Jump to content
Sign in to follow this  
Wolfko

Qestion re. Holger's Autogen Tweak

Recommended Posts

Should off course read "Question" :( A quote of Holger Sandmann from another thread:"TERRAIN_MAX_AUTOGEN_BUILDINGS_PER_CELL is a "hard cap" meaning it limits the total number of buildings displayed in one land class cell (~1.4 sq km). To find a good value for your system you need to consider two things: (1) as mentioned above the autogen density slider removes 20% of the overall autogen density with each slider notch to the left. (2) even the most dense urban areas don't have more than 1300 building annotations per cell. Thus, at Dense you'd only see 1300*0.6=780 buildings to begin with and a MAX_AUTOGEN_BUILDINGS value of anything greater would have no effect."My question is, does the autogen slider remove 20 % of the default max. amount of autogen buildings (1300) per notch to the left or 20 % of the amount of buildings one has set via the TERRAIN_MAX_AUTOGEN_BUILDINGS_PER_CELL entry.In other words, if I set TERRAIN_MAX_AUTOGEN_BUILDINGS_PER_CELL=1040 and additionally move the autogen slider from extremely dense one notch to the left, will I then get 1040 (= 80 % of 1300) or 832 (= 80 % of 1040) buildings displayed per cell? Thanks, Wolfgang

Share this post


Link to post
Share on other sites

Hello Wolfgang,someone posted a similar question in the FTX forum - http://orbxsystems.com/forums/index.php?topic=19916.0 - so I repeat my answers here:

The explanation is that the Autogen Buildings and Trees parameter values are theoretical maxima. The actual number of trees and buildings displayed depends on how many autogen footprints each tile has. Our most densely packed urban classes have about 1300 building footprints and 2000 tree footprints, which would translate to 260 buildings and 400 trees at Sparse (20%), which corresponds with your first two screenshots. The industrial areas to the left of your plane have a lot fewer footprints to begin with (the buildings are much larger), so you probably have less than 100 buildings at Sparse setting there.I looked at the equivalent classes in FSX default and GEX. FSX default has no more than 1000 building footprints in the most dense urban class and about 1100 trees and GEX North America has very similar values. That explains why FTX has a higher autogen density at the same autogen density slider display settings though the difference is only about 20-30%, at least for buildings.It's rather odd that ACES chose a default value of TERRAIN_MAX_AUTOGEN_BUILDINGS_PER_CELL=3000. Like I said though, it's a theoretical maximum anyway that only kicks in when the number of the actually existing footprints is larger than this value.
Maybe a less confusing way of looking at it is that the FSX terrain engine first calculates the number of buildings to be displayed based on the actual number of footprints for a given tile times the percentage of the autogen density slider value. In the second step it then checks whether the MAX_AUTOGEN_BUILDINGS value is lower than the calculated value from step one. If it is lower it further removes buildings down to the number of MAX_AUTOGEN_BUILDINGS. If it isn't it simply ignores MAX_AUTOGEN_BUILDINGS.For example: Step 1: at Dense with 1300 footprints = 1300 * 0.6 = 780Step 2: if MAX_AUTOGEN_BUILDINGS is at 600 then that number of buildings is displayed. If the value is >780 then MAX_AUTOGEN_BUILDINGS is ignored and 780 buildings are displayed.The overall advantage of this parameter is that it "evens out" building densities across tiles. In urban areas with lots of houses it reduces the number of buildings displayed while in rural areas, where the tiles have much fewer building footprints, you keep the level chosen with your autogen density slider.
Cheers, Holger

