I've been doing some thinking, and some tinkering, with thoughts concerning the PGA2K course architect, it's strengths, and it's limitations.
Some simple math tells us that a 10 acre plot is formed by a square 220x220 yards across, a total of 48,400 square yards. That means the standard 2000x2000 set at 220x220 grid "squares" gives you a plot with a roughly 9x9 gird, or about 810 square acres. The actual 2000x2000 comes out to just about 826. This is huge. How huge? Consider that the entirety of Chicago's famous Cog Hill Golf & Country Club takes up just over 710 acres. The game plot can also be greatly expanded, to about 5000x5000 square yards. Folks, that's over 5,165 square acres. Even a site like Bethpage State Park, which hosts 5 courses on just under 1500 square acres, is readily doable, then, at least in terms of the game rendering the needed square acreage. But how close could you come to mapping either Cog Hill or Bethpage State Park before the game simply refused to allow you to add another terrain object?
Does anyone know how far any designer has ever pushed this? I know I'm personally still struggling to put together any sort of basic design at all, but I'm curious how far people *have* pushed the architect. How big a modeled world could you put into the golf course architect before it shut down on you?
Architect Plots Based on Acreage
Re: Architect Plots Based on Acreage
TenSixty6
I try not to push the architect too far. It just leads to much frustration if the course crashes which has happened in the past but not lately. I currently have a course only I can play on my computer but anyone else tries to play it crashes. Have no idea why.
I try not to push the architect too far. It just leads to much frustration if the course crashes which has happened in the past but not lately. I currently have a course only I can play on my computer but anyone else tries to play it crashes. Have no idea why.
Re: Architect Plots Based on Acreage
Gotcha. So, at the end of the day, it's still a persnickety piece of turn of the century coding. That's a bit sobering, but good to know.
It's absolutely amazing to me, though, that we can get real terrain inputs via LiDAR, and the game can handle it, but we still don't know what the upper edge of what it can and can't handle is.
Has anyone fit 36 holes into one map before? Does anyone know?
It's absolutely amazing to me, though, that we can get real terrain inputs via LiDAR, and the game can handle it, but we still don't know what the upper edge of what it can and can't handle is.
Has anyone fit 36 holes into one map before? Does anyone know?
-
BrianZ111
- Legend of Golf
- Posts: 1025
- Joined: November 12th, 2010, 12:59 pm
- Location: Wisconsin
- Contact:
Re: Architect Plots Based on Acreage
The only limitation I’ve ever run into that crashes the architect is too many points within a single dropped terrain shape (or on the main plot). Somewhere around 175,000 points per terrain shape is the limit. It's not consistent though so I try to stay under 160,000 to avoid taking chances.
I’ve loaded an architect file up with enough detail to import 1 foot LiDAR spacing on the in play area of an 18 hole course 1 yard spacing on a 1,000 yard buffer around the course. This is probably overkill. 1 yard spacing on the course and 3 yard spacing outside of it is probably not going to look much different, but this was a concept test.
In order to pull that off you have to pre-drop a bunch of terrain shapes to stay under the per shape point limit. I have a semi-automated way of generating and importing those shapes to be dropped.
The total result from that experiment was 19 million total points in the architect file. The architect does slow way down when moving the view around at that point. But no crash.
My first move after that is to optimize. PGA 2000’s optimize tool is incredibly good at reducing point count while still maintaining terrain features. I optimize the entire terrain surrounding the course (the part only seen from a distance) on the 7th notch from the left. I optimize the in play part of the course on the lowest level, first from the left. That alone turns 19,000,000 points into 1,900,000 points and you won’t really see any difference while playing.
I was never over a million with any of the pre-LiDAR courses I made, so 1.9 million points still sounds like a lot, but I’ve successfully compiled and played it in the game from there without issue. The only advantage to optimizing further would be to keep file size down. A 1.9 million point course is about 60 MB which is pretty big for this game, all though by today’s standards that isn’t really that big of a file.
Eventually I will get a tutorial posted on how to do this. Generating the automated shapes and working with the LiDAR requires the GIS software Global Mapper though, which is an expensive program, so I’m not sure how useful the tutorial will be to everyone.
Back in the day I did Bull's Eye and Tri-City on the same plot. That was 27 holes. If I have time this winter, I am going to try to do the entire Sand Valley resort on one plot, which is now four full length courses and one par 3 course. Two of the courses are too new to have LiDAR data at this point though so I may only start with the first two and the par 3 course.
I’ve loaded an architect file up with enough detail to import 1 foot LiDAR spacing on the in play area of an 18 hole course 1 yard spacing on a 1,000 yard buffer around the course. This is probably overkill. 1 yard spacing on the course and 3 yard spacing outside of it is probably not going to look much different, but this was a concept test.
In order to pull that off you have to pre-drop a bunch of terrain shapes to stay under the per shape point limit. I have a semi-automated way of generating and importing those shapes to be dropped.
The total result from that experiment was 19 million total points in the architect file. The architect does slow way down when moving the view around at that point. But no crash.
My first move after that is to optimize. PGA 2000’s optimize tool is incredibly good at reducing point count while still maintaining terrain features. I optimize the entire terrain surrounding the course (the part only seen from a distance) on the 7th notch from the left. I optimize the in play part of the course on the lowest level, first from the left. That alone turns 19,000,000 points into 1,900,000 points and you won’t really see any difference while playing.
I was never over a million with any of the pre-LiDAR courses I made, so 1.9 million points still sounds like a lot, but I’ve successfully compiled and played it in the game from there without issue. The only advantage to optimizing further would be to keep file size down. A 1.9 million point course is about 60 MB which is pretty big for this game, all though by today’s standards that isn’t really that big of a file.
Eventually I will get a tutorial posted on how to do this. Generating the automated shapes and working with the LiDAR requires the GIS software Global Mapper though, which is an expensive program, so I’m not sure how useful the tutorial will be to everyone.
Back in the day I did Bull's Eye and Tri-City on the same plot. That was 27 holes. If I have time this winter, I am going to try to do the entire Sand Valley resort on one plot, which is now four full length courses and one par 3 course. Two of the courses are too new to have LiDAR data at this point though so I may only start with the first two and the par 3 course.