DOOM3 SOURCE RELEASED!!

simulation · **Posted:** Sat Dec 03, 2011 6:29 pm

aardwolf wrote:

Does a compile doom3 run like the old executable? No delays due to the edited Carmack's reverse code?

Oneofthe8devilz did some timings and visual comparisons at 1080p and did get any loss of fps and quality was identical. seven years worth of gfx card advances and driver improvements on obviously make up the difference.

BNA! · **Posted:** Sat Dec 03, 2011 7:56 pm

simulation wrote:

aardwolf wrote:

Does a compile doom3 run like the old executable? No delays due to the edited Carmack's reverse code?

Oneofthe8devilz did some timings and visual comparisons at 1080p and did get any loss of fps and quality was identical. seven years worth of gfx card advances and driver improvements on obviously make up the difference.

Did or didn't ?

rebb · **Posted:** Sat Dec 03, 2011 8:37 pm

You'll probably have to test such things on a low end machine or with a scene that is particularly heavy on shadow calculations. Most modern rigs are just too powerful.

simulation · **Posted:** Sat Dec 03, 2011 8:45 pm

Oops yes I should have typed Did Not - the perils of iPod typing! Like rebb said you'd have to try it on older hardware to see a difference. That said there are people that have reimplemented the algorithm so I guess we'll soon have it back anyway.

_silverpower_ · **Joined:** Sat Dec 27, 2008 2:00 am **Posts:** 6

I haven't done extensive benching, but my Pentium D+G33 box loses a *lot* of performance. Goes from "somewhat playable, sometimes" to "almost completely unplayable". I haven't gotten a build going on my Radeon 9600-powered laptop... but the difference is pretty dramatic. 25-50% perf loss, but G33 has horrible fillrates anyway.

Oneofthe8devilz · **Posted:** Sun Dec 04, 2011 5:39 pm

How did you benchmark ?

I ran the "timedemo demo1" and compared the frames per second result...

On the 580 gtx there was no difference noticeable with an average result of 210 FPS at 1080p at custom High Quality Settings...

I will run it on my 6800go mobile gfx to cross compare with the previous 580 gtx results...

***Update***

Just got the results which are again the same ... 70 FPS with stock and gpl executable at 480p at custom High Quality Settings...

Laptop is a Fujitsu-Siemens Amilo M3438G released summer 2005

Deadite4 · **Joined:** Sun Jan 09, 2005 11:55 pm **Posts:** 598

On a lower end system that only achieves 55-60fps though, a loss of only say 10-15fps(while insignificant at 210fps) is a loss of 15-20% performance on the lower end. Just an example.

My ATI 5900xt box averages 50-60fps on vanilla Doom 3. I'm sure there would be noticeable performance drops if I were to run one of these new exes where every frame counts

gokusuper5 · **Joined:** Sat Sep 18, 2010 6:16 pm **Posts:** 21

my version give me 10-15 more FPS on an onboard Nvidia 6150se getting drops from 30-70 fps

_silverpower_ · **Joined:** Sat Dec 27, 2008 2:00 am **Posts:** 6

Runthrough from the dropship to where you pick up your PDA. I could get timedemo numbers later if anybody's interested.

Douglas Quaid · **Posted:** Mon Dec 05, 2011 4:07 am

Nice to finally see the source. This is a gift! Looking forward to seeing some games!

licmi · **Joined:** Tue Jan 17, 2012 2:02 pm **Posts:** 16

@Kristian Joensen -about disabling cdkey check (and yes thank you for mention that)
Another variant:
just add these two lines before it start to check
cdkey_state = CDKEY_OK;
xpkey_state = CDKEY_OK;

bool idSessionLocal::CDKeysAreValid( bool strict ) {
int i;
bool emitAuth = false;
////////////
cdkey_state = CDKEY_OK;
xpkey_state = CDKEY_OK;
////////////
if ( cdkey_state == CDKEY_UNKNOWN ) {
if ( strlen( cdkey ) != CDKEY_BUF_LEN - 1 ) {
cdkey_state = CDKEY_INVALID;
} else {
for ( i = 0; i < CDKEY_BUF_LEN-1; i++ ) {
if ( !strchr( CDKEY_DIGITS, cdkey[i] ) ) {
cdkey_state = CDKEY_INVALID;
break;
}
}
............
.......
....
...
..

ThetaRabbit · **Joined:** Sat May 05, 2012 8:01 am **Posts:** 1

licmi wrote:

@Kristian Joensen -about disabling cdkey check (and yes thank you for mention that)
Another variant:
just add these two lines before it start to check
cdkey_state = CDKEY_OK;
xpkey_state = CDKEY_OK;

bool idSessionLocal::CDKeysAreValid( bool strict ) {
...

Why you just can't replace all this code with "return true;"?

jmarshall23 · **Joined:** Tue Dec 06, 2011 8:05 pm **Posts:** 418

Its simpler than that, in builddefines ensure macro D_ENFORCE_KEY is defined as 0.

Doom3world

DOOM3 SOURCE RELEASED!!

Who is online