rnyboy

Thought I'd finish this topic with the results after upgrading to 64gb. I'm now running between 175 and 180 mods including MemGraph. The padheap.cfg is fixed at 16384 (probably could lower it actually) with "visible = true" and "applyPadding = true" in the settings.cfg . The total memory used was 82% with KSP using about 28 gb of memory when the image below was saved. I've seen as high as 87% memory used, so between 10% and 15% of total RAM is still free. I'm sure I could have quite a bit less free because that's a lot of free RAM for Windows and the other apps running in the background. To test the new 64gb setup I've played with quite a few craft using as few as 96 parts to over 250 from launch to final orbit. I also have an over 650 part Hab Tech2 ISS with US and Russian parts plus a Dragon-2 cargo ship docked to it in a 100km orbit around Kerbin and so far time between stutters is pretty much always longer than a minute and often more than two minutes. The stutter time for a garbage-clean is under a second and is typically not even noticed unless I happen to be doing something at the time. At the moment KSP is behaving very nicely after the new RAM upgrade. I haven't had a single hiccup or crash due to a memory overflow after playing for many hours straight. Edit: I meant to add that the total loading time is long, about 7 minutes, but that's a onetime thing per playing session. Edit #2: I just loaded up KSP1 and went to the VAB. Shortly after that I looked and % Memory was 88% and Ram used was 47gb. That's a new observed high but it typically drops down to around 80% after a while of playing and very slowly increases with time played. The following image gives the frame-rate and memgraph data while the Artemis Construction Kit's ICPS and Orion is in Kerbin orbit...

Memgraph sounds interesting. I'll have to try that. Thanks. (Edit: I looked up MemGraph and linuxgurugamer pointed to another mod called HeapPadder. It appears to be something like a MemGraph-lite. I think I'll give that a try first. 2nd edit to the 1st edit... Looks like MemGraph is a dependency of HeapPadder in that CKAN installed both. 3rd Edit: After installing MemGraph, and using it for ~30 minutes, I realized this won't solve the issue at the moment. The real issue seems to be bumping up too close to my max available 32gb of RAM and adding to the heap reduces that small headroom even further. My install of KSP1 is using up around 25gb of RAM at the start, so not small. I think I'm going to bite the bullet and double my RAM to 64 gb and then reinvestigate MemGraph because I think that should reduce whatever stutters I do get to a lower number over time. ) I really don't want to lower the texture quality because I like the looks and generally get good fps. I have an ~650 part ship (Habtech ISS with a docked Benjee10 Space Shuttle) in orbit and the physics slows from 1s in ksp time being equal to1s in real time to 1s ksp being less than 2s in real time. I can live with that. Shortly after liftoff from an alphamense launchpad a ~150 part BDB Apollo Saturn V shows around 30 fps the couple of times I've checked and once in orbit I get about 40 fps. I'm still kind of leaning to upgrade the RAM to 64gb too. I like to post process astronomy images and sometimes videos and generally the more RAM the better for that as well. I only upgraded from KSP v1.11.2 to 1.12.5 about two weeks ago because all the mods I was using finally had 1.12+ versions. I'm now in the process of trying to get orbiting craft, landers, rovers, and bases back to the way I like it. Like usual, after a major upgrade, many of my v1.11.2 craft files have missing or invalid parts and I need to redo the important ones.

Yesterday I added two additional mods that resulted in a drastic increase in app and craft load times and a lot of stuttering during play until KSP would lock up. I really don't want to mention the mods by name because I don't think they were at fault but seemed to be the "straw that broke the camel's back" in that before the two new mods were added I could get to the start screen in about 7 minutes and have about 5 gb of my 32 gb of RAM initially still left free and could work in KSP for something like six or so hours with very little stuttering until getting into the mid to upper 90% of RAM used and then things start to quickly bog down and I either quit and restarted or eventually get the "Unexpected Stack Overflow" error warning from Unity Analytics. When KSP is not running the system plus other processes uses around 27% of the available RAM. After installing those two new mods, one with a fair number of new parts and the other with increased animation effects, the load times went to about 10 minutes and about 98% of memory was being used right at the start screen. Loading up a craft would take a few minutes when it used to take maybe 30s up to 45s on a bad day. During the craft loading free memory would often be jumping into the few hundred mb numbers when before it would pretty much stay in the low gbs at worse. When in flight mode there was 98+% of total RAM used with quite a bit of stuttering and after about 15 or so minutes KSP would lock up. This morning, I uninstalled the two new apps and things went back to normal. The memory usage stays around 85% while in flight with no stuttering and smooth graphics with the physics time scale in the green. So, I'm thinking I may bite the bullet and upgrade from 32gb of RAM to 64gb considering what I've seen suggests I'm bumping up against a lack of free RAM ceiling. Just wondering if this RAM upgrade sounds about right based on the above recent observations or if there is a better solution that doesn't involve reducing the number of mods.

