Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

Faster than 40d, no problems

42 (26.1%)

Faster than 40d, problems

72 (44.7%)

No slower than 40d, no problems

14 (8.7%)

No slower than 40d, problems

16 (9.9%)

Slower than 40d, no problems

2 (1.2%)

Slower than 40d, problems

3 (1.9%)

Doesn't work (please explain)

12 (7.5%)

Total Members Voted: 160

[Xgamer4]

Laptop Model: Acer Aspire 5100
Windows XP SP3
Radeon X1100 running Omega Drivers v4.8.442

Y:0/N/N does what I mentioned earlier.
Y:0/Y/N mirrored whatever was open behind it when I first tried to start DF and then went to a blank white screen.
Y:0/N/Y gives me a half-rendered title screen and then it mirrors whatever's behind it. I can grab a screenshot if you want me to.
Y:0/Y/Y horribly broke everything, but I think you knew that already. <_<

[Taniec]

Well the fix made it playable now with partial print. Nice work Toady.

[Tycho451]

Starting DF with [PARTIAL_PRINT:YES:0] or [PARTIAL_PRINT:YES:1] gets me a flickering screen and no proper redrawing. [PARTIAL_PRINT:YES:2] seems to work.

No matter what I do with the other 2 options [FRAME_BUFFER:YES] gets me this:

Code: [Select]

Error: Invalid framebuffer configuration, FBO disabled. No action required.                     

I'm running Kubuntu 8.10 with Radeon x1550 and fglrx 8.12

Also game still crashes after the fullscreen/windowed Dialog if windowed is set to prompt.

[Baughn]

Not My Fault(tm).

Have you tried reinstalling windows? Or installing, say, ubuntu and running the linux version?

Um, no. I did recently (within the lifetime of this thread) upgrade my drivers.

I wouldn't say that I have enough technical skills to install ubuntu unaided.

http://wubi-installer.org/ <-- Use this one. Trust me, you do have the required skill, which is pretty much zero. :-)

[Baughn]

Starting DF with [PARTIAL_PRINT:YES:0] or [PARTIAL_PRINT:YES:1] gets me a flickering screen and no proper redrawing. [PARTIAL_PRINT:YES:2] seems to work.

No matter what I do with the other 2 options [FRAME_BUFFER:YES] gets me this:

Code: [Select]

Error: Invalid framebuffer configuration, FBO disabled. No action required.                     

I'm running Kubuntu 8.10 with Radeon x1550 and fglrx 8.12

Also game still crashes after the fullscreen/windowed Dialog if windowed is set to prompt.

The crash is a library versioning issue. Static linking would be better, but I can't figure out how to do that on my machine (don't have the right libraries, as I'm cross-compiling), and toady isn't familiar enough with linux to do it.  Still, so long as you can set the value in init.txt, it should be fine.

The FBO thing is.. well, not all cards support the configuration I need. As it says, it falls back to not using them.

[Daniel Charms]

I tried a few different settings and this is what I got (on a low-end laptop with integrated graphics; note that all tests were done in windowed mode):

YES:0, YES, NO: blank window.
YES:0, NO, YES: horrible flicker.
NO, NO, NO: everything's running smoothly. The game is less jerky than 40d at the same fps settings. It also seems to be a bit faster.
YES:2, NO, NO: a huge gain in fps in menu, about 10% gain in fps in-game. Changing between windows results in a momentary flicker, however.

I also encountered a keyboard problem: in some cases, switching between windows resulted in DF not responding to keypresses. Switching back and forth between windows again fixed it.

