April 14, 201313 yr I use Win7 / Norton Internet Security 2013, no issues with Couatl or FSDT sceneries, I don't even disable Norton services, it just runs great. I love FSDT's smooth textures and blends.
April 14, 201313 yr I use Win7 / Norton Internet Security 2013, no issues with Couatl or FSDT sceneries, I don't even disable Norton services, it just runs great. I love FSDT's smooth textures and blends. Same here, Norton runs clean on all my files, except Aerosoft products. They swear it's false positives, but they are the only ones I get flagged on. The thing is they are being flagged for multiple conditions. From Malware to Trojan's. I'm trying to believe it, but I have to also think where's there's smoke there is usually fire!! Whatever it is, it doesn't appear to infect anything else. Thanks Tom My Youtube Videos! http://www.youtube.com/user/tf51d
April 15, 201313 yr Commercial Member Couatl is NOT a spyware, under ANY definition of it. It doesn't install in your system (is launched with FSX and it quits with FSX, it DOES NOT run in the background if FSX is not running) and it doesn't connect to the internet when in use, once a serial is activated, no internet connection is ever made to verify it, until you move to another PC or upgrade the current one, or reinstall Windows from scratch. Couatl is NOT "just" an anti-piracy measure. Yes, it verify the license and it doesn't display a scenery in case it's not valid, but this feature just a couple of lines in its source code and not more: 99.9% of Couatl doesn't have anything to do with anti-piracy. Couatl is a INTERPRETER for the Python language, which is used to create any kid of interactivity and special features in our sceneries. ENTIRE products we made, like XPOI and GSX, are written in Python, and are running entirely under the Couatl interpreter. This means, since they run externally from FSX, they don't steal precious memory space from the limited FSX VAS. Such programs can grow as complex as we want to, but their code will never take a single byte from the FSX memory. And, thanks to the operating system scheduling, they will run on spare cores on a multi-core system, meaning their calculations will never slow down FSX, regardless how complex they might be. Some examples of things done by Couatl are: - The whole GSX program - The whole XPOI program - The calculation needed for the Dynamic Shadows in CYVR (there's a full blown astronomic calculation framework inside Couatl...) - Things that appear/disappear depending on seasons, dates. - Any form of interactivity in the scenery, such opening hangars, special jetways, working clocks (at PHNL, for example), the lightshow at KLAX which run different sequencing programs, buildings that change appearance depending on the date (PHOG terminal was red before August 2011 and it was painted in green after that, and the scenery does it too...) - Special runway lights such as PAPI with FAROS, THLs, RWSLs and RELs, that interacts with user airplane and AI too. - Docking systems, the parking evaluator, are also Python scripts which runs under the Couatl interpreter. Again, this is done entirely external from FSX, so nothing of such programming logic is taking away from FSX memory. Couatl has its own separate up to 4GB addressable space. What is causing problems TO Couatl, are those antivirus products that (as someone also posted here), create similar issues with other developers modules, and to claim to be able to "catch more threats", use questionable heuristic methods, and instead to flag only what is *known* to be a threat, they try to guess basing on certain patterns. Since a common pattern for real virus is to be encrypted and using measures to defend against tampering, but this is obviously a legit case use for any product that ALSO have some anti-piracy measures, this is flagged as a possible threat by such antivirus products. The first offender is Norton: we sent samples of the Couatl,exe file to them countless of times, and their reply was always they acknowledge their error and would fix it in the next live update. But the issue is, they are so slow fixing it, and we are so fast constantly updating Couatl.exe itself, that by the time they finally whitelist it, we released a new version, starting the cycle again. So, what is really causing you a problem is the antivirus, and it doesn't affects *just* us, if affects Aerosoft, Flight1 and others too. The real solution would be, either learn to configure the antivirus to exclude the affected files from scanning, or turn the pesky heuristic detection off OR to switch to a more reliable antivirus, which doesn't mistakenly flag legit products in the first place. There are plenty of them, like the free, lightweight and always updated Security Essentials from Microsoft, which doesn't mistakenly detect any of our software as a threat, and doesn't require any special configuration either. It just works. Another, non flightsim-related, rant against Symantec here: http://www.codeandweb.com/blog/2012/06/23/how-symantec-ruins-independent-developers Umberto Colapicchioni http://www.fsdreamteam.com FSDT on Facebook
April 15, 201313 yr Security Essentials from Microsoft, which doesn't mistakenly detect any of our software as a threat, and doesn't require any special configuration either. It just works. Tipped from here, I now use it exclusively instead of AVG which hogged resources, and correctly stated, it does not affect any FSDT/Cloud9 products. Rick Almeida
April 19, 201313 yr KLAS is very good. Go and get it. Avoid KJFK. It is horrible, for two reasons: 1. Very bad (tight) positioning of the gates -> parked planes have their wings overlapping in each other. 2. The most important: Abysmal frame-rates. It couldn't have been worse.
