Thank you for your time and effort of these test! I don`t want to spend money on FS Traffic, as i find the freeware ones completly fine. I would choose FSLTL over the AIG, because it has similair performance (as we see in the vid) and FSLTL is much easier to set up.
They need to work hard to make it much better as is it now . For me now in not worth at all . All videos show just the beginning . After some time all airplanes are gone and the airport is empty etc . Still same issue with go around etc because is limited by asobo . If you set up FSLTL properly is much much better then FS Traffic . I can write much much more about how bad it is I mean FS Traffic
@@michakwiatkowski4527 This is not my experience. There is a slight initial rush after loading in and that is due to the Asobo system not the addons. After that the traffic departs/arrives based on the schedule in the flight plans. In my FST video some of the time lapse clips were nearly 45 mins long and traffic was still going strong then for example. This video wasn't a comparison of that however. I'm aware of how to set up FSLTL properly - I've been using it since well before it was released to the public, this test was about trying to make things fair across the board. Also what may constitute 'better' can be a subjective thing 🙂
@@BobbyFuzzy did 1 flight with FS Traffic and for now FSLTL work better for me personal . One think happend to me that never happened before with Fenix . Lost connection with Fenix all systems . Tablet ,mcdu etc not working . Never happened to me with any addon .
I think JF did this comparison on their Facebook but that being said I would be interested in seeing those results, would be an easy test too, perhaps one for a short later in the week!
@@Minon157 Entirely up to you! You can match the FSLTL injector to similar levels of the Asobo system too and likely see the same benefits. The problem with the Asobo system is it leaves very few parked aircraft in the right places which makes airports feel quite empty.
Interesting, but there might be some issues with the test. First, Dublin is a medium-large airport. I have noticed major differences in large-very large airports (LFPG, EGLL, EDDF, etc.). That's where I saw FS Traffic truly shine between AIG and FSLTL in terms of performance and stability. 2. AIG, when fully installed, offers much more airlines than FSLTL and FS Traffic. I get that you chose AIG easy install, but there is a difference when you go "all out" with them (and what a pain it is to get that installed, which should be underlined). 3. Your system specs will make a big difference. For someone (like me) who struggles to keep framerates at around 30 in large/very large airports with AI, that's where those extra framerates will really shine for FS Traffic. If I had your system, I wouldn't bother with FS Traffic since the value wouldn't really be there. But with my system (i7 7700k, RTX 2070, 32GB Ram), going from 25fps with iniBuild EGLL with either FSLTL or AIG, to 32fps with FS Traffic... And that's 100% traffic with quick turnaround option. That's impressive to say the least. 4. Software stability. That should also be underlined. AIG is the worst of the three and is very susceptible to crash, especially with GSX running in the background - not withstanding it's load time... FSLTL is rock solid stable, and loads up fast. FS Traffic, very stable, doesn't need an injector, but you need to keep an open eye to sim objects limitations (maybe because I am running at 100 with FS Traffic, whereas AIG or FSLTL I couldn't). ...also, unless I misunderstood your stats, but seems to me that a decrease of 15% framerates with FS Traffic vs 25% with FSLTL or AIG... that's a significant statistical difference.
Hey Phillipe! Thanks for your comment. 1. The test still shows FS Traffic on top regardless but yes I'm sure different airport may give slightly different results but remember these results are based on a normalised injection and FS Traffic injects a lot less by default than FSLTL. 2. This wasn't a test of the amount of airlines etc on offer or a usability test. 3. Of course there's going to be difference given a different system and there's more value to be had on different machines - perhaps I should have touched on that in the conclusion but the graph does speak for itself. You see the same increase in performance as I do. 4. Again this wasn't a test of stability or any other factors than performance. There's always going to be variables one can't account for in one test on one machine in one scenario but the idea was to try and remove as many variables as possible so that at least in this one scenario to give a fair test. Definitely not the be all end be all but it does line pretty much with what I expected the results to be and I imagine the overall conclusion would still hold up given a different machine which I think you proved with your comments! Thanks again!
Thank you for sharing the info and the video. Great video!!! I can't fly without AI traffic independently of who provides it,it's get boring up there without any aircrafts or any atc communications around. Thanks!!
This video was way more complicated to make than I ever would have thought! By no means conclusive but interesting nevertheless! Bear in mind different locations may product different results. What do you guys think to my test?
after reading so many different comments on performance it's great to have something a little more thought out in hand for comparison. I don't think performance alone tells the whole story obviously but this information is very helpful!
I wish someone would make one of these that focuses on smaller GA aircraft at small airports. Cessnas, pipers, bonanzas, and such. There seems to be two camps of flight simmers. Those who almost never fly airliners, and those who never fly anything else. All the traffic addons seem to cater to the bus drivers.