[Baughn]

Notebook with windows XP, videocard: Mobile Intel 945GM
Tried with small fortress, FPS_CAP:200, G_FPS_CAP:100 (set different values without any seen result)
In full screen mode all looks like max FPS set to 60.
(NO, NO, NO) FullScreen - 60 FPS, WindowMode ~85
(YES, NO, NO) strong flicker  FS - 60FPS, WM > 100 (unable to see)
(YES:1, NO, NO) FS fliker - 60FPS, WM > 100 (no problem)
(YES:2, NO, NO) FS - 60FPS, WM > 100 (no problem)
(YES, YES, NO) flicker FS - 60FPS, WM > 100 strong flicker
(YES:2, YES, NO) FS - 60FPS, WM > 100 (no problem)
(YES:2, NO, YES) FS => 100FPS, WM  - blank screen or random signs, music plays
(YES, NO, YES) almost same as above. Only buffer is seen

Did seen limit of 60 frames in D5 or earlier
D5 same fort, PARTIAL_PRINT:NO:2   Full Screen: ~130 FPS, Windowed mode 200 FPS

G_FPS of 100 is far, far too high. You should not under any circumstances increase it from the default of 50; decreasing it is fine.

I'll suggest putting this information in init.txt.

[Rift]

Heres some tests with using a higher resolution and grid size.

2.8ghz hyperthreaded p4, 1 gig of ram, geforce 7600GS fullscreen with 1240:1024 resolution and fullgrid 80:64, with 7 dwarves busy about tasks. [All settings like temperature/weather are ON and using a graphics set]

I tested a bunch of different partial print/G_fps caps with a 500fps cap.
for the "frame buffer" option.
With Yes:0:yes:no, and the f_gps did not change the fps very much. 14 was only a few fps faster then 50.
outside with cliff: 75fps
outside: 90fps
inside with quarter explored: 100fps
inside with tiny explored: 120fps

Comparitively, it ran in the middle of all my results [but a hella lot less optimized when there is unexplored space]. The others, using just partial print/gfps cap manipulation.

Spoiler (click to show/hide)

G_fps/PP format.
15/NONE:
outside with cliff: 58fps
outside: 68fps
inside with quarter explored: 87fps
inside with tiny explored: 93fps
15/20:
outside with cliff: 80fps
outside: 95fps
inside with quarter explored: 130fps
inside with tiny explored: 140fps
50/55:
outside with cliff: 73fps
outside: 90fps
inside with quarter explored: 124fps
inside with tiny explored: 130fps
1/2:
outside with cliff:73fps
outside: 90fps
inside with quarter explored: 124fps
inside with tiny explored: 130fps
15/4:
outside with cliff: 83fps
outside: 98fps
inside with quarter explored: 133fps
inside with tiny explored: 145fps
15/2:
outside with cliff: 77fps
outside: 92fps
inside with quarter explored: 128fps
inside with tiny explored: 140fps
10/4:
outside with cliff: 77fps
outside: 93fps
inside with quarter explored: 127fps
inside with tiny explored: 140fps
15/3:
outside with cliff: 75fps
outside: 90fps
inside with quarter explored: 120fps
inside with tiny explored: 130fps
14/4: [BEST FOUND]
outside with cliff: 87fps
outside: 104fps
inside with quarter explored: 139fps
inside with tiny explored: 151fps
13/4:
outside with cliff: 81fps
outside: 96fps
inside with quarter explored: 134fps
inside with tiny explored: 144fps

In the end, it seems 14 g_fps and 4 PP works best for me, way better then the frame buffer. I would guess any lower causes them to skip or something, but i have no graphical issues at all with 14g_fps and 4PP so its fine to use... I don't know if this is "good" or "bad" comparitevly. But at 7 dwarves it's certainly playable.

[Baughn]

Dwarfed on:

To be clear on the G_FPS thing: While that's not necessarily the case in windowed mode, in fullscreen mode at least it *will* lock G_FPS to a maximum of the monitor refresh rate, and if the game can't reach all the G_FPS it wants then it'll lower the framerate in an attempt to compensate.

That's why you get 60fps in fullscreen. It's perfectly capable of giving more, but you've told it not to.

[Baughn]

Rift: Well, ideally you'd give me the 40d1 numbers to compare to.

[Savok]

I've genned worlds in 40d6 and 33g using totally vanilla DF (except a tileset and init). The world in 40d6 is a 36xy, 40z area with a chasm and a brook. It's running at wonderful FPS at the beginning. More data later; I'm busy now.

