[FIXED] M-61 Vulcan and Gau-8 Avenger dispersion values

In terms of a hud tape (as i've shown earlier in this thread) I use it to get a few details; firing range and the apparant dispersion when those shells hit / and the range whey I see the hits at. This then allows me to compare what I see in the video to DCS.

I wish you luck but I don't think you are going to get too much further on this one.

That's one of the problems in many forums. One can have detailed knowledge, data etc. but all this will very seldom change the mind from people who have a fixed mindset, like nighthawk2174.

This can be frustrating, but don't let this keep you from posting valuable first hand info :)

In terms of a hud tape (as i've shown earlier in this thread) I use it to get a few details; firing range and the apparant dispersion when those shells hit / and the range whey I see the hits at. This then allows me to compare what I see in the video to DCS.

Ok, I will keep trying.

You can't use a hud tape - zoom in or what ever you do and count pixels??? and try to rewrite the laws of physics.

Again, I am more than happy to give you more data about how the gattling system can result in uneven barrel wear and how the aircraft is effected during the firing of the gun.

You can't just post into the "Bug and Problems" forum with a theory you have, and then ignore the reply of every person who is an expert on the subject replying that you are wrong - it works each way - ED is not going to rewrite their code based on a theory you have developed.

I am happy for you to ask me detailed questions via PM but this thread needs to be closed - you are flogging a dead horse mate.

I apologise to everyone for my abrupt reply to this thread.

Gunnie

Ok, I will keep trying.

You can't use a hud tape - zoom in or what ever you do and count pixels??? and try to rewrite the laws of physics.

??? No i'm not, stop strawmaning my argument.

You can't just post into the "Bug and Problems" forum with a theory you have, and then ignore the reply of every person who is an expert on the subject replying that you are wrong - it works each way - ED is not going to rewrite their code based on a theory you have developed.

A) This isn't a theory:

https://forums.eagle.ru/showpost.php?p=4050741&postcount=41

https://forums.eagle.ru/showpost.php?p=4106908&postcount=80

B) me and habu (active duty A10 piolt) are good friends and we are in full agreement that the dispersion values for the guns in-game (gau8 and vulcan) are way too high.

C)Its not a code rewrite its changing a few numbers in one lua file...

edit:

after looking through some F15 manuals found this:

(100% circle as far as I can tell)

I tried to stay away and out of this thread but I have to say something. I'm not saying nighthawk2174 (or his pilot references) are correct or incorrect but......

And other active duty peeps disagreed, so what does that mean?

It means someone is wrong. Maybe Wag’s buddies or nighthawk2174 & habu. The big question is what makes one set of pilots opinions have more weight than another? Did Wag’s pilot references any data or do any testing or did they “feel” it’s correct? How many times in these very forums when someone said “I feel XYZ is incorrect” only for the response to be from ED was something like “without objective evidence we feel it’s correct as is.”?

I know first hand my discussion with pilots in my Reserve unit (windmilling A-10 engine as one example) was dismissed as they’re wrong not to mention my own experience on the A-10 as a crew chief being disregarded because I didn't quote specific tech data references (flaps being lowered with no hydro power or the nose wheel turning to the side by itself with no hydraulic power). We've had multiple examples over the years where verified real world A-10 pilots and/or maintenance (crew chiefs & weapons) state things were incorrect only to be dismissed or flat out told they're wrong.

So, my question to ED is why do pilots that feel something is correct when Wags asks them out way another person's pilot reference or validated by people with real world experience?

??? No i'm not, stop strawmaning my argument.

 

 

 

A) This isn't a theory:

https://forums.eagle.ru/showpost.php?p=4050741&postcount=41

https://forums.eagle.ru/showpost.php?p=4106908&postcount=80

B) me and habu (active duty A10 piolt) are good friends and we are in full agreement that the dispersion values for the guns in-game (gau8 and vulcan) are way too high.

C)Its not a code rewrite its changing a few numbers in one lua file...

 

 

edit:

 

after looking through some F15 manuals found this:

(100% circle as far as I can tell)

Does the F-15's gun in DCS have 8 mil dispersion?