Share this post


Link to post
Share on other sites
Guest firehawk44
Should off course read "Question" :( A quote of Holger Sandmann from another thread:"TERRAIN_MAX_AUTOGEN_BUILDINGS_PER_CELL is a "hard cap" meaning it limits the total number of buildings displayed in one land class cell (~1.4 sq km). To find a good value for your system you need to consider two things: (1) as mentioned above the autogen density slider removes 20% of the overall autogen density with each slider notch to the left. (2) even the most dense urban areas don't have more than 1300 building annotations per cell. Thus, at Dense you'd only see 1300*0.6=780 buildings to begin with and a MAX_AUTOGEN_BUILDINGS value of anything greater would have no effect."My question is, does the autogen slider remove 20 % of the default max. amount of autogen buildings (1300) per notch to the left or 20 % of the amount of buildings one has set via the TERRAIN_MAX_AUTOGEN_BUILDINGS_PER_CELL entry.In other words, if I set TERRAIN_MAX_AUTOGEN_BUILDINGS_PER_CELL=1040 and additionally move the autogen slider from extremely dense one notch to the left, will I then get 1040 (= 80 % of 1300) or 832 (= 80 % of 1040) buildings displayed per cell? Thanks, Wolfgang
I was just looking over at NickN's excellent thread regarding setting up FSX and tuning it. Here's a quote regarding adding lines in the FSX.cfg: "One of the biggest mistakes people make is the thought that lower sliders = better perfomance. FSX is NOT a game, its a simulator and requires thinking a bit outside the box to tickle the hardware by the software settings and get it to run sweet1. DO NOT use the autogen restriction lines in the configuration file for buildings and trees if you are using SP1/SP2. In most cases those lines will throw off the priority system and make the sim run/stutter worse.2. Believe it or not, ZERO autogen slider can make frames drop, how? Same reason as above. Try to maintain at least 50+% (DENSE) however very slow systems may need to limit that to NORMAL/SPARCE or even OFF (off = rare must be very slow hardware)Keeping the autogen slider at DENSE or greater actually tends to raise frames. Going too high for the installed hardware and Scenery Complexity/weather is where perf starts to really dropMany can run a 3/4 (very Dense) Scenery Complexity with a 3/4 (very dense) autogen silder however your mileage may vary.The important part about this slider is if your hardware is not very slow, keep it at a minimum of DENSEThis slider can also be changed mid-flight as well. There are times when you may fly out of a scenery friendly hub, into a very large urban area hub which may require this slider to be dropped a notchGood systems can run 3/4 to 100% depending on the scenery involved and GEX USCAN can go a long way in helping with that performance. Autogen was laid out in a optimal way to reduce CPU overhead."Best regards,Jim

Share this post


Link to post
Share on other sites

Hi Jim,conflicting suggestions, how fun (and unusual in our hobby) :( Fortunately, despite some evidence to the contray, FS isn't a religion so you don't have to believe in anything or anybody. The only thing that counts is what works on your system and that requires that you do the testing and you make the decisions. Since the placebo effect can be such a strong factor I always suggest to set up a standard test flight that you save -- say at 5000ft in your favorite aircraft paused above Seattle -- and then use for systematic testing. Don't just look at fps but also stutters etc. Also, a reference screenshot at the beginning of each flight is often helpful. Only ever change one parameter at once and make notes so that you don't have to rely on memory. Unless you just change a slider by a notch or two you also need to re-start the sim to ensure similar test conditions.Cheers, Holger

Share this post


Link to post
Share on other sites

Holger,These are my autogen settings in the FSX.cfgTERRAIN_MAX_AUTOGEN_BUILDINGS_PER_CELL=600TERRAIN_MAX_AUTOGEN_TREES_PER_CELL=800If I understand you correctly with above settings there should be no difference in the amount of autogen displayed no matter if the slider is at "extremely dense" or "very dense". But on my system it makes a difference! I get definitely less autogen within PNW with the slider at very dense. I have made screenshots which clearly show this. I'll post them tomorrow after I have converted and resized them.Cheers, Wolfgang

Share this post


Link to post
Share on other sites

Hello Wolfgang,I guess it depends where exactly you are. Keep in mind that only the most dense urban areas have ~1300 objects. Suburban and industrial areas, for example, have fewer buildings to begin with so the MAX_AUTOGEN_BUILDINGS=600 cutoff might kick in sooner. Cheers, Holger

Share this post


Link to post
Share on other sites
Guest Nick_N
Hello Wolfgang,someone posted a similar question in the FTX forum - http://orbxsystems.com/forums/index.php?topic=19916.0 - so I repeat my answers here:Cheers, Holger
Holger1. In SP1 and SP2 they switched autogen to batch calls. This has a defined effect on autogen performance which is why during UT Alaska beta Allen developed a

Share this post


Link to post
Share on other sites

Hello Nick,thanks for the information.