Thanks DeadJohn. As mentioned I have BDB and I have NFS but I didn't have Knes. I've added that mod to both my v1.12.5 and v1.11.2 KSPs. I just recently upgraded from v1.11.2 over to v1.12.5 in part because all the mods I used in v1.11.2 were now available for v1.12.5. I was surprised during this latest version upgrade that CKAN upgraded all the necessary mods properly and KSP v1.12.5 loaded and worked on the first try. As usual, for most of the major version upgrades many of my v1.11.2 builds have "missing or invalid parts" and now won't load so I've rebuilt the Starship from the Starship Expansion Project and the Space Shuttle (still loaded but redid it a bit) from the Shuttle Orbital Construction Kit and in the process of redoing Apollo, Gemini, and Mercury from BDB and Artemis from the Artimis Construction Kit. I "cheated" and Hyperedited up Benjee10's HabTech ISS build (usually I do a full rebuild of any included craft) along with the Tantare's Russian bits by Beale now slightly changed by me to be a bit more realistic looking. There's something like 650 parts now floating around up there with the ISS-plus build but the physics only slows from 1s being 1s to 1s now being about 1.5 - 2 s. Beale, Benjee10, Cobalt Wolf, and Kari, along with their various associates, have done super super jobs at what they have provided for KSP players. In hindsight, I think Squad should have used "HG" for any hypergolic fueled RCS/engines instead of MP. That would have been more "realistic" with no added burden on the developers. Too bad about KSP2 being essentially terminated. I didn't purchase it because I didn't feel like dealing with all the frustration from all the constant bugs, bug fixes, and the bug fixes creating new bugs. That's the rest of the story as to why I waited so long to upgrade to v1.12.5. Sadly, a lot of mod developer's efforts related to KSP2 will have been waste if it truly dies.

I've been playing KSP1 for thousands of hours now and over and over again I bump up against what many may consider a silly issue and that issue is we need more monopropellant (MP) engines. The reason being nearly all ships are boosted to orbit and often inserted into a proper transfer by LF/OX or LH/OX engines. But within about 24 hours or so everything transitions over to hypergolic engines because of their high reliability and no need for heavy/expensive cooling/insulation. The closest KSP has to a hypergolic fuel is MP. LCH4 is also a decent fuel for longer duration missions because LCH4 doesn't require Herculean efforts for longer term storage but LCH4/Ox engines are still not as reliable, Musk may disagree , as those using hypergolic fuels. I know, I know, it's just a game but I make as many service-modules and lander builds as I can using nothing but the few monopropellant engines generally meant for the space shuttle type builds from a few mods and sometimes a bit of LF/Ox when using fuel cells. Even for the Artemis Construction Kit, I've substituted the LF/Ox engines on the ESM with scaled monopropellant orbital insertion engines and buried a MP tank and a few of the spherical Space X 400 unit MP tanks to get about the same total Dv. One particular application for suitable MP engines would be for Blue Dog Bureau's Apollo style lunar lander instead of the two provided LF/OX engines. The descent and ascent stages can have their fuels switched to MP with a fuel switcher mod.

Thanks towermom9. Too bad that IRN and ER don't play well together. I'll have to give it a try. In essence I gather one just has to specify the target to capture and EasyRobotics moves the LEE to the target for capturing. Pretty cool. Today I've been modifying the Zaray, Zevezda, and Nauka included in Benjee10s ISS (2023) craft build for better RCS control and a somewhat (maybe?) better comparison to their real counterparts for looks. The Nauka required the most work and I had to do more cheating than I would like to get the COM/ACOM/ECOM close to overlapping and at the correct point for the more or less correctly positioned thrusters to function properly for translations. Of course, this has added parts and I think it's now close to 500 in total right now.

