the darkest and brightest colors possible... i.e., stop at a notch or two short of the most extreme darkness/luminosity. 

Other than the second-to-last Pilot Helmet (Integrated Vision) TI setting wonderfully benefiting from radical color changes (seen in 105 in the Nightstalker scope), the prior TI most needed its dynamic range tweaked rather than all this other stuff done... not that major improvements to how the TI works wouldn't be in order if they could be done well.  I do get what you guys are complaining about with the branches & grass now glowing hot.  They did a bunch of other stuff to TI that was unnecessary, and the contrast curve is now skewed to the other end of the extreme.

For one thing, goggles' imaging back to the eye lack perfect brightness or darkness.  With all NV and TI systems, you're both sensing and imaging back to the user with reduced dynamic range.  You shouldn't be either cramming this reduced dynamic range in Arma 3 to one end of the available curve or trying to stretch it back into the entire dynamic range like some eye-searing cartoon.  it should hover in the middle of the available curve.

I still find some need for using Reshade, but I feel that part of what Reshade previously was needed to do now is already being done by Arma 3's default TI modes.  Still, I prefer the 208 RC branch with Reshade to this new TI with or without Reshade.

I continue to think the use of the black hot mode for the Viper helmet is stupid (not to mention that without the IVAS mod there's no comparable NOD for NATO) and now made even worse blasting out your eyeballs at 2am, but I also think the widest NV mode should be 16:9 ratio (or slightly less) and no wider, and all of the TI goggle modes should be no wider than the ENVG-II(B?) stock TI mode (basically a square in the middle with the same width as the screen is tall, or even narrower).  Even IVAS does not have unlimited FOV for its TI, and its TI is still a narrower FOV than its LL TV NV sensing.  In Arma 3, TI is often unlimited FOV outside of the ENVG goggle, while NV also weirdly tends to be the limited FOV.  That's goofy, both from a game balance standpoint, but most importantly from a realism standpoint. 

Personally, I think the Viper helmet should also be weaker protection.  So, it's got the most overpowered imaging system AND the best protection?  Sort of laughable, and right up there with Arma's 150 round 5.56 dual drum magazine that only weighs 3lbs.  Why wouldn't anyone and everyone just use the Viper helmet with a 150 round mag all day long, then?

[Harzach 2505]

30 minutes ago, Reticuli said:

150 round 5.56 dual drum magazine that only weighs 3lbs

 

Where do you get that number? This was explained to you once.

[Reticuli 14]

2 hours ago, Harzach said:

 

Where do you get that number? This was explained to you once.

ACE Arsenal.

[Harzach 2505]

5 minutes ago, Reticuli said:

ACE Arsenal.

 

ACE is not an official BI product and the process by which the ACE team arrived at that figure is faulty. 

 

They have simply taken the "mass" value, which again is not mass as we know it but an arbitrary representation of an object's weight and volume (and which isn't consistent between different types of objects), and divided it by 10. A standard 30-round magazine full of 5.56 weighs just over a pound, whereas ACE would have you believe it's 0.8lbs.

 

Anyway, I have taken us off topic. If you want to discuss this topic further, let's do it in your other thread.

[Reticuli 14]

23 hours ago, Harzach said:

 

ACE is not an official BI product and the process by which the ACE team arrived at that figure is faulty. 

 

They have simply taken the "mass" value, which again is not mass as we know it but an arbitrary representation of an object's weight and volume (and which isn't consistent between different types of objects), and divided it by 10. A standard 30-round magazine full of 5.56 weighs just over a pound, whereas ACE would have you believe it's 0.8lbs.

 

Anyway, I have taken us off topic. If you want to discuss this topic further, let's do it in your other thread.

Well, however Arma and/or ACE is doing weight or mass to affect sway, aiming response, and aim drift, they need to figure out some way of letting bigger mags influence this more, because now it's mostly just the big guns that have this effect noticeably present, while you can add one of these mega-sized magazines to an AR platform and it has little to no adverse effect... you mostly just get desirable mass quantities of bullets.  Again, poor realism that interestingly causes poor game balance, which is why I brought it up in the context of TI goggles & helmets with a similar pattern.