Does the F-15's gun in DCS have 8 mil dispersion?

All the vulcans are the same values. For now their all based off of the same LUA file: scripts/Database/Weapons/shell_table.lua

Ok so i've done in game tests at a 1000 inches to allow comparison with the project vulcan documentation I showed in my first post. Hopefully this can be the new standard for all testing in this regard as it is what is most often used in testing. I set up an F18 at 1000" (83.333 ft) with active pause on then fired at the target (same as in my last analysis post). After this using tacview I extracted the initial and final positions of each bullet. Each test was done with 100 rounds. I ran this test twice with the third time using the value of .0008 for the dispersion. Units for the graphs below are inches. Additionally I then used the data to figure out what the 80% and 100% circles where for each grouping.

 

Vanilla results:

Spread:

 

 

- Vanilla Trial 1

- Vanilla Trial 2

- Modded

 

 

 

Green = 80% within 10.04mills

100% within 25.78mills

 

Red = 80% within 9.45mills

100% within 22.86mills

 

Orange = 80% within 4.27mills

100% within 10.18mills

 

When all of them are overlayed together along with dispersion pattern 11 you get:

Adding this as this is additional documentation on this matter, thanks to MrTheOx on reddit for this:

This may be of some use or you may have seen this. The DOD freely publishes the specs and standards for both the ammunition and gun mounts. This includes accuracy data.

 

https://books.google.com/books?id=SxOqvylyru8C&lpg=PA5&dq=MIL%20DTL%2045500&pg=PP1#v=onepage&q&f=false

 

The DOD will not accept a M61 from a contractor unless it does the following.

 

“At a range of 1000 inches, 80 percent of a 100 round burst shall be completely within an 8.0 inch diameter circle for accuracy; and the center of the 80 percent group shall be within 2.5 inches elevation or depression, and 4.75 inches azimuth right or left of the true boresight point for targeting.”

 

http://everyspec.com/MIL-SPECS/MIL-SPECS-MIL-DTL/MIL-DTL-45500E_44023/

 

The ammo on the F-16, F-15 and F/A-18 is the PGU-28A/B.

https://www.gd-ots.com/wp-content/uploads/2017/11/20mm-PGU-28-A-B-F-16.pdf

 

Per the spec, Mil C 85717:

 

“The average mean radius of all target impacts resulting from sample cartridges, conditioned at 68 to 72° F, shall be not. greater than 15 Inches at a range of 500 yards.”

 

http://everyspec.com/MIL-SPECS/MIL-SPECS-MIL-C/MIL-C-85717_AMENDMENT-1_25748/

 

Even the M163 the self propelled air defense Vulcan is more accurate than the in game vulcan. The M163 has It’s accuracy listed at 5 mil for the circular clamp.

 

https://books.google.com/books?id=UAQYAAAAYAAJ&pg=PP10#v=onepage&q&f=false

 

According to Yo-Yo the M61 has a dispersion of 17.5 mil. https://forums.eagle.ru/showpost.php?p=3944105&postcount=40 .

 

The only plausible explanation I see for the current in game rating is, That they built the shell table with the air defense version of the Vulcan, the M163, in mind. Which does have a muzzle clamp that makes it fire a 18 mil wide oval, though it’s only 6 mils high. But that shouldn't be the rating for all of the M61 based platforms. It was a special one off for the Army.

 

You can see by some of the Phalanx literature posted earlier that the gun's accuracy can vary based on application and should be further broken in the shell table to reflect that.

Before my post I just want to add more data to the pile, thank you too MrTheOx once again for more information.

Just adding the detailed spec sheet for m61 and m61a1.

1996 spec http://everyspec.com/MIL-SPECS/MIL-SPECS-MIL-DTL/download.php?spec=MIL-DTL-45500_1A.049673.pdf

 

1974 spec http://everyspec.com/MIL-SPECS/MIL-SPECS-MIL-G/download.php?spec=MIL-G-45500_1.049672.pdf

 

The difference in terms of accuracy is, the new spec calls for a 75 round burst. Opposed to the older spec which called for a 100 round burst.