Maybe this is already known and being addressed. I posted this in a reply I made in the Infernal Robotics Next Connection Systems thread too as part of me not understanding how to get the LEE and grapplers working together. At least for me the kerboarm mini servo joint in the HabTech2 mod doesn't show up in the IR Editor. All the rest of the servos do show up.

Hello, yesterday I built the HabTech2 ISS and I was trying to get the Kanadarm2 and Japanese arm working. The LEE doesn't attach/dock with the mating grabber part. From reading this topic I guess I need the IK module, which I guess still isn't released, and I was wondering how things are going since it seemed to work fine in the above vids. Thanks for your efforts around this. Really neat stuff! Edit: Very strange, but in a good way. I'm not sure why, but I switched to a non-power transferring grappler and stuck it in the LEE while in the SPH. When I checked after launch the LEE had a latched grabber. I moved the Japanese arm in position to drop the grappler on the platform and it dropped. I reconnected and latched to it fine and dropped it back on the platform. I then went to a docked to the platform non-powered grappler and the LEE made the connection and latched. I undocked that grappler and lifted it up. Things seem fine but I need to try it with a power transferring grappler and see if that works because that was the problem yesterday and earlier today. I'll get back for a second edit if they work or they don't. Edit #2: Well, silly me. When I reread the info on the LEE I saw the 120-degree part for the powered grappler fixture. Sure enough, after getting lined up I started rotating the LEE and it attached. So, it looks like the Japanese arm and LEE works fine. I haven't tried the Kanadarm2 yet, but I imagine it will work too. Maybe this posting will help somebody else in the same spot I was in. Edit #3: The kerboarm mini servo joint doesn't show up in the IR Editor. I have no idea why, but the other servos show up there.

I'm really not quite sure were to begin here. I'll just start by saying I've been using KSP v1.11.2 as a beta under Steam with no problems and refrained from going to v1.12.3 and was more or less forced to when my v1.11.3 under Steam suddenly developed an issue I couldn't resolve, this just happened to be a few days before Perfect Division added their new launcher, so I moved up to v1.12.3 because all of the 125 or so mods I use were now available for v1.12.2 at least. Things seemed good for about a week under v1.12.3 until Perfect Division decided to add their unneeded launcher and, again, my now modded v1.12.4 threw a new "Houston we have a problem" warning about a serious .dll issue that I can't remember now. So at that point I decided to use v1.12.3 independently from Steam. However, and here is the BIG odd issue I'm having with MJ2 that and I can't solve, for some reason even my earliest saved v1.12.3 has the "Limit acceleration to" and "Limit throttle to" values linked. If I change one the other changes to the same value at the same time! I'm hoping somebody has had this same issue and can give me a solution to the problem. I really didn't notice this issue until a few days ago, but I save the full game about every week or so and this "linked" behavior is in my earliest save of v1.12.3 in early November. Thanks for any help or input here. EDIT: I just obtained a fresh install of KSP v1.12.3 from Steam and copied it to an independent folder just to make sure it works, and it does. I then used CKAN to add MJ2 v2.14.2.0 and its Module Manager v4.2.2 dependency and it still has the acceleration and throttle inputs linked in some weird way. So it seems to be an issue with MJ2 or Module Manger here. I hope this gives somebody a clue for fixing this problem. EDIT #2: Player.log link to Google Drive -> https://drive.google.com/drive/folders/19Ndy7NfR-3NoiPcGfV509SJDvczjXXDO?usp=sharing

Yeap, they "upgraded" the 1.12.3 I just switched to about five days ago to 1.12.4 and I get the warning about tweakscale. The first time I loaded the new 1.12.4 there was a kind of mild warning about tweakscale maybe not being compatible but when I loaded it up later after the game became unstable the first time it became a full out stop warning or you'll endanger your saved craft. Ticks me off because now even if I revert back to 1.12.3 in steam and add my CKAN mods it still doesn't work. I thought I saved my last 1.12.3 working version but it was apparently upgraded to 1.12.4 shortly before I saved it. I wish they would say when they are going to push an upgrade so I could save my last working version of ksp before it potentially gets fubared. KSP is soooooo frustrating if you use any mods or addons. Everytime they do just about anything everything seems go to pot.

