Active Topics
-
The N900 is now 15 years old! (7)
to Nokia N900 by Danct12 - 1 day ago -
Sailfish OS for the Motorola Moto G7 Power (XT1955-5) - (ocean) (4)
to SailfishOS by Maemish - 1 day, 20 hrs ago -
F(x)tec Pro1x (QWERTY) for sale (0)
to Buy & Sell by aln00000 - 4 days, 5 hrs ago - more...
 |
|
2008-02-12
, 15:50
|
|
Posts: 1,107 |
Thanked: 720 times |
Joined on Mar 2007
@ Germany
|
#112
|
Been doing some dosbox build performance tests with various compile-time flags. Finding a reasonable benchmark hasn't been easy. Since dosbox emulates several PC subsystems, I choose to test real-world apps. In this case Doom timedemos.
The results indicate time to start dosbox, run the timedemo and exit dosbox (from script). To compute the mean run time, I discard the slowest score of 5. Because a 'clean' run can be broken by various unknowns, the curve is more of a poisson than a binomial distribution.
dosbox_cvs_01_01_25_a -O3 -fomit-frame-pointer -fforce-addr -fforce-mem -falign-loops=2 -falign-functions=2 -falign-jumps=2 -funroll-loops -mabi=aapcs-linux -mcpu=arm1136j-s -mfpu=vfp -mfloat-abi=softfp
dosbox_cvs_01_01_25_novfp -O3 -fomit-frame-pointer -fforce-addr -fforce-mem -falign-loops=2 -falign-functions=2 -falign-jumps=2 -funroll-loops -mabi=aapcs-linux -mcpu=arm1136j-s
dosbox_feb12 -O3 -fomit-frame-pointer -mabi=aapcs-linux -mcpu=arm1136j-s -mfpu=vfp -mfloat-abi=softfp
dosbox_feb14a -Os -fomit-frame-pointer -fforce-addr -fforce-mem -falign-loops=2 -falign-functions=2 -falign-jumps=2 -mabi=aapcs-linux -mcpu=arm1136j-s -mfpu=vfp -mfloat-abi=softfp
dosbox_feb14b -O3 -fomit-frame-pointer -fforce-addr -fforce-mem -falign-loops=2 -falign-functions=2 -falign-jumps=2 -mabi=aapcs-linux -mcpu=arm1136j-s -mtune=arm1136j-s -mfpu=vfp -mfloat-abi=softfp -ffast-math
dosbox_feb14c same but with --disable-core-inline
While the data are inconclusive on effects of build flags, I think we can conclude that for doom timedemo, variance is reduced when the dosbox cycles significantly exceed what the emulator can achieve in realtime.
It appears that increasing cycles well above the point at which dosbox pegs the N810 CPU effectively defeats doom's internal frameskip so that a faster dosbox build+options will finish the run earlier than a slower one.
A seperate test series would be manually running the timedemos and jotting down doom's internal performance reports, using a very small game window and cycles=max.
Last edited by ArnimS; 2008-02-14 at 16:54.
The results indicate time to start dosbox, run the timedemo and exit dosbox (from script). To compute the mean run time, I discard the slowest score of 5. Because a 'clean' run can be broken by various unknowns, the curve is more of a poisson than a binomial distribution.
dosbox_cvs_01_01_25_a -O3 -fomit-frame-pointer -fforce-addr -fforce-mem -falign-loops=2 -falign-functions=2 -falign-jumps=2 -funroll-loops -mabi=aapcs-linux -mcpu=arm1136j-s -mfpu=vfp -mfloat-abi=softfp
dosbox_cvs_01_01_25_novfp -O3 -fomit-frame-pointer -fforce-addr -fforce-mem -falign-loops=2 -falign-functions=2 -falign-jumps=2 -funroll-loops -mabi=aapcs-linux -mcpu=arm1136j-s
dosbox_feb12 -O3 -fomit-frame-pointer -mabi=aapcs-linux -mcpu=arm1136j-s -mfpu=vfp -mfloat-abi=softfp
dosbox_feb14a -Os -fomit-frame-pointer -fforce-addr -fforce-mem -falign-loops=2 -falign-functions=2 -falign-jumps=2 -mabi=aapcs-linux -mcpu=arm1136j-s -mfpu=vfp -mfloat-abi=softfp
dosbox_feb14b -O3 -fomit-frame-pointer -fforce-addr -fforce-mem -falign-loops=2 -falign-functions=2 -falign-jumps=2 -mabi=aapcs-linux -mcpu=arm1136j-s -mtune=arm1136j-s -mfpu=vfp -mfloat-abi=softfp -ffast-math
dosbox_feb14c same but with --disable-core-inline
Code:
