AV-8B GBU-38 Accuracy/Consistency Testing (LAR 100% x5, 01% x5, 50% x5, 25% x5)

[Tuuvas]

Hey all,

 

Since I've been using the GBU-38 lately, its inconsistency during regular gameplay was driving me crazy. At some points it was hitting my target square in the teeth, but at other points it misses by at least 10 feet. I finally decided enough was enough and decided to run a test with multiple % 5 times each at the same speed/altitude. Below are my results against a T-90 armored unit:

 

Launch Acceptable Range Percentage: 100

1. Destroyed, accurate
   
2. Missed, right 3/4 T-90 length
   
3. Missed, right/short 1/2 T-90 length
   
4. Destroyed, left 1/4 T-90 length
   
5. Destroyed, left 1/2 T-90 length
   

 

Launch Acceptable Range Percentage: 01

1. Destroyed, left/short 1/4 T-90 length
   
2. Destroyed, accurate
   
3. Destroyed, short 1/4 T-90 length
   
4. Missed, right full T-90 length
   
5. Destroyed, right/short 1/4 T-90 length
   

 

Launch Acceptable Range Percentage: 50

1. Destroyed, long 1/4 T-90 length
   
2. Destroyed, accurate
   
3. Destroyed, short 1/4 T-90 length
   
4. Missed, right 1/2 T-90 length
   
5. Destroyed, left 1/2 T-90 length (hit barrel)
   

 

Launch Acceptable Range Percentage: 25

1. Missed, right full T-90 length
   
2. Destroyed, accurate
   
3. Missed, right/short 1/2 T-90 length
   
4. Destroyed, accurate
   
5. Missed, left full T-90 length
   

 

Annnd that's it. Make what you will of the results... but I very much dislike how inconsistent this GPS guided weapon is turning out to be.

Edited February 11, 2020 by Tuuvas

[Captain Orso]

Cool, thanks for going through all the effort :thumbup:

 

Question: was the target pre-designated or TOO?

 

I would make only two suggestions:

 

1. Use Simple Range Script. It will tell you the exact number of meters each bomb hits from the center of the target. Along with the 476th's target range objects, they make doing such testing a breeze.

 

2. I have no idea how the percentage is calculated, neither in Real-Life™ nor by RAZBAM, but with the target always exactly on the fall-line all variation which might occur through the bomb having to turn in toward the target were there a lateral offset, are already eliminated.

 

I find the minimal variations between 1 and 100% more closely resembling general variations any weapon might experience in freefall, for example if you used CCIP and a dumb bomb instead. I cannot believe without real-world evidence that the difference between 1% and 100% is measured in inchest to a very few feet. There should be more difference.

[Zeus67]

Thanks for your tests.

The "percentage" indicates how close you are to the center of the LAR.

100 means you are in the center.

01 means you are in the borders.

 

Regarding accuracy:

The real weapon has a nominal CEP of 13 meters and a realized one of around 7 meters.

A T-90 is 9.63 meters long.

In your tests, the bombs are hitting inside the real weapon's nominal and realized CEP.

Remember that the bomb is flying and subject to DCS physics engine, even when there are no wind parameters at all. It will try to hit the center of the target spot but sometimes it cannot.

 

Regarding damage:

In real life all your targets would have been destroyed. Unfortunately DCS blast damage against tanks is basically nil, you need to hit it head on for it to be considered destroyed.

 

In conclusion, by real world parameters you are hitting the target dead on. The destruction, or lack of, has more to do with how DCS simulates blast damage.

[Tuuvas]

Thanks for your tests.

The "percentage" indicates how close you are to the center of the LAR.

100 means you are in the center.

01 means you are in the borders.

 

Regarding accuracy:

The real weapon has a nominal CEP of 13 meters and a realized one of around 7 meters.

A T-90 is 9.63 meters long.

In your tests, the bombs are hitting inside the real weapon's nominal and realized CEP.

Remember that the bomb is flying and subject to DCS physics engine, even when there are no wind parameters at all. It will try to hit the center of the target spot but sometimes it cannot.

 

Regarding damage:

In real life all your targets would have been destroyed. Unfortunately DCS blast damage against tanks is basically nil, you need to hit it head on for it to be considered destroyed.

 

In conclusion, by real world parameters you are hitting the target dead on. The destruction, or lack of, has more to do with how DCS simulates blast damage.

 

Thanks Zeus - that's fantastic info! I was assuming GPS guided munitions were the end-all to any other form of guided munition with dime-sized precision... but I guess that's what media would have had me to believe. I didn't once consider browsing through the wiki page for JDAMs, but everything lines up with what you said. Thanks!

[Harlikwin]

Thanks Zeus - that's fantastic info! I was assuming GPS guided munitions were the end-all to any other form of guided munition with dime-sized precision... but I guess that's what media would have had me to believe. I didn't once consider browsing through the wiki page for JDAMs, but everything lines up with what you said. Thanks!

 

Yeah no... JDAM's were initially envisioned as cheap replacement for LGB's with some advantages (don't have to see the target to hit it (clouds, bosnia)), and some real disadvantages (can't hit moving targets). Also the accuracy is further degraded if you are using only the INS guidance (in a GPS denied environment). But at the end of the day JDAM's weren't initially meant to hit "tanks" rather buildings. Also, as Zeus said, the current DCS damage modeling is "bad", but thats an ED thing.

[Konovalov]

Excellent painstaking test completed there Tuuvas. Also an excellent response from Zeus67 of Razbam. The video for me really does highlight that the JDAM is not some wonder weapon designed to hit a 50 cent piece every time (choose your own coin/currency). As Harlikwin rightly pointed out, JDAM's were a response to the fallibility and limitations of LGB's during the Balkans in the 90's and also due to the expense of LGB's.

 

Most importantly however is that this video demonstrates the inherent limitation of DCS World and it's simplistic damage model. This is immensely frustrating to me but it is what it is until ED think it is a worthwhile issue to address.

 

So all we can do is to work within the DCS World limitations. Want to have a better than even chance of killing a T-90 with a precision standoff weapon? Use a LGB or other laser guided weapon such as a Maverick or indeed a tv guided weapon. Want to kill a soft skinned vehicle or destroy a building from distance, then by all means use a JDAM.

 

Finally what I would suggest is to start a topic in the DCS World 2.5 section. It is this kind of thing that ED need to look to address.

[SGT Coyle]

Finally what I would suggest is to start a topic in the DCS World 2.5 section. It is this kind of thing that ED need to look to address.

 

 

+1 on that. Great work Tuuvas!

[Harlikwin]

 

Finally what I would suggest is to start a topic in the DCS World 2.5 section. It is this kind of thing that ED need to look to address.

 

You can post it but they already know. ED is slowly working on improved damage models. They are just starting with the WW2 units (don't ask me why, maybe fewer/simpler units), so, eventually I think it will get reworked, but as to when that is, I can guarantee it won't be 2 short weeks.

 

Really the main issues are with blast/frag/cluster damage for some thinner skinned units, and then "mission" kills, for tanks and the like when a big bomb hits nearby.