------------------------------------------------------------------------------------------------------------------------------------------

After discussing with devs once again, we still feel what we are using in-game is the best representation of these weapon systems.

 

For example, we do use 8 mils D80 for the gun, with some additional mills for simulation of an active duty gun, that said, in the above tests, we cannot say that the errors shown are due to these additions, or test errors.[/Quote]

 

Well first off the Vulcun should have a D80 of around 5-6 mills, the avenger D80 at 5 mills. Additionally a full rebuild is supposed to be performed if the gun's dispersion goes above 20% of this value meaning it should never exceed 7.2mills (documentation is 8mills) for the Vulcun and 6 mills for the D80 for the Gau8. So using 8mills and then adding some to account for use is already wrong.

 

Now though... onto the claim it is 8mills in game this can be proved wrong quite easily. Although since I already did tests with the vulcun above i'll do the gau8 today.

Lets do some more tests!

 

Now from the HUD

--------------------------------------------------------------------------------------------------------

 

If this is the case then this should result in a 8 milliradian circle yet the testing does not show this. Feel free to do you own in game testing as well. But lets give you the benefit of the doubt here lets say it is 8 milliradians. Which is still too high as normal maintenance should keep it far closer to the 5 milliradian value. The spread of the remaining 20% rounds outside the 80% circle is still too high. Additionally the GAU-8 should have an 80% circle of 5milliradians. And documentation for the M61 puts the value in the 5.5-6 milliradians range, which coincidentally enough is the value mover gave for the dispersion.

 

It can not be denied that the 100% circles in game, are without a doubt, too high. It is easy to see, at a 1NNmi or a ~6000ft you should get a 100% circle diameter of ~24.08m or 79ft. Yet as you can very easily see for yourself in game this is not the case. As documentation indicates for a M61 this should be in the range of 10 milliradians. Confirmed by Habu23 the spec for the Gau-8 is 13 miliradians for the 100% circle. Beyond this from testing video and documentation it is clear and easy to see that the majority of these 20% of rounds that fall outside the 80% circle should fall *JUST* outside the circle. With one or two random rounds setting the 100% circle value.

 

The .0017 pointed to in the code is not to be taken at face value either. We do have some formulas and math done inside the sim, and in many cases, this is not open for editing or viewing. So while you can edit the .0017, and it will make it "better", chances are you are making it better than reality. And though 0.0017 is proportional to D80 it is not direct D80, D50 or D100. The probability requires its own math to obtain proper comparison as well as test result processing. We use this math to convert any form of available dispersion representation to the values we operate.

[/Quote]

 

Although due to a lack of a understandable or useful answer from Yo-Yo one can't really easily base what is happening in game based on the shell_table.lua. This is why in my reddit post I didn't even quote the actual numbers used in the file I just looked at what it is doing in game.

 

That in terms of actual angular size the dispersion in game is too high. You can clearly see this with the F18 tests and the A10 strafing video.

 

As far as SME's go, and testing with real pilots, Wags has spent many years building relationships with pilots and such across many areas of the military, with this, ED has been putting workstations in at different bases, and active duty pilots are using these as a training tool. So not only has he asked for opinions of the gun modelling in the sim but if this is to be used as a training tool to any extent, Wags would get feedback if there were major issues in areas like this.

And i'll bet a decent sum of money when you ask about dispersion the value the'll give you is 5mills and 13mills, and for the vulcun something like 6mills and 10mills. +- a bit depending on the jet. F15C for example:

 

Additionally what about CIWS? It is my humble oponion no reasonable person would call DCS an accurate representation for the CIWS guns. Any news on that being looked at?

 

In conclusion I have become convinced this is what happened:

According to Yo-Yo the M61 has a dispersion of 17.5 mil. https://forums.eagle.ru/showpost.php...5&postcount=40 .

 

 

The only plausible explanation I see for the current in game rating is, That they built the shell table with the air defense version of the Vulcan, the M163, in mind. Which does have a muzzle clamp that makes it fire a 18 mil wide oval, though it’s only 6 mils high. But that shouldn't be the rating for all of the M61 based platforms. It was a special one off for the Army.