JF are planning to add GA and FSLTL are working on that. That being said the majority of planes in the sky are in fact airliners so it makes sense to work on this first for these types of addons.
@@BobbyFuzzy I appreciate your reply, but, unfortunately, That's not even remotely correct. At least not in the United States, anyway. I'm not sure about worldwide, but in the U.S., from 2005-2022 recorded numbers of active "for-hire air carriers" were in the 6000-7000 aircraft range, annually, over that time period, while active "General Aviation" aircraft numbered in the mid 200,000 range every year over the same time period. I posted this comment earlier with a citation link, but unfortunately RU-vid removed the comment because they have some stupid policy on adding links to your comments. It is pretty easy to find the recorded statistics though.
Apologies - my statement is more based on my impression that facts, but that being said these addons cover the whole world not just the US. Another consideration is the availability of data for commercial flights vs GA.
@@AV8R_1 You got me interested so thought I'd see what I could find and I found on the FAA's website that they are handling 16m flights a year, 10m of which were scheduled passenger flights? I'm not sure more aircraft equate to more flights based on your info?
@@BobbyFuzzy if by "handling flights" they mean flights that are conducted within the ATC system or operating on a filed flight plan, it is true that the majority of those are IFR and commercial aircraft flights, since that is a legal requirement of flights conducted under those conditions. The majority of small GA airplanes operating in VFR will not even file a FlightPlan or use much ATC in route, so those flights would not be "handled" by the FAA. The statistics I quoted were based on simply the number of aircraft known to be actively flying in the United States per year. So that would include aircraft with active registrations, and not aircraft that are sitting derelict on airports.
Bobby, two questions if I may, when running MSFS normally what graphics settings do YOU personally use and which AI system have you settled on now seeing as it's 5 months later? Also, are there any resources to try and help set up fsltl with roughly the same traffic density settings as Asobo lives traffic? Hopefully you're able to answer these, I'd be very grateful. Thank you.
Hey Gregory! I use a mixture of ultra/high/medium you can see what I use here - ru-vid.comUgkxgmiLnS0Thx4VRT-aJL56WRaJaFGFQjYS In all honesty I don't use AI traffic all that much as I mainly fly on VATSIM. That being said I use the FSLTL injector with the JF Models prioritised when I do. I've not really used Asobos live traffic too much of late but I know that doesn't inject any parked traffic so you could start with that at zero. It also doesn't do VFR so again stick that at zero. It does use live traffic data similar to FSLTL so it's probably just a case of playing with the IFR density and radius.
I can't seem to get aircraft on arrival - had it on departure, have setup my arrival ICAO in the in-sim menu.. still no traffic on arrival. Had nothing en route either mind you.
Might be worth dropping JF a line. Not sure on the exact troubleshooting steps with that as I've not had any issues with arrivals. Do you have LittleNavMap? Very useful for debugging!
Hi - not necessarily but I would not run both the injectors at the same time. That being said there's not much to gain in having them both installed unless you are going to manually add the FSLTL models to FS traffic in the control centre.
I've not noticed much of a difference with FS Traffic vs FSLTL stutter-wise. I think the results speak for themselves, the 1% and 0.1% lows seem pretty relative across the board except with AIG meaning that stutters are not really any different between FSLTL and FS Traffic. Yes I do, we streamed it a couple of days ago with FS Traffic model matching on VATSIM. 🙂
@@BobbyFuzzy - well for me that FStraffic is an in-game thing and not a separate program is interesting. I wonder if they won’t keep optimizing it. 7-8 frames for me is a lot. For performance would you recommend FBW or Fenix? I am reading that even though the Fenix is fps heavy, the stutters are not as bad.
Yeah I really liked that too! I don't really fly the FBW enough these days to comment on that but I enjoy the Fenix the most regardless of its performance. It's such a great aircraft!
I like how FSLTL is a fallback for fs traffic if u want. Like fs traffic doesn’t have the 767 and airlines like LATAM. hopefully the fallback system will put it in
+7 FPS... I've seen people upgrading their GPU for a few hunderd bucks (including me....3080 to 3080ti) to get to there... now you can get it for 33 euros. FS Traffic is my choice.
P.S. Could you please make a video explaining how to have FS Traffic as the main traffic addon and then FSLTL as a fall back? I'd also be interested to know how you could get GA traffic working with it too.
It's quite a process if you wanted to do that, I showed a short clip of that in my FS Traffic overview video! There's some info on it (not specifically related to FSLTL) telling you how to add external models to FS Traffic, too.
I think all of these AI addons are actually limited to the mostly wrong aircraft stands on all airports in the sim. If the devs pay more attention to parking accuracy, there would have been airports full with AI aircraft.
Depends on the airport scenery used - parking stand sizes and codes are set by the author of the scenery. In the case of Asobo's sceneries, many are set inappropriately or not at all leading to poor results with AI.