2. If a product has a irregular number of autogen objects annotated to the tiles, that inconsistency can translates to a hi/low frame stammer which in turn, depending on the system in use, can effect the sim render priority system. If the system is negatively susceptible to such changes that can in turn begin a chain of events which may very well and usually does spiral into the

Share this post


Link to post
Share on other sites
Guest Nick_N
Hello Nick,thanks for the information.I'm not sure what "Joe/Jane FSX User" is supposed to derive from this paragraph given its liberal use of "can", "if", "depending", and "usually". If anything it reinforces my point that people should run their own tests and decide for themselves what works and what doesn't. All I did was provide some quantitative information about what those parameters do. I don't ask anyone to make a judgement for or against their use based on my statements alone. Perhaps once you're back from vacation you can share with us the specific tests that led you to the conclusion not to use the max_autogen tweaks. That way everyone interested can run those same tests and we can discuss and compare the results.Cheers, Holger
Thanks for the information? You mean you don

Share this post


Link to post
Share on other sites
I have said too much now. Moderators you may censor me for being direct and honest if you see fit.
Nick, there is frankly no reason to be agressive with Holger or with anyone that advance a different opinion than yours.I really agree that fortunatly FS is neither a religion nor a sect, thus there should be no gospel truth nor gurus.Now, please use PMs if you wish to continue this discussion with Holger.Enjoy your holidays and... relax! :(

Best regards,
David Roch

AMD Ryzen 5950X //  Asus ROG CROSSHAIR VIII EXTREME //  32Gb Corsair Vengeance DDR4 4000 MHz CL17 //  ASUS ROG Strix GeForce RTX 4090 24GB OC Edition //  2x SSD 1Tb Corsair MP600 PCI-E4 NVM //  Corsair 1600W PSU & Samsung Odyssey Arc 55" curved monitor
Thrustmaster Controllers: TCA Yoke Pack Boeing Edition + TCA Captain Pack Airbus Edition + Pendular Rudder.

 

Share this post


Link to post
Share on other sites
Guest Nick_N
Nick, there is frankly no reason to be agressive with Holger or with anyone that advance a different opinion than yours.I really agree that fortunatly FS is neither a religion nor a sect, thus there should be no gospel truth nor gurus.Now, please use PMs if you wish to continue this discussion with Holger.Enjoy your holidays and... relax! :(
Then lock this thread David and let it die naturally! I agree to other communication in PM!I have absolutely no issue with that call or action by this site!Any other action/suggests are bias and you know exactly what I am saying. It falls in line with current events which are also media driven and are BS!!!

Share this post


Link to post
Share on other sites
Why lock it now - i sense a very interesting topic.
Indeed Heiko.Waiting forward to seing Wolfgang's comparative screenshots!

Best regards,
David Roch

AMD Ryzen 5950X //  Asus ROG CROSSHAIR VIII EXTREME //  32Gb Corsair Vengeance DDR4 4000 MHz CL17 //  ASUS ROG Strix GeForce RTX 4090 24GB OC Edition //  2x SSD 1Tb Corsair MP600 PCI-E4 NVM //  Corsair 1600W PSU & Samsung Odyssey Arc 55" curved monitor
Thrustmaster Controllers: TCA Yoke Pack Boeing Edition + TCA Captain Pack Airbus Edition + Pendular Rudder.

 

Share this post


Link to post
Share on other sites
Indeed Heiko.Waiting forward to seing Wolfgang's comparative screenshots!
I'll post them, once I'm at home tonight. It was late evening when I wrote my last post and I was just too tired to adjust the screenies for posting.Wolfgang

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Tom Allensworth,
    Founder of AVSIM Online


  • Flight Simulation's Premier Resource!

    AVSIM is a free service to the flight simulation community. AVSIM is staffed completely by volunteers and all funds donated to AVSIM go directly back to supporting the community. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. Thank you for your support!

    Click here for more information and to see all donations year to date.
×
×
  • Create New...