 

You can see by some of the Phalanx literature posted earlier that the gun's accuracy can vary based on application and should be further broken in the shell table to reflect that.

I have my doubts, serious serious serious, doubts that the bullets not hitting exactly where they should is what we're seeing. And even if we compensate for the massive difference seen. What is the difference? A couple of inches, a foot? That still can't fix how off the dispersion is, we're talking in terms of 100's of percent off not 5 or even 10. In particular for the CIWS vulcun. For the 100% Vulcun and the 80&100% Gau8.

I already did tests where I pulled data from tacview showing how off dispersion is. Beyond that there is plenty of documentation and video evidence to compare too as well.

Lets take this again:

And lets also assume that .0022*8*1000 is infact an accurate representation of the 100% circle... well that's still a 57% difference in 100% circle size for the vulcun. And with the dispersion from my tests that's 220%+ difference. Now with the above stats from the 15's -34-1-1 those numbers rise to 220% and 275%+ respectively. And if we take the CIWS dispersion to be sub 1mill lets call it .8 that's a difference of 2200% and 2750+% respectively.

More pics:

Images from "A comparison of the USAF Projected A-10 Employment in Europe and the Luftwaffe Schlachtgeschwader Experience on the Eastern Front in World War Two"

Well, there's some movement (or, perhaps more accurately, less movement :D) on the topic at least:

Dear Sir, I have personally followed up on this point of dispersion and our engineer made a general mistake pertaining to gun installation rigidity. We are correcting this in those airframes concerned in the coming weeks. On a more general note to all who have valuable information to share, please know this: not all of the information which is written on forums is of a world class nature however we do try to follow up on most issues, especially when they seem plausible, but the nature of the beast is that some if not most software engineers are just human and don’t like their ‘babies’ to be criticized. Furthermore most of them are highly educated and it’s not always easy to get them to verify their stuff in great depth. However, rest assured that I have informed every single one of our guys and gals that they should always listen, analyse and verify and if we make mistakes then TAKE OWNERSHIP, thank the source for the insight and move forward, safe in the knowledge that we have made a community member happy and the program better. Thank you all for your continued support and passion. Yours sincerely Nick

I am really not gonna say much more on this, I discussed all this with Yo-Yo, went over the statement with Wags, the team is happy with what we have.

 

I will point out once again how unscientific using screenshots from the sim are in determining something is wrong as I showed here:

 

https://forums.eagle.ru/showpost.php?p=4099326&postcount=52

 

If you read that post, the effects are not as perfect as real life, they are not a perfect representation of the exact point of impact. Using them as a scientific reference isn't really working. I suggest sitting on this for a while, and let's revisit when the new DM is in the vehicles, in that case, tracking of rounds and points of impact, entering and exiting, etc are supposed to be better tracked, it will give a much better idea of what is going on. Right now using the in-sim effects as an actual measurement isn't probably helping things... and trust me I have stared at a lot of bullet hits over the past year+ when it comes to impact points, effects, etc.

 

Unless Yo-Yo, Wags or another dev want to add more, I have really nothing else to give on this.

Well, do you have anything more to add now?

And many people are bad shot, sorry to say :)

Please record your gun runs with joystick axis window open.

I'll bet you will see a lot of pitch movement :)

I'm guilty of this but practice makes perfect :)

In all of my tests I used active pause to ensure no movement macedk

And my response stands as well it could very well be the case but what if it appears to be 2-3ft less than it actually is as well. If it is 3ft more (worst case), that would only reduce the difference between what is seen and what should be seen by 100% instead of 135% (M61 as an example). Not to mention the in game tests and the extracted data showing the actual dispersion values that I did here.

On another note what has the devs said on CIWS regarding dispersion and the fact ciws uses a different round the MK141 which is going to have not only increased inital velocity over standard M56A3 rounds but it being a saboted round and of a different shape should have better balistic characteristics as well. Not to mention the tracer color as well.

I personally have zero issues with the gun strafe modeling, the dispersion with PGU bullets or the damage effects modeling. I think its very cool as well as seeming to be very realistic.