Comparing average FPS against no traffic. 15% reduction for FS Traffic. 25% reduction for FSLTL. That’s a huge difference. You could argue that it’s not noticeable in normal gaming, but statistically it’s huge.
I did briefly touch on that but at the end of the day this was a comparison of the 3 addons. The majority of people interested in this topic aren't interested in running with no traffic so I don't think it's worth dwelling on that. Further there's a tonne of people that cap the sim normally at 60fps (myself included) or even 30, making the difference negligible.
Thanks for the comparison! For FSLTL, I assume you used their injector? Do you have any idea on the FSLTL performance using the “default sim injection”? For me their injector causes massive stutters, but it runs fairly smoothly with the default sim injection. I wonder how that compares to FS traffic
Hey Ronald, I did yes - you can see the settings for which in the methodology part of the video. Considering the default live traffic does not inject nearly as many aircraft as FSLTL does, along with very few parked aircraft, unless you've spent the time to normalise the two, it makes sense you see better performance with that.
Hi, nice video! many thanks for the comparison. FPS is really important but also how about the number of flight plans (also in remote parts), of aircraft models and of liveries, I wonder how is the difference? Have you seen some difference? Thx
Thanks! I've not really tested this but as with most responses regarding coverage it generally tends to be AIG with the best offline FP's, liveries and models and FSLTL doing well with live data. FS Traffic is a smaller package vs the two at the moment.
Bought this as I got bad performance with fsltl at default settings which I did change and was happy with however wanted to see more traffic unfortunately having ctds with this didn't want to start uninstalling other addons but had no choice so requested a refund and gone back to fsltl.
There is always the risk of some conflicts when running many 3rd party addons, however, I've not had any with FST that I haven't been any to easily sort out. Hopefully JF support can help you out! If you have PMS750, Flight Control Replay and the Heavy Division mod are known to cause CTDs.
@BobbyFuzzy unfortunately you are right about the risk suppose sometimes you expect things to work straight away and it's not always the case. Don't get me wrong eventually fs traffic will work the way it's meant to and with addons that will function properly as well unfortunately that time isn't now maybe I will purchase fs traffic again at some point who knows but for now fsltl all the way.
@@jonathanfrancis1991 I think sometimes it's easy to forget we are essentially just modding a game, in other games this can be a a treacherous and difficult path. It's quite amazing how well the sim does run with so many different and complex changes to it by 3rd parties!
Not bad review, but an experienced PC builder, paring a 5600X to a 3080 is not the best choice. Yes it can work, but a 5600X is an entry level series 5 CPU. I will get a better performance with my higher CPU with more cores to enable a better performance.
Maybe not nowadays but I won the 3080 in a competition about 2 years ago - I didn't want to spend a lot upgrading my CPU so the 5600X was the best bang for buck option at the time that would allow the 3080 to be utilised better (I had a R5 3600 when I won it) that would fit on my current motherboard. Thanks for your comment though!
He took the time to put together the stats and post them. While its easy to see he has a CPU limited setup, firstly it doesn't mean you have to assume he's not an experienced PC builder. That's kind of passive aggressive and a bit rude towards someone who has contributed his time and energy. The reality is a CPU limited setup like his is arguably better for conducting this particular test as it will magnify the variances. So his results probably have more in common with more people than if he had done it with a faster processor. I have a 7950x with 4090 and using his settings and flight setup I lose only 2fps with FSLTL over no traffic. I doubt that FS Traffic would be much different. Even at 3 x 4k with DLSS3 on in the sim pit I only lose 10fps, but I lock that at 30fps so its all a moot point as long as nothing pulls me lower than that. So of course H/W makes a huge difference to test results, but again his represent a lot more people than mine would. People who have a faster processor can scale reasonably easily from his data. For a single display setup his 3080 isn't going to be a bottle neck in this test, so its all CPU.
Man, something is wrong with my PC. I have similar hardware to you and I've never been at an airport of any significance with no traffic getting 60+ fps...I'm lucky if I can start out with 30 fps before starting up FSLTL.
@@BobbyFuzzy are your msfs settings in the video? I'll watch again. I've just noticed that when I upgraded my GPU, I didn't get really any performance boost at airports.
@@BobbyFuzzy alright, I just a did a fresh install of msfs and want to get a no mods baseline...and for your test you were getting 60 at an international airport, yeah? Just trying to figure out where I'm losing performance. My experience since upgrading my GPU has been disappointing. For what I spent I expected a little more of a boost performance wise. Is FSLTL still getting worked on? Optimization what not?
Ah, High preset. Yeah, I think I got too excited with the upgrade that instead of just sticking with high I jumped to ultra. But still not getting those 60+ with no traffic.
I've not tried it honestly since making this video but glad to hear it's made some improvements! I thought I'd read a little while back they were working on some optimisation.