Well sho enough... It took quite a while, maybe 10 - 15 s, but the ascent guidance staged and she took off by herself. Now I'm wondering if CobaltWolf designed his Regors to take about 10s to ramp up to full power since main engine start sequence for the Saturn F-1s was just after T -10s before lift off. If so, kind of cool, otherwise I don't have a clue as to why it taks so long to stage. Of course, I now need to retweak my SI and SII stages since that extra burn time of the Regors before actual lift off makes me use more of the SIV dV and that currently doesn't leave me with enough SIV dV for the TMI. Thanks so much OrbitalManeuvers for that key bit of info.

I solved it during rereading the first page of this addon. I missed that there were craft files and it turned out I was missing the Orion Stage Adapter, the "spacer" I asked about in my post above. If the Adapter would have been included in what I guess was the Janitor's Closet sorting things into the Artemis Construction Kit folder I would have found it straight off.

Really love this SLS mod, best so far. In the past I've kitbashed my own mostly stock version as well as using both the reDirect/reStock "Olympus" and the older huge 7.5m dia. SLS by Sobkos mods when running ksp v1.11.2. I now have my tweaked Block 1 done and I noticed even at 0% decoupling force on the Service Module Adapter that the CM/SM pretty much "rockets" away upon the decoupling. I looked inside the Service Module Adapter and saw that the top of the ICPS tank and plumbing intersects the Viking OME, so this is a parts collider issue during the decoupling that's causing the problem. Am I missing something here, like a spacer maybe as part of the build, or is this something of a bug that needs to be addressed with the ICPS cryo tank? I could kitbash with a hollow cylinder for a spacer but that would lengthen the package pushing it out of proportion to the real ICPS/Orion if my current build is correct. Edit... Well, that was easy. I didn't realize there were craft files in the "Stock" folder until I was rereading the start of this thread. I loaded up the stock Block 1 and saw the Orion stage adapter, which would be the "spacer" I mentioned up above. It wasn't in the folder that I guess "The Janitor's Closet" adds titled "Artemis Construction Kit". I do remember when I was building the SLS that the diameter of the ICPS when joined to the Service Module Adapter seemed just a might bit too small.

Now that's interesting. Thanks! I did let one hang on the pad for a bit over 5s one time thinking maybe it was taking a longer time to build up thrust and nothing happened, so I manually staged it. 15s is a looong time. Btw, was that 15s game seconds or real seconds? For me 15 game seconds is often something like 25s in real time. I'm pretty much always running under a yellow caution. Next time I'll let one set for at least 15s to see if anything happens. I could see 15s for clearing the tower. Apollo 15 took something like 13s to clear the tower, I shoot for 10s or so for that in my Apollo/Saturn V builds. Too much more than 10s to clear tower and the rocket will often start to tilt and then sometimes it's bye bye to rocket, pad, and kerbals if I don't hit abort soon enough.

I happened to find that this morning after I found out the ReStock SLS SRBs immediately decouple after leaving the launchpad. After that episode I went looking for anything new as far as the SLS goes in KSP and hit Benjee 10's mod. I don't know how I missed it the first time I looked at the CKAN mod list for v1.12. This SLS is working fine for me after I switched out the first cryoengine I used for the ICPS. Right now that SLS hits an 80km LKO using only 20 m/s of dV from the ICPS. That's leaving about 1300 m/s dV from the ICPS which seems very unreasonably high given it's only meant for a TLI to the Mun. I'll probably lower the fuel in the ICPS to leave more like 900 m/s dV for a Mun TLI and retweak the rest again. Next is to build the Block 1B and put a monoprop propelled habitat in the payload bay for docking with the Orion/ESM to make the trip into Duna orbit and then I think out or at least partially out of Duna orbit and back to kerbin. That should make the kerbals a lot more comfortable on the way to and from Duna. It's definitely not the root part, the Kane command module is the root. This has happened once before with a ship from another mod but I can't remember which one it was and I just kept manually staging it. If it was a Mechjeb issue I would think lots of others would have experienced this same issue since I can't be the only one using the Ascent Guidance to launch the BDB Saturn V. I think I'll try using only the Saturn V mobile launchpad and see if it stages properly with nothing else happening. Another think I noticed was that there are two green nodes under the center of the Saturn V. Does it make a difference as to which of those nodes the launchpad is connected too?

