Announcement: Why not try our official idea generator for inspiration when making puzzles?


Caravel Forum : DROD Boards : Architecture : Benchmarking hold
New Topic New Poll Post Reply
Poster Message
StuartK
Level: Smitemaster
Avatar
Rank Points: 564
Registered: 06-10-2003
IP: Logged

File: Benchmark.hold (96.6 KB)
Downloaded 94 times.
License: Other
From: Unspecified
icon Benchmarking hold (0)  
The second revision of Benchmark has been added here (also available below) Now with two extra tests provided by trick :)

It's a hold for benchmarking DROD performance on your PC. I've not tested it yet with build 45 (still downloading) The tests can't just be run on their own, unfortunately, and require manual stopwatch timing. If anyone feels inclined to run the tests, please include your CPU and graphics card details with the results. I'll post a new set of build 45 results after I've had some sleep :)

For build 43, all results in seconds, rounded to the nearest (since using a stopwatch isn't so accurate, no point including a decimal point) I'm using the [code] [/code] tags so the results appear aligned.

Duron 800, NVidia Geforce3 TI200, default keyboard refresh rate

        Windowed  Fullscreen
Test 1  47        35
Test 2  49        38
Test 3  49        38
Test 4  51        40

PII 233, ATI Rage IIc, default keyboard refresh rate

        Windowed  Fullscreen
Test 1  90        23
Test 2  92        26
Test 3  94        26
Test 4  99        27


If anyone has good designs for additional tests (basically, something that tests either the graphics card or CPU, but allows Beethro to run through the room unhindered) I'll be happy to add them (no more than 7-8 tests though, and I'll have to include a warp room on level 1 which I've not done yet.

[Edited by StuartK on 09-17-2003 at 09:08 PM]
09-17-2003 at 08:09 AM
View Profile Send Private Message to User Send Email to User Show all user's posts Quote Reply
trick
Level: Legendary Smitemaster
Rank Points: 2580
Registered: 04-12-2003
IP: Logged
icon Re: Benchmarking hold (0)  
Here's my results. Also, i'm interested in seeing how the performance in the linux client is, compared to the same in windows, so if any of you can, run the tests in both windows and linux on the same machine :P

Pentium III 500mhz, NVidia Geforce 3 Ti200, 256 mb ram, build 43 linux, 24 bpp, default repeat rate:

        X Windowed  X Fullscreen  VESA FBCon Fullscreen
Test 1   51          51            39
Test 2   54          53            42
Test 3   55          54            42
Test 4   57          56            46


Interesting that it seems windowed and fullscreen are the same in X, probably because 2d acceleration is "experimental" in the nvidia driver (i do have it enabled, though). Surprisingly (or maybe not), the vesa console framebuffer is better.

- Gerry


[Edited by trick on 09-17-2003 at 03:07 PM GMT: align it]

[Edited by trick on 09-17-2003 at 03:26 PM GMT: add framebuffer test]

[Edited by trick on 09-17-2003 at 03:30 PM GMT: stuff]
09-17-2003 at 03:05 PM
View Profile Send Private Message to User Send Email to User Show all user's posts Quote Reply
trick
Level: Legendary Smitemaster
Rank Points: 2580
Registered: 04-12-2003
IP: Logged

File: More benchmarks.hold (7.8 KB)
Downloaded 53 times.
License: Other
From: Unspecified
icon Re: Benchmarking hold (0)  
I needed a little break from things that require a brain, so i threw together two more benchmarks for you :P . One tests what happens if 714 roaches are born at once (RoachBomb), and the other tests roach killing (RoachKill). Feel free to add them to your hold if you think they're worthy, StuartK :) .

Results (build 45 this time):
Pentium III 500mhz, NVidia Geforce 3 Ti200, 256 mb ram, build 45 linux, 24 bpp, default repeat rate:

           X Windowed  X Fullscreen  VESA FBCon Fullscreen
RoachBomb   57          56            46
RoachKill   57          56            47


- Gerry


[Edited by trick on 09-17-2003 at 08:51 PM]
09-17-2003 at 08:49 PM
View Profile Send Private Message to User Send Email to User Show all user's posts Quote Reply
StuartK
Level: Smitemaster
Avatar
Rank Points: 564
Registered: 06-10-2003
IP: Logged

File: Benchmark.hold (96.6 KB)
Downloaded 53 times.
License: Other
From: Unspecified
icon Re: Re: Benchmarking hold (0)  
trick wrote:
I needed a little break from things that require a brain, so i threw together two more benchmarks for you :P . One tests what happens if 714 roaches are born at once (RoachBomb), and the other tests roach killing (RoachKill). Feel free to add them to your hold if you think they're worthy, StuartK :) .

Oh they're worthy, and have been added. Much more fun than my efforts :)

I've also added working stairs to all the levels in the first room.

This is becoming a rather large hold (biggest so far except for Dugan?) As a side 'benchmark' of sorts, it imports pretty quickly on my Duron, but takes about 3 minutes on my PII. Two or three more levels and we may have the biggest hold ever :cigar

Another thing to mention when reporting benchmarks (I should have considered this before) is whether sound & music are on or off. I'll post my results with build 45 next.
09-17-2003 at 09:48 PM
View Profile Send Private Message to User Send Email to User Show all user's posts Quote Reply
Skylancer64
Level: Delver
Rank Points: 62
Registered: 10-25-2003
IP: Logged
icon Re: Benchmarking hold (0)  
I wonder how long it would take to clear all rooms in the goblins with brains level and the roachkill levels... its possible....

____________________________
Complexity is the source of Life.
11-22-2003 at 03:10 AM
View Profile Send Private Message to User Show all user's posts Quote Reply
New Topic New Poll Post Reply
Caravel Forum : DROD Boards : Architecture : Benchmarking hold
Surf To:


Forum Rules:
Can I post a new topic? No
Can I reply? No
Can I read? Yes
HTML Enabled? No
UBBC Enabled? Yes
Words Filter Enable? No

Contact Us | CaravelGames.com

Powered by: tForum tForumHacks Edition b0.98.8
Originally created by Toan Huynh (Copyright © 2000)
Enhanced by the tForumHacks team and the Caravel team.