core simple, cycles 1800, fs 5, no scaler, -timedemo demo3 dosbox_cvs_01_01_25_a 5:56 4:26 5:07 dosbox_cvs_01_01_25_novfp 4:45 5:06 4:43 dosbox_feb12 4:52 5:03 4:57 core simple, cycles 2500, fs 5, no scaler, -timedemo demo3 dosbox_cvs_01_01_25_a 4:13 4:15 4:32 4:23 4:33 : 4:20.75 dosbox_cvs_01_01_25_novfp 4:31 4:25 4:22 4:27 4:23 : 4:24.25 dosbox_feb12 4:42 4:33 4:27 4:39 4:26 : 4:31.25 core simple, cycles 2500, fs 5, no scaler, -timedemo demo3 'performance' profile dosbox_cvs_01_01_25_a 4:24 4:52 4:23 4:40 4:33 : 4:30.00 dosbox_cvs_01_01_25_novfp 4:33 4:32 4:20 4:39 4:21 : 4:26.40 dosbox_feb12 4:31 4:28 4:30 4:33 5:05 : 4:30.50 core simple, cycles 3600, fs 5, no scaler, -timedemo demo3 'performance' profile dosbox_cvs_01_01_25_a 4:17 5:38 4:17 4:13 4:14 : 4:15.25 dosbox_cvs_01_01_25_novfp 4:19 4:21 4:14 4:17 4:13 : 4:15.75 dosbox_feb12 4:18 4:19 5:00 4:17 4:31 : 4:21.25 dosbox_feb14a 4:55 5:49 4:49 4:58 4:48 : 4:52.50 dosbox_feb14b 4:02 4:41 4:38 4:05 4:01 : 4:11.50 dosbox_feb14c 4:51 4:55 4:54 4:50 4:59 : 4:52.50 core normal, cycles 3600, fs 5, no scaler, -timedemo demo3 'performance' profile dosbox_cvs_01_01_25_a 5:47 5:21 5:36 5:33 5:38 : 5:32.00 dosbox_cvs_01_01_25_novfp 6:01 6:06 6:06 5:59 7:12 : 6:03.00 dosbox_feb12 6:02 5:51 6:01 6:10 5:52 : 5:56.50 core full, cycles 3600, fs 5, no scaler, -timedemo demo3 'performance' profile dosbox_cvs_01_01_25_a 6:25 6:25 6:47 6:33 6:22 : 6:26.25 dosbox_cvs_01_01_25_novfp 5:43 5:43 5:43 5:44 5:52 : 5:43.25 dosbox_feb12 5:57 6:06 5:55 5:55 5:57 : 5:56.00
It appears that increasing cycles well above the point at which dosbox pegs the N810 CPU effectively defeats doom's internal frameskip so that a faster dosbox build+options will finish the run earlier than a slower one.
A seperate test series would be manually running the timedemos and jotting down doom's internal performance reports, using a very small game window and cycles=max.
Last edited by ArnimS; 2008-02-14 at 16:54.
 |
|
2008-02-13
, 10:16
|
|
Posts: 3,811 |
Thanked: 1,151 times |
Joined on Oct 2007
@ East Lansing, MI
|
#113
|
Greetings again ArnimS.
I'm not all too sure what those numbers mean with your testing results for the Doom demo but could you also run them in comparison to the latest version of the OS 2007 software?
I'm still on 2007 because of the unusual feedback by some of the members here on how it's less stable than 2008. I'm just not finding the courage to upgrade since I can't seem to understand the reasons to do so at the moment.
I've only got Dosbox and Xkbd running on a fresh install of 2007 and nothing else, no other applications have been downloaded. Would I see a better performance if I went ahead and upgraded?
I posted this earlier, but your mapper keys for the cycles, for some reason, isn't working again. I don't know what happened because there's nothing that stands out on what I'm doing differently than before.
There's something wrong with the mapper.txt feature but I can't make any connection on what might be causing the problems.
Don't get me wrong on this, your releases are extremely cool. The only thing is, you usually have very poor documentation for extreme newbies, such as myself, to help understand the genius world that you come from.
A little knowledge thrown my way and to others would very much be appreciated (but most likely still misunderstood)
Thanks chief!
I'm not all too sure what those numbers mean with your testing results for the Doom demo but could you also run them in comparison to the latest version of the OS 2007 software?
I'm still on 2007 because of the unusual feedback by some of the members here on how it's less stable than 2008. I'm just not finding the courage to upgrade since I can't seem to understand the reasons to do so at the moment.
I've only got Dosbox and Xkbd running on a fresh install of 2007 and nothing else, no other applications have been downloaded. Would I see a better performance if I went ahead and upgraded?
I posted this earlier, but your mapper keys for the cycles, for some reason, isn't working again. I don't know what happened because there's nothing that stands out on what I'm doing differently than before.