I just started to use this mod (v0.2.11) today to add the USA's, UNITED STATES OF AMERICA, and American flag to a BDB Saturn V. It initially looks GREAT in the VAB but when moved to the launchpad the decals aren't being displayed. When I revert to the launchpad the decals no longer display as well but the greenish shadow panel/box is there if you hover over their locations. Soon as you remove the decal the correct text shows again in the part to be placed but as soon as you place it again nothing is displayed except you can see that greenish panel/box again if mouse is placed over the location. It's weird because they appear to work OK in the VAB for the first addition to a build but after one attempt to launch they don't work after that. Any thoughts as to what maybe wrong here?

Thanks for the reply. Could be MLP issue. I just switched a few days ago to KSP1.12 from 1.11 and this was my first "rebuilding" so I thought it might be an issue with the new stock BDB Saturn V/MLP build. Next up is to start a rebuild of the reStock SLS. If lucky ksp 1.12 will at least load my old 1.11 one. It has the "Invalid or locked Parts" warning and they sometimes load, or if too much missing or changed, it just leaves me stranded for a full build again from the individual parts.

OK, I got everything on the Saturn V , CSM (with SIM instruments), and MLM with collapsible rover all working well for an Apollo 17 like mission. The dVs, TWRs, burn times (given this is a Kerbin launch), time to clear tower (10s), etc. are all quite decent with about 875 dV left in SIV for TLI after a 50 dV burn for perigee orbit circularization. However, I have one odd thing I can't seem to sort out... I'm using the new stock BDB Saturn V on MLP to start with, and with no changes by me, it won't autostage from the MLP using MechJeb 2 Ascent Guidance. I have to manually stage here. Was it planned this way or am I missing something silly again (most likely the case), like the not seeing the "Shielding" button to release the SIM bay panel when I asked for help a couple of days ago? Thanks again for any input.

Thanks, I already figured that out when I looked at the new parts. I was using the "stock" version of the new BDB Saturn V and saw the SM bay and panels for enclosing the LM wouldn't open. I thought something must be wrong so I checked the dependencies and started looking into issues with other mods after posting my problem with the SM here. I haven't gone back to check on what Zorg said but it looks like everything is likely OK and it was just another case of IOC (Insufficient Operator Capability). This is my favorite ship/parts mod. I certainly ranks up there with the other "best of the bests" like MechJeb2, KER, tweakscale, the various procedural parts type stuff, etc 2nd Edit... Thanks Zorg and Entr8899 for help/input. Sure enough, the SM "Shielding" button was the key. I was just expecting it to be the same "Open Payload Bay" as the last version I was using and didn't even notice the shielding bit when looking at the SM description. I've just been tweaking things to get a 80km LKO with having the S-IV engine just barely firing to hit the initial 80km apogee and hopefully using less than 150 dV to circularize. In the KSP v1.11.3 I was using my tweaked BDB Saturn V, CSM, LM with rover would hit the 80km orbit using only about 150 dV on the S-IV leaving about 900 dV for the TLI. At Launch it would take about 11s to clear the tower.

Hi CobaltWolf... This is a great mod and I can't even remember when I started using it. Actually, in my old v1.11.3, I made a small collapsible rover using two small hinges that connected three small panels. The first and last panel had the standard collapsible lunar rover wheels attached and the hinges folded everything down to pretty much the size of a small panel. Because there was no "same vessel interaction" I could then use the small long hinge and a small decoupler attached to the bottom of the middle panel of the rover to move the packed rover out past the side of the descent stage. After decoupling the collapsed wheels would just clear the landing legs and when extending the wheels the rover would essentially push its back end off the hinge/decoupler and freely ride down the landing legs to stop just a bit away from the LM ready for the two kerbals to use. I was quite proud that it actually worked out. Internally it fit pretty well, at least you couldn't see anything from the outside when packed except for a tiny bit of exposed hinge pin sticking out at the bottom of the LM descent stage side. That was kind of nice because I didn't have to assign the extraction phase to an action group.

I finally got around to moving over to KSP v1.12.3 on Steam from v1.11.3 due to some kind of Steam issue when, for some unknown reason to me, they updated my v1.11.3 in the Steam Common folder. Luckily, I saved that last KSP folder the day before that Steam issue occurred and have been running KSP independently from Steam for the last few days. For some reason I can't open the bay on the Kane-11 Service Module and because of all the changes to parts I can't load my old BDB Saturn V where that was working just fine. I'm running 137 mods but I was running nearly all the same mods in v1.11.3 with no issues. I did a search for this issue and came up with nothing to help and I thought I'd try here in case this has happened to anybody else. The new Kane and Sarnus models are really nicely done, and I like the corrections for things like where and how the EVA and Docking lights go and that the Ullage motors on the S4 stage now decouple after use. In my older KSP v1.11.3 I was using the small decoupler for the SubSat to decouple the S4 Ullage motors. I still need to check if there is now a bay on the descent stage of the LM to place a microrover. Thanks in advance for any help.