April 19, 201313 yr KLAS is very good. Go and get it. Avoid KJFK. It is horrible, for two reasons: 1. Very bad (tight) positioning of the gates -> parked planes have their wings overlapping in each other. 2. The most important: Abysmal frame-rates. It couldn't have been worse. I disagree with number 2, JFK even decreased the number of OOM and general crashes I've had with default scenery.
April 19, 201313 yr Commercial Member KLAS is very good. Go and get it. Avoid KJFK. It is horrible, for two reasons: 1. Very bad (tight) positioning of the gates -> parked planes have their wings overlapping in each other. 2. The most important: Abysmal frame-rates. It couldn't have been worse. Both points clearly indicate you might have AI problems, both because if you saw AIs with overlapping wings at gates, might be due to your AI being tweaked for having radius different than real life (common issue with World of AI), Another issue is the scenery obviously have its gates as in real life and, of course, since many gates are never used together, but the FSX ATI/AI engine doesn't support this, a scenery developer is faced with the option to either REMOVE the gate altogether or adjust the radius to an unrealistic small size, so FSX would assign unrealistically small AIs there. Or, just leave the gate as it's supposed to be, hoping the user would just *undertand* the issue, and lower the AI settings a bit. Which, of course, will ALSO have a good impact on your fps, so you won't make the mistake of having too many AI that are bringing your fps down (even more if you are not using native FSX models), and complain with the *scenery*, which is in fact is very well known to be fairly light on fps, especially considering the area. Umberto Colapicchioni http://www.fsdreamteam.com FSDT on Facebook
April 19, 201313 yr I disagree with number 2, JFK even decreased the number of OOM and general crashes I've had with default scenery. Framerates are unrelated to OOM and crashes, so your point is irrelevant. Another issue is the scenery obviously have its gates as in real life and, of course, since many gates are never used together, but the FSX ATI/AI engine doesn't support this, a scenery developer is faced with the option to either REMOVE the gate altogether or adjust the radius to an unrealistic small size, so FSX would assign unrealistically small AIs there. Or, just leave the gate as it's supposed to be, hoping the user would just *undertand* the issue, and lower the AI settings a bit. So you admit that there are overlapping gate radiuses - besides you cannot deny the obvious -> it is clearly visible is someone opens the AFCAD with the latest Airport Design Editor. 1. I can accept this explanation. 2. A user is not obliged to: - adjust AI settings every time he makes a flight in a different region - have access / know all available recent charts for every airport 3. My opinion, given the FSX limitation, is that you should have adjusted the radius every two gates to an unrealistic smaller size, so that they can be used together. PS: The default area of New York is already very heavy in frame-rates (as is in Vancouver). So, in my opinion, you should not have made a scenery there in the first place, as it would be impossible (as it proved to be) to have an even decent frame-rate there. For the same reason, you should have made a scenery in Vancouver, and not started to make a new edition of KJFK. PS2: For the record, I am using WoAI (which is much lighter than the default AI in framerates) and my FSX AI settings are only jetliners set at 100% (everything else at 0%).
April 19, 201313 yr Commercial Member So you admit that there are overlapping gate radiuses - besides you cannot deny the obvious -> it is clearly visible is someone opens the AFCAD with the latest Airport Design Editor. I don't see what's there to "deny" or "admit": I'm saying they have be made like that. 2. A user is not obliged to: - adjust AI settings every time he makes a flight in a different region He should, because AI complexity settings in FSX (the percentages) are NOT dependent by the scenery, but are controlled by the Traffic.bgl file, which is usually a responsibility of the AI Traffic package developer. This means, a certain complexity setting that could generate a certain number of planes in a certain scenery, might generate a different number of planes on the SAME scenery, just because a different AI package has been used. And, there are possible strange issues such as fps drastically going down at a certain place, because some specific fps-heavy models appear, let's say, at 75% complexity, so you might see your fps being good UP TO 74%, then collapsing at 75% and ONLY at a certain place because, due to the combination of AI schedule, parking airline assignment AND AI Traffic package used, THAT specifically fps-heavy model might suddenly appear in quantities in a specific place (and not somewhere else), so you get the WRONG impression the fps drop is CAUSED by the scenery, when in fact is NOT. - have access / know all available recent charts for every airport I don't see how this has anything to do with the issue discussed. 