There's something wrong with the mapper.txt feature but I can't make any connection on what might be causing the problems.
Don't get me wrong on this, your releases are extremely cool. The only thing is, you usually have very poor documentation for extreme newbies, such as myself, to help understand the genius world that you come from.
A little knowledge thrown my way and to others would very much be appreciated (but most likely still misunderstood)
Thanks chief!
 |
|
2008-02-13
, 10:19
|
|
Posts: 5,478 |
Thanked: 5,222 times |
Joined on Jan 2006
@ St. Petersburg, FL
|
#114
|
Originally Posted by Addison
Considering that you get 70 extra megahertz . . . yes.
Would I see a better performance if I went ahead and upgraded?
|
|
2008-02-13
, 11:02
|
|
Posts: 3,811 |
Thanked: 1,151 times |
Joined on Oct 2007
@ East Lansing, MI
|
#115
|
Considering that you get 70 extra megahertz . . . yes.
|
|
2008-02-13
, 11:52
|
|
Posts: 3,841 |
Thanked: 1,079 times |
Joined on Nov 2006
|
#116
|
Speed, not money.. the CPU runs faster (330 MHz -> 400 MHz) with OS2008.
__________________
N800/OS2007|N900/Maemo5
-- Metalayer-crawler delenda est.
-- Current state: Fed up with everything MeeGo.
N800/OS2007|N900/Maemo5
-- Metalayer-crawler delenda est.
-- Current state: Fed up with everything MeeGo.
 |
|
2008-02-13
, 13:58
|
|
Posts: 373 |
Thanked: 110 times |
Joined on Jul 2007
|
#117
|
Originally Posted by Addison
The one thing I find a bit unstable is xkbd in OS2008. To do commands in DOS most times I have to have the CapsLock key turned on or it won't work. And then sometimes it won't work and sometimes it will with games using the alphabetic keys (it's hit or miss). Since your sidescreen-key version doesn't have CapsLock, that was one of the reasons why I haven't used it. Looks nice though. Other than that, DosBox seems to work fine in OS2008.
I'm still on 2007 because of the unusual feedback by some of the members here on how it's less stable than 2008. I'm just not finding the courage to upgrade since I can't seem to understand the reasons to do so at the moment.
|
|
2008-02-13
, 15:27
|
|
Posts: 1,107 |
Thanked: 720 times |
Joined on Mar 2007
@ Germany
|
#118
|
Originally Posted by Addison
There's no reason to. I posted these as references to people who might want to do similar tests. One thing that does seem clear is that for Doom at least, cpu=simple is much faster than either 'normal' or 'full'.
could you also run them in comparison to the latest version of the OS 2007 software?
Originally Posted by Addison
Yes
Would I see a better performance if I went ahead and upgraded?
Originally Posted by Addison
I just confirmed that latest hosted build http://pupnik.de/dosbox_N8x0.tgz works with the mapped keys (cycle up/down and fullscreen toggle). I don't know why they stopped working for you. If you inadvertently changed something then
I posted this earlier, but your mapper keys for the cycles, for some reason, isn't working again. I don't know what happened because there's nothing that stands out on what I'm doing differently than before.
- make sure you are launching the latest dosbox
- redownload the mapper.txt
- make sure that .dosboxrc or dosbox.conf points to the mapper
Originally Posted by Addison
Acknowledged.
Don't get me wrong on this, your releases are extremely cool. The only thing is, you usually have very poor documentation for extreme newbies, such as myself
These newbie difficulties occur because there is no gui frontend.
There are things standing in the way of a gui-frontend.
DosBox as-is runs too slow on the N8x0s for everything but cga/pcspeaker era games.
An effective gui frontend will need configuration profiles and mappers for various games.
Creating these profiles with a broken/slow core dosbox build is useless because when that is fixed, they will all need to be redone.
so....
I'm trying to fix the sound/performance problem so that at least 286-class games using vga/ega and sound blaster can run playably.
|
|
2008-02-13
, 22:43
|
|
Posts: 151 |
Thanked: 14 times |
Joined on Dec 2007
|
#119
|
Originally Posted by ArnimS
Something must have gone seriously wrong...
No f4ck it, i don't give a f4ck about anybody.
i don't owe anyone anywhting/ eat rice.
|
|
2008-02-13
, 23:10
|
|
Posts: 373 |
Thanked: 110 times |
Joined on Jul 2007
|
#120
|
Originally Posted by spirytsick
Hope not. I've heavily relied on ArnimS's expertise on helping me get this cool gaming stuff running on my N800 and owe a ton of thanks. I hope ArnimS is okay and everything is fine.
Something must have gone seriously wrong...
I didn't read the fine print that it was the middle key plus the +/- keys at the same time.
What's my I.Q. again?
Last edited by Addison; 2008-02-08 at 22:03.