How to install Ubuntu in 4 easy steps.
1: download image of latest version.
2: burn image to a CD
3: insert CD in CD-ROM drive.  Autorun will pop a window up.
4: choose the "Install Ubuntu" option.(I dunno how it works, but it installs a second OS and partition from within Windows.)

Could that mess up my comp? If it did, what would happen?

http://wubi-installer.org/ <-- Use this one. Trust me, you do have the required skill, which is pretty much zero. :-)

Again, could that mess up stuff?

[brainfire]

Running 40d6 with YES:0,YES,NO on my intel 945 in linux (opensuse 11.0) and all the problems I was having before (weird-looking tiles like the guy with screenshots had, partial print flickering/black spaces) are fixed.

It doesn't even show through on other virtual desktop now! Although it still shows through other windows on the same desktop, but only parts of the window that are changing (probably because I can now use partial print). That means, not only is it faster than 40d vanilla, but it also has *fewer* problems. Woo.

There's still one problem- if I drag the window around it leaves bits and pieces of the display lying around. But I've always gotten that problem wining it on this machine too, and I don't get it on my other linux machine, so I'm thinking it's something on my end, especially because I don't think anyone else has mentioned it that I've seen.

[Baughn]

How to install Ubuntu in 4 easy steps.
1: download image of latest version.
2: burn image to a CD
3: insert CD in CD-ROM drive.  Autorun will pop a window up.
4: choose the "Install Ubuntu" option.(I dunno how it works, but it installs a second OS and partition from within Windows.)

Could that mess up my comp? If it did, what would happen?

http://wubi-installer.org/ <-- Use this one. Trust me, you do have the required skill, which is pretty much zero. :-)

Again, could that mess up stuff?

Installing Ubuntu from CD certainly could, as it'll want its own partition, and it's somewhat easy to end up overwriting windows. Wubi is safer - it doesn't, and you can just delete the directory it ends up in if you want. Though it does have lower disk performance.

[Razordw]

Running through all the options seems to do little to no difference between each and it's hard to tell as currently suggests slightly slower but it may be running the same. Just need to setup a reliable means to insure a better comparison (and have the time to do it).

Quick stats (not enough time to gather more).
Old 40d save (world genned who knows when 3x3 area with underground magma pipe, aquifer, 4 levels of ground terrain and 105 dwarves) runs about 15-25 in 40d normmaly hangs around 20 or higher. In 40d6 seems to hang in the same range but more likely on the lower end.
Completely flat embark 16x16 with brook (genned 40d and ran only in 40d) gave 80fps
Almost flat 16x16 with brook with about 6 region tiles at +1 lev and 2 at +2 gives 60fps (40d6)

In the end this goes with everything else I ever do graphics wise that the CPU is likely to be the bottleneck. Course probably a good thing now given the GPU fan like's to act up. Keep up the great work as it seems many are at least getting great FPS boost out of this.
 

[G-Flex]

Testing on an 83-dwarf, 81-animal fortress with a fair bit of water flowing around, I get:

.40d: ~22fps with PP:YES:5 (lower numbers cause flicker)
.40d6: ~22fps (more or less the same) with PP:YES:0 and single-buffer on (regular framebuffer doesn't work)

In other words, absolutely no performance difference, at least in that test.

For the record, I'm running it using an Intel integrated graphics chipset from about 2004, and a 2.8GHz P4 with 2GB of RAM.


EDIT:

Further testing shows that, on a NEW fortress, all same settings as before, with less water flow and only 7 dwarves:

.40d - 78-120 fps, jumps around a *lot*
.40d6 - 99 fps, give or two take one or two; pretty much constant (not reaching FPS cap)

.40d6, in this case, seemed a lot more stable in terms of FPS, and maybe slightly faster on average.


In the menus, however, I can get about 4000 fps on .40d, and only about 1500 on .40d6.