3. My opinion, given the FSX limitation, is that you should have adjusted the radius every two gates to an unrealistic smaller size, so that they can be used together. I disagree with this opinion, and I believe it's better to see some wing overlapping, instead than having Learjets parked in place of a 737. PS: The default area of New York is already very heavy in frame-rates (as is in Vancouver). So, in my opinion, you should not have made a scenery there in the first place, as it would be impossible (as it proved to be) to have an even decent frame-rate there. For the same reason, you should have made a scenery in Vancouver, and not started to make a new edition of KJFK. All the discussions about Vancouver, instead, clearly proven something entirely different, that we pointed out to many users that one should be careful how much stuff is installed, we proven the feasibility and the advantages of DX10, and with the 1.1 update, we allowed users to configure CYVR as they like, allowing to customize its memory footprint to be even lower than KLAX, so OOM are now really unlikely. And most of users didn't had any fps problems with CYVR, not even in the first release. Of course, since your JFK problem is caused by AI, it's to be expected you would see it everywhere, ESPECIALLY in places were BOTH NATIVE FSX scenery (CYVR, and JFK) and LEGAGY FS9 CODE is used (Word of AI) PS2: For the record, I am using WoAI (which is much lighter than the default AI in framerates) and my FSX AI settings are only jetliners set at 100% (everything else at 0%). I were prepared to BET you were using WoAI. Fact that WoAI is faster than the default AI doesn't mean much: default AI are not really "AI", they are the same models as the default flyable airplanes, they are not made as AI in the firat place. WoAI is mostly FS9 imported planes, and they cause a serious impact on fps, mostly because they are still FS9 .MDLs. It's very well known that FS9 MDLs are slow in FSX, you can see the difference easily just if the airplane YOU are flying is FS9 (and how fast native FSX models are), now imagine that difference multiplied for a large number of AIs. You should use a modern AI package, such as Ultimate Traffic 2 or MyTraffic X, which are made almost exclusively of native FSX models, and they are carefully optimized for fps, with very well done LODs levels. On top of that, AI models made for FS9 slow down the fps EVEN MORE, when using a 100% native FSX scenery, such as Vancouver, and of course the upcoming JFK update: both don't contain a single line of FS9 code, but this advantage is lost if you force FS9 to activate its "compatibility" code path, and it's enough 1 single AI made for FS9 to enable it. Umberto Colapicchioni http://www.fsdreamteam.com FSDT on Facebook
April 19, 201313 yr The most important: Abysmal frame-rates. It couldn't have been worse. Actually that is not true for me, either. I admit, FPS are noticably lower than at most other airports, but given the proximity to other major airports - and thus quite a lot of planes around - and to Manhattan, this impact is not that hard at all. Besides for me the addon has in fact improved FPS over the default airport. However, if it were me to decide between either LAS or JFK again, I'd go for LAS. Florian
April 20, 201313 yr Moderator I'd have to chime in as well and say that FSDT KJFK well out performs the default KJFK. Not only do I get a solid 20 to 25 FPS in the VC of the NGX with 80% to 100% AI (600+ packages installed) and ~35FPS in external view, but also no OOM's. Pretty good considering the area is a known FPS beater by default. I actually went back to using WOAI for the time being and even though the models are not optimized for FSX, I still am able to maintain the said FPS using WOAI at this airport and that is a ton of traffic for JFK. Avsim Board of Directors | Avsim Forums Moderator
April 20, 201313 yr Of course, since your JFK problem is caused by AI, it's to be expected you would see it everywhere Only that I don't see it everywhere. Only at your KJFK and your CYVR.
April 21, 201313 yr Commercial Member Only that I don't see it everywhere. Only at your KJFK and your CYVR. If you read my explanation carefully, you would understand why a scenery that is made with native FSX code, could be particularly penalized by using FS9 AIs, which is the case with CYVR, and a scenery that lies in an area that is already very slow due to the surrounding scenery, such as JFK, will ALSO be slow BECAUSE OF THE AI, NOT because of the scenery itself. FSDT JFK is universally known to be not slower than default JFK but, of course, its AFCAD have many more parking spaces than the default JFK, is able to attract way more AI compared to the default, so you are easily mislead thinking the scenery is slow, when in fact it's you are AI that are the problem. A proper comparison of the fps of the *scenery*, would be comparing FSDT JFK against default JFK, with 0% AI and 0% ground traffic. That would really compare the fps of the *sceneries* against each other, not the fps or your AI or the FSX default ground vehicles. And no, this doesn't obviously mean you "have to fly with 0% AI", it means you have to learn to adjust their settings depending on the scenery, and not expecting to have a setting that works everywhere because, as I've clearly explained in my previous message, the same percentage setting might have a VERY different result on different airports, because of schedules, number of parkings, different models called, and even different *types* of models called, because 50 AI different AI modules, and each one with a different livery are WAY slower than 50 AI all with the same models and liveries. UT2, for example, allows to adjust the traffic density in real-time with an hotkey, you would think there wasn't any need for such feature, if there was some "universal" AI density setting that would work everywhere. Umberto Colapicchioni http://www.fsdreamteam.com FSDT on Facebook
Create an account or sign in to comment