So I've been setting up a Munar base (Base Armstrong but AKA Hamster House due to how it looks with the tunnels connecting things) and I have a grabber jr. on the front of a Buffalo equipped with a Buffalo ISRU set for Ore>LF and Ox and two packs of the Buffalo drills. The drills are stuck on harvesting alumina so the ISRU is missing ore. When I attempt to change resources on the drills it says I need to unlock the Biome but I have no idea how to unlock the Biome. I was just hoping the drills functioned like the stock drills do and harvesting ore was the default mode. So how do I unlock the biome AND get the Buffalo drills to harvest ore? I searched on "KSP how to unlock the biome for the buffalo drills" but didn't get any useful hits that helped here. Edit: So I just noticed the small "Ore" container I have under the Buffalo is filling up with "Ore" when the drills indicate they are harvesting "alumina". The ISRU says it is missing "Ore". So I assume the "Ore" in the container is actually "alumina" or else something is screwed up somewhere if it's actually "Ore" and the ISRU isn't utilizing it to make LF and Ox. Any help here is greatly appreciated.

To Linuxgurugmer... First off, thanks so much for all the work you put into KSP. Love the game but absolutely hate all the mod issues at every KSP version change though. Seems like every time all the mods are working, or mostly working, a major version change comes along and it's all FUBAR again! Really this post is about the SpaceX ITS ship though. I've been fooling around with the build a couple of days now and for the most part managed to attach thinks to make it able to not just hold kerbals but to allow them to do an EVA as well. I also added a covered CoT Mr. to allow for docking. EDIT BUT, the largest landing legs designed for the ITS Ship in this mod are too short to have both proper mounting in the three leg pylons, because the ends of the flaps extend well past were the extended landing legs end. (Well "stupid" applies here, or maybe more like BLIND... I looked for built legs in the VAB to see what they looked like and while landing several times and didn't see them, and now, of course after posting on this topic, I finally saw they were built-in. Sheesh!). Oh yeah, the ITS doesn't have any kind of payload bay, maybe an equipment/payload package is supposed to be sent to Mars first and it wasn't meant for any payload but I thought I'd bring the topic up anyway. I have used one of the Dragon capsules to make a 2nd stage ascent or descent lander CoT Sr docked to a 1st stage with three radially attached spikes for descent/ascent also and a storage bay with a small rover (I think stock or mostly stock) and two symmetrically attached uISRU (Wild Blue Industries)/drill/ore-tank/radiator units to allow for self-contained refueling. Both stages can be piloted remotely and used independently with the 1st stage having top and bottom CoT Sr ports and the 2nd stage having a bottom CoT Sr and top NASA docking port. It's heavy when fueled up but it makes a decent multi-mission reuseable lander with a rover and how it gets used depends on DV needed and if an atmosphere is present. In case of emergency the Dragon is capable of being used as a typical heat-shielded reentry vehicle. The lander can be used as a complete CSM/lander, or as I tend to do, mount it like in the Apollo program below a separate CSM. I setup the final payload stage to be used as something of a refueling ship that I keep redocking the CMS and/or lander too and, if done carefully, it can be used for interplanetary/moon transfers and refueled from a separate refueler when near empty.

FANTASTIC!

I was extremely happy to see this mod again for the latest KSP. I used it a lot in v1.8 Before I had the BDB Leo-Medicis (Gemini) Service Module (SM) I would add four Sepratrons or other small engines in a four fold symmetry centered at the back of the Service Module and then offset those to the inside front of the SM and then fire them off in opposing pairs to do the deorbit burn as was done back in the day for project Gemini; I also did the same thing for your SM. So I was wondering, since your models are so freakin' great, if you could actually at some time add the four proper red spherical retrorockets on an X-frame in the SM as was done in real life? This would raise the level of realism, at least I think it would, for your SM. Thanks for at least considering the idea if you haven't already and dismissed for some reason as being too hard to implement.