[jgaz-uk 130]

To get back on topic;

 

Apparently the Thermal update was supposed to be ambient temperature related. this image was set to Dec6 2am is there some reason for this obvious thermal downgrade or are the developers too proud to admit they might have got it wrong?

 

[mickeymen 322]

On 8/28/2022 at 11:58 PM, jgaz-uk said:

Apparently the Thermal update was supposed to be ambient temperature related.

Ideally yes! But I think, initially before thermal repair, they did not plan to bind to the current temperature and time of year. I always dreamed about it, because this is how a thermal imager should work. 

Seems to me, at this late stage of Arma3 life in this they do not care about the correct operation of the thermal imager, they only care about a new thermal picture, not more.

On 8/28/2022 at 11:58 PM, jgaz-uk said:

this image was set to Dec6 2am is there some reason for this obvious thermal downgrade

obvious thermal downgrade? In fact, there is no obvious thermal downgrade in your picture. There temperature downgrade in curent landscape, but on the thermal imager the grass and trees are still burning.

Let's take a closer look. What do we see here? The grass still glows, the tree trunks also burn. the foliage on the trees is also light, but still thanks for not burning!

This is the wrong result for Dec6 2am!

Really, by Dec6 2am the grass and trees should be cool by now from even if not great, but the influence of the Sun. 

Since there is no sunshine for a long time and the night in December is cold. So the whole landscape must be dark, while the human body should be brighter than all this landscape. That would be the correct result for Dec6 2am.

On 8/28/2022 at 11:58 PM, jgaz-uk said:

developers too proud to admit they might have got it wrong?

I don't think they even tried to link to the current temperature/time of day/time of year.

All this repair of the thermal imager does not make any sense. I think so - It would be better if they directed their forces to other issues that are still present in the game, and the thermal imager would be left the same

Sad story

[Harzach 2505]

7 hours ago, mickeymen said:

In fact, there is no obvious thermal downgrade in your picture

 

It looks worse than before, that's what he means by "downgrade."

[KeyCat 131]

Just tested the new 2.10 update (briefly) and must say (like others) that I'm not impressed by the thermal upgrade 😞

 

Was hoping for something more like this....

 

https://www.dropbox.com/s/hjrzfssbxk4ox1a/DCS_Thermal1.jpg

 

https://www.dropbox.com/s/9b9t5o6faqmurl1/DCS_Thermal2.jpg

 

https://www.dropbox.com/s/rrf1ss210249iei/DCS_Thermal3.jpg

 

As it is now after the 2.10 update TI is just bad but I hope the dev's will continue to fix and improve on it. It would be interesting to hear the dev's view/reasoning about this topic and what issue/problem they where trying to solve with this update and how 2.10 is more realistic than previous version....

 

PS: I know that DCS is another "level of realism" but I'm sure you get my point.

 

[Reticuli 14]

https://dev.arma3.com/post/spotrep-00106
 

BI finally added parameters to adjust TI contrast, but IMO they ought to keep that while reverting the rest of the TI until they figure out how to gauge object temperatures more reasonably and also have objects respond to the ambient temperatures in a lagging way.  Again, I am absolutely not against the ideas behind attempting TI improvements, but other than 106 finally adding adjustable contrast, I will agree most of it wasn't yet ready for primetime.

Also see this thread for more discussion on this topic:
 

Anyway, I'm again going back to 208 RC.  Nice to see a quick response to some of this stuff, though.

[mickeymen 322]

4 hours ago, Reticuli said:

Also see this thread for more discussion on this topic:

 

My friend, this topic is no less detailed.

I created this before update 2.10 for only one reason - so that the BIS takes into account the current time of day and the current weather, because in my deep conviction these nuances should have a decisive influence on the result of the thermal imager.

In fact, I have only one claim to the new thermal image - Everything I have listed above has no effect on the thermal result. 

Many users are not happy with overly luminous parts of the landscape, but they do not specify why exactly they should not be in specific scene.

I sure, the luminous parts of the landscape can be, but only when they are heated by the sun and only at certain times of the day.

 

I ask all who is interested to leave a comment or discuss in this thread exactly the influence of current weather/current day time/current year time on the result of a thermal image.

[Reticuli 14]

Besides ambient temperature allowing objects to pop out more or not so that there's differing disparity between air and objects, making the TI more or less sensitive to differences, I'm wondering if perhaps the lagging object absorption and release of thermal energy could simply be tied to materials & objects' bullet penetration and durability in concert with prior conditions.  It wouldn't be perfect, obviously, but it would allow softer and/or thinner stuff like trees, bushes, long-deceased beings, and thin wooden walls to basically remain a kind of thermally transparent (at least across time) material that doesn't show much lagging infrared emissions change depending on prior time of day, weather, and ambient temperature. 

In other words, the soft stuff would effectively not be storing energy and releasing it later, but rather appear to change much more quickly or even immediately with the ambient conditions.  This is separate from if an engine is running or a warm-blooded being is alive.   If it's now sunset and the simulation says the weather's been clear and warm previously, then the hard stuff like tarmacs, pavement, concrete buildings, and deactivated vehicles would be giving off more stored infrared radiance for at least the next couple hours than most other objects, and in combination with the now-dropping ambient temperature would pop out visually more at least a little compared to those bushes.  Again, this is besides and in addition to if something's on fire, a vehicle's engine is on, or a person is alive. 

I'm just trying to think in terms of parameters that currently exist, right?  So, we know different materials have different durability and hardness in some way, we know weather is tracked, we know there's exposure to daytime, and ambient temperatures.  The hardness & durability characteristics utilized for crudely & efficiently modeling thermal energy absorption & release could even be modified based on the lightness or darkness of the object's color, and thus a black metal object could be influenced more greatly by daytime sunlight exposure than a white metal object, for instance. 

Just as I've been trying to think how ACE could do more intricate dispersion modeling and differences based on ballistic coefficients between ammo, I think the core Arma 3 might be capable of doing useful, quasi-realistic (or at least 'truthy' in a crude way) stuff with TI in the long run easily compared to some ground-up rework.  Nonetheless, prior to these novel changes, I still think simply reverting the TI to mostly the way it was in 208 while keeping this new manually-adjustable contrast is a good first move. BI seems to have a genuine motivation to improve the TI, even if this was partly a step back with 210.

Edit:

 

Info on future improvement paths to NV, TI, and smoke...
 

Standard smoke and dust can block night vision just as well as normal vision, but it takes rain, thick water-based fog, and heavy dust to at least partially block thermal imaging.  

 

Smokes using sulfur grenades and/or oil cracking dispensers alone are apparently almost completely transparent to the band of IR light that TI sensors detect.  
 

Graphite-based obscurant added to cracked oil dispensers can be very effective against TI, but necessitate people in the cloud wear respirators/masks.

[leonidmanylov 12]

Hello comrades. I have my own server, the thermal imager is too bright, is there any information on how to fix the brightness?

[jgaz-uk 130]

Still no change! expect the Developers are just waiting for everyone to get fed up of complaining about their screw up. Why they just cant change it back I dont know? how hard would it Be?

 

[Harzach 2505]

19 hours ago, jgaz-uk said:

how hard would it Be?

 

Have you submitted a ticket?

[Reticuli 14]

On 9/17/2022 at 10:11 AM, Harzach said:

 

Have you submitted a ticket?

I have.

[kolt.1911 0]

Why thermal can't see through smoke?

irl: https://www.youtube.com/watch?v=BlacCQZpYe0

Arma: https://steamcommunity.com/sharedfiles/filedetails/?id=2866165056

[jgaz-uk 130]

On 9/17/2022 at 3:11 PM, Harzach said:

 

Have you submitted a ticket?

Yes I did. still no change after last patch. :(

 

[Harzach 2505]

14 minutes ago, jgaz-uk said:

Yes I did. still no change after last patch.

 

Link? And it's not magic, there's a process.

[EO 11225]

23 minutes ago, jgaz-uk said:

Yes I did. still no change after last patch. 😞

 

It's also worth remembering that process is handled by a very small crew of developers now, have some patience 

[Reticuli 14]

On 9/22/2022 at 7:45 AM, kolt.1911 said:

Why thermal can't see through smoke?

irl: https://www.youtube.com/watch?v=BlacCQZpYe0

Arma: https://steamcommunity.com/sharedfiles/filedetails/?id=2866165056

I have also done a request for that, but in the meantime you can use Reshade and I guess turn it off if you don't want smoke transparency with just light amplifying or LLTV.